The Routledge Handbook of Philosophy, Politics, and Economics (Taylor Francis Group)

THE ROUTLEDGE HANDBOOK OF
PHILOSOPHY, POLITICS, AND

ECONOMICS
This handbook advances the interdisciplinary field of Philosophy, Politics, and Economics (PPE) by
identifying thirty-five topics of ongoing research. Instead of focusing on historically significant texts,
it features experts talking about current debates. Individually, each chapter provides a resource for
new research. Together, the chapters provide a thorough introduction to contemporary work in
PPE, which makes it an ideal reader for a senior-year course.
The handbook is organized into seven parts, each with its own introduction and five chapters:
I. Frameworks
II. Decision-Making
III. Social Structures
IV. Markets
V. Economic Systems
VI. Distributive Justice
VII. Democracy
The “Frameworks” part discusses common tools and perspectives in PPE, and the “Decision-
Making” section shows different approaches to the study of choice. From there, parts on “Social
Structures,” “Markets,” and “Economic Systems” each use tools from the three PPE disciplines to
study and distinguish parts of society. The next part explains dominant theories and challenges to the
paradigm of “Distributive Justice.” Finally, a part on “Democracy” offers five challenges to current
democratic practice.
C.M. Melenovsky is Assistant Professor of Philosophy and Director of the PPE program at Suffolk
University, USA. His research focuses on social practices, institutions, moral conventionalism, and
Rawlsian political philosophy. He is currently working on a book, Kantian Conventionalism, that
reconciles the social contingency of obligations and rights with a Kantian moral framework.
ROUTLEDGE HANDBOOKS IN PHILOSOPHY
Routledge Handbooks in Philosophy are state-of-the-art surveys of emerging, newly refreshed, and im-
portant fields in philosophy, providing accessible yet thorough assessments of key problems, themes,
thinkers, and recent developments in research.
All chapters for each volume are specially commissioned, and written by leading scholars in the
field. Carefully edited and organized, Routledge Handbooks in Philosophy provide indispensable reference
tools for students and researchers seeking a comprehensive overview of new and exciting topics in
philosophy. They are also valuable teaching resources as accompaniments to textbooks, anthologies,
and research-orientated publications.
Also available:
The Routledge Handbook of Philosophy and Improvisation in the Arts
Edited by Alessandro Bertinetto and Marcello Ruta
The Routledge Handbook of Idealism and Immaterialism
Edited by Joshua Farris and Benedikt Paul Göcke
The Routledge Handbook of Philosophy of Economics
Edited by Conrad Heilmann and Julian Reiss
The Routledge Handbook of Logical Empiricism
Edited by Thomas Uebel and Christoph Limbeck-Lilienau
The Routledge Handbook of Philosophy of Agency
Edited by Luca Ferrero
The Routledge Handbook of Propositions
Edited by Adam Russell Murray and Chris Tillman
The Routledge Handbook of Liberal Naturalism
Edited by Mario De Caro and David Macarthur
The Routledge Handbook of Philosophy, Politics, and Economics
Edited by C.M. Melenovsky
For more information about this series, please visit: https://www.routledge.com/Routledge-
Handbooks-in-Philosophy/book-series/RHP
THE ROUTLEDGE
HANDBOOK OF
PHILOSOPHY, POLITICS,
AND ECONOMICS
Edited by C.M. Melenovsky

Cover image: © Getty Images
First published 2022
by Routledge
605 Third Avenue, New York, NY 10158
and by Routledge
4 Park Square, Milton Park, Abingdon, Oxon, OX14 4RN
Routledge is an imprint of the Taylor & Francis Group, an informa business
© 2022 Taylor & Francis
The right of C.M. Melenovsky to be identified as the author of the editorial material, and of the
authors for their individual chapters, has been asserted in accordance with sections 77 and 78 of
the Copyright, Designs and Patents Act 1988.
All rights reserved. No part of this book may be reprinted or reproduced or utilised in any form
or by any electronic, mechanical, or other means, now known or hereafter invented, including
photocopying and recording, or in any information storage or retrieval system, without
permission in writing from the publishers.
Trademark notice: Product or corporate names may be trademarks or registered trademarks, and
are used only for identification and explanation without intent to infringe.
Library of Congress Cataloguing-in-Publication Data
A catalog record for this title has been requested
ISBN: 978-0-367-40767-4 (hbk)

ISBN: 978-1-032-25169-1 (pbk)
ISBN: 978-0-367-80898-3 (ebk)
DOI: 10.4324/9780367808983
Typeset in Bembo
by MPS Limited, Dehradun
CONTENTS
List of Figures ix
List of Tables x
Notes on Contributors xii
Preface xvii
Introduction xix
PART I
Frameworks 1
Introduction to Part I 1
1 PPE as an Intellectual Enterprise 3

Geoffrey Brennan and Geoffrey Sayre-McCord
2 On Models and their Uses 17

James Johnson
3 Complexity 28
Fred D’Agostino
4 PPE in Marx’s Materialist Conception of History 43

Vanessa Wills
5 Feminist Theory 52
Ann E. Cudd
v
Contents
PART II
Decision-Making 65
Introduction to Part II 65
6 Game Theory 67
John Thrasher
7 Four Structures of Intransitive Preferences 81

Luc Bovens
8 Theories of Choice Behavior 94

Sudeep Bhatia
9 Rule-Following 108
Erik O. Kimbrough and Bart J. Wilson
10 Implicit Bias and Decision-Making 121

Lacey J. Davidson
PART III
Social Structures 135
Introduction to Part III 135
11 Social Norms 137

Ryan Muldoon
12 Institutions and Institutionalism 149

C.M. Melenovsky
13 Property 163
Bas van der Vossen
14 Corporations in our Polity 176

Amy J. Sepinwall
15 Polycentricity 186
Vlad Tarko
PART IV
Markets 201
Introduction to Part IV 201
16 The Advantages of Markets 205

Matt Zwolinski
vi
Contents
17 Exploitation 217
Vida Panitch
18 The Meaning of Markets 227

Brookes Brown
19 Gender and the Division of Labor 239

Gina Schouten
20 Housing Markets 252

Kristina Meshelski
PART V
Economic Systems 265
Introduction to Part V 265
21 Capitalism 267
Peter Boettke
22 Socialisms 276
Samuel Arnold
23 Property-Owning Democracy 289

Alan Thomas
24 Social Democracy 300

Jeppe von Platz
25 Corruption 314
Michael C. Munger
PART VI
Distributive Justice 325
Introduction to Part VI 325
26 Property Rights and Justice in Holdings: A Libertarian Perspective 329

Eric Mack
27 High Liberalism 341

Samuel Freeman
28 Institutionalism, Injustice, and Personal Responsibility 354

Kok-Chor Tan
vii
Contents
29 Social Justice 365

Maeve McKeown
30 Justice Across Borders 377

Serena Parekh
PART VII
Democracy 389
Introduction to Part VII 389
31 In Defense of Epistocracy: Enlightened Preference Voting 391

Jason Brennan
32 Voting Rules 401

Itai Sher
33 Enabling Informed and Equal Participation 416

Thomas Christiano
34 What, If Anything, Can Justify Limiting Workers’ Voice? 428

Lisa Herzog
35 Social Trust 439

Karen S. Cook and Jacob Reidhead
Index 451
viii
FIGURES
3.1 Searching the Social Science and Humanities Article, Conference, and Book
Indices 30
6.1 Coordination Game 68
6.2 Extensive Form Coordination Game 69
6.3 Coordination Game with Strict Dominance 69
6.4 Zero-Sum Baseball Game 70
6.5 Battle of the Sexes 71
6.6 Prisoner’s Dilemma 71
6.7 Centipede Game 73
6.8 Trust Game 74
6.9 Stag Hunt 74
6.10 Battle of the Sexes Game 75
21.1 Share of the World Population Living in Absolute Poverty, 1820–2015 268
ix
TABLES
7.1 Schumm’s Christmas Ornaments 82

7.2 Missing Christmas Ornaments 82
7.3 Bar-Hillel and Margalit’s Missing Test Grades 83
7.4 Broome’s Mountaineering Example 83
7.5 Kamm on Duty, Predilection, and Supererogation 84
7.6 Choice by a Majority of Features 84
7.7 The Condorcet Voting Paradox 85
7.8 The Condorcet Paradox for Choice Affected by Multiple Features 85
7.9 Boxes with Mantle Ornaments with # Markers for Showiness and Tree
Ornaments with * Markers for Shininess 86
7.10 Three-Option Choice Problem over Boxes with Mantle and Tree Ornaments 86
7.11 An Alternative Representation of the Four-Option Choice Problem with Boxes
A, …, D 87
7.12 An Alternative Representation of the Three-Option Choice Problem with
Boxes A, …, C 87
7.13 Sen’s Lady Chatterley’s Lover Example Illustrating the Libertarian Paradox 87
7.14 A Sen’s-Libertarian-Paradox Style Cycle for Policy Evaluations 88
7.15 Sorites Cycle with Christmas Balls 89
7.16 Sorites Cycle for Tversky’s Job Candidates 89
7.17 Sorites Cycle Based on the Original Sorites Paradox 90
7.18 Sorites Cycle of Quinn’s Self-Torturer with the Starting Point in Boldface 91
7.19 Sorites Cycle of Temkin-Rachels Torture-to-Hangnail with the Starting Point
in Boldface 91
7.20 Sorites Cycle of the One-Person Variant of Harrison’s Lawn-Crossing Example 91
8.1 Examples of Modifications to Theories of Choice Behavior 103
15.1 Types of Governance Systems 188
26.1 Three Alternative Distributions for a Society Consisting of Individuals (or
Representatives of Income Classes) A, B, and C. 332
26.2 The Likely Reality of the Ongoing Application of the Difference Principle in
the Ceaseless Endeavor to Achieve Distributive Justice in a Four-Party (or Four-
Income Class) Society. 334
32.1 Table 32.1 406
x
Tables
32.2 Table 32.2 406

32.3 Table 32.3 406
32.4 Table 32.4 407
32.5 Table 32.5 407
32.6 Table 32.6 408
32.7 Table 32.7 408
32.8 Table 32.8 411
xi
CONTRIBUTORS
Samuel Arnold is Associate Professor in the Department of Political Science at Texas Christian
University. His research interests include liberal egalitarian theories of justice, especially the work of
John Rawls; the division of labor and work; consumerism and voluntary simplicity; and socialism
and alternatives to capitalism.
Sudeep Bhatia is Assistant Professor of Psychology at the University of Pennsylvania, where he

studies how people think and decide. Using interdisciplinary methods applied to diverse
experimental and real-world data, he hopes to build computational models that can deliberate
over and respond to a large variety of everyday decision problems in a human-like manner.
Peter Boettke is University Professor of Economics and Philosophy at George Mason University
and Director of the F. A. Hayek Program for Advanced Study in Philosophy, Politics, and
Economics at the Mercatus Center. He is the author of over 15 books including F.A. Hayek:
Economics, Political Economy and Social Philosophy (2018), Challenging Institutional Analysis and
Development (2009, with Paul Dragos Aligica), and Why Perestoika Failed (1993). He has also
edited volumes on James Buchanan, Vincent and Elinor Ostrom, Israel Kirzner, and Alexis de
Tocqueville.
Luc Bovens is Professor in the Department of Philosophy and core faculty in the PPE program at
the University of North Carolina at Chapel Hill. He has published in philosophy of probability,
epistemology, voting theory, philosophy and public policy, philosophy of economics, and moral
psychology. He is a co-author of Bayesian Epistemology (2003) and the author of Coping (2021).
Geoffrey Brennan was originally trained – and has spent much of his academic career – as an
economist. In mid-career, he was a colleague of Nobel Laureate James Buchanan, with whom he
co-authored two books and about a dozen articles. He is now Professor Emeritus in the Philosophy
Department at the Australian National University’s Research School of Social Sciences, and retains
an ongoing research connection with both the Philosophy Department at UNC-Chapel Hill and
the Political Science Department at Duke University (whose joint PPE program he helped establish
in 2005 and taught in for over a decade).
xii
Contributors
Jason Brennan is the Flanagan Family Professor of Strategy, Economics, Ethics, and Public Policy
at the McDonough School of Business, Georgetown University. He is the author of 15 books,
including Debating Democracy, with Hélène Landemore (Oxford University Press, 2021), Against
Democracy (Princeton University Press, 2016), and The Ethics of Voting (Princeton University Press,
2011). He is currently writing Democracy: A Guided Tour of an Idea for Oxford University Press. He
specializes in the intersection of politics, philosophy, and economics.
Brookes Brown is Assistant Professor of Philosophy at Clemson University and the Director of the
Program in Law, Liberty, and Justice. She is the author of a number of articles on civic ethics and the
nature of the state.
Thomas Christiano is Professor of Philosophy and Law at the University of Arizona. He has been
a Laurence Rockefeller Visiting Faculty Fellow at the Princeton University Center for Human
Values, a Visiting Fellow at All Souls College, Oxford, and a Fellow of the National Humanities
Center. He is the author of The Constitution of Equality: Democratic Authority and Its Limits (Oxford:
Oxford University Press, 2008) and The Rule of the Many (Boulder, CO: Westview Press, 1996). He
is Co-Editor in Chief of Politics Philosophy and Economics. He has written many papers in the areas of
democratic theory, distributive justice, philosophy of international law, and economic justice.
Karen S. Cook is the Ray Lyman Wilbur Professor of Sociology and Director of the Institute for
Research in the Social Sciences at Stanford University. She is the co-author of Cooperation Without
Trust? (2005) and is the editor and author of a number of other books and articles on trust.
Ann E. Cudd is Provost and Senior Vice-Chancellor and Professor of Philosophy at the University
of Pittsburgh. She is the author of Analyzing Oppression (Oxford, 2006) and co-author with Nancy
Holmstrom of Capitalism: For and Against, A feminist debate (Cambridge, 2011). She has co-edited
several books, including Theorizing Backlash: Philosophical Reflections on the Resistance to Feminism with
Anita Superson (Rowman and Littlelfield, 2002) and Feminist Theory: A Philosophical Anthology with
Robin Andreason and (Blackwell, 2005).
Fred D’Agostino is Emeritus Professor in the Institute for Advanced Studies in the Humanities at
The University of Queensland, where he was Executive Dean of Arts and President of the Academic
Board. A Fellow of the Australian Academy of the Humanities, his recent work includes Free Public
Reason, Incommensurability and Commensuration, and Naturalizing Epistemology. With Gerald Gaus, he
edited The Routledge Companion to Social and Political Philosophy. He was an Editor of PPE: Politics,
Philosophy and Economics.
Lacey J. Davidson is Assistant Professor of Philosophy at the University of Indianapolis. Her

research focuses on the social, cognitive, and epistemological mechanisms of oppression, and she is
the author and co-author of several book chapters and articles exploring the intersections of these
topics.
Samuel Freeman is Professor of Philosophy and of Law at the University of Pennsylvania, where
he has taught since 1985. He teaches and publishes in social and political philosophy, moral
philosophy, and philosophy of law. He has written books on Liberalism and Distributive Justice (2018),
Justice and the Social Contract (2007) and on John Rawls (2007). Freeman edited the Cambridge
Companion to Rawls (2003), as well as John Rawls’s Lectures on the History of Political Philosophy (2007)
and his Collected Papers (1999). He is currently working on a manuscript on liberalism and
libertarianism.
xiii
Contributors
Lisa Herzog is a Professor of political philosophy in the Faculty of Philosophy and Center for
Philosophy, Politics, and Economics at University of Groningen. Her latest English monograph was
Reclaiming the System. Moral Responsibility, Divided Labour, and the Role of Organizations in Society
(OUP 2018). Her current work focuses on workplace democracy, organizational ethics, and the
relation between markets, democracy, and knowledge.
James Johnson is Professor of Political Science at the University of Rochester and long-time
Instructor at the ICPSR Summer Program in Quantitative Methods of Social Research (Ann
Arbor). His current research traverses pragmatist political thought, democratic theory, philosophy of
social science, and political economy. Most recently he has published (with Susan Orr) Should Secret
Voting Be Mandatory? (Polity 2020).
Erik O. Kimbrough is Associate Professor of Economics in the Smith Institute for Political Economy
and Philosophy at Chapman University’s Argyros School of Business and Economics. His research uses
experiments to study the economics of social norms, conflict resolution, cooperation, and institutions,
among other topics. He sometimes follows rules, even when doing so is costly.
Eric Mack is Emeritus Professor of Philosophy at Tulane University. He is the author of many scholarly
articles on natural rights, property rights, economic justice, and libertarian theory and its history. He is
the author of John Locke (2013), Libertarianism (2018), and The Essential John Locke (2020).
Maeve McKeown is Assistant Professor in Political Theory at the University of Groningen. She is
writing her first book With Power Comes Responsibility: The Politics of Structural Injustice (London:
Bloomsbury Academic).
Chris Melenovsky is Assistant Professor of Philosophy and Director of the PPE Program at Suffolk
University in Boston, MA. His writing focuses on social practices, conventionalism, institutions, and
Rawlsian political philosophy. He is also the Editor of the Routledge Handbook that you are
currently reading.
Kristina Meshelski is Associate Professor of Philosophy at California State University, Northridge.

She is the author of articles about non-ideal theory, affirmative action, and Spinoza’s metaphysics.
She is a former co-chair of the Housing & Homelessness Committee of Democratic Socialists of
America, Los Angeles, a member of LA Tenants Union, Streetwatch LA, and Reclaiming our
Homes.
Ryan Muldoon is Associate Professor of Philosophy and Director of the Philosophy, Politics, and
Economics Program at the University at Buffalo. He is the author of Social Contract Theory for a Diverse
World: Beyond Tolerance (2016) co-authored the 2015 World Development Report: Mind, Society and
Behavior, and co-edits the Philosophy, Politics, and Economics series for Oxford University Press.
Michael C. Munger is Professor of Political Science, and Director of the PPE Program, at Duke
University. He is the author of Tomorrow 3.0: Transaction Costs and the Sharing Economy (2018), and is
a past president of the Public Choice Society.
Vida Panitch is Associate Professor of Philosophy and Director of Ethics and Public Affairs at
Carleton University. She is the co-editor of Exploitation: From Practice to Theory (2017) and of the
forthcoming Routledge Handbook of Commodification (2022). She has authored numerous articles on
exploitation, distributive justice, and the moral limits of markets.
xiv
Contributors
Serena Parekh is Professor of Philosophy at Northeastern University in Boston, where she is the
Director of the PPE Program and co-editor of the journal, Feminist Philosophy Quarterly. She’s the
author of three books and numerous articles, including her most recent book, No Refuge: Ethics and
the Global Refugee Crisis (Oxford University Press, 2020).
Jacob Reidhead is a post-doctoral researcher at National Chengchi University. He recently

received his doctorate in sociology from Stanford University where his dissertation contrasted
patronage and factionalism in the local party politics of South Korea and Taiwan.
Geoffrey Sayre-McCord is the Morehead-Cain Distinguished Professor of Philosophy at the

University of North Carolina at Chapel Hill and the Director of UNC’s PPE Program. He is also
the Founding Director of the PPE Society. He works in moral philosophy and its history (especially
David Hume and Adam Smith), meta ethics, and social and political philosophy.
Gina Schouten is Associate Professor of Philosophy at Harvard University. She writes on issues of
justice and political legitimacy. Her 2019 book, Liberalism, Neutrality, and the Gendered Division of
Labor, assesses the adequacy of a neutrality framework of political legitimacy and considers its
capacity to approve political intervention aimed at eroding the gendered division of labor.
Amy J. Sepinwall is Associate Professor in the Department of Legal Studies and Business Ethics at
Wharton, University of Pennsylvania. She is trained in both law and philosophy, and she has
authored numerous articles addressing corporate responsibility for wrongdoing and corporate
constitutional rights.
Itai Sher is Associate Professor of Economics at the University of Massachusetts Amherst. His
research is at the intersection of ethics and economics. His primary focus is on incorporating broader
ethical criteria into formal economic evaluation. Sher’s recent work has concerned normative
assessment of tax policy, freedom of choice, voting institutions, and value pluralism in normative
economics. Sher is a co-Editor of the journal Economics & Philosophy and also of the Oxford
University Press Philosophy, Politics, and Economics Book Series. He is a founder and co-organizer
of the interdisciplinary conference series “Normative Ethics and Welfare Economics.”
Kok-Chor Tan is Professor of Philosophy at the University of Pennsylvania. He is the author of

Toleration, Diversity and Global Justice (2000), Justice Without Borders (2004), Justice, Institutions and Luck
(2012), and What Is This Thing Called Global Justice? (2017).
Vlad Tarko is Associate Professor of Political Economy and Moral Science at University of
Arizona. He writes primarily on polycentric governance and varieties of capitalism. He is the author
of Elinor Ostrom: An Intellectual Biography (2017) and co-author of Public Governance and the Classical
Liberal Perspective (2019)
Alan Thomas is Professor of Philosophy at the University of York. He is the author of Value and
Context (2006), Thomas Nagel (2008), and Republic of Equals (2017) and has written papers and book
chapters in the fields of political philosophy, ethics, and the philosophy of mind.
John Thrasher is Associate Professor in Philosophy and in the Smith Institute for Political
Economy and Philosophy at Chapman University. He is the co-author, with Dan Halliday, of The
Ethics of Capitalism (2020) and, with Gerald Gaus, Philosophy, Politics, and Economics: An Introduction
(2021).
xv
Contributors
Bas van der Vossen is Associate Professor in the Philosophy Department and Smith Institute for
Political Economy and Philosophy at Chapman University. He is the author of several books,
including In Defense of Openness (2018), and is Associate Editor of Politics, Philosophy and Economics
and Social Philosophy & Policy.
Jeppe von Platz is Associate Professor of Philosophy and the Politics, Philosophy, Economics, and
Law Program at the University of Richmond. He is the author of Theories of Distributive Justice: Who
Gets What and Why (2020) and several articles on questions of economic justice.
Vanessa Wills is Assistant Professor of Philosophy at The George Washington University in

Washington, D.C. Her areas of specialization are moral, social and political philosophy, 19th-
century German philosophy (especially Karl Marx) and the philosophy of race. Wills is a founding
member of the editorial board of Spectre: a biannual journal of Marxist theory, strategy, and analysis.
Bart J. Wilson is the Donald P. Kennedy Professor of Economics and Law and Director of the
Smith Institute for Political Economy and Philosophy at Chapman University. He is the author of
The Property Species: Mine, Yours, and the Human Mind (2020), co-author of Humanomics: Moral
Sentiments and the Wealth of Nations for the Twenty-First Century (2019), and author of a number of
articles in experimental economics.
Matt Zwolinski is a Professor of Philosophy at the University of San Diego and director of USD’s
Center for Ethics, Economics, and Public Policy. He is the editor of Arguing About Political Philosophy
and, with Benjamin Ferguson, the forthcoming Routledge Companion to Libertarianism and Exploitation:
Philosophy, Politics, and Economics. With John Tomasi, he currently writing A Brief History of
Libertarianism, and with Miranda Fleischer, Universal Basic Income: What Everyone Needs to Know.
xvi
PREFACE
When I was an undergraduate, my majors were Economics and Philosophy. I would tell people
what I was studying, and I got a common reaction: “What a weird combination. Those couldn’t be
further apart!” People seemed to think that I was combining a practical area of study with one that
was wholly impractical. Maybe they thought that I was building stock portfolios while learning
classical greek or that I was simultaneously committed to focused avarice and high-minded
asceticism. Regardless, I was surprised by the initial reaction. The two seemed complementary to
me. If you’re reading this, I expect that you agree.
My interest was in evaluating policy, and I couldn’t imagine doing that without both disciplines. I
was also interested in understanding the structure of society, in international affairs, in the role of
large companies in a democracy, in civic responsibility, and in predicting progress. I saw both
disciplines as essential if I were to investigate any of these issues. So, the double major never seemed
odd to me.
At the time, there were very few undergraduate Philosophy, Politics, and Economics (PPE)
programs. Only five schools in the United States had one, and I was only vaguely aware of it as an
area of study. Today, there are over 50 PPE programs in the United States, including one at my alma
mater. In this way, the path of PPE in the United States mimics its path in the United Kingdom and
Europe. What started at Oxford in 1920 has spread to become a primary area of study for those
interested in public service across the continent. There are now more than 60 PPE programs across
Europe and over 150 programs across the world
It is because of the growing popularity of PPE that I thought this Handbook would be worth
developing. For students in these programs, it is useful to have a single volume that showcases
interdisciplinary topics and research. A collected set of interesting readings can provide a shared
background for students at the many different programs around the world, and these chapters
together provide a sound foundation for students to pursue their own projects.
As PPE grows as an educational program, it also grows as a research program. However, the same
professional incentives that encourage interdisciplinary undergraduate programs do not encourage
advanced interdisciplinary research. Each academic wants to show that their own work is meaningful
to those in their established discipline, and it can be hard to do that with research that edges into
other disciplines. Yet, the research side of PPE is growing despite these obstacles. More
interdisciplinary researchers are available to referee articles and to review tenure files. Professional
risks remain, but the field is unmistakably larger and more stable than even ten years ago. It is my
hope that this handbook can also advance this research program. By identifying interdisciplinary
xvii
Preface
topics and methods, interested scholars have an entryway into the literature. Those that are already
familiar with a topic will have a useful reference.
Nowadays, I’ll occasionally find myself under a banner that reads “Philosophy, Politics, and
Economics” as prospective students and their parents walk around to tour the university where I
teach. Occasionally, I still get the same reaction about this combination of disciplines as I got when I
pursued my double-major. However, I also hear from students with older friends and siblings that
have majored in PPE. They come to my banner eager to hear more. As more students graduate with
PPE majors and more research demonstrates the value of an interdisciplinary approach, the
complementarity of the three PPE disciplines will become obvious to more people. Perhaps that
time has already come.
As with any collection, the finished product is the result of hard work from many writers. I thank
all the authors for contributing chapters and meeting deadlines during a hectic year. This book was
composed during the COVID-19 pandemic, and so many had to balance writing and research with
the demands on online teaching, learning, homeschooling children, and personal tragedies. Because
of their hard work, the process went more smoothly than I ever expected.
I would also like to thank two research assistants. Brittany Brown reviewed first drafts for most
chapters and Clare Thomson reviewed second drafts. Both offered invaluable advice on style and
substance. They were always thoughtful, engaged, and extremely helpful. I would never have hit the
deadlines or understood the chapters as well without them. I also want to thank the students in my
Spring 2021 PPE Capstone class at Suffolk University. We discussed many of the chapters there, and
these students helped me to see how each chapter would be received by an undergraduate class.
Additionally, I would like to thank Jeppe von Platz and Barry Maguire for advice along the way and
two anonymous reviewers from Routledge for help in initial planning.
C. M. Melenovsky
xviii
INTRODUCTION
No single handbook can represent everyone’s perspective towards what PPE is and why it is
worthwhile. For any discipline, there is disagreement about what establishes the boundaries and aims
of that discipline. When we combine three disciplines, those disagreements only multiply. Those
with similar research will cluster with one another, and the nature of the field will be viewed
differently from one cluster to the next.
There are at least six views of PPE. One sees the purview of PPE as public policy. You cannot
support strong judgments about any policy without understanding its economic and political
features. These features only matter when you have the clear normative foundations studied by
philosophy. While other disciplines are also important for understanding the impact of public policy,
the PPE disciplines form a core foundation for engagement. When PPE started at Oxford, for
example, it sought to leverage a classical style of education to train a new generation of civil servants
(White et al., 2020, 3).
According to a second view, PPE is the unified study of social institutions. Since the major
institutions that structure social life are political and economic, it makes sense to leverage methods
and traditions from the disciplines that have focused on these institutions. PPE can identify tools and
insights from different disciplines to give a broader perspective on social institutions while
Philosophy aids in the conceptual refinement and critical revisions needed for good social science.
A third view looks at PPE more historically. Before the segmentation of the three disciplines,
there was the field of “Political Economy.” Great figures like Adam Smith, David Ricardo, Anne-
Robert-Jacques Turgot, Karl Marx, and J.S. Mill would not recognize the distinctions between the
disciplines as we currently practice them. Their work was simultaneously political, economic, and
philosophical. More recently, figures like James Buchanan, Amartya Sen, and Elinor Ostrom occupy
similar territory. The third view sees PPE as reestablishing this – more naturally demarcated – area of
study. It unites disciplines whose insights are only obscured by their separation.
Three other views are more expansionist in nature. The first uses the methodological
individualism and mathematized models of Economics as an ideal for studying political structures
and human values. The second uses the discursive analysis and direct argumentation of Philosophy to
orient social science and identify moral foundations. The third uses Political Science as a model for
how a variety of approaches and perspectives can analyze social phenomena. Each of these last three
views uses the methods and traditions of one discipline and applies it to subject matter studied by the
others.
xix
Introduction
It should be clear that none of these six views represent a “true” aim of PPE. Some agendas seem
more fruitful to me than others, but I will not argue for that here. This handbook is not intended to
plant a flag. Instead of starting from a theory of PPE, this handbook starts from practice. It is a topics-
first approach that identifies issues currently receiving interdisciplinary engagement. It gathers
authors that are interested in speaking to interdisciplinary audiences. Some will not view certain
topics as emblematic of true PPE, but I suspect that such judgments will be fairly distributed across
the topics. In this way, the handbook does not aim to achieve a complete consensus about what PPE
is. Instead, I seek – in the words of one oft-cited author – an overlapping consensus.
The chapters are organized into seven parts. Each sets up what follows. Part I, “Frameworks,”
introduces different methods and perspectives that inform interdisciplinary analysis. A chapter on
model theory, for example, gives a perspective towards models that are often used in later chapters.
A chapter on feminist theory previews common themes. From there, the next few parts mimic an
individualist method, building ever more complex social structures by starting with individual
choice (though the affect that these structures have on individual choice is not ignored). Part II,
“Decision-Making,” showcases PPE approaches to how choices get made, how choices should be
made, and how such choices should be modeled. Part III details different kinds of social structure
that arise from – and give context to – individual choice. In order, the chapters explain social norms,
institutions, property systems, corporations, and polycentric systems. Part IV focuses on one
particularly important social structure, the market, from different perspectives. It identifies why
markets can be advantageous, then discusses how markets affect the meaning, and how they can lead
to exploitation. The part ends with particular attention to labor markets and housing markets.
Part V turns towards the study of larger social structures by focusing on “Economic Systems.” The
first four chapters of this part discuss rival proposals for how to structure an economy; Capitalism,
Socialisms, Property Owning Democracy, and Social Democracy. The final chapter discusses how
corruption can persist in economic systems, and why it can be so difficult to root out. Part VI then
turns to examine the ideals used to assess such economic structures. Chapters explain libertarianism,
high liberalism, institutional egalitarianism, social justice, and global justice. The final part,
“Democracy,” offers an assessment of the democratic institutions that have the potential to
change the economic structure and promote ideals of justice. The part does not, however, only
focus on democratic control of the government. It includes a discussion about democracy in the
workplace and the relationship between democracy and social trust.
Modern publishing methods do not yet allow for the publication of infinitely long books. So, not
all topics that deserve attention are here. In most cases, this is the fault of the editor. Chapters on key
PPE authors, on international political economy, on Public Choice theory, and on climate change
would have been welcome if space allowed. In some cases, potential chapters were cut because
authors were no longer able to contribute. After all, this handbook was mostly written during an
international pandemic. Chapters on “Measuring Well-being” and on “Race and Labor Markets,”
for example, were on the original table of contents. Many other topics deserve equal attention, but I
hope that the current contents still demonstrate a range of PPE topics and give a sense of a PPE
perspective.
Reference
White, Bethany, Samuel Wainwright, and Lilly Schreiter (2020). “One Hundred Years of PPE: 1920–2020.”
https://www.humanities.ox.ac.uk/files/ppe100yearsreportpdf (accessed: June 18, 2021).
xx
PART I
Frameworks
Introduction to Part I: Frameworks

In uniting three different disciplines, Philosophy, Politics, and Economics (PPE) has an array of
traditions, tools, and methods it can draw from. However, there are certain perspectives and ap-
proaches that are most commonly used. To set the stage for the varied chapters to come, the first
section covers broader themes and perspectives used in PPE research. Instead of being united around
a central subject matter, these chapters introduce ideas used to analyze different social matters.
In the first chapter of this section, Geoffrey Brennan and Geoffrey Sayre-McCord defend a
particular view of PPE as a research agenda. For them, PPE is a necessary complement to existing
disciplines. It does not seek to replace or displace Philosophy, Political Science, or Economics.
Instead, there is recognition that we understand social phenomena best when we can view them
simultaneously from different perspectives. They develop this idea in two ways. First, they distin-
guish specialization within intellectual pursuits from specialization in economic production. This
contrast shows why we cannot assume that the division of epistemic labor into distinct disciplines
yields the same positive results as the division of productive labor. Second, they emphasize that the
three PPE disciplines each have certain tendencies that can create tensions between them. To best
understand complex social phenomena, these tensions need to be resolved and interdisciplinary
investigation forces that resolution. They catalog seven points of contention. Altogether, the chapter
motivates PPE as a truly interdisciplinary academic venture, complementing each of the existing
disciplines as it continues to grow.
Across the social sciences, understanding advances through the use of models. This is especially
obvious in economics, where models of human decision-making that were used to model markets
are now used to model broader political and social interactions. While the proliferation of model-
based thinking predates the use of computers, we now have a greater capacity to build sophisticated
models with more complex interactions. Questions remain, however, about exactly what we learn
from a model. Models enable control over variables and can therefore seem similar to experimental
methods, but models are not studying the real world. They are representations of the real world.
Perhaps they are abstractions, or generalizations, or simplifications. Regardless, it is not obvious
what conclusions we can draw about reality from a model. In the second chapter, “On Models and
Their Uses,” James Johnson provides a general account of what models do. He argues against a
standard rationale that models are used to generate hypotheses that can be tested in the real world.
Instead, he argues that models are fables. They are fictions that condense reality down to the core
DOI: 10.4324/9780367808983-1
Frameworks
ideas needed to make a point. In this way, what models do is direct our thought to something about
the world that is best presented in the simpler format that the model provides.
Often, social phenomena are analyzed in a simple way. We might look to an authority and
suppose that people act in accordance with that authority, or we might imagine a representative
agent and suppose that everyone acts in the way that the agent acts. Of course, reality is often much
more complicated than this. For much interaction, different agents have different information and
different motivations. They each respond to the unique circumstances they face, which include the
choices made by other similarly situated agents. To analyze the dynamics of this kind of interaction,
we need new tools. In “Complexity,” Fred D’agnostino explains the newer field of Complexity Studies.
Many different fields study complex phenomena – evolutionary biology, economics, neurology, and
sociology to name a few. Across these fields, there are commonalities and tools. D’agnostino introduces
the history of this field and then identifies its object of study, complex systems. He shows the commonly
used ideas to understand complex systems and then identifies the places in political science, economics,
and philosophy where this approach can be fruitfully applied.
In the fourth chapter, Vanessa Wills emphasizes the views of a particularly influential PPE
scholar. There are, of course, many important figures associated with all three PPE disciplines. John
Stewart Mill, Friedrich Hayek, Amartya Sen, and Elinor Ostrom are mentioned frequently in this
handbook. However, the most influential writer must be Karl Marx. It is hard for anyone to rival
Marx in terms of the impact on Politics, Philosophy, Economics, and world history. In her chapter,
Wills asks what Marx’s own attitude towards PPE as an interdisciplinary project would be. While he
started writing on Philosophy, his later work was focused on Economics. Was there one field that he
thought had greater importance? Wills argues that Marx himself sought to unify study and this
required taking up a particular perspective. His approach was historical materialism; to recognize the
course of history as best explained by the development of productive capacity. So, rather than simply
unifying Philosophy, Politics, and Economics, Marx sought to unify all social science as a form of
history. He also argued against any “objective” analysis that was divorced from a social perspective.
Instead, he explicitly argued from a proletarian perspective because it had the potential to free all of
mankind from alienation and exploitation.
In the last chapter of the section, Ann Cudd explains recent work on four topics that have been
advanced from an explicitly Feminist perspective. Each topic draws on and contributes to inter-
disciplinary work in PPE. First, she offers a general account of the nature of oppression and how it
manifests itself. Rather than being a single force, oppression in an interlocking network of norms,
incentives, and beliefs that might be individually insignificant but together cause deep harm. Second,
she emphasizes the ways that the distribution of caring labor manifests injustice. This provides a clear
example of how myriad social causes, some obvious and some subtle, produce systematic differences
in opportunities for men and women. Third, she examines the ways that our preferences are in-
fluenced by the social structure. Without knowledge or intention, the preferences we form can
reinforce existing patterns of patriarchy. Finally, she turns to the question of whether capitalism is
best understood as a force for alleviating or enforcing the oppression of women.
1
PPE AS AN INTELLECTUAL
ENTERPRISE
1 The General Case for (Some) Interdisciplinary Work

Anything that throws light on its subject matter also casts shadows. So much is just in the nature of
light. And those shadows cannot be dispersed simply by increasing the intensity of the light: doing
that just makes the shadows deeper. Of course, if the intensity of the light is reduced, so is the
capacity to see anything clearly. If you want to illuminate the shadows, you have to come at things
from different perspectives: you have to shine some light, simultaneously, from different angles. And
as every theatre-lighting engineer knows, in order to establish a shadow-free stage, you have to
secure an appropriate balance among the various light sources.
This picture animates our view of PPE – that is, Philosophy, Politics, and Economics – as an in-
tellectual enterprise. Bringing philosophy, political science, and economics together in the study of
social and political institutions sheds light and banishes shadows in ways that no subset of these three
disciplines could possibly accomplish. For example, economists shed invaluable light on markets,
efficiency, incentives, and trade-offs, but regularly ignore the formal and informal institutions that
structure choices and social interdependence; and economists (almost as a matter of professional
pride) steer clear of questions of rights, or justice, or value. Meanwhile, political scientists bring a
deep understanding of institutions that structure choices and create social interdependence, along
with their origin and structure and effect. Political scientists also explicitly consider positive rights
and values as they find expression through and in those institutions. Yet, political scientists often
leave to one side perverse incentives and markets (except as phenomena to be regulated or left alone)
and questions concerning institutionally transcendent rights, justice, and values. Philosophers, of
course, talk ad nauseam about rights, justice, and value, and they are fully engaged with questions
concerning ideal political institutions and perfect markets, but they regularly ignore actual social and
political institutions and how they work, or fail to work, and the incentives, perverse and otherwise,
that they create.
It is worth emphasizing at the outset that this view of PPE is both pro- and counter-disciplinary.
It is pro-disciplinary in that it insists that disciplines are crucial sources of illumination and are
indispensable.1 However, because the picture highlights what each of the disciplines (inevitably)
occludes, it foregrounds the importance of reaching across disciplinary boundaries.
Reaching across these boundaries, though, is a huge challenge. Often, the philosopher will be
drawing distinctions that seem pointless or fetishistic to the economist and political scientist. Or the
economist will be making points that will seem to the philosopher and the political scientist totally
DOI: 10.4324/9780367808983-2 3
beside the point. Or political scientists will be insisting on pieces of empirical reality or aspects of the
current political process that do not seem to the economists or the philosophers to fit into any larger
picture of what really matters.
In our experience, genuine interdisciplinary communication requires a certain creative imagi-
nation and patience as well as mutual respect and trust from all parties: otherwise, there is simply a lot
of talking at cross-purposes (which frequently hides mutual misunderstandings and disagreements
that might, once recognized, be fruitfully explored). And if that is the outcome, there will be a
consensus among all participants that, whatever else, the exercise of trying to talk to one’s dis-
ciplinary neighbors is not worth the time.2 As a matter of empirical fact, the amount of intellectual
engagement between disciplinary departments (at least within the Humanities and Social Sciences)
seems, in most places, to be negligible. It would simply not occur to the average economist to attend
a seminar in the Philosophy department for example. Of course, time and intellectual effort are
scarce – and the professional rewards are greater within, and senses of obligation greater towards,
one’s own patch. If practice is any guide, it seems that prevailing norms within the modern uni-
versity do not encourage – and in many cases actively discourage – serious interdisciplinary en-
gagement. If that is a fact, as we believe it to be, then it is one that deserves recognition, at the very
least because it forms part of the environment within which PPE is to be pursued.
We should note that our particular view of PPE is not the only one available. One might, for
example, advocate PPE as the “right way” to do normative social theory – that PPE ought to replace
Political and Social Philosophy, or that it ought to replace Economics and/or Political Science as
separate activities. Such a view harks back to an earlier period when disciplinary boundaries were less
clearly drawn – and sees PPE as recapturing the mental landscape of giants like Smith and Hume.
The folk version of this conception is that Philosophy brings to Economics a broader set of mo-
tivations and a richer (and more plausible) normative framework; while Economics and Political
Science bring to Philosophy a greater recognition of the relevance of facts and a more practical
impulse; while Economics and Philosophy both bring to Political Science a greater analytic structure
and hence more theoretical rigor. All these additions are desirable, but such an approach treats PPE
as a direct rival to the component disciplines. It treats PPE as operating, in the final analysis, on the
same level as, and competing with, the individual disciplines, rather than as depending crucially on
them. We hope it will be clear that this is not how we see things.
There is a third view of PPE that is not really interdisciplinary at all. This view is that each
discipline has some things that it can learn from its neighbors – a mathematical technique that, for
example, Philosophy might usefully borrow from Economics; or robust empirical findings about
politics that Economics might absorb from Political Science; or a probing distinction from
Philosophy that Political Science might embrace. To the extent that such borrowing makes for
better Philosophy or better Economics, or better Political Science, there can be no serious objection
to it. But this third view involves a resolutely disciplinary perspective, in that the test of success is
whether something is added to the borrowing/raiding discipline. PPE, on our reading, appropriately
has broader aspirations.
Our title is designed to draw a distinction between PPE as an intellectual enterprise and PPE as a
pedagogical one. PPE as an undergraduate program (or as an element in a postgraduate one) has many
virtues, most of which relate to the capacities it encourages and develops in students – capacities, for
example, to think analytically; to synthesize diverse perspectives into a coherent view; and to temper
idealism with realism, without losing sight of the ideals. The justification for PPE as a scholarly activity
by contrast lies in its usefulness in illuminating its subject matter rather than in the intellectual qualities
it develops in and requires of its practitioners. Yet, there is an important sense in which the intellectual
and the pedagogical enterprises, as we understand them, have common ground: a commitment to
amalgamating and synthesizing the separate disciplines in a way that a mere juxtaposition of disciplinary
specializations fails to do. What that process of amalgamation and synthesis involves, and secures, and
4
PPE as an Intellectual Enterprise
why juxtaposition of differing specializations falls short, are interesting questions in their own right. We
shall say a little about this aspect of PPE in what follows.
Of course, PPE qua PPE need not be seen as a substitute for juxtaposing disciplinary perspectives
on some topic but rather as a complement to them. The PPE project of integrating disciplinary
perspectives provides, as we see it, an enriched understanding of the single, shared, phenomenon –
social and political institutions.
2 Disciplines and the Division of Epistemic Labor

One might think of disciplines simply as markers in the epistemic division of labor – as mere
reflections of the specializations that disciplines instantiate. And in that spirit, it is tempting to refer
to the early chapters of Adam Smith’s Wealth of Nations in which specialization and the division of
labor play such a central explanatory and normative role.3 Such temptation reflects both Smith’s
status as a proto-typical PPE scholar and Smith’s explicit reference to specialization within in-
tellectual enquiry. As Smith points out:
In the progress of society, philosophy or speculation becomes, like every other employment,
the principal or sole trade and occupation of a particular class of citizens. Like every other
employment too, it is subdivided into a great number of different branches, each of which
affords occupation to a peculiar tribe or class of philosophers; and this subdivision of em-
ployment in philosophy, as well as in every other business, improves dexterity, and saves time.
Each individual becomes more expert in his own peculiar branch, more work is done upon the
whole, and the quantity of science is considerably increased by it. (Wealth of Nations I.1.ix)
We have no quibble with the thrust of Smith’s claim: specialization and division of labor in in-
tellectual enquiry does offer considerable advantages. Yet Smith’s reference to specialization in
enquiry as an instance of the operation of the more general engine of wealth generation raises three
topics worth discussing here.
In the first place, the explanation of specialization that Smith offers in the Wealth of Nations does
not, we think, really fit intellectual enquiry, even as intellectual enquiry exhibits significant speciali-
zation. Specifically, that account treats efficiency and a desire for trade as the engines of specialization.
While efficiency and the desire for trade are clearly implicated in some intellectual specialization, we
think a lot of intellectual specialization has other sources, which must be independently explained.
In the second place, Smith’s own treatment allows, as we think it should, that there may be more
general limits to specialization’s advantages and indeed leaves room for the possibility that the degree of
specialization may be carried too far. That possibility will be, on its face, of special interest to the
interdisciplinary scholar. Much of the interest of interdisciplinary work lies in the recognition that
overspecialization often leaves underexplored issues and opportunities that fall between the speciali-
zations that happen to have developed.
And finally, in the third place, while Smith understandably highlights the advantages of spe-
cialization, the costs (as Smith recognizes) deserve attention. We think these costs, when it comes to
intellectual enquiry, are often significant. In many cases, they are no doubt worth paying. Yet in
others, they strike us as worth taking account of and addressing (more or less directly).
We seek in what follows in this section to address each of these three issues in turn.
2.1 Why the Origins of Specialization in Intellectual Enquiry Are Different?

In the Wealth of Nations, Smith’s explanation of specialization and the division of labor and their
attendant advantages appeals to our interest in exchange. As he puts it (at the outset of chapter 2):
5
This division of labour, from which so many advantages are derived, is not originally the
effect of any human wisdom, which foresees and intends that general opulence to which it
gives occasion. It is the necessary, though very slow and gradual, consequence of a certain
propensity in human nature which has in view no such extensive utility; the propensity to
truck, barter, and exchange one thing for another. (Wealth of Nations I.2.i)
For Smith, this “propensity to truck, barter, and exchange” finds expression in making and taking
offers of the kind: “Give me that which I want and you shall have from me that which you want.”4
The picture is one in which producers of knowledge “sell” their expert knowledge to individuals
who want to use it, and the engine of specialization is the desire to have more to trade. But much
“knowledge” is not produced in this mode. Often, enquiry starts with curiosity, not thoughts of
advantage, and its fruits are often made available free of charge. In Adam Smith’s day, the progress of
intellectual enquiry depended on “gentlemen scholars” (they were mostly men, sadly) driven mainly
by intellectual curiosity. Many had private means; some had university posts, but these were not
particularly well remunerated and certainly the compensation attached bore little relation to the
extent or significance of the incumbent’s research output.
In the contemporary world, intellectual enquiry is often pursued within universities – where the
relation between salary and research output – and still less the usefulness of research output – is vague
and indirect. To a significant extent, a scholar’s salary depends on whether others in the field regard
that person as a “great scholar, doing first-class work.” As a result, academic scholars write primarily
for other academic scholars – the professional standing of each depends on what the academic
audience makes of what that scholar writes. This is not to deny that the “usefulness” of research may
be one criterion in determining whether the “work” in question is “first-class”; but it is clearly not
the only criterion and in at least some fields usefulness may not be relevant at all. Nor is it to deny
that there is considerable serious research going on in pharmaceutical companies, and among
medical technology manufacturers, and computer manufacturing giants and so forth – research that
is evaluated in terms of its capacity to contribute to commercially viable “goods.” Much of this latter
research will be obedient to the exchange principles that govern Smith’s trading nexus: yet a good
deal of science – physical and otherwise – is not.
In short, it seems clear that a major motivation for enquiry has been (and remains) intellectual
curiosity as an end in itself – either on the part of the scholar herself or her intended audience (or
both). Scholars pursue lines of enquiry because they find them interesting and because others whose
opinions they care about find them interesting.5 Specialization in this setting is pursued precisely
because of its advantages in sharpening skills needed to pursue interesting questions and solve
challenging intellectual puzzles. Training within a discipline is in many ways akin to the aspiring
tennis player committing hours to practice his serve, or the golfer spending time and effort in trying
to improve her putting, without thought of turning professional.
Our point here, we should emphasize, concerns only the influential “market explanation” of
specialization offered in the Wealth of Nations. Elsewhere, Smith himself notes the inadequacy of
such explanations when it comes specifically to intellectual specialization:
Wonder, therefore, and not any expectation of advantage from its discoveries, is the first
principle which prompts mankind to the study of Philosophy, of that science which
pretends to lay open the concealed connections that unite the various appearances of
nature; and they pursue this study for its own sake, as an original pleasure or good in itself
without regarding its tendency to procure them the means of many other pleasures. (The
History of Astronomy, III.3, 1795)
6
Taking account of this important motive for intellectual enquiry and the specializations that
it generates is crucial to understanding, among other things, how and why intellectual labor does
manage, so often, to make substantial contributions to the technical progress and the “wealth of
nations.”
There will doubtless be some demand from governments (and the electorate at large) for publicly
funded universities to demonstrate their usefulness; and some internal impulse from scholars
themselves to justify the value of their activities to a wider audience. Exploring how the non-
commercial interests that operate on intellectual enquiry actually work and what their upshots are, is
an important topic for enquiry in itself (perhaps in a PPE mode). For our purposes here, though, we
are content to note that the intellectual specializations are not well explained without recognizing
the extent to which those who engage in enquiry are motivated by curiosity and wonder and not
simply market considerations. To ignore this fact risks crushing the spirit that makes progress, in-
tellectual and otherwise, possible.
2.2 The Limits of Specialization

In chapter 3 of the Wealth of Nations, Smith argues that one prime constraint on the extent of
specialization is the “size of the market.” As he makes clear, the size of the market can effectively be
increased in a number of ways – by reducing transport costs so that a producer can reach a larger
number of customers; by increasing aggregate population; and/or by increasing the density of
population. But these alternative ways of increasing numbers (and thus the degree of specialization)
indicate the relevance of additional considerations in determining how far specialization extends. For
example, Smith contrasts agriculture with manufacturing: farmers, Smith thinks, must have multiple
skills – and this because the land cannot sustain enough workers to allow each agricultural worker to
fully specialize. Increasing the density of the agricultural population in the interests of greater
specialization would reduce, not increase, the per capita income of agricultural workers.
Furthermore, as Smith explicitly noted, specialization is likely to impose intellectual and personal
costs. Concentrating on a particular narrow activity, Smith thought, made many workers “as stupid and
ignorant as it is possible for a human creature to become… Dexterity at his own particular trade …
seems to be acquired at the expense of his intellectual, social and martial virtues.”6 It is not entirely clear
whether Smith thought that workers gave up their “intellectual virtues” unconsciously – becoming
preoccupied with the minutiae of their particular trades – or that they just endured the boredom for the
sake of the additional material benefits their specialized jobs delivered. The former is the more sinister –
at least for those specialized in intellectual enquiry itself because it suggests that boredom may not
operate as an effective constraint against excessive concentration on particular questions or particular
methods.
It is worth noting that there are some arenas in which specialization does not seem to promise
significant advantages. Take a simple example. Suppose there is a difficult sum. We can imagine that
many people might exercise their minds with this sum, but adding additional people may not reduce
the difficulty or promise a more rapid solution. Each additional person will simply replicate the cal-
culations made by all the others. Contrast that case with a mathematical exercise that can be broken
down into separate parts: in this case, there is a potentially profitable division of labor – with each
individual concentrating on a different aspect of the problem. Even here, however, although there are
advantages in additional numbers, there may be no scope for specialization as such: the advantages of
specialization that Smith is focusing on derive from the development of different skills – ones that arise
from the repetition of a distinctive activity. Only if the different aspects of the puzzle involve re-
cognizably different skills and only if those different skills are applicable to a wider range of puzzles,
might there be advantages in the participants in puzzle-solving specializing.
7
2.3 Disciplines as Certifiers

One important constraint on the operation of Smith’s “community of exchange” is the extent of
trust and trustworthiness within the trading nexus. Bee Wilson (2008) illustrates the point nicely in
relation to food additives through the 19th century. When individuals lived on small rural holdings
and prepared their own food from raw ingredients, they did not need to rely on others for the
quality of their food. Once the economy had developed to the point that their diet consisted largely
of “prepared” foods – cheeses, pies, commercially prepared sausages, etc. – consumers were then
vulnerable to adulteration of the foods they consumed. If people believed that suppliers of prepared
foods were untrustworthy – that food additives were as likely to be damaging as advantageous – then
the division of labor in food preparation would never have developed in the way that it did. The
point generalizes. Commercial society of the kind that Smith describes depends on trust. The di-
vision of labor is limited not just by the size of the market but also by the amount of trust operative
within the market system.
Although “enquiry” is not exactly an exchange process (for reasons discussed earlier) the problem
of trust extends to specialization in intellectual enquiry. Securing the benefits of intellectual division
of labor requires two means of coming to know – one by direct evidence (say, by conducting an
experiment oneself); and one by the testimony of someone else who one can trust is reporting the
results accurately. If the second route to knowledge were not available, each would have to derive
and examine all of the evidence individually, thus undermining the point of division of intellectual
labor. The advantages of intellectual division of labor depend on the trustworthiness of testimony.
Such trust is not exclusively a matter of the dispositions of participants. It is also mediated by
institutional arrangements – ones that certify the properties of the information that is available, that
reward trustworthiness and punish trustworthiness, and that detect and make public which is which.
On our view, this is a valuable role that disciplines play in academia. After all, a large amount of time
and energy within the organization of enquiry is taken up not just in training individuals to the tasks
at hand but also: (i) in certifying whether or not particular individuals have achieved an appropriate
level of competence; (ii) in certifying (via refereeing procedures) the quality and reliability of specific
research output; and (iii) in certifying the credentials of other scholars by writing referees reports for
persons whose qualities as a researcher you are judged to be in a position to assess. Professional
reputation is a critical asset for all those for whom “speculation … is a chief occupation or trade”; and
both formally and informally, people monitor the performance of others and contribute to that
reputation. But each does so within a highly partitioned domain. A department Head in Chemistry
would not normally rely on a reference relating to a new hire from a geologist – still less from an
economist or a philosopher. She would (rightly) look to the judgment of acknowledged experts in
the relevant discipline, and probably in the relevant subfield within that discipline. Indeed, each
discipline might plausibly be understood in terms of the set of mutual certifiers.
That established disciplines provide this important “certification” service needs to be borne in mind
when people call for tearing down the silos and dismantling the disciplines. Some commentators rail
against the hegemony/tyranny of disciplines in a manner that makes their attitude seem not so much
interdisciplinary or multidisciplinary but rather anti-disciplinary. Or at least, non-disciplinary. That is
not in our view a proper attitude. Disciplines have some valuable features, not least that they are “trust-
justifying” institutions.
It is time to summarize the burden of this section. Adam Smith’s few throw-away lines about
specialization in the organization of enquiry, early in the Wealth of Nations, encourage seeing enquiry
as just another example of the specialization that is characteristic of market economies.7 Seeing
things this way is, we think, a serious mistake. It underestimates what is distinctive about specia-
lization in enquiry; and, more to the point for our purposes, misses much that is important about the
structure and operation of intellectual disciplines. We have tried to catalog here what some of the
8
relevant aspects of intellectual disciplines might be. All told, they leave us thankful that there are
disciplines, especially for the specialization and trust they make possible, even as we advocate for
intellectual work that actively reaches across disciplines on the grounds that staying within disciplines
has its own serious costs.
3 Shadows and Conflicts

Earlier we suggested that PPE, conceived as a domain of enquiry, involves commitment to the idea
of PPE as an enterprise that integrates the three disciplines. One upshot of this thought is that we
ought to be especially attentive to the ways in which philosophy, political science, and economics
are often in tension, if not direct conflict: cases where what is a mainstream view in philosophy is
widely held to be eccentric or outright wrong in economics or political science. And vice versa. We
shall offer a few examples shortly. In each case, we see the tensions (and conflicts) as intellectual
opportunities that consistently hide in the shadows so long as one operates only within one or
another of the disciplines.
Our first-round hypothesis in relation to such apparent tensions and conflicts is that they reflect the
(necessarily) partial vision of the disciplines involved. The broader view that PPE offers enables either a
reconciliation or a modification in all disciplinary positions that provides a better fix on the matter at
stake. But of course, we cannot rule out the possibility that a particular discipline is just wrong on a
substantive matter8 – or that we should suspend judgment on some proposition on which there is
overwhelming consensus in some discipline because that proposition is staunchly rejected in another.
Whatever response is appropriate, we think PPE as an area of study requires us to be especially attentive
to the relations among the disciplines – to the places and manner in which they seem to be at odds. Such
areas strike us as constituting a core piece of the PPE enterprise. Or at least, the tensions often reveal
especially promising opportunities for PPE research. And on that basis, we conclude this chapter with a
number of examples that illustrate apparent conflicts (sometimes of emphasis, sometimes of substance)
and which might be thought of as primary elements in an agenda for PPE scholarship.
This list of examples is by no means exhaustive – nor are we thinking that they reflect the most
significant points of tension within PPE. They are simply some that strike us as both significant and
ripe for exploration.9
3.1 Categorical vs. Comparative Thinking

John Broome draws a contrast between philosophers and economists in terms of a distinction be-
tween categorical and comparative thinking. He points out that philosophers tend to draw dis-
tinctions and think in terms of categories – true/false or right/wrong or good/bad, for example,
while economists tend to traffic less in distinctions and to think more in terms of comparisons – say
in terms of credence (which presumably comes in degrees) rather than belief or, in the ethical
domain, in terms of betterness rather than rightness.
Of course, not all economists think comparatively in all areas. For example, Arrow’s famous
impossibility theorem and the social choice framework of which that theorem is the most famous
fruit, typically deal in terms of stipulations (categorical specifications). Nor do all philosophers think
categorically. For example, many formal epistemologists deal with their subject matter in terms of
metrics of epistemic warrant rather than belief; and consequentialists in ethics and political philo-
sophy think in terms of trade-offs and optimal outcomes. Still, one might concede, there is a dif-
ference in approach and attitude. And that difference, to whatever degree it exists, might invite the
questions which is right? or which is better?
What seems clear to us, is that forcing a choice here, or thinking one mode of thinking has simply
gone off the rails, is a mistake. Economics can, for example, hardly dispense with its categories. The
9
claim that A is “more X than Y” or that “A is more X than B is” – which are on their face comparative
claims – seem to depend on categories X and Y and for that matter on the A and the B being well-
defined. Just what is at stake in Broome’s distinction and in what domains it matters raise interesting,
and we think unresolved, issues – ones that would seem to be of special relevance to PPE.
3.2 Motives vs. Incentives

Economists are (or purport to be) interested primarily in human behavior – or (in fact more
commonly) in changes in that behavior in response to changes in external circumstances. The
external circumstances they focus on in most cases are changes in relative prices – in the “incentives”
people have to act in particular ways. Given the basic behavioral model deployed in economics (the
“rational” maximization of preference satisfaction subject to relevant scarcity constraints), changes in
beliefs are also a possible source of behavioral change; but as a matter of general practice, belief
change (and preference change) are generally backgrounded by economists in favor of explanation
based on incentives/relative prices. One of the contributions that so-called “behavioral economics”
has made to this general exercise has been to emphasize the role of additional behavioral triggers –
ones that lie outside the standard “rational choice” model. To the extent that such additional triggers
put pressure on that standard model, behavioral economics is seen as a challenge to the mainstream
approach – but behavioral economics remains economics in that it exhibits mainstream economics’
preoccupation with behavior.
Economists tend not to be interested in the mental processes individuals go through in deciding
how to act – or in the considerations that induce them to act as they do. Although economists talk
loosely about “revealed preference,” as if behavior did genuinely reveal something about the mental
processes that produce the behavior in question, there is actually little interest among economists in
those mental processes as such.
Philosophers and political scientists by contrast are often vitally interested in the motives that
induce action – in the kinds of beliefs and values and attitudes that combine to produce behavior.
And they are interested in such things in part for their own sake and not merely as tools for
predicting behavior. Often, they are interested in behavior because of what the behavior reveals
about beliefs, values, and attitudes, not the other way around.
Now, one might respond that there is no real conflict here – that there are just two different (if
somewhat related) objects of study. But that underplays the contrast. Economists traditionally see beliefs,
values, and attitudes as inferable from a specification of the (non-psychological) conditions and the
resulting behavior, whereas philosophers and political scientists traditionally see them as pre-existing and
independent causes of the behavior they allow us to predict and explain. Traveling with this difference
is a contrast in the weight given, or not, to how beliefs, values, and attitudes are influenced or changed.
The issues here strike us as partly methodological, and partly normative. We are not confident that
we ourselves fully understand what is at stake, but there do seem to be a variety of questions here that it
would be helpful to unravel. And such unraveling seems like a good project for the PPE scholar.
3.3 Motivational Assumptions

Of course, economists do famously regularly make some assumptions about values or attitudes. Ask
for an “economic” explanation of some phenomenon and the assumption of material self-interest
will not take long to appear. So, although it is technically the case that utility functions can include
all kinds of considerations, in practice the tendency of economists is to assume that material self-
interest governs most of the decisions. Homo Economicus is both “rational” and self-interested – with
self-interest here understood not in terms of the fact that A’s preferences are indeed A’s, but rather
that the content of A’s preferences involves A’s material consumption, relatively narrowly construed.
10
That A’s preferences might include the well-being of others (both positively and negatively valued)
or that it might include obedience to certain norms or habits – or even the requirements of virtue –
are possibilities from which economists tend to think one can safely abstract in most applications.
Political scientists and philosophers tend to import a considerably richer psychology.
It is worth noting here that the issues at stake are basically empirical ones (at least on their face).10 But
there are matters of debate about how fine-grained motivational assumptions should be. In work done
by one of us (Brennan and Pettit, 2004), the assumption that individuals desire the esteem of their
fellows (a long-standing assumption in social theory, endorsed by Adam Smith among many others) has
been explored within an analytic frame that borrows strongly from economics. The significant feature
in this work is that individuals care about the judgments and attitudes of their fellows (and have
incentives to behave in ways of which those others approve and avoid behaving in ways of which they
disapprove) without any departure from egoism: each is taken to care primarily about the esteem she
herself accrues. There is no necessity to invoke concern for the esteem in which others are held.11
One might make a related point in the contrast between economics and political science. The
“economic” approach to politics has, for the most part, attributed to voters (and to political agents
such as politicians and bureaucrats) the same self-interest that those same individuals exhibit in
markets. We think that that particular motivational assumption is misplaced – that the direct ex-
trapolation of behavior from market settings to the political fails to take adequate account of the
differing circumstances of political and market choices (see Brennan and Lomasky, 1993). Political
Science is much more open concerning the things that might figure in the motivations of voters (and
of those who are ultimately subject to voters’ approval): political scientists tend to come at this issue
inductively – as an empirical matter to be explored “without prejudice” rather than something
where market behavior provides the relevant presumptions. As a result, allegiance, party mem-
bership, identity, or values come on the scene as relevant to political choices in ways that resist
reduction to a person’s concern with herself, let alone her own material well-being.
There is a further (distinct) question about how much allowance should be made for motivational
heterogeneity. Philosophers and political scientists are more disposed to allow for differences in
motivation within the population; whereas economists are prone to assume homogeneity (or treat
heterogeneity as irrelevant or as safely ignored for explanatory purposes).12 The effects for analysis
are significant. Absent a recognition of heterogeneity, there is no scope for selection effects in the
analysis of institutions or policies (and these, unfortunately but not inevitably, play a minor role in
economic analysis).
The bottom line is that such motivational issues arise as a core topic in PPE – partly because each
of the disciplines, as it seems to us, tend to ignore the richness of the phenomena they are all
exploring. Our own view here is that the assumption of predominant self-interest uniformly across
the population is of rather limited usefulness even in traditional “economic” applications – and
becomes increasingly strained as economists stray further from those traditional applications.
3.4 Normative Considerations

One conspicuous difference between economics, political science, and philosophy lies in their
different conceptions of “normative desirability.” Of course, economists often declare that judg-
ments of value have no place within economics at all.13 But, in practice, economists seem quite
uninhibited about making policy recommendations based on their judgments about normative
desirability. Yet those judgments often reflect a narrow (and, in our view, implausible) conception
of what counts as important, both when it comes to ethics and when it comes to what is rational. On
the view that we see as common within the standard economic framework, the only thing that
counts normatively is preference satisfaction. Accordingly, liberty, justice, equality, virtue – these all
count exactly to the extent that individuals have preferences for them (which is shown directly by
11
how much the individuals are prepared to sacrifice to secure them). Moreover, preference sa-
tisfaction is routinely treated within economics as coterminous with “well-being.” So, for example,
it is simply a conceptual impossibility for a mother to sacrifice her own well-being in order to
increase the well-being of her child because any such sacrifice must – thanks to the assumptions that
choice reveals preference – increase the mother’s own preference satisfaction and therefore her own
well-being. Equally, soldiers who heroically sacrifice their lives to save their fellows – by throwing
their bodies on exploding grenades, say – are revealing their preference that their fellows’ lives
should be saved at the expense of their own. In that sense, the soldiers are entirely “preference-
satisfying.” We are prepared to allow this conception of revealed preference and (revealed) pre-
ference satisfaction. It is surely a stretch, however, to declare that those who sacrifice their lives for
others are always increasing their (own) well-being.
None of these criticisms commits one, of course, to the view that preference satisfaction is not a
consideration in normative evaluation or that perhaps, in some settings, a decisive one. It is simply to
register dissatisfaction with the idea that preference satisfaction (revealed or otherwise) is the only
normatively relevant metric in town.
Political scientists and philosophers do, in contrast, regularly appeal to a richer understanding of
well-being, justice, equity, and rights that are treated as important independent of peoples’ actual
preferences. We are both on this side of the contrast, when it comes to the normative evaluation of
behavior, preferences, policies, and institutions. But our concern here is not to defend our side, but
to highlight the extent to which thinking about the contrast is relevant to PPE.
3.5 Discounting for Time

Discounting for time is a more specific instance of a case where what is taken as given in one
discipline is routinely rejected in the other. Economists treat as self-evident that one should discount
for time – that it makes sense to value something less the further it is in the future, so that, say, it
makes sense to value getting $5 today over getting the same amount, or even little more, tomorrow.
For philosophers,14 it is equally widely held that discounting is irrational, that the mere fact that some
payoff is in the future, rather than now, should not have an impact on how much one values it. Of
course, usually the fact that some payoff is in the future means that waiting for it carries some risks –
that one will not actually get it, or that when one gets it the $5, say, has less purchasing power – and
those risks mean it may make sense to value future payoffs less. But here what makes how much one
values the payoff is not merely the fact that it is in the future.
Timing issues arise in political science as well of course. Incumbents make choices involving the
spreading of costs and benefits between the current electoral period and future ones. And voters who
in their “economic roles” discount the future might be expected to do so at the ballot box as well.
Do democratic norms about the authority of the “voice of the people” require a corresponding bias
towards benefits in the present (and postponing costs to the future)?
In addressing questions like this, it is reasonable to suspect that there is a lot of “talking at cross-
purposes” – mutual misunderstandings and some confusion on all sides. Do the issues here collapse into
ones already mentioned relating to the normative status of individual preferences or the appropriate way
to understand “rationality”? Or are there some extra considerations in play? We do not aim to sort out
this confusion here15: we simply note that this is another issue with claims to be on the agenda of the
PPE specialist.
3.6 Ideal vs. Non-Ideal Theory

There is a debate – rather confused, as we see it – around so-called ideal vs. non-ideal theorizing in
normative political theory. It seems clear that one thing that is at stake in this debate is the relevance
12
of facts about agent motivation, and what is feasible in light of those facts, or more particularly the
point at which such considerations matter in figuring out which social and political arrangements are
best, and why.
An attractive thought is that there is a distinction between the normative and the positive or
between values and facts, that allows for, and maybe demands, treating questions of value as fun-
damentally independent of facts (including facts about what people are willing to do or accept or
believe). This idea travels comfortably with the thought that it is a philosopher’s job to worry about
the “true” meaning of justice or to specify the domains in which inequality most matters and the
political scientist’s and economist’s job to work out what is feasible in the pursuit of greater justice or
greater equality, so defined. This picture suggests a neat separability between the normative and
positive and one that falls conveniently along disciplinary lines. But that picture is at least open to
question.16 More than a few students of politics think that the idea of doing normative analysis
before considering the basic facts about how politics actually works is either flying blind or (at best) a
matter of building castles in the sky.
3.7 Methodology and the Philosophy of Science

Philosophy and social analysis (whether of an “economic” or “political” kind) meet at two points: in
relation to applications in which ethical considerations are in play; and in relation to general
methodological questions. In this latter connection, one might expect that philosophy of science
might have things to say to both economics and political science that would be helpful to those
disciplines – and that issues in those disciplines would provide grist for philosophers of social science.
To take a specific example from the political setting, there is an issue as to how causality is to be
understood in settings of collective determination. In what sense, for example, do individual voters
“cause” the outcome of elections in a democratic polity, when typically each individual voter’s vote
has no influence on who wins?17 This kind of issue is especially relevant for economics where
“methodological individualism” is standardly taken to be a core commitment of the basic method.
What status do collections of individual agents have in any explanatory scheme that insists on some
kind of ultimate reducibility of social phenomena to individual action? Alternatively put, how
exactly is “methodological individualism” to be understood in the social sciences?
It is an interesting feature of economics in particular (at least in its mainstream variant) that it is
committed to what, on the face of things, seems a quite specific method – with rather little in the
way of methodological reflection. George Stigler, Nobel Laureate in Economics, used to insist that
no economist should turn to methodological questions before the age of 60 – and this appears to be a
widespread view within the profession. One of the advantages of PPE, as we see it, is that the
evident variety of methodologies used to understand social and political institutions encourages
methodological self-consciousness.
The situation in Political Science and its connections with methodological issues is different from
that in Economics: there certainly seems to be no prejudice against a concern with method within
political science. The difficulties in political science reflect more the fracturing of the discipline itself.
The absence of a unified method in political science18 means that a conversation on methodological
issues between philosophy or economics on the one hand, and “political science” as a whole, on the
other is hard to imagine. It seems clear however that issues of method are in play in the economics/
political science intersection – and specifically in the application of economic methods to the study of
political processes (or what we might term “rational actor political theory”). Philosophers of science
may be useful brokers in sorting out some of the issues at stake in that particular intersection – and that
would seem a natural place for PPE to start.
However, it would have to be conceded that a good deal of philosophy of science is so abstract
that it is hard to see how it would be much help in that task. As an example of something to be
13
striven for, we are impressed by the extent to which many prominent “philosophers of biology” are
also players in biological/evolutionary theory: this area seems to be one where the philosophy of
science and the practice of a discipline come together in mutually profitable ways. That kind of
relation seems to us a reasonable aspiration concerning the interplay of the philosophy of social
science and the social sciences.
The general thought we have wanted to advance in this section is that significant points of
tension between the participant disciplines represent a natural point of departure for PPE scholar-
ship. There are in fact lots of such “points of tension” – and the object in this section has been just to
offer a few examples that strike us as interesting and that we think make some claims on the attention
of PPE scholars. The list is not supposed to be exhaustive. One thing however is clear: such issues
are not likely to be a high priority for those who see their audiences as lying resolutely within any of
the component disciplines taken separately.
4 Conclusion
In this chapter, we have offered a picture of PPE as an intellectual enterprise that is decidedly pro-
disciplinary while, at the same time, being committed to transcending disciplinary boundaries and
paying special attention to the questions and issues that are cast in the shadows of one or another (or
even all) of the disciplines. The pro-disciplinary features arise from a recognition of the important role
that disciplines play in the organization of enquiry and the presumptive authority that each of the three
disciplines involved in PPE carry as separate enterprises. One upshot of this approach is that we reject a
conception of PPE as itself a rival discipline – an alternative to philosophy, political science, and
economics as traditionally pursued. We are not denying, though, that PPE is an area of specialization
and might be, or might become, a “discipline” – albeit one that will always be somewhat parasitic on its
component disciplines.
At the same time, PPE is clearly a “counter-cultural” activity in the organization of enquiry. It
recognizes that disciplines come at a cost – they have their blind spots and their downsides – and part
of the role of PPE is to try to offset those costs and exploit the opportunities the disciplines leave
unexplored.
Our picture involves the thought that PPE, if it is to be properly pursued, must be undertaken (in
large part, at least) as an integrative enterprise – one that helps itself to the tools of philosophy, political
science, and economics to look carefully into the shadows left unexplored by philosophy, by political
science, and by economics, left on their own. Yet, in addition to exploring the shadows, PPE (as we
understand it) finds opportunities in negotiating and trying to reconcile tensions among the disciplines,
taking those tensions as, themselves, signs of what we do not yet understand. Indeed, as we see it, those
points of tension – operating at all sorts of levels – constitute a natural point of departure for PPE
scholarship; and, in that sense, reflection on where those points of tension lie represents a natural point
of departure for what is, after all, still a somewhat nebulous and ill-defined niche in the overall division
of intellectual labor.
Notes
1 In that sense interdisciplinarity is to be contrasted with “nondisciplinarity” or “antidisciplinarity” with
which it is sometimes confused.
2 Talking to one’s disciplinary colleagues about neighboring disciplines (and the apparently ridiculous things
that preoccupy them) is, of course, not at all the same thing. That kind of conversation can be fun. But it
hardly meets the central purpose of PPE.
3 Importantly, specialization and division of labor are separate phenomena, at least to the extent that people
might divide up jobs without there being any real specialization required – people might be able to do any
of them and might even cycle through different jobs over time.
14
4 As Smith puts it later in chapter 2, the issue at stake is not whether you and I want the objects of exchange
for their own sake or for some further purpose, but whether exchangers must give up something to get what
they want (for whatever reason).
5 There is in these cases a market of sorts for what is of interest to others, especially once interest is recognized
as the coin of the realm. Yet even allowing that, there is, we think, a large amount of intellectual activity,
and specialization in the process, that is driven by the scholar’s interest without regard for, or sensitivity to,
whether others are or will be interested.
6 Here is the full passage: “The man whose whole life is spent in performing a few simple operations, of
which the effects are perhaps always the same, or very nearly the same, has no occasion to exert his
understanding or to exercise his invention in finding out expedients for removing difficulties which never
occur. He naturally loses, therefore, the habit of such exertion, and generally becomes as stupid and ignorant
as it is possible for a human creature to become. The torpor of his mind renders him not only incapable of
relishing or bearing a part in any rational conversation, but of conceiving any generous, noble, or tender
sentiment, and consequently of forming any just judgment concerning many even of the ordinary duties of
private life.” An Inquiry into the Nature and Causes of the Wealth of Nations, Book 5, Chapter 1.
7 In some ways, that view is encouraged by Smith’s general posture towards the older Aristotelean idea that
specialization arises out of differences in natural talents: that view, Smith describes in terms of the “vanity of
the philosopher.” Smith’s general skepticism about natural differences was a feature that 19th-century
economists broadly shared: it was this (and not Malthus’ population theory) that induced Carlyle to invent
for economics the “dismal science” tag.
8 Otherwise progress in that discipline would be ruled out a priori.
9 On the other hand, we would be interested if readers have suggestions as to examples we have not included
and might have (and some indication of why they are significant); and would be appreciative of com-
munications to us in that connection.
10 In some cases, it may be appropriate to make “worst case” or “best case” assumptions about motivation. In
political settings, for example, Hume famously assumes that “everyman is a knave and has no purpose in all
his actions other than self-interest” not because this is so, but because, as Mill puts it, “this is the special
circumstance against which constitutions are designed to guard.”
11 The exception would be others with whom one’s reputation is associated. So one might care about the
esteem of one’s academic department and thereby of one’s fellows because of the borrowed glory one might
thereby obtain.
12 After all, all subjects face the same changes in relative prices and so all face the same kinds of incentives to
alter their behavior “at the margin.”
13 Such was the claim in Robbins’s influential treatment (1932).
14 And some economists – notably Pigou and Ramsey.
15 Note though Broome’s suggestion that there is a misunderstanding about what it is that is being discounted –
economists are talking about discounting goods or money values; philosophers are talking about discounting
utilities. This strikes us as a useful point, but we do not think it exhausts the confusion in this domain! There is
for example some evidence that individuals do discount utilities. See Greene and Barron.
16 We have dipped into this topic, together, in a paper prompted by David Estlund’s Utopophobia. See Brennan
and Sayre-McCord (2020).
17 My vote is neither necessary nor sufficient to secure the victory of the candidate/party for which I vote in
say the US presidential election. On this see Goldman (1999), Tuck (2008), and Brennan and Sayre-
McCord (2015).
18 It is implausible to think of a basic textbook in Political Science (if such could be imagined) carrying the title
“The Political Science way of Thinking”: identifying a common perspective across the discipline, beyond
the conviction that politics somehow understood “matters,” presents as a serious challenge.
References
Brennan, G. and L. Lomasky. (1993). Democracy and Decision. New York: Cambridge University Press.
Brennan, G. and P. Pettit. (2004). The Economy of Esteem. Oxford: Oxford University Press.
Brennan, G. and G. Sayre-McCord. (2015). “Voting and Causal Responsibility.” Oxford Studies in Political
Philosophy 164: 235–250.
Brennan, G. and G. Sayre-McCord. (2020). “Real-World Theory, Complacency and Aspiration.” Philosophical
Studies 178. doi: 10.1007/s11098-020-01531-x.
15
Goldman, A. (1999). “Why Citizens Should Vote.” Social Philosophy and Policy 16: 201–217.
Robbins, L. (1932). The Nature and Significance of Economic Science. London: Macmillan.
Tuck, R. (2008). Free Riding. Harvard, MA: Harvard University Press.
Wilson, B. (2008). Swindled: The Dark History of Food Fraud, from Poisoned Candy to Counterfeit Coffee. Princeton,
NJ: Princeton University Press.
16
2
ON MODELS AND THEIR USES
James Johnson
1 Introduction
Start with the obvious: Politics, Philosophy, and Economics (PPE) is an interdisciplinary under-
taking. One way to think about how the three-component disciplines converge is to point out that
they deal with contiguous, sometimes overlapping, substantive matters. I think it is also plausible to
point out that the three disciplines converge not just around questions of what we know but, in
important ways, about how we know and how we know it. Models are central to the way the three
disciplines address those matters.
It hardly is news that economists devote much of their intellectual energy to making and ex-
amining models (Solow, 1997). Indeed, Rodrik (2015, 5) suggests, that precisely this is “what makes
economics a science.” Similarly, noting that using models has emerged as “the dominant feature” of
contemporary political science, Clarke and Primo (2012, 1) depict it too as a “model discipline.”
Like Rodrik, they are concerned to discern how models put the science in political science. This, of
course, is an instance of what is a much broader, perennial question for philosophers of science
(Contessa, 2011; Frigg and Hartmann, 2020). And, importantly in my view, it prompts us to ask
how we use models to facilitate the dissemination of information, knowledge, and insight derived
from specialized inquiry in democratic politics (Anderson, 2007, 2011; Kitcher, 2006). But having
noted that substantive concern in what is among the central subfields of contemporary philosophy, I
also note that in political philosophy, seminal thinkers such as John Rawls and Michel Foucault use
models at crucial junctures in their work (Johnson, 2014).
I begin with two caveats. First, the intellectual terrain we face is vast. I approach it from my own
disciplinary home. Moreover, I focus my discussion on the most common families of formal model –
game theory and social choice theory – used in political science. Austen-Smith and Banks (1998,
1999, 2005), McCarty and Meirowitz (2007), Gelbach (2013), and Mershon and Shvetsova (2019)
all provide useful, detailed treatments of this work. I spend nearly no time talking about standard
models in microeconomics although I believe my arguments extend to such work as well. Likewise,
I deal only glancingly with “statistical,” “causal,” or “empirical” models of various sorts or with the
diverse relationships between these and models of the formal sort. In what follows, “models” is
shorthand for formal models.
Second, the argument I offer is not neutral. As they say, I “have a dog in this fight” and I do not
try to hide my own view. This is so in two ways. I have a particular view of how we use models in
political-economic inquiry ( Johnson, 2014, 2019, 2021) and much of what I argue aims to articulate
DOI: 10.4324/9780367808983-3 17
James Johnson
that view. I do, nevertheless, refer along the way to those who articulate competing views on
various matters large and small. I also have a particular view of the importance of democratic politics
(Knight and Johnson, 2011), and what I write here takes that view as background.
2 What Is a Model?
This question gets us off on the wrong foot. Consider Page’s (2018) masterful tour of models and the
ways we use them in social, political, and economic inquiry. Among the considerable virtues of his
study, Page forcefully conveys a basic point – the intellectual terrain here is exceedingly various. Put
otherwise, “there are lots of kinds of models serving lots of different purposes” (Cartwright, 1999,
180). It, therefore, is perfectly plausible to doubt whether there exists a single set of definitional
criteria that any “model” must meet. Second, asking what a model is distracts us from arguably more
germane questions: What are models for? What do we use models to do?
As I noted above, my arguments are animated primarily by thinking about formal models used in
political inquiry. We can classify two families of model by reference to their “methodological ap-
proaches” ( Austen-Smith and Banks, 1998, 263).1 The first consists of models of “direct preference
aggregation,” often relying on “methods” of social choice theory. Here one can think most ob-
viously of research initiated by Arrow’s impossibility theorem and the Gibbard-Satterthwaite the-
orems showing that any aggregation mechanism is susceptible to strategic manipulation ( Johnson,
2015b). List (2013) and Myerson (2013) offer concise surveys of developments in this domain of
inquiry. The second analyzes the vicissitudes of strategic interaction and relies on “the methodology
of game theory” ( Austen-Smith and Banks, 1998, 260–261). Anyone with even a rudimentary
exposure to contemporary social science will have encountered simple games with prosaic names
like Prisoners Dilemma, Chicken, and Battle of the Sexes. But as our games incorporate more
complex environments – made so, because players interact repeatedly over time or because they
confront informational asymmetries – so too do the models we construct to capture their interac-
tions ( Johnson, 2015a). Kreps (1990), Rubinstein (1991), and Gibbons (1997) each address fun-
damental issues in this domain of inquiry in succinct, accessible ways.
The boundary between these two families of models is a porous one (Austen-Smith and Banks,
1998). Consequently, while political scientists commonly speak of game theory or social choice theory
it seems more accurate to dispense with the honorific and see these undertakings as consisting simply
of loose “families of models” (Schelling 1978; Austen-Smith and Banks, 1998). To help get our
bearings, let’s turn to Page once more. Consider his opening definition:
Models are formal structures represented in mathematics and diagrams that help us to
understand the world. Mastery of models improves your ability to reason, explain, design
communicate, act, predict, and explore. (Page, 2018, 1)
We use models, in other words, as we undertake various tasks. And we assess any given model by
how useful it proves in accomplishing the task we have set ourselves. Conventionally such assess-
ment is seen as a more or less directly empirical matter. Realists claim to assess models by asking if
their assumptions are empirically plausible. Instrumentalists claim to assess models in terms of their
success in generating empirical predictions. Thus, both prejudge the matter of how we should assess
formal models. Both presume that models and hence our assessments of them are in the first place
empirical (MacDonald, 2003). That, I think, is a mistake.
What, then, is the primary point of using formal models in political-economic inquiry? Here I
follow Schelling (1978, 90): “A model is a tool. … A shared model is a help in communicating,
especially if the model has a name.” This view resembles conventional instrumentalism but does not
link performance to empirical prediction. I will argue instead that we use formal models not for
18
On Models and Their Uses
directly empirical but conceptual purposes. As a start toward defending that view, we must cast our
net more widely than Page allows. Specifically, it is a mistake to insist as he does that “Models use
mathematics not words” (Page, 2018, 6). Typically, a model will use both. There is, indeed, no
sharp distinction to be drawn here. I adopt the view that couching our inquiries in mathematical
terms is continuous with rather than sharply distinct from doing so in everyday language. It consists
in the “progressive sharpening and regimenting of ordinary idioms” (Quine, 1986, 150). So, when
we abandon one or the other idiom, it typically will be the mathematics rather than the words that
we discard. Think of how Rawls (1971) uses the Original Position or of how Foucault (1979) uses
The Panopticon. Both are models in the straightforward sense we will discuss here even though
neither takes a mathematical form (Johnson, 2014). If we start with such verbal or discursive models,
it is easier to grasp the ways we use models that in fact are couched in mathematical language.
Political theorists differentiate general and abstract concepts from more particular and concrete
conceptions (Rawls, 1971, 5, 9–10; Dworkin, 1986, 70–72). I press that distinction into service here.
Rawls is concerned with the abstract concept of justice. He relies on the original position as a model
to help specify how the artificial agents who populate it might come to endorse the particular
conception of justice as fairness he himself articulates to govern the basic structure of their society.
Likewise, Foucault is concerned with the abstract concept of power. He relies on the Panopticon as
a model to help specify how the mechanisms that constitute his particular conception of disciplinary
power function to generate docile bodies across multiple institutional settings.
Rawls and Foucault, then, use their respective models for conceptual purposes.2 They aim to
show what an abstract concept – justice and power, respectively – means in a more concrete setting.
For Rawls that is society conceived as an ongoing scheme of cooperation. For Foucault, it is a
disciplinary or carceral regime. I think that we use the more mathematical families of models that
political scientists deploy primarily for conceptual purposes too (Johnson, 2021). We use them to
establish what abstract concepts such as “rationality,” “institution,” or “deterrence” mean in some
more concrete or particular context – namely the one captured in the model. The latter amount to
particular conceptions of the more abstract concepts.
In other instances, models are conspicuous by their absence. This is the case with the concept of
“spontaneous order” central to Hayek’s (1960, 1973) version of what is called classical liberalism.
Any such order emerges, Hayek claims, from a non-Darwinian process of cultural evolution ani-
mated by mechanisms of group selection. Yet Hayek provides no persuasive model of how that
evolutionary process operates. Indeed, he offers nothing we might call a model at all. That, in turn,
imparts a hand-waving quality to the analytical apparatus that provides the scaffolding for his en-
twined defense of classical liberalism and criticism of “socialism.” Even sympathetic readers are left
unsure what Hayek means when he talks about spontaneous order ( Johnson, 2020).
Models, as Rawls and Foucault use and Hayek neglects them, are understood best as interpretive
tools rather than in representational terms. The same holds for mathematical families of models
familiar to political scientists and economists (Cartwright, 1999; Johnson, 2021). This claim will
perhaps seem contentious to many who use formal models in political-economic inquiry. It runs
counter to what I call “the standard rationale” for constructing and solving formal models. On this
view, we rely on formal models for more or less directly empirical purposes. Hence, we (i) use
formal models to generate predictions that we (ii) treat as empirical hypotheses, and that we (iii)
proceed to test against (preferably quantitative) evidence derived from the “real world.” Many of
those who use formal models embrace the standard rationale only tacitly or partially. For others it is
definitional. Hence, Dowding (2016, 88) insists “a formal model is a set of statements related for-
mally or analytically to generate testable hypotheses or predictions.” He is not alone in doing so
(e.g., Morton, 1999).3
The standard rationale suffers from numerous problems – logical, historical, and practical.
Logically, there is the problem of how one might derive “true” conclusions from what are readily
19
James Johnson
acknowledged to be false assumptions or premises (Clarke and Primo, 2012). A related, historical
problem emerges from the recognition that we cannot in any case conclusively confirm or verify the
conclusions of a model. And if we opt instead for a weaker criterion – falsification – ironies emerge.
Popper, who proposed falsification as a criterion of empirical performance in the first place, ex-
plicitly insists that formal economic models are not falsifiable (Popper, 1967). Advocates of the
standard rationale minimally owe some account for why Popper is mistaken or why his strictures do
not apply to themselves. Finally, there is the practical reality that it is easy to find canonical formal
models in political science and economics that make no predictions whatsoever (Johnson, 2021).
This is the case, for instance, with Schelling’s checkerboard model of residential segregation. It is
true as well of Akerlof’s market for lemons, of both the Nash and Rubinstein bargaining models, of
Arrow’s impossibility theorem, and of the McKelvey-Schofield models of the general indeterminacy
of majority rule. This list is far from exhaustive.
I believe these difficulties are fatal to the standard rationale. In the remainder of this chapter, I
recommend an alternative interpretation of why we create models and what we use them for.
3 Fiction, Fables, Models

If we find the standard rationale wanting, what is the alternative? I already have suggested that we
use formal models for conceptual rather than directly empirical purposes. This hardly is an original
suggestion. Rubinstein (1991), Hausman (1992), and Elster (2007), for instance, all argue that we do
just that. It remains to spell out what this might mean.
Discussion of models often proceeds metaphorically. We regularly are asked to think of formal
models as something else, for instance, as caricatures (Gibbard and Varian, 1978), analogies (Gilboa,
2014), novelettes (Leamer, 2012), or maps (Clarke and Primo, 2012). I will proceed in the same way
here. Some prominent economists and philosophers suggest that constructing and examining a
formal model is like creating and interpreting a fable (Rubinstein, 2012; Myerson, 2007, 2013;
Cartwright, 1999, 2010). I think this is just right ( Johnson, 2019, 2021). But if such metaphors are
to be useful, they cannot remain merely ornamental; we must pursue them and establish the sug-
gested resemblance in some detail. Pursuing the models-as-fables metaphor allows us to make sense
of a number of claims that emerge in recent discussions of how and why we use models in the
political economy. That includes the claim that we use models for conceptual purposes.
First things, first. Treating formal models as fables presupposes that we categorize them as fiction.
This is common enough (Frigg and Hartmann, 2020, sec 2.2) even if there are those who vigorously
contest this practice on philosophical and pragmatic grounds (e.g., Giere, 2009).
So, let’s start with this more encompassing metaphor. Novelist Walter Mosley (2007, 2019) who
perhaps is best known for his crime fiction provides a useful point of departure. As part of the advice
he offers to aspiring writers, Mosley makes a telling remark.
There’s one last thing to say before we get into the main body of this disquisition, and that
is – condensation. Even though I have not used this word in the main body of text, it is a
major unspoken element of fiction writing. That’s how you write a novel; you take a small
section of the larger world … and then crush the subject down to only those elements that
are salient to the story being told. Once you’ve achieved this you add as little of the
commonplace as possible to make a story that seems large and real and pedestrian, and
hopefully, revealing.
(Mosley, 2019, vii–viii)
Thus, on Mosley’s account when we craft fiction, we discard, omit or leave things out. We con-
dense.4 The same holds, I believe, when we create formal models. One might think of the way
20
Schelling (1960) describes how, in making models, we eliminate or jettison “incidental detail.”
Mosley characterizes this process in a helpful way. He prompts us to abandon the common image of
making a model as a process of abstraction or idealization and embrace instead an image of making
things more compact, concentrated, or concrete.
For writers of fiction, the “job is to create a concrete world populated by unique, well-defined,
and yet shifting characters” (Mosley, 2019, 40). So too with models. Thus, in the original position,
the “world” is defined to a considerable degree by the constraints of the veil of ignorance. And the
“characters” are the “artificial creatures” (Rawls, 2001, 83), the free, equal, moral agents who
populate the original position where, from behind the veil, they choose principles of justice to
govern the basic structure of society. Likewise, in game-theoretic models, the “world” consists in
the various specified parameters, strategies, payoffs, and equilibrium solutions. The “characters” are
the players, themselves modeled as utility functions, who inhabit that world. Whereas Rawls’s ar-
tificial creatures are not just rational but reasonable, players in game-theoretic models are essentially
sociopaths. Conventionally understood, these players are not just incessantly calculating and in-
sistently self-interested but care for the wellbeing of others or for the truth solely insofar as those
particulars bear on their own payoffs.
Crucially, in these models both the world and the characters who inhabit it are imaginary.5 As
will become clear shortly, this shift in focus is especially helpful. It allows us to appreciate that in
making and using formal models we are not, as is commonly suggested, abstracting from reality. We
are moving in nearly the opposite direction from an abstract concept to a more concrete conception.
Crucially, as Mosley describes it, the sort of condensation involved in creating fiction aims to be
“revealing.” In this regard, fiction aims to “show” rather than to “tell” (Mosley, 2007, 30–40). It is,
on his account, “more experiential than informational.” While writers teach us things, when readers
engage with stories and tales they learn not by being told this thing or that but “by what they are
shown about the lives and circumstances of the characters therein” (Mosley, 2007, 40). For our
purposes, this observation is important as it allows us to start to narrow our focus from fiction
generally to fables as a particular fictional device and to see some of the latter’s characteristic features.
A fable is itself condensed in comparison to other literary forms. It is a narrative device stripped of
literary niceties (e.g., character development) and driven primarily by plot (Blackham, 1985, 195). In
fables, as in fiction more generally, the plot provides “the structure of revelation” by which a story
emerges and discloses whatever truths or possibilities it holds (Mosley, 2007, 56–62). This is fine for
starters. Two things, however, bear emphasis. First, like models, fables are remarkably various:
“what makes a fable is not the form of the material, but its orientation and the use made of it”
(Blackham, 1985, 136, 202–208). So, as is the case with models, we are pressed to ask not what a
fable is but rather to ask how we use it.
Second, the emphasis here on revelation is fundamental. Both fables and models show us things. They
aid us in discerning features of our concepts that otherwise remain obscure. Thus, Berlin (1969 [1958])
articulates a defense of liberty grounded in the tragic nature of value pluralism. He tells us that not all
good things go together without remainder. Despite its importance to his larger undertaking, this basic
point is simply asserted, and not entirely clearly. By contrast, in his formal impossibility proof Arrow
shows forcefully and precisely how uncompromisingly this tragic feature haunts our normative con-
cepts. The form his axiomatic proof takes – first showing that if some coalition is decisive over some
pair of alternatives it will be decisive over all pairs and then that there is always a decisive coalition –
illustrates nicely how the idea that models provide a structure of revelation. His proof reveals how any
aggregation procedure violates the axiom of non-dictatorship. And this, given that Arrow sets out to
explore the character of “capitalist democracy” relative to authoritarian and theocratic modes of politics,
is what leads readers to find his result so disturbing ( Johnson, 2015b).
Cartwright recommends that we think of interpretive models as fables. We use them to show what
some abstract concept or other “amounts to” or “consists in” in a more particular setting, namely that
21
James Johnson
captured in the parameters of the model (Cartwright, 1999, 189, 41). The model allows us to say, in
precise terms, “here in this setting, and for these purposes, this is what ‘X’ consists in” and to consider
the implications of doing so. Here “X” could be justice, power, institution, rationality, deterrence, and
so forth. On Cartwright’s account the need to somehow “fit out” abstract concepts in more concrete
terms “marks an entirely commonplace feature of language” (Cartwright, 1999, 40). She insists that our
theories – whether scientific or philosophical – inescapably rely on “abstract concepts … whose re-
lation to the world must be mediated by more concrete concepts.”6 Adopting the convention among
political theorists mentioned earlier, I refer to the latter as conceptions. These are “very specific in their
form: the forms are given by the interpretive models of the theory.” Exploring conceptions in this way
renders our theories more precise by circumscribing the range of application of the abstract concepts
they invoke. The abstract concepts we use in our theories have purchase as far as, and no further than,
our interpretive models extend (Cartwright, 1999, 3).
Here we see the value of thinking of formal models, like fiction, as a process of condensation.
Formal models allow us to render abstract concepts more concrete. The process I just described
clarifies what Rodrik (2015, 46, 63–67, 71) must have in mind when he suggests that economics
advances by proliferating models. They do not typically test models empirically and reject those that
prove faulty with a more complex version of the same model. Instead, they move “horizontally” and
construct a new relatively simple model, adjacent to the last, aimed at “explaining aspects of social
outcomes that were unaddressed earlier.” And here explanation involves the conceptual task of
identifying and exploring mechanisms (Rodrik, 2015, 12, 17).7 The aim is to grasp and explore
possibilities.
4 The Moral of the Story

We have been tacking back and forth between models and fables. Both are narrative devices. They
have in common being useful for deriving truths, lessons, principles, or, in short, morals. It is
important to see how this works.
Typically, we rely on fables as “a narrative device, to provoke and aid concrete thinking, focused
on some general matter of concern” (Blackham, 1985, xi, xvii). How do fables accomplish this?
With fables we aim to disclose truths that are easily overlooked or poorly acknowledged. We do so
not by advancing arguments or providing evidence but by picturing, showing, or demonstrating
how the general or abstract works itself out in some particular circumstance (Blackham, 1985, 175).
“Thus, a fable is a story that represents a particular case that is immediately recognizable as a general
principle” (Blackham, 1985, 102). The principle is broader and more enduring than the specific tale
the fable tells. It embodies whatever truths the fable reveals – the lesson, message, or moral it
imparts. Yet, because the moral or lesson a fable conveys retains a connection to the general it is not
merely an illustration. We use fables to provoke thought. Rather than instruct us what to think in
some didactic way a fable reveals something. It shows us something, reminding us thereby what we
need to think about (Blackham, 1985, xi, 223).
An example no doubt will help us see how we use formal models in just the same way as we do
fables. A lecture on game-theoretic models of deterrence Roger Myerson (2007) delivered to active
duty military officers at the US Army War College, is especially apt. Myerson transports formal
modeling out of the academy to an audience who operate “in the wild.” He demonstrates how
game-theoretic models work as tools of practical reasoning.
Myerson guides his audience through a series of simple, familiar game-theoretic models. He
explicitly depicts these models as stories we can use to make sense of problems that we encounter in
the “dangerous” world. Specifically, he insists these models offer broad “lessons” about the need to
balance force and restraint in conflict situations.8 He is clear that the key features of his models –
players, information, payoffs, strategies, equilibrium – provide an impossibly thin description (the
22
“plot”) that captures everything that can possibly happen in his story and that those constraints
provide whatever structure the story has.9 He acknowledges that thinness but remains unperturbed
by it precisely because the world in his models is an imaginary one. His aim is not to describe the
“real” world. Instead, he uses the models to define resolve and restraint, to show us what they mean
in particular circumstances. In the process, he is defining deterrence and drawing lessons from the
models. The lessons he draws are not just about the importance of restraint but specifically its relative
importance for powerful countries given the temptation to demonstrate resolve. Specifically, his
lesson is a warning about how misjudging the ratio of resolve and restraint can lead to unintended
consequences, namely exacerbating the very problems we hope to address. And, importantly, he
acknowledges that other models, capturing other imaginary worlds, might suggest other lessons. His
conclusion? “On such important policy questions, we need a debate in which different models and
views are compared” (Myerson, 2007, 19). He does not claim certainty but calls for ongoing inquiry
and argument.
Defenders of the standard rationale might well complain that the position Myerson articulates
here is an outlier, just a quirky example exploited solely to make my point. It is easy enough to
deflate such suspicions. Myerson is hardly so idiosyncratic as skeptics might imagine. Humphreys
(2017) offers a compendium of close to fifty familiar “political games.” And while it might appear
that what he offers is simply a textbook-like digest, that appearance is deceptive (Johnson, 2019). For
each of the four dozen models he analyses, Humphreys demonstrates in short compass how to derive
lessons (he calls them “principles”) of just the sort Myerson identifies in his lecture on deterrence. In
other words, Humphreys makes clear that formal models work like fables in a generalized fashion.
And the lessons, principles, or morals we derive from any model work in entirely familiar ways too.
In each case, they disclose truths and prompt reflection. Humphreys himself makes this point in a
refreshingly frank way. He encourages readers to explore each model and the refined lesson or
principle he derives from it in a refreshing way. “Use them,” he implores, “as a starting point for
conversation, not as problem sets” (Humphreys, 2017, xxvi). That is some sound advice.
5 Using Models in Public Places

We have arrived at a view distant from the standard rationale, but proximate to the position
Schelling articulates. It is a view eminently more suitable for the enterprise of PPE. Myerson uses a
series of models to communicate with military officers. Some might complain that his is a tech-
nocratic undertaking, a Nobel Prize winner interacting with unelected military personnel and,
moreover, that that is inherent in any use of models. I disagree (Orr and Johnson, 2018).
Myerson illustrates the crucial task of disseminating the results of systematic political-economic
inquiry – of science – to popular audiences of many sorts situated at various locations in a democratic
polity. This is a fraught undertaking to which Dewey (1927) called our attention early in the last
century. Contemporary political theorists indebted to his arguments have revived his concern
(Anderson, 2007, 2011; Kitcher, 2006). A serious discussion of this matter would take us far afield. I
want simply to conclude with an example that might give pause to those who deem Myerson’s
undertaking suspect.
Let’s start with the model central to this example, the so-called “chaos” theorem established by
McKelvey (1976, 1979). This model establishes that in a multi-dimensional spatial model, when the
core is empty (no Condorcet winner), when voters have Euclidean preferences, and when they vote
sincerely, preference aggregation by simple majority rule will not generate an equilibrium so long as
voters retain agenda-setting power. Under those conditions, from any status quo, there exists a finite
path of majority decisions leading to any other alternative in the policy space. In other words, “when
social preference breaks down, in the sense of not admitting a core, it breaks down completely in
that one can get from any one alternative in the space to any other via the social preference relation”
23
James Johnson
(Austen-Smith and Banks, 1999, 178). This is what it means to talk of the generalized instability of
majority rule.10 The model highlights the importance of agenda-setting in political decision-making.
In the fall of 2011 McKelvey’s model turned up in an unexpected place. At the time, politics in
the United States were convulsed by the Occupy movement that originated in New York and
spread coast to coast. Authorities responded to the Occupy protests with a determined, coordinated
campaign of violent repression (Wolf, 2012; Knuckey et al., 2012). In Oakland, police violence
elicited renewed protest. In that context, an ad hoc group, “Mathematicians Against Police
Violence,” held a teach-in at UC Berkeley. Among the speakers was a young post-doctoral fellow,
Felix Breuer, who spoke to those assembled on “Geometry, the Majority Vote and the Power of
Agenda Control.”11 The focus of this talk – given outdoors on campus, using whiteboards propped
up on benches – was McKelvey’s model, sketched above, demonstrating the generalized instability
of majority rule in an institutionally sparse context. Here is how Breuer describes his presentation:
The title of my own talk was “Geometry, the Majority Vote, and the Power of Agenda
Control” and its topic was McKelvey’s theorem. McKelvey’s theorem is a surprising result
with a beautifully intuitive proof that can be made accessible to everyone. Its morale [sic] in
this context is this: Protest, i.e., the struggle for the control of the political agenda, is a much more
important element of democracy than voting! Ergo: the police should not beat up protesters!
(Breuer, 2011)
Indeed, what McKelvey shows is that the process of setting the agenda is crucial in democratic
politics. And that – making extreme political-economic inequality captured by the slogan “We are
the 99%!” central to the political agenda in American politics – is what the Occupy protesters were
all about. What Breuer demonstrates is how we can use formal models and the lessons they impart to
talk reflexively about democratic politics as we partake of it. His undertaking is in that sense of a
piece not just with Myerson’s lecture but with the wide array of models Humphreys examines. All
of these exemplify what, while largely unexpected, is perhaps the most important use of models,
helping us communicate in what often are fraught political circumstances.
Notes
1 For collections of “classic” contributions to each of the two families see Kuhn (1997) and McLean and
Urken (1995), respectively.
2 Gaus (2000, Part I) offers a brief, accessible overview of some of the broader philosophical background and
political implications of the ways political theorists engage in conceptual analysis and disagreement.
3 I elsewhere establish that the standard rationale is indeed standard (Johnson, 2021). It is useful to note that
not all political scientists or economists embrace the standard rationale. Hence Rodrik insists: “This is a nice
story but it bears little relationship to what economists do in practice and how the field really makes
progress” (Rodrik, 2015, 64).
4 Some might take offense at my comparing modeling in political-economic and philosophical inquiry to so
putatively “unserious” a genre as crime fiction. Consider instead what may hopefully be seen as an ap-
propriately “weighty” source. Kundera (1988, 71–73) insists that in order to encompass “the complexity of
existing in the modern world” fiction “demands a technique of ellipsis, of condensation.” He embraces this
imperative: “I see the art of ellipsis as crucial. It insists that we go directly to the heart of things.” Hence, in
his own writing, he aims to dispense as far as possible with literary conventions and the “verbalism” they
engender. In his fiction, he aims instead “to make it dense.” And, of course, as Kundera (1988, 95) himself
insists: “entertainment doesn’t preclude seriousness.”
5 Once again, Kundera (1988, 34) is helpful. He notes: “A character is not a simulation of a living being. It is
an imaginary being. An experimental self.” For both Rawls and game theorists the characters in their models
are distillations. They bear no more resemblance to actual, complex human beings than do the characters in
Orwell’s Animal Farm.
24
6 For Cartwright (1999, 179–180) “models mediate between theory and the world.” This view is subject to
considerable argument and contestation (e.g., Morgan and Morrison, 1999). Yet Cartwright insists that it
illuminates how models play an essential role in how our theories allow us to make substantive claims about
and, more importantly from her perspective, intervene in the world.
7 Conceptual in the sense of being not directly empirical – examining basic causal mechanisms such as rationality
(beliefs and preferences), institutions, etc., that are not themselves observable in any straightforward sense.
8 Myerson’s lecture is part of an ongoing lineage. He acknowledges the formative intellectual influence of
Schelling (1960) on the development of game theory generally. Indeed, he insists Schelling set the in-
tellectual agenda to which more technically oriented thinkers (e.g., Harsanyi, Selten, etc.) responded
(Myerson, 2009). He also explicitly locates his specific story in that intellectual lineage (Myerson, 2007, vi).
Slantchev (2017, 2020) takes up the broad approach Myerson sketches and draws out its implications for
ongoing inquiry into the international political and economic conflict.
9 “A story is a simple thing. It is a narrative that has a beginning, middle, and end” (Mosley, 2007, 48). The
game-theoretic structure helps us keep Myerson’s story simple, hence clear and useful.
10 McKelvey’s result is often referred to as a “chaos theorem.” It generalizes Arrow’s impossibility theorem in
the context of a spatial voting model. It is commonly recognized that like Arrow’s theorem, it makes no
predictions (Austen-Smith and Banks, 1998, 270–271, 1999, 84; McCarty and Meirowitz, 2007, 79–80).
11 For links to various materials on this event see: Berkeley Occupier (2011).
References
Anderson, Elizabeth. (2007). “The Epistemology of Democracy.” Episteme 3: 8–22.
Anderson, Elizabeth. (2011). “Democracy, Public Policy, and Lay Assessments of Scientific Testimony.” Episteme 8:
144–64.
Austen-Smith, David and Jeffrey Banks. (1998). “Social Choice Theory, Game Theory, and Positive Political
Theory.” Annual Review of Political Science 1: 259–87.
Austen-Smith, David and Jeffrey Banks. (1999). Positive Political Theory I. Ann Arbor: University of Michigan Press.
Austen-Smith, David and Jeffrey Banks. (2005). Positive Political Theory II. Ann Arbor: University of Michigan Press.
Berkeley Occupier. (2011). “Occupy UC: Math Against Police Violence.” San Francisco Bay Area
Independent Media Center (November 18). https://www.indybay.org/newsitems/2011/11/18/18700055.
php (Accessed 1 March 2021).
Berlin, Isaiah. (1969/[1958]). “Two Concepts of Liberty.” In Four Essays on Liberty. Oxford: Oxford University Press.
Blackham, H. L. (1985). The Fable as Literature. London: Athelone.
Breuer, Felix. (2011). “Teach-In: Mathematicians Against Police Violence.” Felix Breuer’s Blog (November
17, 2011). http://blog.felixbreuer.net/2011/11/17/protest.html (Accessed 3 August 2021).
Cartwright, Nancy. (1999). The Dappled World: A Study of the Boundaries of Science. Cambridge: Cambridge
University Press.
Cartwright, Nancy. (2010). “Models: Parables v. Fables.” In Beyond Mimesis and Convention, edited by Roman
Frigg and Mathew Hunter (Boston Studies in the Philosophy of Science, Volume 262). New York: Springer.
Clarke, Kevin and David Primo. (2012). A Model Discipline: Political Science and the Logic of Representations.
Oxford: Oxford University Press.
Contessa, Gabrielle. (2011). “Scientific Models and Representation.” In Continuum Companion to the Philosophy
of Science, edited by Steven French and Juha Saatsi. London/New York City: Continuum International
Publishing.
Dewey, John. (1927). The Public & Its Problems. Athens, Ohio: Swallow Press.
Dowding, Keith. (2016). The Philosophy and Methods of Political Science. London: Palgrave Macmillan.
Dworkin, Ronald. (1986). Law’s Empire. Cambridge, MA: Harvard University Press.
Elster, Jon. (2007). Explaining Social Behavior. Cambridge: Cambridge University Press.
Foucault, Michel. (1979). Discipline & Punish: The Birth of the Prison. New York: Pantheon.
Frigg, Roman and Stephan Hartmann. (2020). “Models in Science.” Stanford Encyclopedia of Philosophy. https://
plato.stanford.edu/archives/spr2020/entries/models-science/.
Gaus, Gerald. (2000). Political Concepts and Political Theory. Boulder, CO: Westview Press.
Gelbach, Scott. (2013). Formal Models of Domestic Politics. Cambridge: Cambridge University Press.
Gibbard, Allan and Hal Varian. (1978). “Economic Models.” Journal of Philosophy 75: 664–77.
Gibbons, Robert. (1997). “An Introduction to Applicable Game Theory.” Journal of Economic Perspectives 11:
127–49.
25
James Johnson
Giere, Ronald. (2009). “Why Scientific Models Should Not Be Regarded as Works of Fiction.” In Fictions in
Science: Philosophical Essays on Modeling and Idealization, edited by Mauricio Suarez. London: Routledge.
Gilboa, Itzhak. (2014). “Economic Models as Analogies.” The Economic Journal 124: F513–33.
Hausman, Daniel. (1992). Essays on Philosophy and Economic Methodology. Cambridge: Cambridge University Press.
Hayek, F. A. (1960). The Constitution of Liberty. Chicago: University of Chicago Press.
Hayek, F. A. (1973). Rules and Order. Vol. 1 of Law, Legislation and Liberty. Chicago: University of Chicago Press.
Humphreys, Macartan. (2017). Political Games: Mathematical Insights on Fighting, Voting, Lying, and Other Affairs of
State. New York: W.W. Norton.
Johnson, James. (2014). “Models Among the Political Theorists.” American Journal of Political Science 58: 547–60.
Johnson, James. (2015a). “Game Theory.” In The Encyclopedia of Political Thought, edited by Michael T.
Gibbons. Oxford: Blackwell.
Johnson, James. (2015b). “Arrow’s Impossibility Theorem.” In The Encyclopedia of Political Thought, edited by
Michael T. Gibbons. Oxford: Blackwell.
Johnson, James. (2019). “Review Essay – Formal Models in Political Science: Conceptual, Not Empirical.”
Journal of Politics 81: e6–10.
Johnson, James. (2020). “Is the Concept of ‘Spontaneous Order’ Simply Libertarian Fairy Dust?” In Politics,
Philosophy & Austrian Economics edited by Daniel D’Amico and Adam Martin. Bradford, UK: Emerald
Publishing.
Johnson, James. (2021). “Models-As-Fables: An Alternative to the Standard Rationale for Using Formal Models
in Political Science.” Perspectives on Politics 19: 874–89.
Kitcher, Phillip. (2006). “Public Knowledge and the Difficulties of Democracy.” Social Research 73: 1205–24.
Knight, Jack and James Johnson. (2011). The Priority of Democracy: Political Consequences of Pragmatism. Princeton:
Princeton University Press.
Knuckey, Sarah, Katherine Glenn, and Emi MacLean. (2012). Suppressing Protest: Human Rights Violations in the
U.S. Response to Occupy Wall Street. NYU School of Law and Fordham Law School.
Kreps, David. (1990). Game Theory & Economic Modelling. Oxford: Oxford University Press.
Kuhn, Harold, ed. (1997). Classics of Game Theory. Princeton: Princeton University Press.
Kundera, Milan. (1988). The Art of the Novel. New York: Harper.
Leamer, Edward. (2012). The Craft of Economics. Cambridge: MIT Press.
List, Christian. (2013). “Social Choice Theory.” Stanford Encyclopedia of Philosophy. (Winter 2013 Edition),
edited by Edward N. Zalta. https://plato.stanford.edu/entries/social-choice/.
MacDonald, Paul. (2003). “Useful Fiction or Miracle Maker: The Competing Epistemological Assumptions of
Rational Choice Theory.” American Political Science Review 97: 551–65.
McCarty, Nolan and Adam Meirowitz. (2007). Political Game Theory: An Introduction. Cambridge: Cambridge
University Press.
McKelvey, Richard. (1976). “Intransitivities in Multidimensional Voting Models and Some Implications for
Agenda Control.” Journal of Economic Theory 12: 472–82.
McKelvey, Richard. (1979). “General Conditions in Formal Voting Models.” Econometrica 47: 1085–112.
McLean, Iain and Adam Urken, eds. (1995). Classics of Social Choice. Ann Arbor: University of Michigan Press.
Mershon, Carol and Olga Shvetsova. (2019). Formal Modeling in Social Science. Ann Arbor: University of
Michigan Press.
Morgan, Mary S., and Margaret Morrison. (1999). Models as Mediators. Cambridge: Cambridge University Press.
Morton, Rebecca. (1999). Methods & Models. A Guide to the Empirical Analysis of Formal Models in Political Science.
Cambridge: Cambridge University Press.
Mosley, Walter. (2007). This Year You Write Your Novel. New York: Little Brown & Company.
Mosley, Walter. (2019). The Art of Fiction. New York: Grove Atlantic.
Myerson, Roger. (2007). Force and Restraint in Strategic Deterrence: A Game-Theorist’s Perspective. U.S. Army War
College, Strategic Studies Institute.
Myerson, Roger. (2009). “Learning from Schelling’s Strategy of Conflict.” Journal of Economic Literature 47:
1109–25.
Myerson, Roger. (2013). “Fundamentals of Social Choice Theory.” Quarterly Journal of Political Science 8:
305–37.
Orr, Susan and James Johnson. (2018). “The Entanglement of Methods of Policy Inquiry with Democratic
Possibilities.” In Routledge Handbook of Ethics and Public Policy. New York: Routledge.
Page, Scott. (2018). The Model Thinker. New York: Basic Books.
Popper, Karl. (1967/1985). “The Rationality Principle.” In Popper Selections, edited by David Miller. Princeton:
26
Quine, W. V. O. (1986). “Success and Limits of Mathematization.” In Theories and Things. Cambridge, MA:
Harvard University Press.
Rawls, John. (1971). A Theory of Justice. Cambridge: Harvard University Press.
Rawls, John. (2001). Justice as Fairness: A Restatement. Cambridge: Harvard University Press.
Rodrik, Dani. (2015). Economics Rules: The Rights and Wrongs of the Dismal Science. New York: W.W. Norton.
Rubinstein, Ariel. (1991). “Comments on the Interpretation of Game Theory.” Econometrica 59: 909–24.
Rubinstein, Ariel. (2012). Economic Fables. Cambridge: Open Book.
Schelling, Thomas. (1960). The Strategy of Conflict. Cambridge: Harvard University Press.
Schelling, Thomas. (1978). Micromotives and Macrobehavior. New York: W.W. Norton.
Slantchev, Branislav. (2017). “On the Proper Use of Game-Theoretic Models in Conflict Studies.” Peace
Economics, Peace Science and Public Policy 23: 1–14.
Slantchev, Branislav. (2020). “The Simple, the Trivial and the Insightful: Field Dispatches from a Formal
Theorist.” In The SAGE Handbook of Research Methods in Political Science and International Relations, edited by
Luigi Curini and Robert Franzese. London: Sage Publications.
Solow, Robert. (1997). “How Did Economics Get That Way and What Way Did It Get?” Daedalus 126:
39–58.
Wolf, Naomi. (2012). “Revealed: How the FBI Coordinated the Crackdown on Occupy.” The Guardian
(December 29).
27
3
COMPLEXITY
Fred D’Agostino
[W]e nowadays find … a flourishing project of accepting the world’s complexity and devising the
cognitive instrumentalities needed to come to grips with it.
Rescher (1998: 207)
1 Introduction
Complexity arises in situations where numerous diverse individual agents change their behaviors and
their characteristics, including their reactions to the behaviors of others, in response to the outcomes
that they jointly create through their interactions. Sometimes, (mirco-level) individual behaviors
(e.g., buying and selling behavior in a market) give rise to a relatively stable macro-level pattern or
order, whose characteristics (e.g., the adjustment of supply and demand for objects of exchange) are
intrinsic to that level of description. These characteristics emerge from the underlying behaviors
without being directly reducible to them. No one buying or selling a particular commodity intends
that supply and demand across a great range of commodities be adjusted in the particular way that
the market secures.
Complexity studies illustrate the point that a fuller understanding of human social phenomena
may depend upon new analytical tools. It was the development, over the past thirty years, of
computer-based modeling that finally permitted complex human social phenomena to be under-
stood via simulation. Many phenomena of relevance to politics and economics could thereafter be
analyzed in terms of their complexity. To some, this heralded a new era in the social sciences.
Previously, while the abstract idea of complexity had been in play – in the work of Hayek and
Simon, for example – various tools of simplification were applied to complex phenomena to secure
their analytic tractability.
2 Three Stages
According to Lane (2016: 9), complexity was, at the time of its introduction into scholarly discourse,
“an honestly new idea in the world” that arose in specific settings and involved specific people.
There are, nevertheless, three recognizable stages in the development of complexity studies. These
are as follows: (Stage 1) introduction of the idea, (Stage 2) systematic thinking about complex
phenomena, and (Stage 3) theoretical modelling of these phenomena (cp. Kauffman, 1993: 173).
28 DOI: 10.4324/9780367808983-4
Complexity
2.1 Stage 1
In the wake of World War II, Weaver (1948) distinguished (a) organized simplicity as in the physics of
medium-sized objects, (b) disorganized complexity as with the behavior of gases (which could be and
has been dealt with using statistical tools), and (c) organized complexity, which involves the tangled
interactions of numerous diverse agents and would require other “cognitive instrumentalities” to
understand and manage. He thought (1948: 540) that addressing complexity would require a great
scientific advance, or, as we later learned to say, a paradigm shift in our approach to certain phe-
nomena. Wiener (1948) and Ashby (1952), working in the then-emerging area of cybernetics, were
other notable contributors at this stage.
2.2 Stage 2
In the 1960s, thinkers began to work through the idea of complexity and its role in economic and
other systems of interactions. This work provided some general and abstract ideas about the conditions
for and potentialities of complex systems. The General Systems Theory of Bertalanffy (1968) and the
economics of Simon (1962), Hayek (1967b), and Boulding (1966) were key texts at this stage.
2.3 Stage 3
Proper theorizing about complexity depends on the availability, from the early 1990s, of high-powered
computing systems that permit in silico simulation of complex phenomena (cp. Colander, 2014a: 4;
Lane, 2016: 9; Thurner et al., 2018: 25). The most important tool is agent-based modeling, which was
pioneered by Holland and Miller (1991) and Epstein and Axtell (1996). Such a computational approach
is necessary, according to Hooker (2011: 37), because the equations that we might otherwise use to
model complex systems typically cannot readily be “solved” for the crucial factors. As Pierpaolo
suggested (2011: 3), in this respect complexity studies add to the repertoire of science the use of
computer simulation, to supplement already well-established tools of experiment and calculation. (This
nicely illustrates a general point Galison (1997) made, namely that instrumentation can play an im-
portant role, comparable to that of theory and experiment, in scientific advances.)
The earlier work of Schelling (1969) captured some aspects of complexity dynamics with desktop
tools, showing that patterns in social behavior (e.g., residential segregation) could arise from in-
dividual behaviors not directed towards the creation of those patterns. Along with work by
Lindblom (1959), Demsetz (1967), Rittel and Webber (1973), Loasby (1976), and Weick (1976),
Schelling’s work was transitional between Stages 2 and 3.
Institutionalization of complexity studies during the late 20th and early 21st centuries is con-
spicuously illustrated by the establishment in 1984 of the Santa Fe Institute (German, 2021). Also
notable is the establishment of specialist journals: Complexity (1995); Journal of Complexity (1985);
Emergence: Complexity and Organization (1999); and Complexity, Governance & Networks (2014).
However, as Boyd and Richerson (2005: 415) remind us, Campbell (1979: 184) had already
wondered whether the kind of between-discipline work that is required for effective complexity
studies will be generated inside scholarly communities, which incentivize attention towards what
can be learned within specific disciplines and their even narrower research specializations
(D’Agostino, 2019). In this way, Campbell anticipates a point that Brennan (2010: 388) made about
PPE studies, namely that undertaking interdisciplinary work involves risks to career advancement.
While studies in complexity theory are relatively recent, as indicated by Web of Science (WoS)
publication data, they do show a “take-off,” in the literature of the social sciences and humanities,
dating from around 1990, which is the beginning of Stage 3. After that, the intensity of discussion
increases rapidly, as shown in Figure 3.1.
29
Fred D’Agostino
WoS "complexity" mentions per decade

80000
70000
60000
50000
40000
30000
20000
10000
0
1940-49 1950-59 1960-69 1970-79 1980-89 1990-99 2000-09 2010-19
Figure 3.1 Searching the Social Science and Humanities article, conference, and book indices.
3 Defining Complexity
There is considerable diversity in definitions of complexity (on which see, e.g., Horgan, 1995; Lloyd,
2001).
One reason for the proliferation of definitions is that thinking about this concept both originates
from and informs work in a number of disparate disciplines and takes on different associations
depending on its disciplinary placement.
Another reason for such proliferation is that the theoretical framework for complexity studies
remains, even now, a work in progress. This can show itself (Kauffman, 1993: 3) in the “poorly
articulated form” shown by some complexity studies where scholars are (Waldrop, 1992: 9) “trying
to grapple with questions that defy all the conventional categories.” This may result, as Cohen is
quoted as saying (Page, 2015: 23), in “a festival of metaphors,” rather than the kind of mature
science that provides off-the-shelf tools (cp. Delli Gatti et al., 2010: 114) and exhibits high degrees
of consensus about core issues, thus sustaining normal science research à la Kuhn (1970). Indeed,
Clayton (2013: 334) rejects the idea that there can be a unified science of complexity and Arthur
(2015: 90) suggested that complexity thinking shows, in its various manifestations, a family re-
semblance rather than constituting “a single, coherent vision.”
Nevertheless, some ideas about complexity are widely acknowledged across various disciplines. It
is helpful in making sense of complexity to distinguish (§4) conditions giving rise to and (§5) features of
complex phenomena (as do Page, 2012: 7; Williams, 2020: 22).
4 Conditions for Complexity

Complexity in social situations arises under certain conditions that include the following:
1. numerous and diverse agents that

2. interact dynamically, in the sense that
3. the behavior of any given agent is a function
a. according to an internal system of rules
b. which may itself change as a result of the agent’s interactions
30
Complexity
4. of the behaviors of other (“local”) agents

5. and of the representations of the environment that each agent constructs
6. with the possibilities of both negative and positive “feedback” relationships among the various
behaviors
(Cp. Arthur, 1999: 107; Cilliers, 1998: 94–5; Durlauf, 2012: 46; Epstein, 2006b: 1588; Gaus, 2006:
234–5; Hanseth, 2007: 81–2; Holt et al., 2011: 361; Page, 2012: 7; Page, 2015: 24–5; Spier, 2015: 24.)
Although these conditions may seem banal, they represent, as Ashby (1958: 10) recognized, a
departure from assumptions underpinning other theorizing about human social phenomena, as-
sumptions that, as Rescher pointed out (1998: ch. 3, §2), are, as a cognitive matter, easier to work
with and hence become defaults. For example: Because of the intuitive appeal of homeostasis as a
guiding idea, early cybernetics dealt mainly with negative feedback systems (contra condition 6).
Prior to the development of structuration thinking in sociology (cp. Giddens, 1984), relations
between structure and agency were (contra conditions 4 and 5) largely seen as unidirectional (though,
of course, in either direction, depending upon the theoretical approach), whereas, as Arthur puts it
(2013: 2), “there is a recursive loop” between the macro-level patterns arising from the behaviors of
agents and the behaviors that give rise to these patterns. Classical macroeconomics makes “nor-
malizing” assumptions (contra condition 1) about microeconomic agents (cp. Hooker, 2011: 74),
effectively masking their obvious empirical diversities and, at the same time, reducing to a “re-
presentative” singleton the numerous agents whose interactions ground the relevant phenomena.
(McKelvey (1999: 297) calls this “[t]he uniformity assumption,” which was central to Rawls’s veil of
ignorance method in A Theory of Justice (1971)). And Hayek (1967b: 29) suggested that much social
and economic modeling of collective behavior treats the individual agents (contra condition 3.b) as if
the behavior of each is independent of that of the others.
As Feyerabend already realized (1975: 295), models and methods always presuppose the validity
of certain assumptions, and their reliability therefore can’t be guaranteed when the assumptions are
not correct. Methods of analysis that are appropriate to situations of simplicity or unorganized
complexity may not work in situations of organized complexity. Hardin (1963: 162) had already
seen this in relation to planning in a complex situation; Smith (2007: 115) acknowledges that
“methods developed for [simple] systems give the wrong answer when applied to [complex] sys-
tems”; and Page notes (2008: 116) that “new methodological tools” may be needed in the analysis of
complexity (cp. Thurner et al., 2018: 20).
5 Complex Orders and Their Features

Several features of complex situations are analytically important, even if they do not constitute an
entirely satisfactory definition of complexity even when taken together (cp. §3 above). These include
(i) order; (ii) basins of attraction; (iii) path-dependency; (iv) density; (v) modularity; (vi) interfaces;
(vii) stability; (viii) novelty; (ix) increasing returns; (x) tipping-points; (xi) generative entrenchment;
(xii) fat-tailed distributions; and (xiii) the specific kind of order that Hayek called catallaxy.
i. Order. Because of their interactions with one another, agents’ behaviors in the conditions of
complexity may, but, as Hayek already saw (1973: 43–4), may not constitute a system or, in his
words (1973: 45), “secure an overall order,” which exhibits some stability in its main features.
When an order does exist, an agent can plan how to act with some confidence because the
features relevant to their action’s success will not change too abruptly before the intended out-
come of the action is secured. When behaviors give rise to a macro-level order, this is an emergent
property of these behaviors (Page, 2015: 32), not least because, typically, none of the agents
intended to produce an order with that property (cp. Axtell, 2016: 78; Epstein, 2006a: 32).
31
Fred D’Agostino
Understanding how and in what particular circumstances order emerges from complex
interactions is a crucial issue, according to Hayek (1973: 45) and McKelvey (2003: 105). Hayek
(1976: 107) thought that “the market” is an order in this sense. Typically, the market has been
modeled by economists as an equilibrium (Page, 2015: 28), where, however individual agents
might behave, their behaviors will continue to sustain an orderly overall structure of interactions.
It is therefore important to be clear that, in situations of complexity, the notion of a unique
equilibrium is no longer central (Arthur, 2013: 3, 18; Beinhocker, 2013: 340).
ii. Basins of attraction. Certainly, even without a unique equilibrium state, an overall order may be
established. And that order may be (relatively) self-sustaining in the sense there will be in-
centives, through feedback loops, that define what is called a “basin of attraction” (Kauffman,
1993: 212). This striking wording refers to situations where interactions will, mostly, move the
system around within a certain restricted range, much as what might occur in a genuine
equilibrium. Most “normal” combinations of behaviors will not move the overall system as a
whole irretrievably away from its basin. So, the system is “self-adjusting.” What differs, in the
case of a complex system, is that the basin of attraction into which the system may have settled
at any point is only one of several others into which it might have settled if the system had
developed differently (Colander, 2014b: 9; Hanseth, 2007: 81–2). There are multiple potential
equilibria, rather than the unique one assumed by classical economics. (It is important not to be
misled by the metaphor. The attractor, where it exists, is itself an emergent endogenous product
of the interactions of agents and becomes a “force” shaping agents’ subsequent behaviors only
after its establishment (cp. Felin et al., 2014: 10)).
iii. Path-dependency. This multiplicity of potential attractors is why we say, in cases of genuine
complexity, that the system configuration may be path-dependent (cp. Hooker, 2011: 33;
Williams, 2020: 25). Because there are many potential attractors, the matter of which of these is
established will depend on the path that the system traces in its historical development through
the space of configurational possibilities. (Notice that, in a proper equilibrium, the destination is
path-independent; all paths lead to, and after disequilibrating events lead back to the unique
equilibrium state.)
iv. Density. Whether agents’ interactions will indeed create an order depends, as Anderson noted
(1999: 222), on the nature and density of these interactions and of the feedback loops that link
agents together. If interactions and feedbacks are too dense, then the situation will become
chaotic in the sense that it will not settle to a stable configuration but bounce around among
various potential basins of attraction, none of which becomes significant enough to stabilize the
interactions. If, however, the density is too low, the situation will be, as Arthur put it (2013:
12), “dead” because it lacks any strong potential attractor. Order arises, rather, in systems whose
“components are partially, not fully, connected” (cp. Cilliers, 1998: 97; Ethiraj and Levinthal,
2004: 161; Hooker, 2011: 51; Kauffman, 1993: 36)).
Such a configuration – interactions and feedbacks in “the Goldilocks zone” – is, according to
many theorists, a configuration that will be favored by selective or self-organizing processes (cp.
Frenken et al., 1999: 6, commenting on Kauffman, 1993: 250; cp. also Thurner et al., 2018:
16–17; Wilson, 2016: 32). In particular, systems of interaction that develop to a point of sta-
bility organized around a basin of attraction will take on a form in which (Anderson, 1999: 222)
“elements are loosely coupled with one another” (cp. Weick, 1976: 3).
v. Modularity. A complex situation that organizes itself into a (relatively) stable order (characterized
by a particular attractor) will commonly exhibit a modular structure. This was already un-
derstood by Simon (1962: 468), and is illustrated by “the market” with its numerous, diverse,
and loosely connected buyers and sellers (each of them a module). Other examples of modular
32
Complexity
structures that manage complexity include federal, multi-ministry, and checks-and-balances

systems of government; divisions of labor in organizations; and the numerous competing and
cooperating firms that constitute a sector in the global economy. In all these cases, the mod-
ularized order is stable in the sense that it is relatively insensitive to ordinary fluctuations in
operational variables and may be robust in the sense that the key features of its order of actions
(Hayek, 1967a) are resilient in the face even of changes to organizational architecture (cp. Jen,
2003: 15). It must be noted, as a proviso, that, of course, modularized systems can, under certain
circumstances, “demodularize,” as with technologies and institutions that acquire more and more
dense connections between their components as they develop (cp. Arthur, 2015: 149–50).
vi. Interfaces. With modularization, as Baldwin and Clark (2000: 63–4) understood, interactions
among the various agents are managed by interfaces, which limit the influences passing between
one module and another (cp. Langlois, 2012: 5–6). For example, the conventions of exchange
in the market, which connect buyer and seller, limit the information that needs to be shared
between the participants and, in particular, does not require that they know or respond to a
great many of each other’s various and varying characteristics (cp. Gaus, 2016: 203; Hayek,
1976: 108; Langlois, 2012: 6).
More generally, for individual participants in an open society, “jurisdictional rights” (ac-
cording to Gaus, 2016: 199, 200; cp. Langlois, 2002: 20) enable individuals to ignore many of
the ways in which their fellows conduct themselves, thus buffering the system as a whole from
changes in the attitudes and actions of individuals whose behaviors give rise to it. These rights
modularize the realm of social interactions, thus ensuring that they are manageable by parti-
cipants (cp. D’Agostino, 2003: §19; Loasby, 1976: 142).
vii. Stability. These tools and others reduce the density of inter-agent transactions so that the system
as a whole is not overly sensitive to these transactions, and can thus maintain some stability as an
“order,” where, as Hayek emphasized (1973: 36), individual agents can form reliable ex-
pectations about how their behavioral choices may play out, notwithstanding the constant
adjustment of agents’ behaviors to the changing circumstances that these behaviors jointly create
(cp. Arthur, 2015: 183). Some theorists call this “universality,” which, according to Durlauf
(2012: 52), means that the emergent macro-level properties of a system remain relatively stable,
and thereby support reliable expectations despite continual change at the micro-level (cp.
Axtell, 2016: 78; Tao, 2012: 25; Wimsatt, 2013: 314).
viii. Novelty. A further concern is whether a stable order, supporting reliable expectations, also
provides opportunities for novel behaviors (which might, sometimes, provoke a shift to a
new basis of order characterized by a different attractor) and hence for the adaptation of the
order to changing circumstances (cp. Simon, 2000: 753). Again, the modularity of self-
organizing complex systems is important (Kauffman, 1993: 67; Ladyman et al., 2013: 60;
Langlois, 2012: 7) and for two reasons.
First of all, modularity supports diversity among modules (e.g., firms or individual agents)
and diversity is a crucial factor in innovation, as Page (2011: 9) has argued, and as Ashby
(1958: 86) and Loasby (1976: 56) had, in Stages 1 and 2, already recognized.
Second, modularity buffers the system because an innovation, responding to changed
circumstances, can be initiated within a module (e.g., a firm) and, because of the density-
reducing character of the interfaces between that module and others, the system as a whole is
not critically exposed if that innovation fails (cp. Perrow, 2008: 165; Weick, 1976: 7). On
the other hand, if innovation is successful, and if there are even a few connections between
agents in different modules, there will be channels using these “weak ties” in “small-world”
networks (cp. Bechtel and Richardson, 2010: xxxvi; Caldarelli and Catanzaro, 2012: 30)
33
Fred D’Agostino
through which the innovations can be dispersed (cp. Ethiraj and Levinthal, 2004: 165) –
perhaps via exchange, migration, and imitation – to other agents (cp. Schneider, 2012: 8),
thus altering the modular array.
ix. Increasing returns. If there should be “increasing returns” or “positive network externalities”
(Arthur, 1994: 114; Ma, 2007: 65), where imitation is rewarded and increasing levels of imi-
tation are increasingly rewarded, this will produce a cascade of reproductions of the successful
innovation (Gaus, 2018: 16–17), leading, in some cases, to a new overall order of actions in the
system, characterized by a different attractor.
x. Tipping-points. Such a development might show itself as a “tipping-point” phenomenon or
phase transition where, as Durlauf put it (2012: 52), there is an abrupt change rather than a
smooth transition. Bubbles and crashes in commercial contexts are good examples of tipping-
point phenomena; this would be a case where the system was not stable but was robust: the
changes shifted it away from its “basin,” but the system as a whole, thus transformed, survived
around a new basin of attraction (cp. Page, 2015: 33).
xi. Generative entrenchment. Indeed, a crucial dynamic, in relation to complex systems, is main-
taining the potential for innovation, for one consequence of cascades mediated by increasing
returns is “generative entrenchment” (Wimsatt, 2013: 318–319). Sometimes an innovation’s
institutionalization in a system is consolidated to the extent that it becomes the substrate for a
range of interactions and feedbacks (Williams, 2020: 154). When this happens, further radical
innovation may become difficult because “[t]o get rid of that foundation would require the
wholesale replacement of all the elements based upon it.” Metaphorically, the basin of attraction
becomes deeper and harder to escape; there is too much at stake. Hanseth illustrated this with
examples of technological innovation, noting that (2007: 82), when a particular technology is
deeply embedded in a larger social system that depends crucially upon it, the “accrued in-
vestments in the installed base” make it difficult to abandon that technology even when an
otherwise “better” alternative is available.
xii. Fat-tailed distributions. Certainly, we can expect, in relation to the dialectic of innovation and
entrenchment, a characteristic feature of complex systems, namely the so-called “fat-tailed” or
power-law distribution of adjustments to the system (Arthur, 2015: 15), with a plethora of
minor variations on an already established (entrenched) order, characterized by a particular
(very strong) attractor, and very few radical innovations establishing fundamentally different
orders of action, characterized by different attractors. Page (2015: 24) called this a “tension in
balancing exploration with exploitation,” alluding to work by March (1991). Kauffman and
Levin (1987: 35) hypothesize that this tension may play itself out, in specific domains, in three
distinct phases, corresponding, crudely, to the pre-paradigmatic, paradigmatic, and post-
paradigmatic stages in Kuhnian philosophy of science (Kuhn, 1970: chs. II & VIII). In other
words, we can expect to find an early exploratory stage, settling in due course in a specific basin
of attraction, followed by a second exploitative stage conducted largely within that basin. When
exploitative opportunities are exhausted there may then be a “long jump” exploratory mod-
ulation to a new basin of attraction (cp. Arthur, 2015: 10). Perhaps the evolution of the mobile
phone illustrates this, with much second-phase tinkering after the initial establishment of
the technological system, followed by the (relatively abrupt) transition to the “smart phone” as
the new standard. (See Thurner et al. (2018: 16–17) who posit “self-organized criticality”
where “systems… endogenously organize themselves” to a state where the possibilities of ro-
bustness and adaptive change are both accessible. Such systems are adapted but also adaptable.)
xiii. Catallaxy. Further questions about complex social orders include, perhaps most importantly,
whether any particular order is a “good” one or whether it is, instead, a “perverse spontaneous
order,” as Axtell put it (2016: 79).
34
Complexity
Hayek argued (1976: ch. 10) that the market order or catallaxy is exemplary in this regard
and in various ways.
First of all, the order established in the market, via exchange regulated by contract and
property law, is one in which each individual may freely seek to advance their own values in
accordance with their own understanding of the nature of their situation (Hayek, 1976: 109).
Even when individuals’ values and understandings differ, different individuals, exchanging and
contracting with one another, will each still be able to advance their different ends.
Second, in catallaxy, individual agents have incentives, in the form of potential transactions
favorable to themselves, to use the local knowledge that they may uniquely possess, and in this
way alert others to possibilities for action (Hayek, 1976: 115, 117), thus resulting, at least
sometimes, in a general increase in the efficient and effective use of resources (Hayek, 1976: 118).
Third, because the rules establishing the catallaxy do not prescribe ends, but, rather, proscribe
means (such as theft and fraud), they support the development of The Great Society, as Hayek
called it (Hayek, 1976: 109). In other words, they support the extension of peaceful and mutually
beneficial interactions beyond small groups whose members have the same values and beliefs and
whose cooperation depends on that commonality.
These mechanisms are so central to the social order that they might be said to be deeply
entrenched, in the sense that (Williams, 2020: 154) they, especially the principles of property,
contract, and exchange, are “acting as a platform for other systems to operate upon or within.”
Threats to their stability may well be threats to the continued existence of the social order itself.
6 Implications of Complexity for PPE

How do the disciplines of politics, philosophy, and economics engage with the ideas associated with
complexity thinking?
6.1 Complexity in Political Science

Williams (2020: 2, 53) believes that complexity theory can lead to a transformation of political science
by enabling us to answer a question raised by Harvey and Reed (1994: 391), namely, “how do the
mechanisms producing social order periodically produce chaos and pave the way for radical social
transformation.” In light of The Global Financial Crisis (GFC), 9/11, Brexit, and the like, this question
is no longer merely hypothetical. It is one for which complexity mechanisms, especially increasing
returns, and complexity features, particularly tipping points, provide models. (With the GFC, the more
firms failed the more likely it became that even more firms would fail, with the risk that, without
intervention, the whole financial system would have tipped over into failure and chaos.)
Methods. Astill and Cairney point out (2015: 147) that, in situations of complexity, where nu-
merous variables interact in producing overall patterns, it will not typically be true that the relations
among these variables can be successfully modeled by the kinds of quantitative methods familiar to
political scientists (cp. Gaus, 2021: §27). They also questioned the use of qualitative methods relying
on comparative case studies. As Jervis noted (1998: 582), comparative methods in political science
rely on the assumption that we can “estimate the role of one element or action” by “making
comparisons between two situations that are identical except for one variable.” That assumption will
not be valid in situations of complexity where “everything else cannot be held constant” because of
the tangled relations among the various factors (cp. Astill and Cairney, 2015: 175). On the other
hand, Homer-Dixon et al. (2013: 342) believe that using the tools of complexity science is a
promising approach to understanding how ideologies emerge and spread through social systems.
Topics. The concept of risk has become central to the ways in which contemporary societies
organize themselves because of the unpredictability and difficulties of accountability that complexity
35
Fred D’Agostino
generates (Beck, 1992; Hanseth, 2007: 80). In a complex system, such as a globalized, modernized
world system, “the unintended consequences of actions reverberate throughout the whole of society
in such a way that they … become intractable” (Latour, 2003: 36). This proposition is all too easy to
believe at the time of writing, which is in the midst of the COVID-19 pandemic and its socio-
economic consequences. With a more narrow focus, Perrow notes (2008: 162, 165) that, because of
their complexity, many contemporary technological and/or governance arrangements are vulner-
able to failures that could “cascade and bring the system down,” thus creating a pressing risk
management agenda for political actors.
Psychology. What human agents are to make of the complexity they encounter, and the risks it
brings in its train, has been considered in psychological studies that seem to show that decision-
makers often find complexity overwhelming and exhibit a variety of dysfunctional behaviors in-
cluding problem-hopping, response rigidity, and focusing on proximate events. As Connolly and
Koput say (1996: 288), “[h]ighly dynamic, complexly looped environments are certainly difficult to
understand and operate” (cp. Chernev et al., 2015: 335). More worryingly, there is evidence that
complexity might be a trigger for conspiratorial thinking. Bessi et al. (2014: 2, 3) argued that
conspiracy theories are appealing to those who need cognitive closure in situations where, because
of complexity, such closure is not easily found. On the other hand, a certain degree of complexity
may be needed for a satisfactory human experience. As Janlert and Stolterman report (2010: 24), in
order to be satisfied with their situations, human agents need a balance between too much and too
little order, between situations that are too safe and those that are too risky; they seem to need
orderly complexity.
Paradigms. With somewhat different interests, Ma (2007: 57) identified historical institutionalism
as an emerging paradigm for political analysis, alongside behavioralism and rational choice theory.
Crucially, he, and others such as David (1994), see in historical institutionalism some of the me-
chanisms, such as increasing returns, and features, such as path dependence, that are associated with
complexity. Pierson (2000: 257) mentions key ideas associated with complex systems, including
generative entrenchment (using the phrase “sunk costs”), and makes the point that (2000: 260) the
usual political mechanisms of trial and error may not be effective and that a sense of agency may be
difficult to support where there are numerous and complicated links (with feedback loops) between
actions and their consequences.
6.2 Complexity in Policy-making

What Pierson calls (2000: 259) “the complexity and opacity of politics” will be relevant to policy
development and implementation, where new complexity-respecting approaches have been developed.
As Colander put it (2014b: 15–16), the standard policy framework is defined by assumptions that are
not appropriate if the systems for which policies are being designed are complex systems. As Maughan
said (2020), “No one is driving” in situations of complexity; there are, in effect, no elements in a larger
system where intentional and directed action can reliably deliver stable macro-level change or, as
Holland puts it (2014: 25), no “levers” for policy. As Colander sees it, this suggests a method of “laissez
faire activism,” which aims, not at specific policy outcomes, but at the development of a social eco-
system in which “bottom-up” policy initiatives form the basis of the evolution of that system. There is
resonance here with ideas about “experimentalist governance” (cp. Sabel and Zeitlin, 2012: 175;
Sanderson, 2009: 707–708), where policy-making becomes an iterative learning exercise, perhaps
exhibiting some analogies with polycentric governance arrangements (Ostrom, 1999) that depend on
central oversight, though not direction, of diverse agencies acting under a principle of subsidiarity, with
periodic reviews of the results of their different experiments. As Majone put it (1989: 183), public
policy-making should be directed to enhancing the capacities of the various social elements to learn
how to adapt to changing circumstances, with less attention to the pursuit of specific outcomes.
36
Complexity
Laissez-faire activism and experimentalist governance presuppose certain capabilities on the part
of social agents, a matter to which Head (2019: 190) has given attention, with particular emphasis on
the capabilities “to deal with multiple frames …, to adjust actions to uncertain changes …, to
respond to changing agendas and …, to unblock stagnations.” This list mirrors that of Connolly and
Koput’s dysfunctional responses, and it particularizes Barrett’s (2020: 125) exhortation that, in the
face of complexity, the policy-maker’s aspiration should be to pursue “progressiveness” per se, that
is, the capacity of systems to improve themselves. Also relevant here is Hayek’s notion (2014 [1974]:
371–372) that the policy-maker, since they cannot confidently aim at the production of specific
results, ought to “cultivate growth by providing the appropriate environment, in the manner in
which the gardener does this for his plants” (cp. Levinthal and Warglien, 1999: 342).
While Bankes also acknowledges (2005: 2) that, with a complex situation, we cannot be con-
fident of our ability to predict the consequences of our policy initiatives (cp. Williams, 2020: 180),
he still sees a role for modeling potential outcomes of policy initiatives, with the proviso (2005: 6)
that policy-makers should look to limit the possibilities for failure, especially catastrophic failure,
rather than pursuing optimal outcomes. This approach to policy analysis focuses on the robustness of
policy initiatives or those that are “highly immune to failure” (Bankes, 2005: 13–14). Planners
should consider ensembles of alternative initiatives and the alternative scenarios in which those
initiatives might be implemented, aiming to find policy options that provide (relatively) satisfactory
outcomes across a range of possible scenarios (Bankes, 2005: 9).
The discussions by Bankes and Head suggest the potential for other approaches, for instance,
those that, in the face of complexity, emphasize the importance of purpose-independent resources
and strategic primary goods that would be of use to agents coping with complexity regardless of the
specific direction of social change.
Writing earlier, at Stage 2, Lindblom (1959: 86) had already grappled with some of these issues,
advocating that, in the face of complexity, contestability of values, and limitations on knowledge,
policy-making should proceed by a method of “successive approximation.” This kind of approach
shows itself again in Arthur’s suggestion (2015: 184) that what policy-makers need is “[n]ot a heavy
hand, not an invisible hand, but a nudging hand” (cp. Thaler and Sunstein, 2009), that seeks “to
push the system gently toward favored structures.” Lindblom (1959: 84) specifically advocated that
comparison of policy options be limited to “those policies that differ in relatively small degree from
policies presently in effect,” not realizing that even this modest aspiration might be difficult to fulfill
in situations of complexity.
6.3 Complexity and Political Philosophy

Some of these issues are related to a basic difficulty in contemporary political philosophy, the
viability of “ideal theory” or the identification of a specific political formation as “ideal” and of
reformist steps that could be taken to realize that ideal. According to Gaus (2016), ideal theory is
problematic in principle because we can’t identify the ideal and we couldn’t track progress towards it
even if we could. Barrett puts this point more strongly when he says (2020: 117), “ideal theory is an
impossible enterprise.”
There are two difficulties for ideal theory arising from the complexity of human social and
cultural formations. First of all, as pluralists have long maintained (D’Agostino, 2004) and parti-
cularists have thematized (Dancy, 1993), the evaluative standards underpinning identification of an
ideal are themselves plural and complexly interrelated. So, there are multiple, sometimes in-
compatible, balancings of these various standards, and the various balancings define peaks on an
evaluative landscape, none of which is unequivocally optimal. In other words, “the ideal” is not
uniquely defined. Second, however, because of the complex interrelations of agents, institutions,
and the like, the evaluative “landscape” traversed by specific reformist steps is so “rugged” (in the
37
Fred D’Agostino
language of NK fitness modeling) that even a policy that differs very little from that presently in
place could lead to a significantly worse, rather than a slightly better, situation, thus putting in
question Lindblom’s proposed “successive approximation” method of policy development. As Gaus
put it (2016: 68–69), “[i]n … a maximally high-dimensional landscape … there is no point in getting
close to the ideal point … but not achieving it: its near neighbors may not be at all just.”
6.4 Complexity and Economics

The influence of complexity thinking on economics is disputed, with some scholars (Holt et al.,
2011: 357) maintaining that “[t]he neoclassical era in economics has ended and has been replaced …
[by] ‘the complexity era’,” while others (Durlauf, 2012: 53) think that these statements are grandiose
and that those who make them do not properly understand the ways in which economists analyze
relevant situations. Whatever the existential issues, it is clear that complexity-based economics has
facilitated a discussion of (a) neglected authors, (b) macro-economic modeling, (c) methodological
issues, and (d) new analyses of familiar phenomena.
(a) Neglected Authors. Within mainstream economics, the ideas of Schumpeter and the Austrian
economists constitute heterodox positions, but Foster (2005: 873) argues that these ideas are more
intelligible when considered within complex systems theory. Arthur believes (2013: 28) that
“complexity meets up with and revives the grand tradition of political economy,” and attributes
(2013: 2) the avoidance of complexity by mainstream economics to analytical difficulties that
could only really be addressed with the development of agent-based modeling.
(b) Macro-Economic Modelling. An important simplifying assumption is much commented on
by advocates of complexity economics, namely, the use of the “representative agent” in
macroeconomics, where theorists postulate a single individual with certain characteristics and
then simply multiply that individual’s behavior in order to derive the macro-level economic
indicators they are interested in (cp. Baddeley, 2017: 107; Delli Gatti et al., 2010: 116; Holt
et al., 2011: 359–60). In complexity economics, on the other hand, these indicators emerge
from an agent-based simulation in which the empirical diversity of individuals is more properly
represented.
(c) Methodological Issues. While modeling such diversity only becomes feasible with the devel-
opment of the necessary computational infrastructure in the 1990s, there are non-trivial metho-
dological issues associated with even moderately realistic modeling. Anderson (1999: 227) notes that
it is not enough to explain an observed order to show that it can be generated by a particular agent-
based simulation, because “a given outcome can be explained equally well by a host of simulations
with very different assumptions.” Foster (2011: 525) and Epstein (2006a: 9) have given considerable
thought to how such modeling might nevertheless be profitably conducted.
(d) Analyses of Phenomena. Markose identified (2005: 159) a number of well-known phenomena
that remain puzzling within orthodox economics but are grist to the mill for complexity economics,
namely, “innovation, competitive co-evolution, persistent heterogeneity, increasing returns, the
error-driven processes behind market equilibria, herding, crashes and extreme events such as in the
business cycle and in stock markets.” Arthur (2015), a founding figure in complexity economics,
focuses his analyses on central topics such as rational expectations and innovation and has pioneered
the case for the importance of increasing returns. And, without, indeed in advance of, sophisticated
modeling techniques, Hayek (1967b) provided a complexity informed account of the fundamentals
of market behavior, connecting this analysis with more general topics such as The Great Society, the
use of dispersed information by economic agents, and prices as a signaling mechanism.
38
Complexity
7 Conclusions
It is salutary to recall H.L. Mencken’s observation (quoted in Colander, 2014a: 1): “For every
complex problem there is an answer that is simple, clear, and wrong.” With the development of
“new cognitive instrumentalities,” there is now some prospect of analyzing complexity in a better
way than was previously possible, though it is important to note that, even with new techniques
such as agent-based modelling, we may not be able reliably to predict the consequences of
thoughtful, empirically-based policy implementation. As Rescher said (1998: 195), “[t]he answer to
the question of how to conduct life in a complex world is – very carefully.”
Notwithstanding these reservations, there is much for the PPE scholar to draw on from the
toolkit of complexity science, including, especially, its understanding of such recurrent phenomena
as generative entrenchment, tipping-points, and path-dependency.
Acknowledgments
The Editors asked my friend Jerry Gaus to write this article for the Companion. He asked whether I
could be added as a co-author. The Editors graciously agreed. He and I then decided that I would
draft the article for discussion, in Brisbane, in August of 2020. The COVID pandemic made that
meeting impossible and, in any event, Jerry died during the very week we had hoped to meet in
Australia. He never saw the draft from which this chapter is descended, but, clearly, his thinking is
everywhere evident. Thanks to his students and associates Jacob Barrett, Alex Motchoulski, Ryan
Muldoon, and Alexander Schaefer for their comments. Thanks too to editor Chris Melenovsky for a
careful reading of an earlier draft that led to considerable improvement of the article.
In memoriam Jerry Gaus
References
Anderson, P. 1999. “Perspective: Complexity Theory and Organization Science,” Organization Science, 10,
216–232.
Arthur, W. B. 1994. Increasing Returns and Path Dependence in the Economy. Ann Arbor; University of Michigan
Press.
Arthur, W. B. 1999. “Complexity and the Economy,” Science, 284, 107–109.
Arthur, W. B. 2013. Complexity Economics: A Different Framework for Economic Thought. Santa Fe, NM; Santa Fe
Institute.
Arthur, W. B. 2015. Complexity and the Economy. Oxford; Oxford University Press.
Ashby, W. R. 1952. Design for a Brain. London; Chapman & Hall.
Ashby, W. R. 1958. “Requisite Variety and Its Implications for the Control of Complex Systems,” Cybernetica,
1, 83–99.
Astill, S. & Cairney, P. (2015) “Complexity Theory and Political Science,” in Astill, S. & Cairney, P. eds.
Handbook on Complexity and Public Policy. Cheltenham, UK; Edward Elgar Publishing.
Axtell, R. (2016) “Hayek Enriched by Complexity Enriched by Hayek,” Advances in Austrian Economics, 21,
63–121.
Baddeley, M. (2017) Behavioural Economics: A Very Short Introduction, Oxford; Oxford University Press.
Baldwin, C. Y. & Clark, K. B. (2000) Design Rules, Cambridge, MA; MIT Press.
Bankes, S. (2005) “Robust Policy Analysis for Complex Open Systems,” Emergence: Complexity and
Organization, 7, 2–10.
Barrett, J. (2020) “Social Reform in a Complex World,” Journal of Ethics and Social Philosophy, 17, 103–132.
Bechtel, W. & Richardson, R. C. (2010) Discovering Complexity: Decomposition and Localization as Strategies in
Scientific Research, Cambridge, MA; MIT Press.
Beck, U. (1992) Risk Society: Towards a New Modernity, London and Newbury Park, CA; Sage Publications.
Beinhocker, E. D. (2013) “Reflexivity, Complexity, and the Nature of Social Science,” Journal of Economic
Methodology, 20, 330–342.
39
Fred D’Agostino
Bertalanffy, L. V. (1968) General System Theory: Foundations, Development, Applications, New York; G. Braziller.
Bessi, A., Coletto, M., Davidescu, G., Scala, A., Caldarelli, G. & Quattrociocchi, W. (2014) “Science vs
Conspiracy: Collective Narratives in the Age of (Mis)information,” arXiv.org, 10.
Boulding, K. E. (1966) The Impact of the Social Sciences, New Brunswick, NJ; Rutgers University Press.
Boyd, R. & Richerson, P. J. (2005) “Simple Models of Complex Phenomena,” in Boyd, R. & Richerson, P. J.
eds. The Origin and Evolution of Cultures, New York and Oxford; Oxford University Press.
Brennan, G. (2010) “PPE: An Institutional View,” Politics, Philosophy & Economics, 9, 379–397.
Caldarelli, G. & Catanzaro, M. (2012) Networks: A Very Short Introduction, Oxford; Oxford University Press.
Campbell, D. (1979) “A Tribal Model of the Social System Vehicle Carrying Scientific Knowlege,” Knowledge,
1, 181–201.
Chernev, A., Böckenholt, U. & Goodman, J. (2015) “Choice Overload: A Conceptual Review and
Meta‐analysis,” Journal of Consumer Psychology, 25, 333–358.
Cilliers, P. (1998) Complexity and Postmodernism: Understanding Complex Systems, London and New York;
Routledge.
Clayton, P. (2013) “On the Plurality of Complexity-producing Mechanisms,” in Lineweaver, C. H., Ruse, M.
& Davies, P. C. W. eds. Complexity and the Arrow of Time., Cambridge; Cambridge University Press.
Colander, D. (2014a) “Complexity a New Kind of Science?,” in Colander, D. & Kupers, R. eds. Complexity
and the Art of Public Policy, Princeton; Princeton University Press.
Colander, D. (2014b) “The Complexity Policy Framework,” in Colander, D. C. & Kupers, R. eds. Complexity
and the Art of Public Policy, Princeton; Princeton University Press.
Connolly, T. & Koput, K. (1996) “Naturalistic Decision Making and the New Organizational Context,” in
Shapira, Z. ed. Organizational Decision Making, Cambridge; Cambridge University Press.
D’Agostino, F. (2003) Incommensurability and Commensuration: The Common Denominator, Aldershot &
Burlington, VT; Ashgate.
D’Agostino, F. (2004) “Pluralism and Liberalism,” in Gaus, G. & Kukathas, C. eds. Handbook of Political Theory,
London; Sage.
D’Agostino, F. (2019) “The Situational Logic of Disciplinary Scholarship,” in Sassower, R. & Laor, N. eds. The
Impact of Critical Rationalism, Cham; Palgrave Macmillan.
Dancy, J. (1993) Moral Reasons, Oxford and Cambridge, MA; Blackwell.
David, P. A. (1994) “Why are Institutions the ‘Carriers of History’?: Path Dependence and the Evolution of
Conventions, Organizations and Institutions,” Structural Change and Economic Dynamics, 5, 205–220.
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.
Demsetz, H. (1967) “Toward a Theory of Property Rights,” The American Economic Review, 57, 347–359.
Durlauf, S. N. (2012) “Complexity, Economics, and Public Policy,” Politics, Philosophy & Economics, 11, 45–75.
Epstein, J. M. (2006a) “Agent-Based Computational Models and Generative Social Science,” in Epstein, J. M. ed.
Generative Social Science: Studies in Agent-based Computational Modeling, Princeton; Princeton University Press.
Epstein, J. M. (2006b) “Remarks on the Foundations of Agent-Based Generative Social Science,” in Tesfatsion,
L. & Judd, K. L. eds. Handbook of Computational Economics, Amsterdam; Elsevier.
Epstein, J. M. & Axtell, R. (1996) Growing Artificial Societies: Social Science from the bottom up, Washington, DC;
Brookings Institution Press.
Ethiraj, S. & Levinthal, D. (2004) “Modularity and Innovation in Complex Systems,” Management Science, 50,
159–173.
Felin, T., Kauffman, S., Koppl, R. & Longo, G. (2014) “Economic Opportunity and Evolution: Beyond
Landscapes and Bounded Rationality,” Strategic Entrepreneurship Journal, 8, 269–282.
Feyerabend, P. (1975) Against Method: Outline of an Anarchistic Theory of Knowledge, London; NLB.
Foster, J. (2005) “From Simplistic to Complex Systems in Economics,” Cambridge Journal of Economics, 29,
873–892.
Foster, J. (2011) “Economic Systems,” in Hooker, C. ed. Philosophy of Complex Systems, Amsterdam; Elsevier.
Frenken, K., Marengo, L., Valente, M. & Brenner, T. (1999) “Interdependencies, Nearly-decomposability and
Adaptation,” in Brenner, T. ed. Computational Techniques for Modelling Learning in Economics: Advances in
Computational Economics, Boston, MA; Springer.
Galison, P. (1997) Image and Logic: A Material Culture of Microphysics, Chicago; University of Chicago Press.
Gaus, G. F. (2006) “Hayek on the Evolution of Society and Mind,” in Feser, E. ed. The Cambridge Companion to
Hayek, Cambridge; Cambridge University Press.
Gaus, G. F. (2016) The Tyranny of the Ideal: Justice in a Diverse Society, Princeton; Princeton University Press.
Gaus, G. F. (2018) “Self-organizing Moral Systems: Beyond Social Contract Theory,” Politics, Philosophy &
Economics, 17, 119–147.
40
Complexity
Gaus, G. F. (2021) The Open Society and Its Complexities, Oxford; Oxford University Press.
German, J. (2021) “History” Santa Fe Institute, https://www.santafe.edu/about/history.
Giddens, A. (1984) The Constitution of Society, Cambridge; Polity Press.
Hanseth, O. (2007) “Complexity and Risk,” in Hanseth, O. & Ciborra, C. eds. Risk, Complexity and ICT,
Cheltenham; Edward Elgar Publishing Limited.
Hardin, G. (1963) “The Cybernetics of Competition: A Biologist’s View of Society,” Perspectives in Biology and
Medicine, 42, 58–84.
Harvey, D. L. & Reed, M. H. (1994) “The Evolution of Dissipative Social Systems,” Journal of Social and
Evolutionary Systems, 17, 371–411.
Hayek, F. A. (1967a) “Notes on the Evolution of Systems of Rules of Conduct,” in Hayek, F. A. ed. Studies in
Philosophy, Politics, and Economics, Chicago; University of Chicago Press.
Hayek, F. A. (1967b) “The Theory of Complex Phenomena,” in Hayek, F. A., Studies in Philosophy, Politics and
Economics, London; Routledge & Kegan Paul.
Hayek, F. A. (1973) Law, Legislation, and Liberty, Vol. 1: Rules and Order, London and Melbourne; Routledge
and Kegan Paul.
Hayek, F. A. (1976) Law, legislation and liberty, vol. II: The Mirage of Social Justice, London; Routledge & Kegan
Paul.
Hayek, F. A. (2014) [1974]. “The Pretence of Knowledge,” in Caldwell, B. ed. The Market and Other Orders:
The Collected Works of F.A. Hayek, volume 15, London and New York; Routledge.
Head, B. W. (2019) “Forty Years of Wicked Problems Literature,” Policy and Society, 38, 180–197.
Holland, J. (2014) Complexity: A Very Short Introduction, Oxford; Oxford University Press.
Holland, J. & Miller, J. (1991) “Artificial Adaptive Agents in Economic Theory,” The American Economic Review,
81, 365–370.
Holt, R. P. F., Rosser, J. B. & Colander, D. (2011) “The Complexity Era in Economics,” Review of Political
Economy, 23, 357–369.
Homer-Dixon, T., Maynard, J. L., Mildenberger, M., Milkoreit, M., Mock, S. J., Quilley, S., Schröder, T. &
Thagard, P. (2013) “A Complex Systems Approach to the Study of Ideology,” Journal of Social and Political
Psychology, 1, 337–363.
Hooker, C. (2011) Philosophy of Complex Systems, Amsterdam; Elsevier.
Horgan, J. (1995) “From Complexity to Perplexity,” Scientific American, 272, 104–109.
Janlert, L.-E. & Stolterman, E. (2010) “Complex Interaction,” ACM Transactions on Computer-Human Interaction
(TOCHI), 17, 1–32.
Jen, E. (2003) “Stable or Robust: What’s the Difference?,” Complexity, 8, 12–18.
Jervis, R. (1998) “Complexity and the Analysis of Political and Social Life,” Political Science Quarterly, 112,
569–593.
Kauffman, S. A. (1993) The Origins of Order: Self Organization and Selection in Evolution, New York; Oxford
University Press.
Kauffman, S. A. & Levin, S. (1987) “Towards a General Theory of Adaptive Walks on Rugged Landscapes,”
Journal of Theoretical Biology, 128, 11–45.
Kuhn, T. S. (1970) The Structure of Scientific Revolutions, Chicago; University of Chicago Press.
Ladyman, J., Lambert, J. & Wiesner, K. (2013) “What is a Complex System?,” European Journal for Philosophy of
Science, 3, 33–67.
Lane, R. (2016) “The Simplicity of Complexity,” in Lane, R. ed. The Complexity of Self Government,
Cambridge; Cambridge University Press.
Langlois, R. N. (2002) “Modularity in Technology and Organization,” Journal of Economic Behavior and
Organization, 49, 19–37.
Langlois, R. N. (2012) “Design, Institutions, and the Evolution of Platforms,” Journal of Law, Economics & Policy,
9, 1–13.
Latour, B. (2003) “Is Re-modernization Occurring - and if so, how to prove it?,” Theory, Culture & Society, 20,
35–38.
Levinthal, D. & Warglien, M. (1999) “Landscape Design,” Organization Science, 10, 342–357.
Lindblom, C. (1959) “The Science of ‘Muddling’ Through,” Public Administration Review, 19, 79–88.
Lloyd, S. (2001) “Measures of Complexity: a Nonexhaustive List,” IEEE Control Systems, 21, 7–8.
Loasby, B. J. (1976) Choice, Complexity, and Ignorance, Cambridge and New York; Cambridge University Press.
Ma, S.-Y. (2007) “Political Science at the Edge of Chaos? The Paradigmatic Implications of Historical
Institutionalism,” International Political Science Review, 28, 57–78.
Majone, G. (1989) Evidence, Argument and Persuasion in the Policy Process, New Haven; Yale University Press.
March, J. G. (1991) “Exploration and Exploitation in Organizational Learning,” Organization Science, 2, 71–87.
41
Fred D’Agostino
Markose, S. M. (2005) “Computability and Evolutionary Complexity: Markets as Complex Adaptive Systems
(CAS),” Economic Journal, 115, F159–F192.
Maughan, T. (2020) The Modern World Has Finally Become Too Complex for Any of Us to Understand [Online].
OneZero. https://onezero.medium.com/the-modern-world-has-finally-become-too-complex-for-any-of-
us-to-understand-1a0b46fbc292 [Accessed 1 December 2020].
McKelvey, B. (1999) “Self-Organization, Complexity Catastrophe, and Microstate Models at the Edge of
Chaos,” in Baum, J. A. C. & McKelvey, B. eds. Variations in Organization Science: In Honor of Donald T.
Campbell, Thousand Oaks, CA; SAGE.
McKelvey, B. (2003) “Emergent Order in Firms,” in Mitleton-Kelly, E. ed. Complex Systems and Evolutionary
Perspectives on Organisations, Oxford; Elsevier Science.
Ostrom, E. (1999) “Polycentricity, Complexity, and the Commons,” The Good Society, 9, 37–41.
Page, S. E. (2008) “Uncertainty, Difficulty, and Complexity,” Journal of Theoretical Politics, 20, 115–149.
Page, S. E. (2011) Diversity and Complexity, Princeton, NJ; Princeton University Press.
Page, S. E. (2012) “A Complexity Perspective on Institutional Design,” Politics, Philosophy & Economics,
11, 5–25.
Page, S. E. (2015) “What Sociologists Should Know About Complexity,” Annual Review of Sociology, 41, 21–41.
Perrow, C. B. (2008) “Complexity, Catastrophe, and Modularity,” Sociological Inquiry, 78, 162–173.
Pierpaolo, A. (2011) “Complexity and Innovation,” in Allen, P., Maguire, S. & McKelvey, B. eds. The Sage
Handbook of Complexity and Management, London; SAGE Publications Ltd.
Pierson, P. (2000) “Increasing Returns, Path Dependence, and the Study of Politics,” American Political Science
Review, 94, 251–267.
Rawls, J. (1971) A Theory of Justice, Cambridge, MA; Belknap Press of Harvard University Press.
Rescher, N. (1998) Complexity, London and New York; Routledge.
Rittel, H. & Webber, M. (1973) “Dilemmas in a General Theory of Planning,” Policy Sciences, 4, 155–169.
Sabel, C. F. & Zeitlin, J. (2012) “Experimentalist Governance,” in Levi-Faur, D. ed. The Oxford Handbook of
Governance, Oxford; Oxford University Press.
Sanderson, I. (2009) “Intelligent Policy Making for a Complex World,” Political Studies, 57, 699–719.
Schelling, T. (1969) “Models of Segregation,” The American Economic Review, 59, 488–493.
Schneider, V. (2012) “Governance and Complexity,” in Levi-Faur, D. ed. The Oxford Handbook of Governance,
Oxford; Oxford University Press.
Simon, H. (1962) “The Architecture of Complexity,” Proceedings of the American Philosophical Society, 106,
467–482.
Simon, H. (2000) “Public Administration in Today’s World of Organizations and Markets,” PS: Political Science
and Politics, 33, 749–756.
Smith, L. A. (2007) Chaos: A Very Short Introduction, Oxford; Oxford University Press.
Spier, F. (2015) Big History and the Future of Humanity, Chichester; Wiley Blackwell.
Tao, T. (2012) “E pluribus unum: From Complexity, Universality,” Daedalus, 141, 23–34.
Thaler, R. H. & Sunstein, C. R. (2009) Nudge, Camberwell; Penguin Group Australia.
Thurner, S., Klimek, P. & Hanel, R. (2018) Introduction to the Theory of Complex Systems, Oxford; Oxford
University Press.
Waldrop, M. (1992) Complexity: The Emerging Science at the Edge of Order and Chaos, New York; Simon &
Schuster.
Weaver, W. (1948) “Science and Complexity,” American Scientist, 36, 536–544.
Weick, K. (1976) “Educational Organizations as Loosely Coupled Systems,” Administrative Science Quarterly,
21, 1–19.
Wiener, N. (1948) Cybernetics, or, Control and Communication in the Animal and the Machine, New York; Wiley.
Williams, A. (2020) Political Hegemony and Social Complexity, Cham; Palgrave Macmillan.
Wilson, D. S. (2016) “Two Meanings of Complex Adaptive Systems,” in Wilson, D. S. & Kirman, A. P. eds.
Complexity and Evolution: Toward a New Synthesis for Economics, Cambridge, MA; MIT Press.
Wimsatt, W. C. (2013) “The Role of Generative Entrenchment and Robustness in the Evolution of
Complexity,” in Lineweaver, C. H., Ruse, M. & Davies, P. C. W. eds. Complexity and the Arrow of Time,
42
4
PPE IN MARX’S MATERIALIST
CONCEPTION OF HISTORY
Vanessa Wills
1 Introduction
What relationships emerge among philosophy, politics, and economics in Marxist thought? Karl
Marx’s earliest academic writings were firmly located within specifically philosophical debates and
scholarly traditions. His 1841 doctoral dissertation, The Difference Between the Democritean and Epicurean
Philosophy of Nature, analyzed the relationships among materialism, determinism, and free will in the
atomistic theories of ancient Greek philosophers, Democritus and Epicurus. His next major writing, A
Contribution to the Critique of Hegel’s Philosophy of Right, analyzed G. W. F. Hegel’s approach to the
concept of justice and also interrogated topics such as alienation and the character of religious belief.
Famously, however, after skewering what he took to be the major shortcomings of that which he and
his collaborator Friedrich Engels dubbed “the German Ideology,” Marx more centrally addressed
political economists as his interlocutors in later works such as his magnum opus, Capital. He began his
scholarly trajectory with a discussion of a philosophical abstraction – the atom – and ended it with an
investigation into an economic one – the commodity.
This path raises the question of how we ought to understand Marx’s relationship to philosophy over
the course of his career. Can we not reasonably conclude that Marx executed – or at least, attempted to
execute – a hard turn away from philosophy as a mode of investigation and towards politics and
economics as fields with far greater explanatory power? After all, one might (wrongly, I think, but
more on that later) understand Marx to regard philosophical theory as a mere epiphenomenon, a pale
reflection of hard political and economic reality, such that philosophy is simply altogether obviated by
political and economic study. This is the reading of Marx offered by Louis Althusser in his For Marx.
A parallel and occasionally competing line of argument runs as follows for the claim that, having
declared in his 11th thesis on Feuerbach that, “The philosophers have merely interpreted the world, in
various ways; the point is to change it,” the later, post-1845 Marx cannot properly be understood to
have been a philosophical thinker. Philosophy as a discipline generally takes itself to investigate that
which is true, objective, universal, and even eternal. Marx, however, openly advertised that what he
and Engels called “the materialist conception of history” is an analysis of human social existence from
the specific point of view of the working class and its struggle against the bourgeoisie under capitalism.
Famously, Marx aimed at the victory of the proletariat and its assumption of what he took to be its
rightful place as the class to lead humanity out of its alienated and degraded condition.
Marxist theory – historical materialism – takes a definite side in the political struggle between
proletariat and bourgeoisie that it analyzes. If that is so, then it might seem reasonable to suppose that
DOI: 10.4324/9780367808983-5 43
Vanessa Wills
Marxism is inherently subjective and particular in a way that renders it non- or even anti-
philosophical and calls into question any respectability it might have as a mode of real scientific
inquiry. It might seem reasonable then to regard it as fundamentally mistaken to think of Marxist
theory as having any pretensions to philosophy at all. Marxism, this line of argument goes, dis-
dainfully rejects philosophy altogether as a kind of fool’s errand in search of necessarily false and
misleading universalism, as odious false consciousness, and as an attempt to obscure and mystify
human social relations and make seem harmonious what is fractured into a sharp political battle.
So the question, “What is the relationship among philosophy, politics, and economics in Marxist
thought?” leads us here to investigate two related subquestions. The first is whether Marx’s turn to
economic study in his later career is at odds with, and represents a departure from, the philosophical
concerns of his earlier writings. The second is whether Marx’s method of taking up a proletarian
“standpoint” as his lens, and his seeming indifference to any supposed rigid division between analysis
and advocacy, makes Marxist theory particular rather than universal, and subjective rather than
objective, in a way that disqualifies it from being philosophical (or, for that matter, truth-tracking or
scientific at all). Once we establish that the answer to the second question is “No,” one key mo-
tivation evaporates for thinking of the “later” Marx as departing from the concerns present in his
earlier work.
We will take these questions up each in turn. But first, let us make an observation about the
present-day field of Philosophy, Politics, and Economics (PPE), and how it relates to the state of
he academy. It is well-known that as separate disciplines, Philosophy, Politics, and Economics are
frequently the separate provinces of academic departments that are too often functionally siloed from
one another, with often paltry interaction and little cross-pollination of methods and ideas. Precisely
that state of affairs produces the continued demand for PPE as a field and as an attempt at trans-
formative intervention into how scholars carry out academic research into these areas. (In fact, it is
just over a century now since the first PPE degree program opened its doors at Oxford in 1920, as a
way of training British civil servants to navigate a world made increasingly complex and inter-
connected by the then-recent events of World War I, the 1918 flu pandemic, and of course, the
Russian Revolution of 1917 (White et al., 2020).)
Marx also recognized, and regarded as a kind of intellectual ill, the tendency for academic study to
be fractured into a proliferating and increasingly distanced array of intellectually isolated specialties and
subspecialties. He saw this phenomenon as one that represents and helps further the expansion of
human knowledge and subject matter expertise, but that also makes the relationships among these
various areas of knowledge harder to keep in view, so that we increasingly lose sight of the whole.
Marx thought of philosophy – a discipline that concerns itself with pure abstract thought as one of its
chief subject matters – as particularly susceptible to this. It fails to recognize its incapacity to resolve its
own questions while isolating itself from the resources made available by more empirical fields. Partially
in response to the academy’s tendency to be splintered, Marx and Engels wrote in their 1846 Critique of
the German Ideology, “We know only a single science, the science of history” (1976a, 28). Here, their
point was to bring into question the treatment of “the history of nature” and “the history of men” as
though natural and social science were two unrelated areas. Marx and Engels regard them instead as
components in a holistic understanding of the mutually inseparable constituents within one dynami-
cally developing and interrelated whole of existence.
It can be tempting to characterize the development of Marx’s ideas over the course of his career
as a departure from one field of study and wholesale abandonment of it for another. However, to do
so is to fail to grasp the central significance of Marx’s and Engels’s “materialist conception of history”
as a theoretical method. Historical materialism – Marxist theory – is a unifying, integrative approach
to comprehending a reality whose character can be fully grasped only as its elements come to be,
develop, and cease to be, as the result of more or less rational and conscious human intervention
carried out in the course of time.
44
Marx’s Materialist Conception of History
Strikingly, for both Marx and for the Oxford dons of the 1920s, one key aim of a more unified
theoretical approach is not merely to understand the world, but to change it. Obviously, they had
mostly very different ideas about how the world ought to be changed. Our aim in the present
discussion is twofold: to better understand Marx’s continuing significance for PPE as a field, and to
better understand the nexus of philosophical, political, and economic study in Marx’s thought.
In the sections that follow, we will pursue this aim by first investigating Marx’s relationship to
philosophy especially in the persons of G. W. F. Hegel and his followers, and then examining the
notion of a “proletarian standpoint” and its relationship to ideals of objectivity and universality in
empirical scientific study. We will later conclude with a discussion of the specific relevance of these
questions for the study of PPE.
2 Philosophy and History in Marx and Hegel

We of course cannot speak of Marx’s relationship to philosophy without speaking of his relationship
to one preeminent philosopher in particular. That figure is Hegel, whose system of dialectical
idealism loomed large over German philosophers of the late 19th century as a method to be either
adopted and extended or refuted and overcome.
Hegel’s dialectical idealism seeks to account for and explain all aspects of existence as parts or
“moments” of one unified, developing processual whole. This process is driven forward by inner
contradictions within and among concepts contained in one grand, all-encompassing concept,
which Hegel calls, “the Absolute Idea.” Since this totalizing ideal abstraction is all existence, the
typical opposition between form and content holds little meaning in Hegel’s dialectical idealist
system.
Put differently: “Being” is fundamentally conceptual for Hegel. Objects have determinate ex-
istence in virtue of conceptual boundaries that distinguish what an object is from what it is not. But
those determinate boundaries themselves shift over time as the object goes through a process of
development and interaction with the world around it. Hence, Hegel’s claim that all things contain
their own negation: they exist as they are in virtue of what they are not, over the course of time they
assume qualities and characteristics they did not have before and lose ones that they previously did
have, and just as they emerged in the course of history, so they may fade away in the course of it
even while being preserved in a sublated form.
Apprehending a thing in static isolation tells us only how that thing appears at that particular
time. To know the real nature of that thing, we have to understand how its present appearance is a
negation of its prior appearances and we must appreciate that its future appearances may annihilate its
present one. Thus, taken historically, a thing might be identical with what, if taken only in static
isolation and rigidly opposed to it, might appear as a completely different object, sharing no par-
ticular appearance with it at all.
One consequence of this for Hegel is that logic can be fully understood not a priori but only in
connection with substantive, empirical content. “A=A” is true enough as a tautology of formal
logic, but no object is simply identical with itself. It is also always what it is not. The logical
relationship of ideas, which is also the actual relationship of things in existence, is revealed only once
the content within the abstraction of pure “Being” is unfolded and can be known as history.
This doctrine appears throughout Hegel’s writings. In the preface to his Elements of the Philosophy
of Right, Hegel counsels his reader,
Form in its most concrete significance is reason as conceptual cognition, and content is
reason as the substantial essence of both ethical and natural actuality; the conscious identity
of the two is the philosophical Idea. (Hegel, 1991, 22)
45
Vanessa Wills
In The Science of Logic (the so-called “Lesser Logic” which appears as Part One of his Encyclopedia of
the Philosophical Sciences in Basic Outline), Hegel, who expressed the essential oneness and inter-
relatedness of all existence as its identity with “Spirit” or the “Absolute Idea,” wrote the following:
As far as its essential content is concerned, the history of philosophy deals not with the past,
but with what is eternal and absolutely present, and its result must be compared not to a
gallery of errors in the human spirit, but rather to a pantheon of divine figures
[Göttergestalten]. These divine figures are the various stages of the ideas as they emerged
successively in the dialectical development. Now it is left to the history of philosophy to
demonstrate in greater detail the extent to which the unfolding of its contents that takes
place in it agrees with the dialectical unfolding of the pure, logical Idea, on the one hand,
and diverges from it, on the other. All that needs to be mentioned here is that the be-
ginning of the logic is the same as the beginning of the history of philosophy proper. […]
This is to be regarded as the proper beginning of philosophy because philosophy is,
generally speaking, a process of knowing by way of thinking [denkendes Erkennen], but here
for the first time pure thinking has been taken hold of and become an object [gegenständlich]
for itself. (Hegel, 2010, 138)
Philosophy and its history, for Hegel, constitute a kind of coming to self-consciousness of the
Absolute Idea, to Spirit’s recognition of itself as itself, by way of abstract indeterminate Being taking
on determinate form such that, in the course of history, Being is concretely realized as a dynamic
process of “Becoming.” One consequence of this is that in Hegel’s view, philosophy cannot be
thought of as one particular, bounded mode of inquiry into what is. Philosophy, the study of “pure
thinking,” is at once also the study and the history of all that is and has come to be.
In an afterword to the second German edition of Capital, Marx described his relationship to
Hegel’s philosophical method in the following oft-cited terms:
“My dialectic method is not only different from the Hegelian, but is its direct opposite. To
Hegel, the life process of the human brain, i.e., the process of thinking, which, under the name of
‘the Idea,’ he even transforms into an independent subject, is the demiurgos of the real world, and
the real world is only the external, phenomenal form of ‘the Idea.’ With me, on the contrary, the
ideal is nothing else than the material world reflected by the human mind, and translated into forms
of thought.
The mystifying side of Hegelian dialectic I criticised nearly thirty years ago, at a time when
it was still the fashion. But just as I was working at the first volume of “Das Kapital,” it was
the good pleasure of the peevish, arrogant, mediocre Epigonoi [Epigones – Büchner,
Dühring and others] who now talk large in cultured Germany, to treat Hegel in the same
way as the brave Moses Mendelssohn in Lessing’s time treated Spinoza, i.e., as a “dead
dog.” I therefore openly avowed myself the pupil of that mighty thinker, and even here
and there, in the chapter on the theory of value, coquetted with the modes of expression
peculiar to him. The mystification which dialectic suffers in Hegel’s hands by no means
prevents him from being the first to present its general form of working in a compre-
hensive and conscious manner. With him it is standing on its head. It must be turned right
side up again, if you would discover the rational kernel within the mystical shell.”
(1996, 19)
Here – and hopefully we might be forgiven for also coquetting with a few Hegelian modes of
expression – Marx insists that Hegel’s philosophical idealism ought not to be simply discarded
without care as some tired old entry in a “gallery of errors,” but rather merits a treatment more
46
closely befitting an honored member in the “pantheon of divine figures” that represent stages of
dialectical development. What, in Marx’s view, is mystical in Hegel and what is rational?
It is worth noting that in drawing a distinction between Hegel’s method and his own, Marx
describes each in somewhat exaggeratedly stark terms. Concrete materiality for Hegel is not only or
merely the external form of Spirit, as though it were a kind of passive, inert shadow of some realm of
Forms. What Marx calls the “real world” is, in Hegel, also simultaneously a necessary condition for
the possibility of Spirit becoming truly realized not only as the abstract concept of “Being” but as
what Hegel calls “existence.” “Existence” for Hegel is a kind of unfolding, deepening, and coming
to fruition of the inner content of “Being,” which comes about only in the course of historical
development.
Marx, conversely, regards it as “mystical” to take up abstract Being as the chief object to be
analyzed; “mystical” to concern oneself chiefly with the careers of concepts as though their devel-
opment were not themselves entirely determined by the histories of living human beings interacting
with one another and their physical environment. Against this, he and Engels insist in The German
Ideology that
The premises from which we begin are not arbitrary ones, not dogmas, but real premises
from which abstraction can only be made in the imagination. They are the real individuals,
their activity and the material conditions of their life, both those which they find already
existing and those produced by their activity. These premises can thus be verified in a
purely empirical way. The first premise of all human history is, of course, the existence of
living human individuals. (1976a, 31)
This statement captures the key lesson of Marx’s and Engels’s historical materialism: that human life
is a product created by human beings intervening into their natural and social environment, under
definite natural and social circumstances, to satisfy their needs. Thus, to understand thought itself,
much less particular concepts and ideas, one must first proceed by investigating the practical human
activity that gives rise to it, and the particular human needs and conflicts to which it gives ex-
pression. To attempt to make sense of abstract ideas without thinking of them as conceptual ex-
pressions of concrete human circumstances is, for Marx, always a nonstarter. He embraces dialectics,
in the form of a Hegelian emphasis on process, dynamism, conflict, and history, and especially
Hegel’s insistence that one cannot truly know a thing until one has observed its development over
time. However, where Hegel tends to apply these to the conceptual realm, Marx’s method is to
apply these to the study of human beings to explain their development as concrete, material beings
living in, manipulating, and theorizing their environment.
Similarly, Marx’s criticism of Hegel’s followers, whom he generally regards as pale reflections of
their hero, is not that they do philosophy; it’s that they don’t do anything else. He contends that this
intellectual narrowness renders them unable to make real sense of their own dogmas and debates or
of the puzzles and theoretical culs-de-sac into which they are led. He charges, “It has not occurred to
any one of these philosophers to inquire into the connection of German philosophy with German
reality, the connection of their criticism with their own material surroundings” (1976a, 30).
In other words, Marx’s critique is that these philosophers have mired themselves in what is
“mystical” in Hegel – his staunch philosophical idealism, his emphasis on the ideal, conceptual, and
abstract as the most fundamental and determining aspects of existence. Yet, Marx argues, they have
failed to embrace Hegel’s emphasis on movement, dynamism, process, activity, and interaction – in
a word, history – as the starting point for real insight into the nature of what is. Once one does this –
salvages the “rational kernel” and sets the dialectic on its feet, upright – only then is one truly
prepared to address the problems of philosophy. And these are themselves revealed to be none other
47
Vanessa Wills
than the problems of human beings seeking to satisfy their needs and realize their aims through
interaction with their natural and social environment.
In this sense, Marx’s turn towards increasingly empirically informed analysis in his works fol-
lowing The German Ideology is not a wholesale refusal of philosophy and philosophical questions as
such, but rather an insistence that such questions need not remain abstract intellectual exercises. His
turn towards a growing focus on economics in his later works is the logical conclusion of his views
regarding how it is that philosophical questions and the conditions of their emergence could actually
be comprehended and solved. Pure, abstract philosophical contemplation of abstract concepts, in
isolation from apprehension of the empirical, concrete world in which human beings develop their
conceptual frameworks, cannot even answer the philosophical questions that it sets before itself. For
that, one needs a method that can analyze how these questions emerge in the course of history, out
of definite circumstances that arise at particular historical moments as the practical problems of living
human beings.
Let’s return to the question that set us upon this inquiry into Marx’s relationship to Hegel; “what
is Marx’s relationship to philosophy?” We observed that one reason often given for arguing that
Marx abandoned philosophy is that in his “mature” works, he seemed to make a turn towards
empirical economic study and away from the kinds of abstract philosophical questions that animated,
for example, his doctoral thesis. But to characterize this as a “rejection” of philosophy is so overly
simplistic as to be mistaken. Marx’s critique is rather a specific denial of the claim that philosophy on
its own offers a method capable of allowing us to understand the nature of what is. It is most
certainly not a claim that some other discrete academic discipline can take over what seemed to be
the role of philosophy. Significantly, it is not economics, but history Marx and Engels name as
the single science that can make sense of the relationship between the social and the natural, and
between the material and the ideal. This is not a simple replacement of one discipline of inquiry as
we know it with another, but the supersession of both and their subsumption within one novel,
general, and unifying theory of all change and development.
More on this later, but for now, let us turn toward the role of the “proletarian standpoint,” and
its relationship to particularity and universality, and to subjectivity and objectivity, in Marxist theory.
3 “Picking Sides”: Marx and the Proletarian Standpoint

We have discussed already that in addition to the economic turn in Marx’s later work, there is a
second reason one might doubt that Marxist theory speaks to the concerns of philosophy, much less
that Marxism is “scientific,” as it purports to be. That reason is that Marxism seems to pick a political
side in a way that one might argue impugns its claim to scientificity.
At the outset of the Communist Manifesto, Marx and Engels claim that “The history of all hitherto
existing society is the history of class struggles” (1976b, 482). The liberal state, they explain, is led by
an executive branch that “is but a committee for managing the common affairs of the whole
bourgeoisie.” While the state touts itself as a neutral, democratic institution, in fact, it functions as a
political weapon in the hands of a ruling capitalist class that wields the state to establish and maintain
its dominance over its national proletariat. While this power relation is obscured by liberal theory
with its abstract assurances of freedom, equality, and individual rights, Marx’s and Engels’s materialist
conception of history reveals the class character of the liberal state; only, not from some point of
imagined neutrality and indifference. For Marx, it is only when we analyze the political history of
liberal states from the specific point of view of their working classes, and only when we set aside
abstract liberal pieties to observe how these states function not in word but in deed, that we gain
insight into the real character of civil society, stripped of all bourgeois mystification.
In this way, Marxist theory does not stop at simply denying that bourgeois ideology presents a
correct, objective view of the world it is about. Marxism further denies the idea that within class
48
society there are anything but perspectives thoroughly shaped by class position and by each class’s
relationship to the central class struggle that drives the historical development of that society at that
moment. But if this is true, then what lends the working class’s point of view any greater validity or
claim to objectivity than the perspective of the bourgeoisie? How can Marx claim to have unveiled
the “real” character of bourgeois institutions by applying a proletarian perspective?
As if to drive the point home even further, in his 1880 Socialism: Utopian and Scientific, Engels
coined the term, “scientific socialism,” which he described as “the theoretical expression of the
proletarian movement” (1989, 326). There he also wrote, “These two great discoveries, the ma-
terialistic conception of history and the revelation of the secret of capitalistic production through
surplus-value, we owe to Marx. With these discoveries Socialism became a science” (1989, 305).
That is to say, in Marxist thought, socialism is not merely an ethical ideal, but a serious theoretical
and practical method for bringing about human emancipation.
We see, in Marx’s and Engels’s descriptions of the materialist conception of history, what might
seem to be two competing claims: on the one hand, that historical materialism is specifically a theory
of working-class revolution and on the other, that historical materialism is a science offering true and
universally valid insights into the objective character of natural and social human existence.
The key here, as is so often the case when working to make sense of Marx, is to keep in view the
centrality of history and dialectical movement in Marxist theory. Marx saw himself as superseding
what he took to be Hegel’s conceptualization of the universal totality of existence as an Absolute
Idea – an abstraction playing the role of demiurge in history. Universality, for Marx, is not already
given – even in a highly abstract sense – as a fully realized fact of human existence. Nor (contra Kant’s
Categorical Imperative to act as though one’s individual will had validity as a universal maxim) does
he satisfy himself with an ethics of behaving as though it were universal.
If we may flirt again with Hegelian modes of expression, in a class society, the human species and
a human perspective both has and does not have universality. It exists in a state of becoming.
Ushering it into full realization is a human task yet to be completed. That task, Marx argues, falls to
the working class. The proletariat’s class interests lead them to wage political struggles that have the
potential to liberate all humanity, abolishing humans’ alienation from their own productive and
transformative potential. From the proletariat’s epistemic standpoint, and in their political and
economic class interest, antagonistic particularities dividing humanity against itself can be overcome.
In Marx’s vision, this would occur not only as a kind of intellectual exercise, spiritual longing, or
philosophical abstraction but also as the working class’s real historical achievement in producing a
society in which “In place of the old bourgeois society, with its classes and class antagonisms, we
shall have an association, in which the free development of each is the condition for the free
development of all” (1976b, 506).
Marx earlier made a similar point in The Holy Family, with characteristic poetry. There, he
described the proletariat as
a class with radical chains, a class of civil society which is not a class of civil society, an
estate which is the dissolution of all estates, a sphere which has a universal character by its
universal suffering and claims no particular right because no particular wrong but wrong
generally is perpetrated against it; which can no longer invoke a historical but only a
human title; which does not stand in any one-sided antithesis to the consequences but in an
all-round antithesis to the premises of the German state; a sphere, finally, which cannot
emancipate itself without emancipating itself from all other spheres of society and thereby
emancipating all other spheres of society, which, in a word, is the complete loss of man and
hence can win itself only through the complete rewinning of man. This dissolution of
society as a particular estate is the proletariat. (1975, 186)
49
Vanessa Wills
The movement from fracture to wholeness, from particularity to universality, is something that must
be eventually produced; and that product will be the result of a political project accomplished by the
proletariat as a revolutionary subject in the course of human history, seeking emancipation from its
own exploited, alienated, and degraded condition. In the Communist Manifesto, Marx and Engels
refer to the proletariat as “the class that holds the future in its hands” (1976b, 494). It is the pro-
letariat’s specific capacity to dissolve social antagonisms and produce a society in which the flour-
ishing of each conduces to the flourishing of all, that lends the character of universality to its
perspective. In the proletariat’s historical task, the opposition of particular and universal interest is
not merely theoretically and philosophically, but practically and politically overcome.
So Marxism has the following to say in response to the well-worn canard that it is problematically
“ideological” because it picks a side in the class struggle and is for that reason inherently “interested”
and unscientific. Ironically, Marx counters, to claim impartiality with respect to class interest – to fail
to adopt the proletariat’s class standpoint and fail to engage in the pursuit of knowledge as a the-
oretical expression of the proletarian movement which seeks to reconcile social antagonisms and
bring about true human emancipation – is to flee real, concretized universality and embrace mere
partiality and subjectivity draped in universalism’s illusory shadow. It is only in reconciling social
antagonisms and producing ourselves as a species that is aware of itself as having conditions of
flourishing that belong to us all jointly as members of a single species, that human beings can
produce the kind of perspective on themselves and on their relationship to nature that is a pre-
condition for fully scientific knowledge of what is.
To return to Marx’s 11th thesis on Feuerbach, it is not that we must cease interpreting the world in
order to start changing it. Rather, understanding the world and transforming it are joined together in a
single unity of praxis. To characterize Marx as discarding the concerns of philosophy altogether in favor
of purely economistic explanations is mistaken, as is the assumption that because Marxist theory picks a
side in political struggle, it is not also committed to objectivity and universality as epistemic values. The
first error, as we already saw in our discussion of Marx’s relationship to Hegel, is a consequence of
disregarding Marx’s insistence that all areas of disciplinary study are to be superseded by and subsumed
within “a single science of history.” The second is to think that political science is best done by
dispassionate observers and not as a way of theorizing lived, active struggle in the first-person, as it
were, and not consciously in order to seek the victory of one of society’s contending classes, whose
historical task makes a really objective and universal perspective possible.
4 Conclusion
While Marx of course never described himself as pursuing a program of “Philosophy, Politics, and
Economics” – and indeed it would be incorrect and quite misleading to characterize his intellectual
project in exactly this way – we can draw some key connections, as his program is more like an
integrated PPE approach than it is a wholesale rejection of philosophy and replacement of it with
politics and economics.
We launched this discussion with the question, “What relationships emerge among philosophy,
politics, and economics in Marxist thought?” A completely thorough answer well outstrips the con-
straints of the present space. However, a key theme of Marx’s materialist conception of history is the
insight that all of existence moves in a state of restless flow and is knowable, not in static isolation or
abstraction, but only through apprehension of its motion and processual development. This is the lesson,
for example, of Capital, where Marx describes his analytical standpoint as one “from which the evolution
of the economic formation of society is viewed as a process of natural history” (1996, 10).
Moreover, for we as humans to attain knowledge of what is, we must not attempt (vainly, one
might add) to stand apart from the world as impartial observers of it. There is no true apprehension
of what is without acting to bring about what we wish for there to be – this insight is the secret of
50
labour, of conscious production, as the essential human activity. In our attempt to know the world,
to attain an objective and universally valid consciousness of it, we must intervene in it. We must
thereby come to know, produce, and reproduce ourselves – that is, ourselves as a human species – as
an active, self-determining, self-realizing, collective agent of historical change. Insofar as capitalism
frustrates this process of conscious self-determination at the level of the species, scientific knowledge
of the world is therefore an inherently political, contentious, and liberatory project. It requires for its
success the emergence of real political and economic democracy, so that there may be a resolution of
the alienating antagonism between those who carry out socially necessary productive labor and those
who direct it. It is only through such consciously transformative activity that we can truly come to
realize ourselves, and know ourselves, within the greater unifying flow of history in which we are
enmeshed as its conscious element. From the standpoint of historical materialism, to bring into
existence, and come into possession of, our fully realized productive powers would be to practically
abolish the historically contingent contradiction between subject and object and make scientific
knowledge of the world truly possible.
Obviously, the vast majority of scholars working in PPE today do not think of themselves as
historical materialists – far from it! And my primary aim here is not to produce converts although of
course, they should know they are welcome. Investigating Marx’s relationship to the concerns of
PPE is essential because it is part and parcel of the PPE approach to agree that inquiry in philosophy,
politics, or economics is needlessly hamstrung when scholars in these fields attempt to resolve the
core questions of their discipline without availing themselves of the resources and methods of the
other fields. Moreover, scholars in PPE tend to agree that theoretical inquiry is most successful when
it is informed by practical efforts to bring about a better world, and most valuable when it can
inform such efforts in turn. For PPE scholars, there is significantly much to gain from a deep, critical
engagement with Marx’s writings and theoretical system, and much to be lost from overlooking
Marx’s efforts to synthesize multiple modalities of human inquiry into a single science of historical
development and change.
Acknowledgments
Thank you to Liam K. Bright for your comments on an earlier draft. I also wish to thank Olúfẹ́mi
O. Táíwò and Brandon Hogan for our many riveting discussions and debates about Hegel over the
summer of 2021.
References
Hegel, G. W. F. (1991) Hegel: Elements of the Philosophy of Right. New York; Cambridge University Press.
Marx, Karl, and Friedrich Engels. (1975) Marx and Engels: Collected Works, Vol. 3: Karl Marx March 1843–August
1844. New York; International Publishers.
Marx, Karl, and Friedrich Engels. (1976a) Marx and Engels: Collected Works, Vol. 5: Marx and Engels, 1845–47.
New York; International Publishers.
Marx, Karl, and Friedrich Engels. (1976b) Marx and Engels: Collected Works, Vol. 6: Marx and Engels, 1845–48.
Marx, Karl, and Friedrich Engels. (1989) Marx and Engels: Collected Works, Vol. 24: Marx and Engels,
1874–1883. New York; International Publishers.
Marx, Karl, and Friedrich Engels. (1996) Marx and Engels: Collected Works, Vol. 35: Marx, Capital Volume 1.
White, Bethany, Samuel Wainwright, and Lilly Schreiter. (2020) “One Hundred Years of PPE: 1920–2020.”
https://www.humanities.ox.ac.uk/files/ppe100yearsreportpdf (Accessed June 18, 2021).
51
5
FEMINIST THEORY
Ann E. Cudd
1 Introduction
Feminist theory uses gender as a lens through which the social world can be examined and critiqued.
Feminist theory makes descriptive claims, such as that gender is socially constructed, variable, and
ubiquitous in human life, that sex and gender are different, non-overlapping categories, and that
women and men live segregated lives, and have unequal access to power, income, and wealth. Feminist
theory is not focused entirely on gender or sex, however, because these categories cut across other
social groups and identities and affect those observations. This intersectionality of identities constructs
different categories of genders, though “women” and “men” remain coherent cluster concepts.
Feminist theory is also self-consciously normative. Although feminist thinkers do not endorse the
same claims or share the same reasons, feminist theory generally supports normative claims about
power, value, well-being, and the comparative access of men, women, and non-binary persons
across intersecting identities. For example, claims such as:
• Gender is essentially a social hierarchy in which men dominate women.

• Women are oppressed, though not equally or in the same ways, throughout history and across
societies.
• Women are not equal to men and are oppressed as women, even more so as members of
dominated race, ethnicity, class, disability, sexual orientation groups.
Feminist theory explores what justice and freedom, which have never existed universally, would
look like and how we can get to those ends.
Feminist theory in the interdisciplinary study that is Philosophy, Politics, and Economics (PPE) as
well as the separate disciplines that make it up, is a tool for critiquing specific social norms or
economic systems. Feminist theory contributions to PPE are practically boundless when understood
as feminist political theory or feminist philosophy or feminist economics, as feminism has penetrated
all the main topics in each of the disciplines. This essay focuses on what I see as the feminist issues
central to the intersection of the three disciplines of PPE. In particular, I will explore how feminist
theory answers the following questions: what does it mean to say that women are oppressed as
women? How does the gendered division of labor and specifically the feminine gender of caring labor
enact women’s oppression? How are women socialized/coerced into choosing caring labor? Finally,
are women oppressed by capitalism?
52 DOI: 10.4324/9780367808983-6
Feminist Theory
2 Gender Inequality and the Oppression of Women

In every place and time in human history, men and women live and have lived differently even
while existing together in kin groups and in society. The gender divide structures and cuts across
nearly every aspect of life and society, from family and work to religion, education, and health. This
divide not only determines different outcomes from decisions and behaviors, but those outcomes
also unequally empower men and women to make and execute decisions over their lives and that of
others and their communities. Feminist theory aims to establish that not only are these significant
differences or inequalities but that women constitute an oppressed social group. That is, that women
are oppressed as women.
In order to establish this claim, we need to first consider what it means to say that some group is
oppressed and what it means to be a member of the group women. Feminist theory offers multiple theories
or accounts of social oppression, its causes, and consequences as well as different theories of what a group
is and what makes women a group. Feminist theory claims that the oppression of women is not only
enacted in intentional ways but also unconsciously and implicitly through social norms that often go
unrecognized and undiscovered. Feminist theory thus aims to provide a critical lens for discovering
harms and their silently systematic nature in the everyday social norms that we take for granted.
Marilyn Frye’s classic essay “Oppression” (Frye, 1983) defines oppression as a system of interrelated
barriers and forces which reduce, immobilize, and mold people who belong to a certain group, and
subordinate them to another group. Women, she claims, are oppressed and subordinated to men. Frye
also posed an unforgettable metaphor that serves the feminist aim of accounting for the often silent or
invisible nature of the oppression of women. She likened oppressive forces to the wires of a birdcage,
which viewed as a whole was readily seen as confining the bird to a small space, but when any one of
the wires is viewed up close and without any of the connecting bands it would seem to be easily flown
around. Likewise, the wires of women’s birdcages, the stereotypes or normative expectations of
feminine behavior, would seem to be relatively minor taken one by one. A similar metaphorical
description of women’s oppression is as a ton of feathers – each feather or sexist norm a seemingly light
load, but when weighed overall the aggregation is overwhelmingly heavy.
In Analyzing Oppression (Cudd, 2006), I elaborated on Frye’s definition to give a theory of
oppression that may be used, in the first instance, to establish that a social group suffers the social
injustice of oppression. Because feminism makes a normative claim about gender injustice and lack
of freedom, it is important to have an account of oppression that enables us to identify the actual
cases of oppression: they have to be unjust not simply bad conditions or differences in life outcomes
that are due entirely to accidental, natural, or avoidable, self-caused harms.
In my theory, there are four conditions that must be met for us to claim that a social group is
oppressed. First, there must be some sort of harm that is inflicted on or suffered by persons who belong
to the group, which I call the harm condition. Second, the group must be an identifiable, non-accidental
social group, whose identity can be described apart from the harm that they suffer. This condition, the
social group condition, eliminates from consideration as oppressed a set of people who just happened to
suffer harm due to an accidental coming together, such as being on the same airplane when it crashes.
Third, to establish that the harm is unjustly inflicted, the harm must satisfy the coercion condition, namely
that there is some kind of coercion involved in how the individuals in the group are subjected to the
harm. The harm cannot be self-inflicted in the face of other, coercion-free, harmless alternatives that
the individuals of the group can freely choose. Coercion can take the form of a forced choice between
two bad options, such as when women are talked over in meetings by men and are then forced to
either appear aggressive and unfeminine or to acquiesce to being marginalized and their interests or
ideas discounted. Finally, the privilege condition requires there to be a correlative social group that
benefits in some ways from the harms inflicted on the oppressed group. This creates the inevitable
subordination of one group to another that is a hallmark of social oppression.
53
Ann E. Cudd
With the four-condition model, we can ask whether any group is oppressed by examining
whether the group faces each of the conditions. In order to focus on the claim that women are
oppressed as women, I will start by showing how women form a social group. While that may seem
obvious, it turns out to be complicated and controversial to define what a gender group is. Yet,
feminist theory can’t really get off the ground without some basic notion of what constitutes the
groups “women,” “men,” and other gender groups. Furthermore, one cannot deny the reality of
gender – everything from public bathrooms to Olympic track events are segregated and policed
according to a binary gender system that divides men from women. This is not to condone or justify
gender distinctions, but simply to observe their social power.
Without going into all the complexities in this brief essay, I offer an externalist model for social
groups that is particularly appropriate for exploring oppression. On this account, a social group is a
collection of persons who share (or would share under similar circumstances) a set of social constraints
on action, whereby “constraint” I mean either rewards or penalties consequent on actions. On this
account, one belongs to the group women if one is treated as a woman in at least some circumstances.
This externalist account thus distributes the identification of members of the group to both the person
(who may or may not self-identify with the group women) and to others (who treat the person as if
they were part of that group). Women are a social group precisely because there are characteristic
constraints placed on a collection of persons. Self-identifying as or being identified by others as a
woman offers a set of common opportunities and constraints and consequences for acting on them.
Regardless of how they characterize gender as a social group, feminist theorists take pains to show
that gender is socially constructed and not simply natural or unchangeable. Simone de Beauvoir was
perhaps the first feminist philosopher to note this when she wrote that “one is not born, but rather
becomes, a woman” (Beauvoir, 1973: 301). While the biological sexual characteristics that one is
born with typically determine how one will be gendered, gender is much more than a merely
observational fact of bodily form. How one is gendered (first by others and later by oneself) de-
termines, along with other salient social group memberships, how one will be raised, talked to,
educated, what social roles will be open to one, how one will be included or excluded from other
social groups, activities, and the like. And these facts much more than one’s biological sex determine
one’s life outcomes and outlook.
Among the characteristic constraints on women are the harms they face, such as inequalities in
power, wealth, income, and well-being when compared with men who share other social group
memberships. One might object that such inequalities vary widely within the group “women” by
other aspects of their identities. Black women in the US face much higher rates of maternal mortality
than White women, for example. While that is true, there are commonalities across race and
ethnicity groups for persons who identify as or are identified by others as women. Men are not
denied the ability to control their reproductive lives in the way that women of all races and eth-
nicities can be and are on average. Furthermore, the inequalities that exist among women are for the
most part simply amplified when comparing the prospects between women and men, particularly
between women and men of the same racial or ethnic group. There is a gender wage gap that favors
Black men over women just as it does White men over women, for instance. While this account of
social groups does not capture all of the interesting features of gender difference, it does allow us to
argue that women are oppressed as women.
In her influential book, Justice and the Politics of Difference, Iris Marion Young describes a set of
related but irreducible, socially systemic harms that characterize oppression. Oppression “names vast
and deep injustices some groups suffer as consequence of frequently unconscious assumptions and
reactions of well-meaning people in ordinary interactions… of everyday life” (Young, 1990: 41).
Young offers five categories of oppressive harms – the “five faces” of oppression – and claims that
with these categories we can describe the oppression of any group. These five faces are exploitation,
marginalization, powerlessness, cultural imperialism, and violence. These harms are jointly necessary
54
Feminist Theory
conditions for oppression for Young; if a group does not suffer from any of these categories of harm,
then it cannot claim to be oppressed. Women can be seen to suffer from each of these categories of
oppressive harm.
The first three of the categories are facets of the way capitalism structures modern society and are
matters of concrete power differences. A group is exploited if its labor is systematically used for the
benefit of another group through a process that maintains the power of the benefitted group to
continue to benefit from the labor of the oppressed. While exploitation was conceived by Marx as a
primarily economic matter, Young’s definition supports the feminist argument that women’s sexual
and reproductive labor is exploited by men. In the next section, this essay will explore how caring
labor oppresses women. Exploitation confers status benefits, as well as wealth and income benefits,
on groups who benefit from the exploited group’s oppression.
Marginalization occurs when a social group is rendered invisible by the economic structure of
society. In a capitalist economy, both capitalists and workers have status, even if it is unequal, but
one who neither works nor owns capital is completely marginalized. Because marginalization entails
invisibility, groups who suffer from this harm are in danger of being completely neglected or even
exterminated. This is not a ubiquitous condition for women, but it certainly exists such as in
Afghanistan under the Taliban or for women of certain castes in India.
Young’s third category, powerlessness, afflicts all dominated groups to some degree, including
women. This can be seen in the degree to which men are predominant in political and economic
institutions all over the world. Political rank and wealth are two of the most important forms of
power, which enables persons to benefit themselves and their fellow social group members in all
kinds of ways. Men also dominate other forms of power, such as religious status and professional
status. Women are often said to be dominant in household decision-making, but public power
inevitably trumps private power, as can be seen in the unequal outcomes for men and women when
private, familial relationships fracture. This public inequality leads to women having less bargaining
power in private family life (Okin, 1989).
Cultural imperialism exists when a dominant group establishes its experiences, tastes, and beliefs
as norms for all. Anyone for whom these norms do not obtain is seen as the Other by the dominant
group. The dominated group is forced to learn the dominant group norms as well as their own so
that they can codeswitch when interacting across groups. Cultural imperialism is not only a form of
disrespect, it can be even more damaging if the dominant group forcefully represses alternative forms
of cultural expression, as happened to Jews in Nazi Germany, or to gender and sexual minorities in
nearly every society. Although women exist in both dominated and dominant cultural groups,
women suffer a related harm, namely androcentrism. Men’s experiences and male bodies are often
taken to be the norm against which women’s experiences are seen as unusual and female bodies
inferior. Women are forced to conform to masculine norms if they wish to access power and equal
opportunities in public life. Yet, most women cannot perfectly emulate masculine norms for ap-
pearance, of course, and there arise various feminine versions of norms – some that attempt to
emulate masculine norms as closely as possible and some that attempt to please masculine sensibilities
about appropriate or attractive femininity – that make for ambiguities about the standards that
women are supposed to meet.
The final category of oppressive harm – violence – is perhaps the most obvious form. It is the
most directly damaging of the harms and can bring about the others. Oppressive violence is sys-
tematic because it disproportionately targets members of certain social groups, even as it may
randomly target individuals within those groups. Women are disproportionately victims of intimate
partner violence and are far more likely to be victims of violence by men than the other way around.
Generally speaking, women who are murdered are likely to have been murdered by their family
members, as physical violence is a common way that men maintain power and domination over
women and keep other men away from “their” women. Honor killings are a type of murder where
55
Ann E. Cudd
women are executed by their male family members because the women are seen to have damaged
the males’ honor, even by being the victim of rape or incest.
Women thus suffer harm consequent on their belonging to the social group of women. They are
exploited for their unpaid domestic and reproductive labor, rendered powerless by their lack of
status relative to men who are otherwise similarly socially situated with respect to their social group
status, they suffer from androcentric assumptions and cultural norms, and they suffer from systematic
gender-based violence. If we understand coercion as causing non-autonomous change in one’s
behavior, each of these kinds of harm are coercive and unjust. For example, because they face a
gender wage gap, women often are coerced into doing more than their fair share of unpaid domestic
labor. Their powerlessness causes them to change their goals and comes about because of an un-
merited lack of social recognition relative to men.
Androcentric norms, especially in public life, are enforced by men’s relative power and status,
which is enforced by invidious discrimination, bias, and violence. And the credible threat of gender-
based violence is practically by definition coercive. Violence can only be justified when it is a
defense against proportionately similar violence. Yet women are often sexually assaulted or violently
attacked just for being women. Women who are subjected to violence by family members and
intimate partners are typically forced to choose between the vulnerability to that violence or being
shunned (often along with their dependent children) by the only or main sources of social support
that their societies have made available to them.
Finally, women’s oppression meets the privilege condition in that men are benefitted from at least
some of the systematic harms that befall women. Men are relatively more powerful than otherwise
similarly socially situated women. There is a gender wage gap that favors men, and this in turn
benefits men in intrahousehold bargaining over who will perform various chores, make choices
about joint activities, and who will control the joint benefits of division of unpaid and paid labor in
and outside the home. Men have more political and economic power than women in the aggregate.
Men’s activities, performances, and accomplishments receive more recognition and reward. There
are ways in which women are better off than men in the aggregate. For example, women’s life
expectancy is longer than men’s. To the degree that this is socially controlled, however, it is largely a
result of what other men do to them and to each other in the pursuit of masculine power and
authority, and not what women do to men. Still, feminist theory makes the point that the oppression
of women is not good for anyone’s long-term interest, even individual men’s.
It is also important to note that many women have more power, status, and are freer from
systemic, group-based violence than many men. But it is not as women that these women are better
off than those men are as men. Rather, it is due to other intersecting social groups’ relative status and
power. Men from oppressed social groups can be significantly worse off on a variety of measures of
well-being or social advantage than women who are White, cis-gender, heterosexual, able-bodied
professionals. Feminist theories of oppression must recognize the intersection of oppressive forces as
it envisions a future in which women are free. Women are not free until all women are free of
oppression (Combahee River Collective, 1983), and since that would entail freedom from all other
forms of social oppression, all men too would be free in such a future.
3 The Gender of Caring Labor

This section explores caring labor as a fundamentally gendered social construct by means of which
women are exploited and otherwise oppressed. Feminist theory describes the gendered division of
labor and establishes that caring labor is gendered feminine. Caring labor is an essential kind of labor
for human survival, given our long periods of dependency, and for well-being, because of its im-
portance to our mental, physical, and emotional health. Yet, caring labor is poorly compensated in
contemporary society, largely because so much of it is provided without pay by women, making for
56
Feminist Theory
what has been called the “paradox of caring labor” (Folbre, 2003). Feminist theory reveals that
through the feminine gendering of caring labor, women are systematically placed in a position
where they are either exploited or forced to choose to behave non-normatively and face the social
penalties that come with such resistance.
We can characterize caring labor objectively by the kinds of activities involved, as the
International Labour Office (ILO) does. “Care work consists of two overlapping activities: direct,
personal and relational care activities, such as feeding a baby or nursing an ill partner; and indirect
care activities, such as cooking and cleaning” (International Labour Office, 2018: xxix). Using this
characterization, the ILO determines that women are the primary providers of unpaid caring labor,
performing 76.2% of the unpaid caring labor, or 3.2 times as much unpaid caring labor as men. And
while the division of caring labor differs between countries, in no country in the world do men do as
much unpaid care work as women. Even as women have entered the paid workforce in greater
numbers over the past few decades, women still perform almost two hours more unpaid care work
per day than do men.
Women’s unpaid care labor reduces their paid labor force participation and affects the type and
quality of jobs that women take. Women have less time to devote to paid work and are more likely
not to work for pay when they are mothers of small children, while men are more likely to work for
pay when they are fathers of small children. By the same token, women pay a penalty in their wages
when they have dependent children at home compared to childless persons, while men experience a
wage premium compared to all other groups of workers when they have dependent children
(International Labour Office, 2018: xxxiii).
Women also dominate the paid care labor force. The global care workforce, according to the
ILO definition (International Labour Office, 2018: 5–10), numbers 381 million workers, of which
249 million are women and 132 million are men (International Labour Office, 2018: 169). The paid
care workforce includes a heterogeneous group of occupations, from education to domestic work to
healthcare. Some of this work can be well paid (e.g., professional educators and health care workers)
but much of it is poorly compensated (e.g., domestic workers, attendants for the elderly and dis-
abled, uncredentialed daycare and healthcare workers, and paraeducators). In general, the more
women there are in the occupation, the more likely it is to be poorly paid (Bergmann, 2005: Ch. 6).
Caring is a distinct kind of labor that involves empathy and centers the needs of others, and
essentially involves one in a personal relationship between caregiver and the receiver of care. The
word “care,” as Elizabeth Brake (2017: 136) notes, “is ambiguous between affect and activity.” It
can refer to the attitudes or warm feelings of wishing someone well, or to the activity of providing
for the needs of another. It is mainly in the latter sense that I am going to talk about it here, but the
two senses cannot be entirely separated. The feminist economist Nancy Folbre builds in the atti-
tudinal aspect of care into the definition of caring labor when she defines it as “labor undertaken out
of affection or a sense of responsibility for other people, with no expectation of immediate pecuniary
reward” (Folbre, 2003: 214). This excludes labor that is done only for wages but includes labor that
is paid as long as the expectation of payment is not the only motivation for the work. Using this
definition of caring labor excludes some of the labor included by the ILO accounting (namely, that
which is done only for pay), and it makes the precise determination of when it is being performed
more difficult. However, this definition allows us to explain how the gender of caring labor makes it
a central force of women’s oppression, and as such is a critical component of the larger category.
Caring labor, when defined as Folbre does, is paradoxical for economists because it need not be
rewarded to be obtained and it cannot be bought. As the song goes: “money can’t buy me love.”
Furthermore, the more one has to pay for care the less one feels cared for, making it a kind of Giffen
good.1 Those who need care to thrive or even to survive often could not afford the care they need if
they or their loved ones had to pay for at a competitive wage. Caring labor is essential for sustaining
human life when persons are dependent due to age, disability, or physical health and cannot obtain
57
Ann E. Cudd
food and drink or tend to their needs for shelter and medical care. The feeling of being cared for,
which comes from caring labor, also plays an essential role in healthy neurophysiological and
emotional development of children (Alstott, 2004: 15–20). Caring is thus a central feature of the
unpaid household labour provided within families.
For there to be enough care for those who need it, then, caring labor must be (at least in part)
either given voluntarily or coercively taken from the caregiver. Paid and unpaid labor, as we have
seen, is much more often done by women. Caring attitudes indicate warmth, fellow feeling, a desire
to make the receiver of care happy, even love, infuse many kinds of affective, domestic, and health
or hygiene-related labor, and particularly unpaid household labor. The appearance of these attitudes
is in fact critical to the value of much caregiving labor and this appearance is nullified if the care
appears to be forced or even given only in exchange for wages. The fact that women do the majority
of care work appears to be a matter of women’s voluntary choices or preferences. And given its
relation to family life and especially to childrearing, this choice appears to be naturally associated
with the female sex and feminine gender. In this way, caring labor is gendered feminine. Women
appear to naturally prefer to do more than their equal share of unpaid caring labor and lower-paid
caring labor. But even if it were true that women are natural caregivers, if women are paid less and
expected to do more hours of labor because of their gendered preference or greater capacity for
caring work, this is problematic from a perspective of justice.
Feminist theory reveals that the gendered nature of caring labor as voluntarily chosen is socially
constructed and maintained through oppressive force. Susan Moller Okin’s (1989) analysis of unpaid
household labor shows how women’s oppression creates coercive conditions that enforce women’s
exploitation through household labor, which in turn reinforces those forces of oppression in a
vicious cycle. Since women face on average lower pay in the workforce (as well as discrimination
and harassment), they will on average bring home lower pay than their male partners. The power to
determine who will work outside the home for pay and who will do more of the unpaid household
labor within families is at least in part determined by who can earn more. Women’s greater share of
unpaid caregiving labor has also been examined by economists who study how heterosexual couples
decide to allocate household labor and consumption by modeling the decision situation as an in-
trahousehold bargain in which one partner faces structurally different outside options (Lundberg and
Pollak, 1996). Since men earn on average better outside options for compensation, women have less
power to negotiate their share of household labor and consumption.
As women have entered the paid workforce, they have gained opportunities to resist unpaid
caregiving. However, the COVID-19 pandemic highlighted the degree to which women are still
coerced into shouldering the burden of unpaid caregiving roles and how devastating that can be for
their careers and the livelihoods of them and their families. Because childcare is often provided for
“free” by women, it is poorly paid and therefore underprovided in the paid economy. When the
COVID-19 pandemic crisis hit, the US and many other countries did not act quickly to provide
childcare or keep schools open but rather relied on women to provide care for their families
(Albanesi and Kim, 2021). Women have been forced out of the workforce to care for children or
relatives at home at twice the rate of men (Bateman and Ross, 2020). This will further hamper
women’s abilities to compete in the workforce when normalcy returns, as they have had less time
than men to invest in their human capital for paid work. This not only engenders a wage penalty for
women, but it also continues the vicious cycle of household unpaid care being shouldered by
women because of their lower wages.
The gendering of caring labor reduces the options for and quality of paid care labor and coer-
cively requires women to give care. As a result, children get less and more uneven quality care and
women are not fully able to participate in the market or develop their talents in the directions they
would choose under conditions of freedom. Furthermore, it must be noted that the feminine
gendering of care is a bar to men’s freedom to choose unpaid caregiving roles. Different countries
58
Feminist Theory
have attempted to solve these problems with varying degrees of success (Morrissey, 2017; Waldfogel,
2001). De-gendering caregiving labor, which is necessary for justice and freedom, remains a long-
term grand social challenge.
4 Adaptive Preferences and Internalized Oppression

One of the reasons that the gender of care is so hard to undo is because girls and women are raised
and expected to be caregivers, that is, to prefer caregiving activities and to develop caring attitudes
toward others. This is not part of male socialization or expectation; indeed, boys are raised to expect
to be given care by women. In this way, women’s preferences are molded by early socialization and
training and reinforced throughout their lives by the expectations expressed by others that they will
prefer to perform caring labor and exhibit caring attitudes. This section explores the work of
feminists to show how such adaptive and molded preferences undermine women’s autonomy and
reinforce their oppression.
Preference is an important concept in all the disciplines of PPE. Philosophers are concerned with
characterizing agency and autonomy, and one way of describing free agency is as being able to act
rationally or deliberately on one’s authentic preferences. Political theorists appeal to preference in
describing and evaluating social choice mechanisms. And economists use the concept of preference
to refer to what motivates choice-making behavior. Many normative political and moral theories
assume that it is better, all other things equal, for people to get what they prefer. That claim is
defeasible in some cases, such as when preferences do not reflect persons’ all things considered
interests.
Adaptive preferences are preferences that adapt or adjust to the agent’s feasible options in a way
that we would find somehow psychologically or morally problematic. Jon Elster first identified
adaptive preferences in the context of discussing freedom and autonomy (Elster, 1983: 117).
Preferences that we are caused to have, but do not intentionally choose to form, because our
freedom has been constrained are, he reasoned, non-autonomous. For Elster, the paradigmatic
example of adaptive preference formation is the story of the fox who, when he found that he could
not reach the grapes that he wanted, changed his preference, declaring the grapes to be sour. In
Elster’s view, the characteristics of adaptive preferences that make them non-autonomous changes in
preference are (1) the change is in response to a (perceived) constriction of the set of states of affairs
that the agent could feasibly bring about, and (2) the change happens by an unconscious process that
happens “behind the back” of the agent. The nature of the process of change is more important than
the constriction of the feasible set; as long as one intentionally chooses to change one’s preferences,
then the change does not compromise autonomy.
Feminist concerns suggest another way to see how adaptive preferences can be criticized, and that
is by looking at their source in socialization. Many of our preferences come to us from our up-
bringing in families and in particular cultures. There is nothing particularly non-autonomous about
that fact, provided that we feel that we can change them if we so choose. The problematic pre-
ferences for feminists are preferences of the oppressed that are morally problematic for the oppressed.
Serene Khader characterizes what she calls inappropriately adaptive preferences as:
(1) preferences inconsistent with basic flourishing (2) that are formed under conditions
non-conducive to basic flourishing and (3) that we believe people might be persuaded to
transform upon normative scrutiny of their preferences and exposure to conditions more
conducive to flourishing. (Khader, 2011: 42)
This definition thus highlights the harmfulness of the preferences, as they are non-conducive to basic
flourishing, and the badness of the conditions under which they are formed. Khader resists saying
59
Ann E. Cudd
that the bad conditions cause the preferences in order to avoid suggesting that those who have such
preferences are mere victims of conditions.
We can agree that persons are not mere victims, yet still criticize the preferences, particularly
when they reinforce the oppression of others and not only oneself. Furthermore, if the preferences
are problematic, it is important to understand their causes and ways that they might be changed. The
most problematic are adaptive preferences that are caused by oppressive conditions and are for states
that reinforce those conditions. I have termed these “preferences that are adaptations to and for
oppression,” or PAOs for short (Cudd, 2015a). PAOs are held by oppressed persons, caused by
oppressive circumstances, and are for the conditions that constitute at least some aspect of the
person’s oppression.
An example of a PAO that undermines autonomy as Elster described is a woman who sincerely
prefers to remain with an abusive spouse when she has no safe or secure option to leave. In such a
case, her set of feasible options has been constrained by the oppressive circumstances for women that
make them vulnerable to domestic abuse. The set does not include a reasonable alternative, so she
reconciles her situation by saying to herself, “well, it’s not so bad after all.” By hypothesis, she
actually changes her preferences in light of her constrained circumstances.2 Then this PAO is an
adaptive preference and autonomy-undermining.
Not all PAOs undermine autonomy, yet they are still problematic. Sandrine Berges draws on the
work of Mary Wollstonecraft to distinguish between the fox and grapes type of adaptive preferences
that undermine autonomy and preferences that are “caused by a systematic stunting of the growth of
their rational capacities, and an excessive nurturing of [women’s] love for comfort and small
pleasures” (Berges, 2011: 87). This kind of preference is a PAO because it is adaptive to oppression,
but it is caused by an autonomous process of socialization. Yet, as Wollstonecraft argued, (class-
privileged) women of her time with such preferences would not change their circumstances even if
they were offered a chance to do so because they enjoy the small pleasures and comforts that come
with them. Their socialization stunted their capacities to the degree that they cannot appreciate a life
with more freedom and its attendant responsibilities. This stunting is at least problematic from a
perspective of human potential for creative or intellectual achievement and productivity, even if it
does not harm their happiness or well-being or that of anyone who rightfully depends on them.
The gendered preference for women to prefer performing the majority of the unpaid household
labor in a family with two adult heads is another example of a PAO that comes about as a result of
socialization. As discussed, it is far more common for women to perform the majority of unpaid care
in the family and the gender of caring labor is an integral aspect of women’s oppression. It seems
likely that women’s preferences (when they do indeed prefer it) for being the household specialist in
unpaid caring labor is caused by the way that girls are raised to care about caring and domestic work.
Their toys, games, and the activities they are encouraged or required to engage in teaching them
how to perform and enjoy caring labor. Their first mentors are their caregivers and unlike boys, girls
are encouraged to see their mothers as role models to be emulated. As they grow up, girls and
women are positively reinforced for expressing empathy and caring attitudes and negatively re-
inforced for showing tendencies to aggression or competition. These PAOs for caring labor that are
caused by socialization may not undermine the autonomy of one who authentically has them.
However, they are still problematic because of the way that they set women up for oppressive
harms, including exploitation, powerlessness, and violence.
Whether or not PAOs undermine autonomy by causing one to change preferences to adapt to
oppression, PAOs that are caused by gendered socialization are morally problematic because of their
role in reinforcing the oppression of others. When oppressed women internalize gendered norms
and act on their PAOs, they reinforce those norms and make the expectation that others around
them will also adhere to them stronger. Rosa Terlazzo writes, “Social norms are as resilient as they
are because persons comply with them, and when members of marginalized or dominated groups do
60
Feminist Theory
so, their preferences often contribute to that marginalization and oppression” (Terlazzo, 2016: 207).
Women thus become complicit with the oppression of women by enforcing and reinforcing norms
for caring labor as appropriate for girls and women.
5 Capitalism and Women’s Oppression

Feminist theory can help us to understand and critique capitalism and decide how and whether to
improve or reinvent it. Feminists critique capitalism on several grounds; I will discuss three criticisms
here. First, capitalism is a system in which workers are exploited, and women are among them. Since
women have even less access to capital than men do and bear a greater burden of reproductive labor,
women are even more likely to have to accept labor that is poorly paid, dangerous, or undignified
(Gimenez, 2005). Second, capitalism encourages the commodification of women’s bodies in sex
work and surrogacy, which in turn reinforces the objectification of women (Pateman, 1988). Third,
against a liberal ideology of meritocracy and legal equality, the inequalities inherent in capitalism
lead to internalized oppression of women and minorities who do not succeed in the competition for
wealth (Cudd and Holmstrom, 2011: 196). Whether capitalism can be reformed or must be
overthrown is a live question for feminists.
Marxist feminists, such as Martha Gimenez, view capitalism as a mode of production that forces
those who do not own capital to work for wages that are determined by the owners of capital, who
have to compete among each other to avoid themselves falling into the working class. Profit or
capital accumulation is derived from the surplus labor of workers. Women, who play the primary
role in reproduction, are exploited to an even greater extent because of their burden of unpaid
household labor.
The capitalist structural constraints affecting how propertyless men and women can make a
living and the likelihood they will be able to form stable unions are the material basis for
the structured inequality between men and women …. This network [of constraints] sets
limits to the opportunity structure of propertyless men and women, allocating women
primarily to the sphere of domestic/reproductive labor and only secondarily to paid (waged
or salaried) labor, thus establishing the objective basis for differences in their relative
economic, social, and political power. (Gimenez, 2005: 24)
Gimenez argues that the capitalist form of production and private property rights makes alternative
forms of reproduction all but impossible. On this view, capitalism is responsible for maintaining the
oppression of women at least in its current form.3
A related critique is leveled by Carole Pateman, who argues that the idea of property in persons
leads to the commodification of women’s bodies in capitalism and thus the normalization of
prostitution and surrogacy. (Pateman, 1988; Anderson, 1990) The idea of property in persons
underlies the Lockean justification of the right to unlimited ownership of property that is necessary
for capital accumulation. Capitalist property rights permit the alienation of property, and thus if one
has a property right in one’s own person, one has the right to alienate it. Of course, men can just as
easily alienate their persons as women. But it is not normative for women, that is, acceptably
feminine, to rent men’s bodies for sex, while it is normative for men, that is, considered masculine,
to rent women’s bodies for sex. Capitalism thus furthers women’s historic subjugation and de-
gradation by men. “When women’s bodies are on sale as commodities in the capitalist market, the
terms of the original contract cannot be forgotten; the law of male sex-right is publicly affirmed and
men gain public acknowledgement as women’s sexual masters” (Pateman, 1988: 208).
Inequality in income and wealth is inherent in capitalism because it permits and promotes
competition for the accumulation of wealth. It is not a fair competition, as persons start with
61
Ann E. Cudd
different endowments from their families and different talents and abilities. Parents with means can
hoard opportunities that all but guarantee their children’s success, as well. Furthermore, invidious
discrimination lowers the wages of women and minorities and also limits opportunities to develop
their skills and talents. Yet, capitalism is a system where one’s status at birth does not determine
whether one can own property or make contracts, and so it appears to be a meritocracy. Those who
are unable to succeed thus tend to blame themselves at least to some degree. For women and
minorities, this self-blame is a form of internalized oppression, in which they come to see themselves
and members of their group as less worthy or able in the competition for wealth (Cudd and
Holmstrom, 2011: 196).
Critiques of capitalism suggest either that changes within capitalism are needed or that an al-
ternative system would be the best way to improve the system. The foregoing feminist critiques each
take the form of arguing that capitalism is oppressive to women because it maintains, reinforces, or
amplifies pre-existing oppressions. While each critique raises significant concerns for women’s
freedom under capitalism, to argue that a different economic system should replace it would be to
show that despite the background starting point of women’s oppression, the system could overcome
whatever social forces have thus far maintained it.
Against such a revolutionary proposal, I have argued for reforms to limit and ultimately curtail
women’s oppression. I argue that the most effective, efficient, and justifiable form of capitalism is a
system in which there are non-discriminatory, legal protections of decentralized, private ownership
of resources, cooperative, social production for all citizens, and free and open, competitive markets
for exchange of goods, labor, services, and material and financial capital (Cudd and Holmstrom,
2011; Cudd, 2015b). Each phrase in this definition obviously requires explanation and justification,
which I do not have space for here. In my view, traditional patriarchy is the enemy of women’s
freedom and capitalism actually opposes oppressive gender socialization and provides options for
women to avoid unpaid labor (Cudd and Holmstrom, 2011: 116–130). Capitalism’s ability to
disrupt tradition and reward innovation aids women in their quest to escape unpaid domesticity and
gain power. Regardless of where one comes down on the question of reform or revolution, feminist
theory provides a much-needed lens for PPE to examine how gender works with and against the
underlying motivational and social forces of an economic system.
Notes
1 A Giffen good is a good that one consumes more of as the price rises and less as the price falls. The analogy I
am drawing with paid care is that if one has to pay for it, it will be less preferred than if it is freely given or you
are charged less for it out of a sense of affection. Given two caregivers: one who charges a lot of money because
they don’t have any affection for you but can work full-time and one who offers it at a low cost because they
do have affection for you but can only devote part-time, the latter could be the preferred option.
2 It is controversial in the domestic violence literature how common this kind of preference reversal is, and
even more so, what should be done about it. The best psychological or clinical approach to helping
someone out of danger is likely not to be to convince a person that their preferences or feelings are wrong,
but rather to offer safe and secure opportunities to separate from the abusive situation to observe non-
abusive situations that could be made feasible.
3 An even more recent take on Marxist feminist critique of capitalism is social reproduction theory
(Battacharya, 2017), which centers the reproductive labor of women in the understanding of capitalism as an
exploitative social system, not simply an economic or market-based system.
References
Albanesi, S. and Kim, J. (2021) “The Gendered Impact of the Covid-19 Recession on the US Labor Market,”
National Bureau of Economic Research, Working Paper 28505. http://www.nber.org/papers/w28505
accessed 13 March 2021.
62
Feminist Theory
Alstott, A. (2004) No Exit: What Parents Owe Their Children and What Society Owes Parents, Oxford; Oxford
University Press.
Anderson, E. (1990) “Is Women’s Labor a Commodity?” Philosophy and Public Affairs 19, 1, 71–92.
Bateman, N. and Ross, M. (2020) “Why Has COVID-19 Been Especially Harmful for Working Women?”
Washington, DC; Brookings Institution. https://www.brookings.edu/essay/why-has-covid-19-been-
especially-harmful-for-working-women/ accessed 21 February 2021.
Battacharya, T. (ed.) (2017) Social Reproduction Theory: Remapping Class, Recentering Oppression, London; Pluto
Press.
Beauvoir, S. (1973) The Second Sex, New York; Vintage Books.
Berges, S. (2011) “Why Women Hug Their Chains: Wollstonecraft and Adaptive Preferences,” Utilitas 23, 1,
72–87.
Bergmann, B. (2005) The Economic Emergence of Women, 2nd edition, New York; Palgrave Macmillan.
Brake, E. (2017) “Fair Care: Elder Care and Distributive Justice,” Politics, Philosophy & Economics 16, 2,
132–151.
Combahee River Collective. (1983) in Smith, B. ed. Home Girls, A Black Feminist Anthology, New York;
Kitchen Table: Women of Color Press.
Cudd, A. E. (2006) Analyzing Oppression, New York; Oxford University Press.
Cudd, A. E. (2015a) “Adaptations to Oppression: Preference, Autonomy, and Resistance,” in Oshana, M. (ed.)
Personal Autonomy and Social Oppression, New York; Routledge.
Cudd, A. E. (2015b) “Is Capitalism Good for Women?” Journal of Business Ethics 127, 4, 761–770.
Cudd, A. E. and Holmstrom, N. (2011) Capitalism For and Against: A Feminist Debate, New York; Cambridge.
Elster, J. (1983) Sour Grapes, New York; Cambridge University Press.
Folbre, N. (2003) “’Holding Hands at Midnight’ The Paradox of Caring Labor,” in Barker, D. and Kuiper, E.
eds. Toward a Feminist Philosophy of Economics, New York and London; Routledge.
Frye, M. (1983) “Oppression,” in The Politics of Reality, New York; Crossing Press.
Gimenez, M. (2005) “Capitalism and the Oppression of Women: Marx Revisited,” Science and Society 69, 1,
11–32.
International Labour Office. (2018) Care Work and Care Jobs for the Future of Decent Work, Geneva; International
Labour Office.
Khader, S. (2011) Adaptive Preferences and Women’s Empowerment, New York; Oxford.
Lundberg, S. and Pollak, R. A. (1996) “Bargaining and Distribution in Marriage,” Journal of Economic Perspectives
10, 4, 139–158.
Morrissey, T. W. (2017) “Child-Care and Parent Labor Force Participation: A Review of the Research
Literature,” Review of the Economics of the Household 15, 1–24. 10.1007/s11150-016-9331-3.
Okin, S. M. (1989) Justice, Gender, and the Family, New York; Basic Books.
Pateman, C. (1988) The Sexual Contract, Cambridge; Polity Press.
Terlazzo, R. (2016) “Conceptualizing Adaptive Preference Respectfully: An Indirectly Substantive Account,”
Journal of Political Philosophy 24, 2, 206–226.
Waldfogel, J. (2001) “International Policies Toward Parental Leave and Child Care,” The Future of Children 11,
1, 98–111.
Young, I. M. (1990) Justice and the Politics of Difference, Princeton; Princeton University Press.
63
PART II
Decision-Making
Introduction to Part II
Any issue in social science will be informed by an understanding of individual decision-making. For
the methodological individualist, we can best explain complex social structures as constituted from
individual choices. For the holist, individual choice is always influenced and contextualized by the
surrounding social structure. With either approach, good social science involves a view towards
individual decision-making.
PPE is especially helpful for studying decision-making because the typical models and assumptions of
one discipline often conflict with the models and the assumptions of another. Economists might focus
on the ways that individuals are responsive to incentives, whereas Philosophers focus on the principles
that seem to justify individual choices. Conflicts like this spark discussion and force reconciliation.
When trying to understand voting behavior and political action, for example, we need to choose one
approach or develop a mutually acceptable alternative. It is not obvious how we should do this.
To give a better sense of PPE’s approach to decision-making, Part II begins with a focus on
interdependent choices. Often, choices are made based on the expectations about the choices of
others. Game Theory has tried to model and understand the dynamics of such interdependent
choices, and it has been used widely across the social sciences and Philosophy. In his chapter, John
Thrasher introduces the core concepts of Game Theory and demonstrates its application to core
problems in PPE. Sometimes we are in conflict with others, sometimes we are trying to coordinate,
and sometimes we have conflicting preferences over how we coordinate. Game Theory models such
strategic choice-making to better understand the dynamics of interaction, whether the relevant
actors are individuals, businesses, states, or evolving phenotypes. Thrasher explains this by reviewing
the basic approach of game theory, its model of rationality, its challenges, its solution concepts, and
the dynamics of bargaining.
The second chapter of Part II scrutinizing a common assumption about choice behavior. If one
prefers mechanical pencils over wooden pencils, and one prefers pens over mechanical pencils, then
it seems like that person should prefer pens over wooden pencils. To do otherwise seems irrational.
In other words, preferences should be “transitive.” People often demonstrate the problem with
intransitive preferences through the example of a money-pump. I could charge a person 1/100 of a
cent to exchange their wooden pencil for a mechanical pencil, charge 1/100 of cent to exchange
their mechanical pencil for a pen, charge 1/100 of a cent to exchange their pen for a wooden pencil,
then start all over again. Soon, the person would have much less money and still be in the same
DOI: 10.4324/9780367808983-7
Decision-Making
cycle. Yet, there are cases when intransitive preferences do not seem so irrational. The existing
literature offers a litany of such examples, so one might conclude that intransitive preferences are
quite typical. However, Bovens aims to show that the extensive number of cases actually only track
four kinds of intransitive preferences; (i) Negligible-Value-Differences and Missing-Values Cycles;
(ii) Condorcet-Voting-Paradox Style Cycles (iii) Sen’s-Libertarian-Paradox Style Cycles; and
(iv) Sorites Cycles. Bovens explains the basic structural features that unite each kind.
The third chapter, by Sudeep Bathia, focuses on theories of choice behavior. The original model for
these theories was utility theory, which was intuitive and elegant but routinely failed to predict how
people would make choices in laboratory settings. To build a better model, behavioral economists
distinguished three different choice contexts; risky choice, multiattribute choice, and intertemporal
choice. Expected utility theory was the standard theory for risky choice, the weighted additive model
for multiattribute choice, and discounted utility for intertemporal choice. Because none of these seemed
to predict behavior well, rival theories like “the tally hueristic,” “prospect theory,” and “hyperbolic
discounting” were offered to better explain observed choices. As none of these were supported by a
wide consensus, a much broader menu of theories has emerged. Bhatia distinguishes these new theories
by how each treats certain basic choice variables. Specifically, Bhatia differentiates theories by the effect
that choice situations have on (a) interactions within options, (b) interactions between options,
(c) transformations of values, (d) transformations of weight, and (d) choice processes.
The central concern of the fourth chapter is the tendency of people to follow social rules. Such
rules constitute much of the social structure, and individual choices are meaningfully impacted by
the rules of their social group. In the fourth chapter, “Rule-Following,” Kimbrough and Wilson
review work on rule-following from three disciplines. First, they review the dominant theories of
rule-following behavior from economics and psychology. For each, they highlight the ways in
which challenges to one theory motivate other theories. Second, they offer two philosophical ac-
counts of what social rules are. By covering these different approaches, Kimbrough and Wilson
show the distinct ideas used to understand social rules and how these different approaches might
complement one another. They conclude with a criticism of the way that theories use “norms” and
“rules” interchangeably. They suggest that this tendency reveals a focus on rules as causing behavior
rather than identifying the reasons why people care about particular rules.
The final chapter of Part II emphasizes how unnoticed features of individual choice can have
broad social ramifications. The idea that people make perfectly rational decisions has, at this point,
been well exposed as a myth. There is clear evidence of biases that predictably lead people to make
decisions that are not rationally justified. To name a few, we recognize effects from confirmation
biases, framing effects, anchoring biases, attribution biases, and bandwagon effects. The phrase
“implicit bias,” specifically references those cognitive biases that unconsciously result in judgments
that harm members of already disadvantaged groups. Such unnoticed stereotypes and prejudices
affect the decisions of even those that work to advance social equality. In “Implicit Bias and
Decision-Making,” Lacy Davidson starts by explaining this bias and how it is typically studied. She
goes on to identify the clear effects it has on Education, Employment, the Medical System, and the
Legal System. Against the view that social structures are more important for explaining systematic
injustice than implicit bias, Davidson argues that implicit biases form a kind of “soft structure” that is
significant for explaining injustice. She ends the chapter by discussing potential ways to address and
change common implicit biases.
6
GAME THEORY
John Thrasher
1 Introduction
The formal study of strategic interaction known as game theory is one of the most important
elements in the PPE toolbox. It is no exaggeration to say that game theory is the fundamental logic
at the heart of the social and behavioral sciences, and it is crucial for understanding social and
political life. Game theory is the core of modern economic theory, formal models in political
science, and the most sophisticated and rigorous approaches to moral and political philosophy. Game
theory is also important in many of the natural sciences, especially in theoretical biology. As Herbert
Gintis (2009, xiii) puts it, “game theory is central to understanding the dynamics of life forms in
general, and humans in particular.” As he goes on to argue, “disciplines that slight game theory are
the worse – indeed much worse – for it” (Gintis, 2009, xiii). Because of this, any serious student of
PPE should have at least some knowledge of and capacity to use game theory.
Game theory is also essential is in experimental, behavioral studies. This area of economics,
political science, and even philosophy has exploded in the last several decades. One of the distinctive
aspects of behavioral experiments in economics and political science as opposed to experiments in
social psychology, for instance, is the use of game-theoretic modeling to identify and formalize
experimental hypotheses. There are a number of ways these models are used in experiments, but
game theory is crucial for experimental behavioral studies in economics and political science.1
Philosophers have also begun to make more use of game-theoretic modeling in their experimental
designs (e.g., Schmidtz, 1991; Bruner et al., 2018; Bicchieri, Xiao, and Muldoon, 2011).
Here I will outline the basic concepts of game theory and then show how game theory is used in
several different areas that are important to PPE, highlighting important applications.
2 The Problem of Strategy

Game theory is the logic of rational strategic choice. Rational choice in this sense is consistent
maximizing, which is merely to say that rational choice aims at choosing the most likely prospects to
one’s most favored outcomes. So, if your idea of a relaxing vacation is lounging on a beach in the
South Pacific and your choice is between a trip to Fiji and one to Alaska, the rational choice is to
choose Fiji. We might say that you “prefer” or “desire” to go to Fiji or that your “values” or
“interests” suggest that Fiji is the right choice, but all of this is imprecise and apt to lead to mis-
understanding. In the theory of rational choice that game theory uses, rationality is a normative
DOI: 10.4324/9780367808983-8 67
John Thrasher
standard for choices, which are over actions. These, in turn, are understood merely as paths or
prospects to outcomes. These possible outcomes can be ranked, and preferences are binary relations
over those outcomes.
So, if we think of my ordering of vacation outcomes as ranking “lounging on the beach” as the
highest and “trudging through artic landscapes” as the lowest, then we can say that I prefer being on
the beach to being in the artic for my vacation. I can’t simply choose to be on the beach, though, I
can only choose an action “fly to Fiji,” which is likely the best prospect to this outcome over “fly to
Alaska,” which is the best prospect to the other. It is still possible, after all, that my flight to Fiji
won’t make it there or that when I arrive, it is unseasonably cold and rainy such that I’m not able to
spend time at the beach after all.
The basic material, the building blocks of game theory are the ideas of choice, preference, actions,
and outcomes. It shares these building blocks with traditional decision theory. However, problems in
traditional decision theory – such as where to go on vacation – are problems that do not require us to
take the preferences and actions of others into account when we are choosing. Many economic
choices are like this. If I have a budget of $20 that I can spend on some bundle of goods, I will need
to determine the optimal mix of that bundle given my preferences and the constraint of my budget.
This is a problem of optimization with respect to constraints and it is at the heart of microeconomic
analysis. When what I can choose depends on what another chooses, though, I must now take into
account their likely choices when I consider my own and they must do the same. In these circum-
stances, we face a strategic choice, and we must incorporate the actions of another into our own
decision calculus.
So, if I am deciding whether to spend my day at the beach or hiking in the mountains, all I need
to do is consider which I want to do more. If I am trying to coordinate with a friend, though, I need
to take into account what they want to do as well. Let’s say two friends are indifferent between
going to the beach or hiking, but they want to do whatever they do together. We can represent
their choice in the matrix below (Figure 6.1).
Here Sam’s choice is represented by the rows, while Alex’s is represented by the columns. The
value that we see in each cell is the payoff of the outcome.
Payoffs are preferences represented as utility. Cardinal representations of utility can be generated
from ordinal preference orderings in several ways, but John von Neumann and Oskar Morgenstern
(1947) showed in their classic work that if choices are represented as lotteries it is possible, given a
number of plausible assumptions, to generate a representation of their preferences as a real number
function. The payoffs in game theory are understood to be utility in the von Neumann-
Morgenstern sense. Further, the payoffs are meant to include all the relevant considerations affecting
the choosers, which is to say that there are no relevant considerations outside the game; all the
payoffs are represented in the game. We also assume that all the payoffs are common knowledge to
everyone in the game.2
A game is defined by the players in the game, the strategies available to them, and the payoffs for
each strategy. A strategy is the choice or the action that the player takes. So, in the coordination game
both Sam and Alex have two strategies, either go to the beach or go on a hike. Each strategy has
two payoffs associated with it depending on what the other player chooses. So, given two strategies for
Alex
Beach Hike
Beach 1,1 0,0
Sam
Hike 0,0 1,1
Figure 6.1 Coordination game.
68
Game Theory
Alex
Beach Hike
Sam
beach hike beach hike
1, 1 0, 0 0, 0 1, 1
Figure 6.2 Extensive form coordination game.
each player, there are four possible outcomes with different payoffs. The table above represents those
outcome payoffs for each player. Representations of games like this are called normal form.
Games can also be represented as decision trees. This is called the extensive form representation of
the game. The same game as can be represented in extensive form (Figure 6.2).
In this representation, Alex makes the first move choosing either to go to the beach or to hike.
Sam chooses at the second node, but since the normal form game assumed that Sam made this
decision at the same time as Alex, Sam wouldn’t know which choice Alex made. This is represented
as a dotted line between decision nodes – an information set – which indicates that Sam doesn’t know
which of the two nodes obtains when Sam chooses. Because of this feature of the game, it doesn’t
matter whether Alex or Sam moves first in this game. But, in games where moves are sequential, it
may matter who moves first. In the same game where Sam knows what Alex chose, the game
becomes a form of “follow the leader” where the only rational option is for Sam to choose what
Alex chose. Alex, knowing this, will effectively choose for both of them. This means that in the
sequential version of the coordination game, it matters who chooses first, whereas it doesn’t in the
simultaneous game. The normal form of the game makes it difficult to represent this difference,
which is why extensive form games are important.
2.1 Rationality in Games

If we think of game theory as the logic of strategic rationality, we need an idea of rationality in
strategic contexts. In the non-strategic context, rational choice is just maximizing expected utility.
In the strategic context, though, the expected utility of a strategy depends on what strategies are
chosen by the other players. Rationality in a game can be generalized in the form of a solution
concept, which specifies the equilibrium strategies given various assumptions about the game and
rationality. One of the most basic solution concepts is built from the idea of dominant strategies. A
strongly dominated strategy is one that is strictly worse than some other available strategy, no matter
what the other player does. In the game below, we can imagine that Alex and Sam would rather be
at the beach alone than to hike, even though they would really like to go to the beach together. In
this trivial example, hiking is dominated by going to the beach since both would rather go to the
beach regardless of what the other chooses (Figure 6.3).
Alex
Beach Hike
Beach 2,2 1,0
Sam
Hike 0,1 0,0
Figure 6.3 Coordination game with strict dominance.
69
John Thrasher
If we eliminate all the strictly dominated strategies for each player, we are left with the set of
rationalizable strategies. For a strategy to be rationalizable, it needs to be a possible best response to
the strategy of another player given common knowledge of the game and common knowledge of
rationality (Bernheim, 1984; Pearce, 1984).
Rationalizability, as a solution concept, is a broader idea than the most commonly used so-
lution concept in game theory, the Nash Equilibrium. A Nash equilibrium is a strategy that
no one has a reason to unilaterally deviate from given the strategies of everyone else. We
can formalize a normal form game as including a set of players i = 1,2, …N , a set of pure
strategies s1…sn Si , and a payoff function ui that transforms each strategy into the player’s von
Neumann-Morgenstern utility for each ui (s ). We can also denote all the other players in the game
as i so that ui (si , s i ) is the utility for i given the strategy that others also play. A Nash equili-
brium is a strategy si si such that ui (si , s i ) ui (si , s i ), which is to say that payoff from playing si
is no worse than any other alternative strategy against s i .
Not every game has a Nash equilibrium in this form, since we are dealing here only with pure
strategies, for example, going to the beach or hiking. But we can also think of strategies that are
probabilistic mixtures of pure strategies, for instance going to the beach and hiking with equal
probability. We can represent the probability mixture of a strategy as i (si ) where 1 0, i i
and ui ( ) is the payoff function for the player’s mixed strategies. The probability mixtures of each
player are assumed to be independent of one another. John Nash proved that every finite game has a
Nash Equilibrium with mixed strategies, though perhaps not with pure strategies (Nash, 1950b). So,
if we generalize the Nash equilibrium concept to cover mixed strategies, we can say that a mixed
strategy i i is a Nash equilibrium if ui ( i , i) ui (si , i ), which is to say that the mixed
strategy played against any other mixed strategy does no worse than any other strategy played against
it. Intuitively, if one finds oneself playing a Nash equilibrium strategy, there is no reason to uni-
laterally deviate from that strategy.
The idea of randomizing over pure strategies in order to generate a mixed strategy may seem odd.
Why would one decide to go to the beach and hike with equal probability? What would such a thing
even mean? One way to interpret the idea of a mixed strategy is to think of the player as having a
randomizing device with a set probability for choosing one option over another. So, imagine that Alex
and Sam each have a six-sided die that they roll to determine whether they will go to the beach or
hike; each hikes when they roll evens and each goes to the beach when they roll odd. This could make
sense in games of conflict or conflictual coordination where the parties do not want the other to be
able to easily guess what strategy they will play. So, in the game below where a pitcher and batter are
squaring off, if the pitcher throws a curve and the batter guesses correctly, he will get a hit. If the batter
guesses fastball when the pitcher throws a curveball, he will whiff at the pitch and look like a fool. If the
pitcher throws a fastball and the batter guesses curve, he will get a strike. If, however, the pitcher
throws a fastball and the batter guesses correctly, he will get a home run and embarrass the pitcher.
These payoffs are represented in the normal form game below (Figure 6.4).
In this game, there are no pure strategy equilibriums, but there is a mixed strategy Nash equi-
librium where the pitcher plays Curve and Fastball with a probability of 1/2, while the batter guesses
curve with a probability of 2/3 and fastball with a probability of 1/3. One could imagine that both
the pitcher and the batter know each other’s tendencies and base their randomization on that.
Batter
Curve Fastball
Curve -1, 1 2, -2
Pitcher
Fastball 1, -1 -2, 2
Figure 6.4 Zero-sum baseball game.
70
Game Theory
The game above is a zero-sum game where the payoffs always sum to zero, which is to say that one
player’s gain is another’s loss. These are games of pure conflict. In them, the Nash equilibrium is
equivalent to each player choosing the strategy that minimizes the maximum expected gain of the
other player or Minimax (Neumann and Morgenstern, 1947, 154). In these cases, the interpretation of
mixed strategies is that one is trying to keep one’s opponent off guard (Luce and Raiffa, 1957, 75).
Although this makes sense for games of pure conflict, it will not apply in games of coordination where
players benefit from coordinating or even in cases of what Peter Vanderschraaf (2018, 33) calls con-
flictual coordination, where the players have differing preferences over coordination equilibriums. A
classic example of the latter is the “battle of the sexes” game, represented below (Figure 6.5).
In this game, we return to our friends Alex and Sam; in this case, Sam prefers going to the beach
to hiking whereas Alex prefers hiking to going to the beach. Both prefer doing either together to
doing either alone. In this game, there are multiple Nash equilibria, two pure strategy equilibria
where both either go to the beach or hike, and one mixed strategy equilibrium where each assigns
their preferred option with 2/3 probability and there less preferred option with 1/3 probability. The
mixed strategy has a payoff of 2/3, while the pure strategies have payoffs of at least 1, which raises
the question of why the mixed strategy would ever be played in a game like this.
One possibility is that we might interpret the probability mixture over pure strategies as re-
presenting an individual’s conjecture or belief over what strategy they expect the other to play. If the
players are Bayesian rational, have common knowledge of the game, and independent consistent
conjectures, then their probability mixtures as conjectures will be equivalent to the probability
mixtures in the Nash equilibrium (Brandenburger and Dekel, 1987). Peter Vanderschraaf (2018, 71)
calls this interpretation equilibrium in conjectures.3
2.2 The Prisoner’s Dilemma

Many games have multiple Nash equilibriums (we will return to this below), but some games have a
unique pure strategy. In these games, mixed strategies are unnecessary.4 Perhaps the most instructive
and well-studied game with one Nash Equilibrium is the Prisoner’s Dilemma. In the canonical version
of this game, two criminals are arrested for a bank heist and put into separate rooms for interrogation.
Each prisoner is told that if he rats out on his partner, he will walk while his partner will receive the
maximum. They also know that if they both stay silent, they will get off with minimal jail time. It
seems like the thing to do is for both to stay silent, but if one of them is silent and the other rats, this is
the worst outcome possible. This game is represented below where the outcomes are ranked assuming
each person wants to reduce their jail time and I>II>>II>IV (Figure 6.6).
Alex
Beach Hike
Beach 2,1 0,0
Sam
Hike 0,0 1,2
Figure 6.5 Battle of the sexes.
Alex
Silent Inform
Silent II, II IV, I
Sam
Inform I, IV III, III
Figure 6.6 Prisoner’s dilemma.
71
John Thrasher
The best option for both of them together is for both to stay silent and the worst is for both to
inform on the other. But the worst outcome for each person is to stay silent while the other informs.
Both Sam and Alex know this, and they know that the other does as well and so on (common
knowledge). Informing dominates staying silent for both Sam and Alex because, whatever the other
person does, one does best by informing. By eliminating the dominated stay silent strategy, the only
possibility left is to inform. As it is rational for each to inform, though, they achieve their worst joint
possibility; they would have been better off had they stayed silent.
This example raises a number of issues. The first is that rationality and mutual benefit (i.e., Pareto
optimality) do not always align in game theory. In this game, the mutually beneficial outcome is for
both to stay silent. This outcome is clearly a Pareto improvement over the outcome where both
inform. Nevertheless, by staying silent, one opens themselves up to the possibility of getting their
worst outcome if the other party maximizes given that option and, hence, both are rationally driven
to the worse outcome.
The Prisoner’s Dilemma is interesting because it shows a tension between the standard of in-
dividual rationality as maximizing and the efficiency (Pareto) standard of mutual benefit. The
question is whether one can reason oneself to mutual benefit in a prisoner’s dilemma, making Pareto
optimality and individual rationality align. To do so would require moving away from the standard
conception of rationality as maximizing expected utility. This is a substantial cost and one that most
are not willing to make. Most game theorists accept that the only rational outcome of the prisoner’s
dilemma is the Pareto inferior outcome. There is no way to reason your way out of a prisoner’s
dilemma, to avoid this outcome, one must avoid prisoner’s dilemmas.
Some philosophers, most notably David Gauthier (1986, 2013b, 2013a), have argued that in-
dividual rationality and mutual benefit must align – it cannot be rational to choose to be worse off
when a better option is on the table. This has led them to develop unorthodox conceptions of
individual rationality in the face of this problem.5 Game theorists working with classical rationality as
the maximizing framework, however, reject the possibility of achieving the most mutually beneficial
outcome in a prisoner’s dilemma.
2.3 Repeated Games and Backwards Induction

The conflict between mutual benefit and rationality only applies to the one-shot prisoner’s dilemma,
however. When the game is repeated indefinitely, new possibilities emerge. Repetition allows
reciprocity and punishment by playing the cooperative (stay silent) strategy with those who tend to
cooperate and the defect (inform) strategy with those who do not cooperate. As Robert Axelrod
(1984) showed in a computer simulation pitting various conditional strategies for repeated prisoner’s
dilemmas against one another, Tit-for-Tat (cooperating in the first round and then cooperating until
your partner defects), is a very successful strategy. When games are repeated, the number of possible
strategies increases substantially since each game becomes, in effect, a stage in a larger game. When
games are repeated infinitely and the players are patient, any feasible outcome is a possible equili-
brium. This general result is known as a “folk theorem,” though there are many different specifi-
cations and characterizations of this “folk theorem.” This means that if a prisoner’s dilemma is
repeated infinitely or at least with an indefinite end, cooperation is a possible equilibrium outcome.
If, however, the game is finite and has a known end, backwards induction makes universal defection
the unique equilibrium outcome.
Backwards induction involves reasoning from the end of a series back through each decision
node that led to it in order to determine the optimal strategy. So, in the Prisoner’s Dilemma, if all
the parties know that the game will end on round n, then it makes sense for all the players to treat
the nth round as a one-shot game and, hence, to defect in that round. But, if the defection is rational
in the nth round, it is also rational in the n-1th round since now this round is effectively the end of the
72
Game Theory
repeated game. This reasoning applies to every previous round until we are led to conclude that
defection is uniquely rational in the first round as well.
Backwards induction introduces a new solution concept in the form of a sub-game perfect
equilibrium. A sub-game perfect equilibrium (SPE) is a refinement of the basic Nash equili-
brium solution concept. The difference is that the SPE specifies that players will not choose
dominated strategies in any of the subgames of a stage game. We can think of subgames as being the
choice nodes of the game. Backwards induction requires that each player choose a Nash Equilibrium
in each subgame, working from the end to the beginning of the game. An equilibrium strategy that
can pass the test of backwards induction is subgame perfect.
Subgame perfection can lead to seemingly paradoxical results. Consider the centipede game,
represented below. In this game, player one has the option of taking $2 or passing in round one. If
Player 1 takes, he gets $2 and Player 2 gets nothing. If, however, Player 1 passes in round 1, Player 2
has the option to take the money or pass in round 2. Even if Player 2 ends the game and takes in
round 2, she gets $3 and Player 1 gets $1. The payoffs of the game are such that Player 1 is better off
taking on his turn and Player 2 is better off taking on her turn, though the payoffs keep increasing as
the game goes on. If both parties can make it to the end, they will each get $100 (Figure 6.7).
It seems like the best outcome is for both players to keep passing until they get to the end of the
game. The subgame perfect strategy at the last node is for Player 2 to take rather than pass. Knowing
this, the subgame perfect strategy in the penultimate round is for Player 1 to take and so on until we
get back to round 1, where the subgame perfect strategy is for Player 1 to take. Subgame perfection
means that this game never really gets off the ground.
As with the Prisoner’s Dilemma, in experiments using finitely repeated games like the centipede
game, subjects don’t tend to play the SPE. In the trust game (Berg, Dickhaut, and McCabe, 1995),
Player 1 can choose to take or send some amount of money in roughly the same way as in the
centipede game, except both players will typically get some amount of money ($20 in the example
below). If Player 1 sends the money to Player 2 (Trusts), then both can potentially get a higher
amount if Player 2 reciprocates Player 1’s initial trust ($25 in the example). Player 2 can also choose
to not reciprocate the trust and take a higher amount ($30), giving Player 1 less money than they
started with ($15) (Figure 6.8).
The only SPE equilibrium in this game, like the centipede game, is for Player 1 to not trust. In an
experiment that has been replicated and varied a bewildering number of times, the basic result is that
most Player 1s trust Player 2 with some amount of money, and many Player 2s are, in turn trust-
worthy.6 This suggests that subjects are doing something other than applying backward induction in
the lab.
The use of game theory has been crucial in the study of the formation and stability of institutions.
Institutions are understood as equilibria of rules in repeated games on this view (Schotter, 2008;
Shepsle, 2006; Guala, 2016). The competitor view of institutions that sees them as structures of rules
that need not be equilibrium solutions to some underlying game can and does use game theory as
well, though in a different way (e.g., North, 1990; North, Wallis, and Weingast, 2009; Wallis,
1 Pass 2 Pass 1 Pass 2 Pass 1 Pass 2 Pass

($100, $100)
Take Take Take Take Take Take
($2, $0) ($1, $3) ($4, $2) ($97, $99) ($100, $98) ($99, $101)
Figure 6.7 Centipede game.
73
John Thrasher
DON’T TRUST TRUST
($20, $20) 2
UNTRUSTWORTHY TRUSTWORTHY
($15, $30) ($25, $25)
Figure 6.8 Trust game.
2011). An important area where the study of institutions and norms overlap is in the work of Elinor
Ostrom and her followers. She argued in a number of works, that informal institutions, which
resulted from negotiation and coordination of members of the community, are crucial to self-
governance (Ostrom, 1990; Ostrom, Walker, and Gardner, 1992). She also developed a theory of
institutions that relies on game theory (Ostrom, 2005).
2.4 Equilibrium Selection

Many games have multiple Nash Equilibriums. A classic example of a game with multiple equilibria
is the stag hunt or assurance game. In this game, two hunters can either hunt stag together or hare
apart. If they take down a stag, there is more meat to go around, but they will only be able to take
down a stag if both parties decide to hunt stag. The game is represented below (Figure 6.9).
As Brian Skyrms (2004) has persuasively argued, this game is interesting for a number of reasons.
It shares some features with the Prisoner’s Dilemma, but in this game, the off diagonals give a lower
payoff than the coordination solutions of both hunting stag or both hunting hare. This game has
three Nash equilibria, two pure strategy equilibria (both hunt stag or both hunt hare), and one
mixed strategy equilibrium (where both parties hunt stag and hare with a probability of ½). hunting
hare guarantees a payoff of at least 2 and possibly 3. As such, it is the safer possibility because,
whatever the other player does, one will at least get something by hunting hare. Not so for hunting
stag. If both coordinate on stag, then each will receive their highest possible payoff, but they will get
nothing if they fail to coordinate. Harsanyi and Selten (1988) described the hare equilibrium as risk-
dominant and the stag equilibrium as payoff dominant.
This game raises the question of how to rationally select between multiple equilibria. Put dif-
ferently, is it more rational for players to coordinate on the stag or hare equilibrium in this game?
The stag equilibrium is more beneficial, but also riskier. In the stag hunt, both parties need some
assurance or trust that the other party will play the payoff dominant strategy. In some games, though,
equilibrium selection is not simply a matter of assurance. Consider an asymmetric coordination game
Stag Hare
Stag 5, 5 0, 3
Hare 3, 0 2, 2
Figure 6.9 Stag hunt.
74
Game Theory
Alex
Beach Hike
Beach 2,1 0,0
Sam
Hike 0,0 1,2
Figure 6.10 Battle of the sexes game.
like the “battle of the sexes” game, which we looked at above. In that game, both parties want to
coordinate, but they receive differential benefits from coordinating (Figure 6.10).
If we look at Figure 6.10, we see that Sam gets double the payoff from the Beach outcome that
Alex does and Alex gets double the payoff that Sam does from the Hike outcome. It not only seems
like there isn’t a uniquely rational equilibrium, Sam and Alex have opposing preferences because the
two pure strategy equilibria are asymmetrical.
But recall from above that there is also a mixed strategy equilibrium in this game where each
party plays their preferred option with a probability of 2/3 and their least favored option of 1/3. This
mixed strategy has the property of being symmetrical in the sense that each player gets the same
payoff. The problem is that the payoff to the symmetrical mixed strategy is less than the payoff of
either of the asymmetrical outcomes, which raises the question of why they would play it. One
reason might be for reasons of fairness of equity since each party gets the same payoff in the
symmetrical solution. Doing so, however, would sacrifice Pareto efficiency for symmetry. If the
game is repeated, it would make sense to Alex and Sam to alternate between equilibria so that each
person gets their preferred option at some point.
In other games, the solution space covers basically all the options in the game. This is the case in
the ultimatum game. The ultimatum game is an asymmetrical extensive form game where one party
is designated as the “proposer,” who acts as the first mover, and the other party is designated as the
“decider.” Both players are given some fixed sum of resources (c) that they must divide. The
proposer can propose any amount (x) between 0 and c to the decider as a “take it or leave it” offer. If
the decider accepts, the proposer gets c-x and the decider gets x. If the decider rejects the offer,
neither gets anything. This is represented in the game tree below.
If c = $20, then any combination from ($0, $20) to ($20, $0) is a feasible outcome of the game.
Each of the feasible combinations that sums to $20 is also a Nash Equilibrium of the game. There is,
however, only one dominant SPE where the proposer offers $1 and the decider accepts the offer
since anything is better than nothing.7 When real people are asked to play this game, however,
75
John Thrasher
almost none of them play the SPE. Most offers vary between 40% and 60% of the value of the initial
endowment and most of those offers are accepted, though these can vary considerably based on the
exact details of the game (Güth and Tietz, 1990; Hoffman, McCabe, and Smith, 2008).
Equilibrium selection is especially important in the study of conventions and norms. David
Lewis, following the lead of Thomas Schelling (1960) instigated a game-theoretic approach to the
study of conventions in Convention (1969). There he developed the idea that a convention is one of
many equilibria in a pure coordination game, where everyone follows the convention because they
expect everyone else to and where that fact is common knowledge. This account of convention
proved to be incredibly useful and influential and has been used to explain things as diverse as the
evolution of meaning and language (Lewis, 1969; Skyrms, 2004, 2010), justice (Vanderschraaf,
2018), and the practice of Female Genital Mutilation (Mackie, 1996).
Much of the contemporary study of social norms also follow Lewis’s account of convention in
some way. The most influential account is developed by Cristina Bicchieri. In her work, social
norms are equilibrium solutions (Nash equilibria) to mixed-motive or impure, asymmetric co-
ordination games that rely on common knowledge of shared normative and empirical expectations
(Bicchieri, 2006, 2016). Herbert Gintis (2010), using a similar approach argues that norms should be
modeled as a correlated equilibrium. Peter Vanderschraaf (2018) also models conventions as cor-
related equilibria. A correlated equilibrium is a solution concept that expands the solution space
considerably beyond the Nash equilibrium by using a coordinating mechanism that is both external
to the base game and common knowledge to the players that they can use to reliably coordinate
(Aumann, 1974, 1987). Regardless of the solution concept that is used, this basic approach has
revolutionized the study of social norms.
3 Bargaining and Cooperative Games

The games we have looked at so far, be they zero or non-zero-sum, are all examples of non-
cooperative games. In non-cooperative games, the players cannot coordinate through commu-
nication or assume a binding mechanism to enforce their agreement. Cooperative games relax these
restrictions and allow direct communication and coordination as well as binding enforcement so that
whatever agreement is reached in the game will go into effect. There are two aspects of cooperative
game theory that are especially interesting to PPE: bargaining games and coalitional games.
The bargaining game was introduced by John Nash (1950a) as a problem of rational fair division.
In the two-person model, we can think of Alex and Sam as having cooperated to produce some
surplus together that needs to be divided between the two parties. In principle, any division is
possible, but rational agents will only consider bargaining outcomes that are both (1) greater than
what they would get if there were no agreement on how to divide the surplus (disagreement point)
and (2) any agreement wherein either party could have done better without making the other worse
off. These two constraints set the feasible space of the bargaining problem that is bounded by the
respective disagreement points of each party and the set of Pareto optimal agreements. The question
then becomes which of the Pareto optimal points rational bargainers would agree to. We can think
of the bargaining problem as a non-sequential version of the ultimatum game where both parties
have to agree to a division of the fixed sum in order to get the payoff from the agreement.
If there is a unique answer to this question, it is plausible to think that problems of fair division
can be reduced to individual rationality in the context of the bargaining problem. The problem,
however, is that there are many “unique” solutions to the bargaining problem depending on what
assumptions we are willing to allow into our solution. John Nash (1950a, 1953) identified the
first major solution, which was later generalized by John Harsanyi (1956, 1963), generally referred
to as the Nash solution. In addition to restricting the scope of a possible bargaining solution
to the disagreement points and the set of Pareto optimal agreements, this solution makes two
76
Game Theory
other assumptions. The first is that in symmetrical bargaining situations, that is, where the parties
have equivalent bargaining power, the solution should be symmetrical. This is known as the
symmetry assumption. The second is Independence of Irrelevant Alternatives. This independence
assumption is different from the one used by Kenneth Arrow and it amounts to an assumption of
expansion consistency. If the bargaining solution is in some subset of the original bargaining space,
then that same solution should hold if the bargaining space is expanded. Given these assumptions,
Nash proved that there is a unique bargaining solution wherein both parties maximize the product
of their gains from cooperation.
The appeal of this solution is that it relies on a minimum of mostly uncontroversial assumptions,
though there are controversies surrounding both the symmetry and independence assumptions.8 It is
also appealing since, as Ariel Rubinstein (1982) showed, the Nash bargaining solution is also the unique
solution to a non-cooperative, sequential version of the bargaining game. It has been criticized,
though, since it gives the party with the greater bargaining power (in terms of the better disagreement
point) more of the cooperative surplus. As John Rawls (1999, 116) argued in relation to the Nash
solution, “each according to his threat advantage is not a principle of justice.” This may make the Nash
solution unappealing as a principle of justice or fairness, but it doesn’t necessarily count against it as a
bargaining solution as such.9 Nevertheless, alternative bargaining solutions abound.
The Kalai-Smorodinsky solution (Kalai and Smorodinsky, 1975) rejects Nash’s independence
assumption in favor of a monotonicity assumption, which basically states that as the bargaining space
grows, the benefits from the bargaining solution should also correspondingly grow. This solution
will change depending on the bargaining space as well as the initial demands of the bargainers. David
Gauthier (1986) adopted this bargaining solution as his primary principle of distribution at the heart
of his contractual theory in Morals by Agreement. Gauthier calls his solution minimax relative con-
cession since each party aims to minimize the concession that they make from their ideal bargaining
outcome. Gauthier (1993) later abandoned this solution in favor of the Nash solution, before
seemingly embracing something like it again in a modified form (Gauthier, 2013a, 2013b).
As should be clear by now, philosophers are especially concerned with the bargaining problem.
This is because they see it as a formal model of fair division. A solution to that problem should link
rationality to fairness in an important way. This is especially important in contractual and con-
ventionalist theories of morality and politics where agreement in some form is the basic justificatory
currency. Many have defended versions of the Nash solution at the core of their larger contractual
theories. Ken Binmore (1994), for instance, has argued that the Nash solution is uniquely rational.
Ryan Muldoon (2017) also uses it in his dynamic theory of distributional justice. Michael Moehler
(2010, 2018) employs a modified version of the Nash solution, what he calls the stabilized Nash
solution, which ensures that each party does not go below a sufficiency threshold in their dis-
agreement points. David Gauthier (1986), as mentioned above, endorsed minimax relative con-
cession, as did Jerry Gaus (1990) in his early work. Peter Vanderschraaf has defended a version of
Braithwaite’s (1955) and Kalai’s (1977) egalitarian solution.
Cooperative game theory is also important in political science in the form of coalitional models.10
In economics, there is extensive experimental literature on bargaining.11
3.1 Conclusion
The formal study of the logic of strategic choice, which we call game theory, has established itself as
a core tool in the PPE toolbox since the middle of the last century. I have tried to introduce the
basic concepts and ideas of game theory, but the field is so large, that I have only really scratched the
surface. Those interested in a more detailed overview of the basics should see (Gaus and Thrasher,
2021) and those interested in a more complete technical overview should see (Fudenberg and
Tirole, 1991; Osborne and Rubinstein, 1994). I have not discussed evolutionary game theory, for
77
John Thrasher
instance, which is an important area of research, though one which is more popular in theoretical
biology and the study of social evolution than in economics or political theory (though see Skyrms,
1996). For a classic work on the topic, see (Maynard Smith, 1982) and for an overview of the field,
see (Weibull, 1997). I have also ignored the important interdisciplinary study of social evolution that
uses game theory as an essential tool. For an excellent overview, see Birch (2017).
Notes
1 For a historical survey see Vernon Smith (1992).
2 This is not true in some signaling games and in other games of imperfect information.
3 For two more interpretations of mixed strategies, see Rubinstein (1991).
4 Technically a pure strategy is just a mixed strategy with a given strategy played with a probability of 1.
5 The behavioral evidence seems to suggest many people reason as Gauthier argues they should, that is, by
choosing the Pareto optimal outcome. For a recent survey of these results, see Holt, Johnson, and Schmidtz
(2015).
6 How much trust and trustworthiness are observed depends on the specification of the game. For a meta-
analysis that looks at many of the important factors see Johnson and Mislin (2011).
7 Technically the option where the proposer offers nothing is also a SPE, but the decider will be indifferent
between that option and rejection since both yield $0. If we compare options of accepting the $0 offer and
rejection, accepting is clearly Pareto superior to rejecting.
8 Although independence is usually cited as the most controversial assumption, the symmetry assumption has also
come under fire. One of the early critics of symmetry was Thomas Schelling (1956, 1959). Harsanyi (1956,
1958) defended symmetry, pointing out that it was the only way to generate a unique solution to the bar-
gaining problem. For a good overview of the Harsanyi-Schelling debate over symmetry, see (Innocenti, 2008).
More recently, this author (Thrasher, 2014) has argued against the plausibility of the symmetry assumption
when bargaining theory is used in the context of contractual models of political and moral justification.
9 H. Peyton Young disputes the idea that the Nash solution is somehow unfair, arguing that it is actually the
most equitable of the bargaining solutions (Young, 1995, 129).
10 For a somewhat dated though still excellent and clear introduction, see Ordeshook (1986).
11 This literature is vast and includes the literature on the ultimatum game and related games. For an early
survey that highlights many of the important findings, see Roth et al. (1991).
References
Aumann, Robert J. 1974. “Subjectivity and Correlation in Randomized Strategies.” Journal of Mathematical
Economics 1, 1, 67–96. 10.1016/0304-4068(74)90037-8.
Aumann, Robert J. 1987. “Correlated Equilibrium as an Expression of Bayesian Rationality.” Econometrica 55,
1, 1–18. 10.2307/1911154.
Axelrod, Robert. 1984. Evolution of Cooperation. 1st edition. Basic Books.
Berg, Joyce, John Dickhaut, and Kevin McCabe. 1995. “Trust, Reciprocity, and Social History.” Games and
Economic Behavior 10, 1, 122–142. 10.1006/game.1995.1027.
Bernheim, B. Douglas. 1984. “Rationalizable Strategic Behavior.” Econometrica 52, 4, 1007–1028.
Bicchieri, Cristina. 2006. The Grammar of Society: The Nature and Dynamics of Social Norms. Cambridge University
Press.
Bicchieri, Cristina. 2016. Norms in the Wild. Cambridge University Press.
Bicchieri, Cristina, Erte Xiao, and Ryan Muldoon. 2011. “Trustworthiness Is a Social Norm, but Trusting Is
Not.” Politics, Philosophy & Economics 10, 2, 170–187.
Binmore, Ken. 1994. Game Theory and the Social Contract, Vol. 1: Playing Fair. The MIT Press.
Birch, Jonathan. 2017. The Philosophy of Social Evolution. OUP Oxford.
Braithwaite, Richard. 1955. Theory of Games as a Tool for the Moral Philosopher. Cambridge University Press.
Brandenburger, Adam, and Eddie Dekel. 1987. “Rationalizability and Correlated Equilibria.” Econometrica 55, 6,
1391–1402. 10.2307/1913562.
Bruner, Justin, Cailin O’Connor, Hannah Rubin, and Simon M. Huttegger. 2018. “David Lewis in the Lab:
Experimental Results on the Emergence of Meaning.” Synthese 195, 2, 603–621. 10.1007/s11229-014-
0535-x.
78
Game Theory
Fudenberg, Drew, and Jean Tirole. 1991. Game Theory. 1st edition. The MIT Press.
Gaus, Gerald. 1990. Value and Justification: The Foundations of Liberal Theory. Cambridge University Press.
Gaus, Gerald, and John Thrasher. 2021. Philosophy, Politics, and Economics: An Introduction. Princeton University
Press.
Gauthier, David. 1986. Morals by Agreement. Clarendon Press.
Gauthier, David. 1993. “Uniting Separate Persons.” In edited by David Gauthier and Robert Sugden,
Rationality, Justice and the Social Contract: Themes from Morals by Agreement, 176–192. Ann Arbor: University of
Michigan Press.
Gauthier, David. 2013a. “Achieving Pareto-Optimality: Invisible Hands, Social Contracts, and Rational
Deliberation.” Rationality, Markets and Morals 4, 191–204.
Gauthier, David. 2013b. “Twenty-Five On.” Ethics 123, 4, 601–624. 10.1086/670246.
Gintis, Herbert. 2009. The Bounds of Reason: Game Theory and the Unification of the Behavioral Sciences. Princeton
University Press.
Gintis, Herbert. 2010. “Social Norms as Choreography.” Politics, Philosophy, and Economics 9, 3, 251–264.
Guala, Francesco. 2016. Understanding Institutions. Princeton University Press.
Güth, Werner, and Reinhard Tietz. 1990. “Ultimatum Bargaining Behavior: A Survey and Comparison of
Experimental Results.” Journal of Economic Psychology 11, 3, 417–449.
Harsanyi, John. 1956. “Approaches to the Bargaining Problem Before and After the Theory of Games: A
Critical Discussion of Zeuthen’s, Hicks’, and Nash’s Theories.” Econometrica: Journal of the Econometric Society,
April, 144–157.
Harsanyi, John. 1958. “Notes on the Bargaining Problem.” Southern Economic Journal, April, 471–476.
Harsanyi, John. 1963. “A Simplified Bargaining Model for the N-Person Cooperative Game.” International
Economic Review, May, 194–220.
Harsanyi, John, and Reinhard Selten. 1988. A General Theory for Equilibrium Selection in Games. MIT Press.
Hoffman, Elizabeth, Kevin McCabe, and Vernon Smith. 2008. “Reciprocity in Ultimatum and Dictator
Games: An Introduction.” Handbook of Experimental Economics Results 1, 411–416.
Holt, Charles, Cathleen Johnson, and David Schmidtz. 2015. “Prisoner’s Dilemma Experiments.” In edited by
Martin Peterson, The Prisoner’s Dilemma, 243–264. Cambridge: Cambridge University Press. 10.1017/
CBO9781107360174.014.
Innocenti, Alessandro. 2008. “Linking Strategic Interaction and Bargaining Theory: The Harsanyi-Schelling
Debate on the Axiom of Symmetry.” History of Political Economy 40, 1, 111–132.
Johnson, Noel D., and Alexandra A. Mislin. 2011. “Trust Games: A Meta-Analysis.” Journal of Economic
Psychology 32, 5, 865–889. 10.1016/j.joep.2011.05.007.
Kalai, Ehud. 1977. “Proportional Solutions to Bargaining Situations: Interpersonal Utility Comparisons.”
Econometrica 45, 7, 1623–1630. 10.2307/1913954.
Kalai, Ehud, and Meir Smorodinsky. 1975. “Other Solutions to Nash’s Bargaining Problem.” Econometrica 43,
3, 513–518.
Lewis, David. 1969. Convention: A Philosophical Study. Harvard University Press.
Luce, R. Duncan, and Howard Raiffa. 1957. Games and Decisions: Introduction and Critical Survey. John Wiley &
Sons.
Mackie, Gerry. 1996. “Ending Footbinding and Infibulation: A Convention Account.” American Sociological
Review 61, 6, 999–1017.
Maynard Smith, John. 1982. Evolution and the Theory of Games. Cambridge University Press.
Moehler, Michael. 2010. “The (Stabilized) Nash Bargaining Solution as a Principle of Distributive Justice.”
Utilitas 22, 04, 447–473.
Moehler, Michael. 2018. Minimal Morality: A Multilevel Social Contract Theory. Oxford University Press.
Muldoon, Ryan. 2017. Social Contract Theory for a Diverse World: Beyond Tolerance. Routledge.
Nash, John. 1950a. “The Bargaining Problem.” Econometrica 18, 2, 155–162. 10.2307/1907266.
Nash, John. 1950b. “Equilibrium Points in N-Person Games.” Proceedings of the National Academy of Sciences of
the United States of America 36, 1, 48–49.
Nash, John. 1953. “Two-Person Cooperative Games.” Econometrica: Journal of the Econometric Society 21, 1,
128–140.
Neumann, John von, and Oskar Morgenstern. 1947. Theory of Games and Economic Behavior (Commemorative
Edition). Princeton University Press.
North, Douglass C. 1990. Institutions, Institutional Change and Economic Performance. Reprinted. Cambridge
University Press.
North, Douglass C., John Joseph Wallis, and Barry R. Weingast. 2009. Violence and Social Orders: A Conceptual
Framework for Interpreting Recorded Human History. Cambridge University Press.
79
John Thrasher
Ordeshook, Peter C. 1986. Game Theory and Political Theory: An Introduction. Cambridge University Press.
Osborne, Martin J., and Ariel Rubinstein. 1994. A Course in Game Theory. 1st edition. The MIT Press.
Ostrom, Elinor. 1990. Governing the Commons: The Evolution of Institutions for Collective Action. Cambridge
University Press.
Ostrom, Elinor. 2005. Understanding Institutional Diversity. Princeton University Press.
Ostrom, Elinor, James Walker, and Roy Gardner. 1992. “Covenants With and Without a Sword: Self-
Governance Is Possible.” The American Political Science Review 86, 2, 404–417.
Pearce, David G. 1984. “Rationalizable Strategic Behavior and the Problem of Perfection.” Econometrica 52, 4,
1029–1050.
Rawls, John. 1999. A Theory of Justice. Revised. Belknap Press.
Roth, Alvin E., Vesna Prasnikar, Masahiro Okuno-Fujiwara, and Shmuel Zamir. 1991. “Bargaining and Market
Behavior in Jerusalem, Ljubljana, Pittsburgh, and Tokyo: An Experimental Study.” American Economic Review
81, 5, 1068–1095.
Rubinstein, Ariel. 1982. “Perfect Equilibrium in a Bargaining Model.” Econometrica 50, 1, 97–109. 10.2307/
1912531.
Rubinstein, Ariel. 1991. “Comments on the Interpretation of Game Theory.” Econometrica 59, 4, 909–924.
10.2307/2938166.
Schelling, Thomas. 1956. “An Essay on Bargaining.” The American Economic Review 46, 3, 281–306.
Schelling, Thomas. 1959. “For the Abandonment of Symmetry in Game Theory.” The Review of Economics and
Statistics 41, 3, 213–224.
Schelling, Thomas. 1960. The Strategy of Conflict. Harvard University Press.
Schmidtz, David. 1991. The Limits of Government: An Essay on the Public Goods Argument. Westview.
Schotter, Andrew. 2008. The Economic Theory of Social Institutions. Cambridge University Press.
Shepsle, Kennneth. 2006. “Rational Choice Institutionalism.” In edited by R. A. W. Rhodes, Sarah A. Binder,
and Bert A. Rockman, The Oxford Handbook of Political Institutions, 23–38. Oxford University Press.
Skyrms, Brian. 1996. Evolution of the Social Contract. Cambridge University Press.
Skyrms, Brian. 2004. The Stag Hunt and the Evolution of Social Structure. Cambridge University Press.
Skyrms, Brian. 2010. Signals: Evolution, Learning, and Information. Oxford University Press.
Smith, Vernon. 1992. “Game Theory and Experimental Economics.” History of Political Economy 24, January,
241–282.
Thrasher, John. 2014. “Uniqueness and Symmetry in Bargaining Theories of Justice.” Philosophical Studies 167,
3, 683–699.
Vanderschraaf, Peter. 2018. Strategic Justice: Convention and Problems of Balancing Divergent Interests. Oxford
University Press.
Wallis, John Joseph. 2011. “Institutions, Organizations, Impersonality, and Interests: The Dynamics of
Institutions.” Journal of Economic Behavior & Organization, The Dynamics of Institutions in Perspectives:
Alternative Conceptions and Future Challenges 79, 1, 48–64. 10.1016/j.jebo.2011.02.002.
Weibull, Jorgen W. 1997. Evolutionary Game Theory. First Paperback edition. The MIT Press.
Young, H. Peyton. 1995. Equity. Princeton University Press.
80
7
FOUR STRUCTURES OF
INTRANSITIVE PREFERENCES
Luc Bovens
1 Introduction
The transitivity of (strict and all-things-considered) preference seems like a requirement of ra-
tionality: if someone prefers option A to B and prefers B to C, then they should also prefer A to C. It
would be unreasonable to prefer A to B and B to C and yet to be either indifferent between A and C
or prefer C to A. And we can extend the length of the chain: If, all things considered, someone
prefers A to B, B to C, …, and Y to Z, then they should also prefer A to Z. Why does this matter?
First, there is the philosophical project of human self-understanding. Humans have preferences.
So, what are preferences like? They have a certain structure. For instance, preferences are asym-
metric. If you prefer A to B, then you can’t at the same time prefer B to A. That is not too
contentious. But what about the transitivity of preferences? Is it an analytic truth? Is it a constraint of
rationality? Or can we say no more than that our preferences are typically transitive, though, in
special circumstances, it is perfectly reasonable to have intransitive preferences?
Second, the standard economic theory rests on a theory of rational choice that allows a re-
presentation of preferences by means of a utility function. A utility function u maps options onto real
numbers so that A is preferred to B if and only if u(A) > u(B). To do so requires that preferences are
transitive. If A is preferred to B, B to C, and C to A, that is, if preferences are intransitive, then we
would be at a loss in representing this preference. This is so because the “greater than” relation on
the real numbers is transitive, and hence it cannot be the case that u(A) > u(B) > u(C) > u(A).
Third, when our preferences are moral preferences, they track goodness: a moral preference for A
over B maps onto a moral judgment that A is better than B. But could A be better than B, B better
than C, and C better than A? We tend to think of goodness (or value) like weight: if A has more of it
than B and B more than C, then A has more of it than C as well. Furthermore, if better-ness is
intransitive, then there are choice situations in which there is no moral choice to be made since
there is no best option. If moral preferences can be intransitive, this would disturb our ordinary
understanding of value, and it would leave us without a moral choice in some choice situations.
And yet, over the last half-century or so, we have seen many attempts by economists, psy-
chologists, and philosophers to spell out scenarios in which one might at least have some appre-
ciation for agents who hold intransitive preferences over option pairs. Whether these preferences
reflect their true preferences, and, if so, whether such intransitive preferences can be reasonable, is a
profound question that I will not pursue in this paper.
DOI: 10.4324/9780367808983-9 81
Luc Bovens
I have set myself two more modest tasks. First, I will taxonomize the many scenarios in the
literature in four structures. Second, considering that this paper is to appear in a volume on phi-
losophy, politics, and economics, I will show how these structures of intransitive preferences are
relevant to policymaking.
A note about terminology. By “preferring” I always mean “strictly and all-things-considered
preferring,” and, similarly, for “preference.” Preferences are defined over options. And options have
one or more features. These features are properties of the options that make them more or less
choice-worthy. And these features have values. It is in virtue of having these values that they become
more or less choice-worthy. For instance, in choosing a car, I have n options (a Toyota Camry, …),
and I care about two features, viz. number of passengers (with values 1, …, 7) and mileage (with
values of x miles per gallon).
2 Negligible-Value-Differences and Missing-Values Cycles

We start with an example by George F. Schumm (1987). You are buying boxes of Christmas tree
ornaments. In the first store, you find a box of brand A with a Red, a Green, and a Blue ball and a
box of brand B also with a Red, a Green, and a Blue ball. The Red ball in brand A is much shinier
than the Red ball in brand B. There isn’t much difference between the Blue balls or between the
Green balls. We could say that there is no discernible difference between the Blue balls and between
the Green balls, or we could say that the differences are just too small to affect your preference. So,
you prefer brand A to brand B. Being a picky shopper, you go to a second store and find a box of the
same brand B and a box of brand C, again with a Red, a Green, and a Blue ball. The Green ball of
brand B is much shinier than the Green ball of brand C. There isn’t much difference between the
Red balls and between the Blue balls. So, you prefer brand B to brand C. You go to a third store and
find a box of brand A and brand C. The Blue ball of brand C is much shinier than the Blue ball of
brand A. There isn’t much difference between the Red balls and between the Green balls. So, you
prefer brand C to brand A.
How is this possible, assuming that two boxes of the same brand are exactly alike? In Table 7.1, I
indicate shininess through *s – the more *s there are, the shinier the ornament is. Now differences
of one * either don’t register for you, or they don’t affect your preferences. Only differences of two
*s matter, and you prefer more to less shiny ornaments. Then you might well come to prefer A to B,
B to C, and C to A.
Here is a variant of this example. In Schumm’s example, the single *-ed balls do not make any
difference since they are only one * removed from the other same-colored balls. Because they are
not difference makers, we might as well remove them. So, suppose that the boxes each have two
balls with the colors as indicated in Table 7.2. Of the two Reds, the one with the # is the prettier
Table 7.1 Schumm’s Christmas Ornaments
Brand A RED** GREEN* BLUE

Brand B RED GREEN** BLUE*
Brand C RED* GREEN BLUE**
Table 7.2 Missing Christmas Ornaments
Brand A RED# – BLUE

Brand B RED GREEN# –
Brand C – GREEN BLUE#
82
Structures of Intransitive Preferences
Red. Similarly, for Blue and Green. Let us assume that you are indifferent between Red, Blue, or
Green balls. Color theory teaches us that color is saturation, brightness, and hue. Suppose you
appreciate saturation in Red balls. More saturation is what makes one Red ball prettier than another.
You appreciate brightness in Blue balls and a shamrock hue in Green balls. When you hold boxes A
and B, you point to the difference in saturation between the two Red balls. You can fill in the story
for comparisons between B and C and between A and C. And again, one can see how you would
come to prefer A to B, B to C, and C to A.
An alternative way of telling the story is that a # indicates a singly pretty-making feature – say,
shininess – and then, to make the example work, we need the additional stipulation that you are
only capable of making intra-color comparisons.
This missing-values structure reoccurs at many places in the literature. Slovic and MacPhillamy
(1972) did an experimental study in which subjects were asked to compare students. Students were
given scores on three performance features. When two students were each assigned a score on
feature F1, while one was assigned an additional score on feature F2 and the other on feature F3,
then subjects were prone to rank the student higher who had the higher score on the common
feature F1. Bar-Hillel and Margalit (1988) use this behavioral trait to generate intransitive pre-
ferences. People will rank candidates X over Y, Y over Z, and Z over X when test grades and
missing grades are presented on academic subjects as in Table 7.3.
John Broome (1991: 100–101) presents the case of Maurice who prefers to go mountaineering
rather than stay home because it would be cowardly to stay home; he prefers to stay home rather
than go to Rome because he finds sightseeing boring, and he prefers to go to Rome rather than
mountaineering because it is more cultured. This example has the same structure of missing values
generating the intransitivity, as we can see in Table 7.4.
Why is no value assigned to the option of Rome on the feature “Courageous”? (Similarly, for
Mountaineering on “Avoidance of Sightseeing Boredom” and for Home on “Cultured.”) It’s not
that one could not assign such a value. One could say that there is nothing particularly courageous
about going to Rome. That is true – one could say that. But for the subject in question, the issue of
courage just does not enter in when comparing Rome to other options.
Frances Kamm (1985) presents the following example. Suppose you promised to show up for a
lunch date that you have a duty to keep (D). On your way to the appointment, you come upon a car
crash. The person injured needs an immediate kidney transplant, and you are the only person who
can help. Donating a kidney would not be a duty – it would be a supererogatory act (S). It would be
Table 7.3 Bar-Hillel and Margalit’s Missing Test Grades
SubjectsStudents English History Math
X A – B
Y B A –
Z – B A
Table 7.4 Broome’s Mountaineering Example
Features Options Courageous Avoidance of Sightseeing Boredom Cultured
Mountaineering YES – NO
Home NO YES –
Rome – NO YES
83
Luc Bovens
Table 7.5 Kamm on Duty, Predilection, and Supererogation
FeaturesOptions Promise-Keeping Protection of Personal Sphere Life-Saving
Duty YES – NO
Predilection NO YES –
Supererogatory act – NO YES
permissible to donate a kidney and miss the lunch date. So, you may want to donate a kidney rather
than keep your lunch date: S D. Given that the act is supererogatory, it is permissible to refuse to
donate the kidney and pursue some predilection (P) of yours – say, you don’t want to lose out on a
vacation that you had planned. So, you may want to satisfy your predilections rather than do a
supererogatory act: P S. But you cannot point to predilections in order to shirk your duty. You
should do your duty rather than satisfy your predilections: D P. Hence it is within the constraints
of morality to hold the following intransitive preferences: D P, P S, and S D. The example is
presented in Table 7.5.
What is striking is that all these seemingly unrelated examples share a common structure. There
are three relevant features, and each feature enters in when ranking a pair of options but does not
affect or engage the remaining option. When the options and features are lined up in a particular
way, we obtain a structure that generates intransitive preferences.
3 Condorcet-Voting-Paradox Style Cycles

We will continue with Schumm’s Christmas balls theme to illustrate the other three structures as
well. In our new setup in Table 7.6, we have three boxes, each containing a Red, a Blue, and a
Green ball. You prefer a super-shiny Red ball (R**) to a shiny Red ball (R*), and you prefer R* to
a matte Red ball (R). The same holds for the Blue and the Green balls. You care equally about each
color and can recognize whether a ball is more or less shiny than another same-colored ball, but you
are not able to make cross-color comparisons, and you are not able to judge how much shinier it is.
Now let’s do the same setup as before. There are three stores, the first with brands A and B, the
second with brands B and C, and the third with brands A and C. The boxes contain balls with the
features in Table 7.6. What might you choose? Given our stipulations, you might choose A over B
in store 1, since two of the three balls in A (namely, the Red and the Green ones) are shinier than in
B. Similar reasoning has you choose B over C in store 2 and C over A in store 3. So once again,
your preferences are intransitive.
This example has the same structure as the Condorcet Voting Paradox, with majority voting
leading to cycles. Table 7.7 presents an example with three voters (1, 2, and 3) and three candidates
(A, B, C). Voter 1 ranks candidate A above B and B above C etc. If we conduct a pairwise majority
vote, then A wins from B with two votes, B wins from C with two votes, and C wins from A with
two votes. Hence, the ranking generated by the pairwise majority vote is intransitive.
In the example of the Christmas balls, the voters in the Condorcet Paradox are replaced by
features of the boxes. On the “Red Ball” feature, brand A ranks above brand B, and brand B above
Table 7.6 Choice by a Majority of Features
Brand A R** B G*
Brand B R* B** G
Brand C R B* G**
84
Table 7.7 The Condorcet Voting Paradox
Voters
Voter 1 Voter 2 Voter 3
A B C
B C A
C A B
Table 7.8 The Condorcet Paradox for Choice Affected by Multiple

Features
Features
Red Ball Blue Ball Green Ball
A B C
B C A
C A B
brand C, etc. Table 7.8 offers this alternative representation of the choice problem in Table 7.6. The
subject then sets their pairwise preferences over the brands by ranking a brand higher when it ranks
higher on a majority of features. And this yields precisely the same cycle.
The example of intransitive preferences is a simple transposition of the Condorcet Paradox in
social choice theory to the construction of preferences by a single person based on which one of two
options ranks higher on a majority of features.
To the best of my knowledge, the first occurrence of this transposition is in an article by the
economist Kenneth May (1954). May asked 62 subjects to rank potential marriage partners A, B, and
C in pairwise comparisons. On Intelligence, A is ranked higher than B and B higher than C. On
Looks, B is ranked higher than C and C higher than A. On Wealth, C is ranked higher than A, and
A higher than B. Seventeen of the 62 subjects ranked A over B, B over C, and C over A.
Majority rule for preference formation by a single person is a strange rule. Think of ranking
restaurant X and restaurant Y. First, it’s not unreasonable to say that you care more about food
quality than ambience, service, and waiting time. So even if Y is stronger on ambience, service, and
waiting time, you may still prefer X over Y because the food is better. Second, even if you care
equally about all these issues, it’s not unreasonable to prefer restaurant X over restaurant Y, just
because the food is so much better in X than in Y, while the ambience, the service, and the waiting
time are only slightly better in Y than in X.
The only reason that majority rule made sense in our example is that we stipulated that the agent
forming a preference (i) cared equally about each of the features, (ii) was only able to make ordinal
judgments of shininess, and (iii) was only able to make intra-color comparisons. So, it’s only given
very atypical stipulations that majority rule made sense in preference formation by a single person. In
May’s experiment, these atypical stipulations seem to have held up for a sizable minority of subjects.
Presumably, they cared roughly equally for Intelligence, Looks, and Wealth, and they were only
provided ordinal information about these features.
In voting, these stipulations are not atypical. There is an egalitarian constraint that everyone’s
vote count for one. We only register ordinal preferences for each voter. Comparing the strength of
preferences across voters would be an open invitation for the voters to misrepresent how much they
care about the issue just to win the vote.
85
Luc Bovens
In conclusion, it is tempting to transfer the Condorcet Voting Paradox from social choice to
preference formation by a single person. But certain normative constraints make majority voting at-
tractive and these constraints are typically absent in the context of single-person preference formation.
4 Sen’s-Libertarian-Paradox Style Cycles

You have a choice between four boxes of Christmas ornaments as outlined in Table 7.9. Each box
contains a mantle ornament (an angel or a pinecone) and a tree ornament (a star or a ball). Stars and angels
are representative of Christmas (since they refer to the bible story), while pinecones and balls are not.
Suppose you have the following preferences. First, you prefer showier to less showy mantle
ornaments and shinier to less shiny tree ornaments. Second, ornaments that are symbolic of
Christmas receive special consideration. If two boxes contain mantle ornaments representative of
Christmas, it would be sacrilegious to take the box with the less showy one. If two boxes have tree
ornaments representative of Christmas, it would be sacrilegious to take the box with the less shiny
one. In other words, showiness and shininess are what matters, but they matter particularly in
comparisons between ornaments representative of Christmas. Finally, the (weak) Pareto principle
holds: you prefer a box containing both a showier mantle ornament and a shinier tree ornament.
The boxes A, B, C, and D come in the following combinations. In Table 7.9, #-signs indicate
the degree of showiness, ranging from no # to three #s, and *-signs indicate the degree of shininess,
ranging from no * to three *s.
A is preferred to B on grounds of Pareto. B is preferred to C because the angel, representative of
Christmas, is showier in B. C is preferred to D on grounds of Pareto. And D is preferred to A
because the star, representative of Christmas, is shinier in D.
We can also do a version with three options – see Table 7.10. A is preferred to B since it has the
showier mantle ornament representative of Christmas. B is preferred to C since it has the shinier tree
ornament representative of Christmas. And C is preferred to A on grounds of Pareto.
We can represent the choice problems in Tables 7.9 and 7.10 slightly differently. We will order
the boxes on how well they do on the showiness and the shininess of the ornaments. Showiness is
based on the mantle ornaments, while shininess is based on the tree ornaments. Here is what the
Table 7.9 Boxes with Mantle Ornaments with # Markers for Showiness and Tree
Ornaments with * Markers for Shininess
Options Mantle Ornament Tree Ornament
A Pinecone### Star*
B Angel## Ball
C Angel# Ball***
D Pinecone Star**
Table 7.10 Three-Option Choice Problem over Boxes with Mantle and Tree
Ornaments
Options Mantle Ornament Tree Ornament
A Angel# Ball
B Angel Star**
C Pinecone## Star*
86
Table 7.11 An Alternative Representation of the Four-Option Choice Problem

with Boxes A, …, D
Table 7.12 An Alternative Representation of the Three-Option Choice Problem

with Boxes A, …, C
four-box (Table 7.11) and three-box (Table 7.12) problems would look like, with the higher-
ranked boxes higher up in the ranking. The dotted circles indicate that the features (showiness and
shininess) are particularly pertinent to the boxes encircled because the features affect ornaments
representative of Christmas.
It is easy to see in this representation that by Christmas Representativeness B C and D A, and
by Pareto A B and C D, yielding the cycle A B C D A.
The three-box case is fully analogous, leading to the cycle A B C A.
This representation reveals the origin of these intransitive rankings in social choice. Sen (1970a
and 1970b) introduces what has come to be known as Sen’s Liberal or Libertarian Paradox.1 The
principle of Minimal Libertarianism states that certain options are strictly within one’s personal
sphere, and one should be fully decisive over these options. That is, the social ordering should
respect one’s personal preference. The Pareto principle and the principle of Minimal Libertarianism
may lead to intransitive social rankings.
Sen’s example is represented in Table 7.13. Prude and Lewd are in a bookstore. They find a
single copy of D.H. Lawrence’s seedy classic Lady Chatterley’s Lover. Prude prefers that this book
remain unread, but if anyone is going to read it, it better be Prude because heaven knows what it
Table 7.13 Sen’s Lady Chatterley’s Lover Example Illustrating the Libertarian
Paradox
87
Luc Bovens
would do to Lewd. Lewd thinks it would be a sad day if this book remains unread, and Prude rather
than Lewd should read it because it would do Prude some good.
By Minimal Libertarianism, Lewd’s preference to read (Lewd Reads Unread) and Prude’s
preference not to read (Unread Prude Reads) should be respected in the social ranking. But both
parties prefer that Prude rather than Lewd read: by Pareto, Prude Reads Lewd Reads. Hence, we
have a cycle: Lewd Reads Unread Prude Reads Lewd Reads.
Sen’s Lady Chatterley’s Lover example has the same structure as our three-box Christmas ornament
case. The analog of the Pareto principle is that if one box is ranked above another on all relevant
features (i.e., showiness and shininess), then the former box is ranked above the latter, all things
considered. The analog of Minimal Libertarianism is that options may have special characteristics that
demand respect for particular features.
Bovens (1990, 1993, 1994) illustrates this idea through an intransitive preference ranking over
vacation destinations and an intransitive moral ranking over more and less worthwhile development
programs. I will here offer an example of how such intransitive rankings could arise in another
policy context. Suppose that we have to rank several proposals for transport projects. There are two
features that planners take into account. First, projects should displace as few people as possible.
Second, projects should minimally destroy existing wilderness areas. Now let us suppose that four
projects A, …, D rank as in Table 7.14 on these two features. Also, suppose that “Minimal
Displacement of People” is a particularly sensitive issue when we are comparing projects in poor
communities, while “Minimal Destruction of Wilderness” carries much weight when we are
comparing projects in locations that have a reputation for their natural beauty. B displaces people in
Poor-Ville and C in Broke-Town. Hence, we should rank B C. A destroys wilderness areas in
Pristine-Nature-Ville and D in Idyllic-Prairie-Town. Thus, we should rank D A. Along with
Pareto, this leads to the cycle A B C D A.
5 Sorites Cycles
You are shopping in Harrod’s and find a fantastic Christmas ball of brand A. It is a super-fancy
Red ball R7* with a price tag of $27 = $128. One shelf down is a ball of brand B that may be a bit
less fancy but really does not look all that different, for half the price – that is, R6* with a price tag of
$26 = $64. You are a price-conscious shopper and prefer B to A. In the next store (one step down
from Harrod’s), you find both brand B and another marginally less fancy ball of brand C for half the
price of $25 = $32. Again, you prefer C to B. Carry on like this until you find yourself in the dollar
store comparing brand G – that is, R1* for $21 = $2 – with brand H – that is, R0* for $20 = $1.
Again, you prefer H to G. But when you compare H with A on the web, H looks so drab against A
that you prefer A at $128 to H at $1.
The structure of this intransitive preference relation is laid out in Table 7.15.
Table 7.14 A Sen’s-Libertarian-Paradox Style Cycle for Policy Evaluations
88
Table 7.15 Sorites Cycle with Christmas Balls
A B C D E F G H A
Quality R7* R6* R5* R4* R3* R2* R1* R0* R7*
Price $128 $64 $32 $16 $8 $4 $2 $1 $128
How did this intransitive relation come about? There are many ways to tell the story. Maybe
one-* differences are simply imperceptible. So why pay double as much? But after several steps, the
difference is pronounced, and you would rather have the more expensive ornament at a higher
price. Or maybe the differences are perceptible but not sufficient to justify the price difference. But
as we repeat the steps multiple times, it becomes a whole different kind of ornament. We move from
an artwork to a piece of junk. And that difference in kind is what more than justifies the $127 price
difference between A and H.
This is a narrative from left to right with starting point A. We are offered a binary choice between
retaining A or trading in A for B. We choose B. This chain goes on until we are left with H,
regretting that we did not retain A. I will call the narrative in this direction “Downhill Slide.”
There is also a narrative from right to left with starting point H. We are offered a binary choice
between retaining H or trading in H for G. We retain H. Had we chosen G, we would have been
offered a binary choice between retaining G or trading in G for F. We would have chosen G. This
counterfactual chain goes on until we would have been offered a binary choice between retaining B
or trading in B for A. Now, if we had been offered a binary choice between retaining H or trading
in H for A we would have done so. But given that the route went over a chain of successive steps,
we were not able to reach A. I will call the narrative in this direction “No Way Up.”
In the real world, there are many examples of both narratives. The route to becoming an addict is
an example of Downhill Slide. You start from a drug-free and respectable life. Each fix offers an
exhilarating experience (analogous to getting a 50% reduction on the price tag) and costs very little
in the way of loss of lifestyle (analogous to getting an ornament that has marginally lower appeal).
But as you say yes to one fix after another, you end up in Junkie Lane, wishing you could go back
where you started from.
Here is an example of No Way Up. You are working a low-wage job in food services. You can
get a promotion, but you just don’t think it’s worth it. You barely get paid more, and it’s a lot of
extra responsibility on your shoulders. This is what the promotion path looks like in your company:
small steps (analogous to an ornament that has marginally more appeal) that are barely worth the
hassle (analogous to paying double). But your co-worker, who did not disagree with you about the
benefits and costs of each step, chose counter-preferentially in each binary choice. They had their
eyes on the prize, accepted each promotion, and are now CEO! You definitely prefer their position
to your low-wage job.
Tversky (1969) presents a case that has this general structure. There are n candidates ordered on
Intelligence and Experience. An employer may give precedence to Intelligence over Experience.
But if the difference between the scores between two candidates on Intelligence is too small, they
will rank them on Experience instead. In Table 7.16, the values of the features in the hiring scenario
Table 7.16 Sorites Cycle for Tversky’s Job Candidates
A B C D E F G H A
Intelligence (IQ score) 170 160 150 140 130 120 110 100 170
Experience (months) 1 2 4 8 16 32 64 128 1
89
Luc Bovens
are transforms of the values of the features in the Christmas ornaments scenario. For IQ, I used the
transform (100 + 10x) with x being the number of stars marking the quality of the Christmas ball,
and for Experience, I reversed the Price scale since Experience is a good-making feature and Price a
bad-making feature. An employer ranks B over A since a difference of 10 points in IQ may be due
to measurement error or, even if correct, would make only a minimal difference. Hence, the greater
experience is what supports B’s hire. The same reasoning holds for the pairs B and C, C and D, etc.
But for the pair H and A, H is ranked over A since Intelligence takes precedence over Experience,
and 70 points is a substantial difference.
Dorothy Edgington (1992, 1996) frames these cases as decision-theoretic variants of the Sorites
Paradox or the Paradox of the Heap. This paradox goes back to the philosopher Eubulides in the
fourth century BC: a single grain of sand is not a heap. Adding one extra grain of sand to a non-heap
cannot make it into a heap. It follows that however many grains of sand we pile up, there can never
be a heap. The argument is valid and seems sound, yet we all know its conclusion to be false. There
certainly are heaps of sand.
Table 7.17 represents a decision-theoretic version of the Sorites paradox that stays very close to
the original paradox. Suppose that you very much value the sublime view from your front window.
Your neighbor asks you whether they can dump a wheelbarrow of sand on your driveway and offers
to pay $1,000. You agree, thinking that one wheelbarrow of sand does not ruin your view. Then
time after time, your neighbor makes the same request. Each time you reason that one extra
wheelbarrow can’t affect your view, and you cash in on the $1,000. After 1,000 episodes, there is a
heap of sand in front of your window that blocks your view, and you prefer a driveway without any
sand and no cash in your pocket to the heap blocking your view and $1,000,000 in your pocket.
This Sorites cycle, which stays close to the original Sorites Paradox, has the same structure as the
Sorites cycles with Christmas balls and with job candidates.
This is a Downhill Slide narrative. But we could also do a No Way Up narrative. Suppose you
have $1M in some spare bank account and a heap of sand in front of your house. You would gladly
spend it on heap removal to secure a sublime view. But you are unwilling to make consecutive $1K
payments to remove one wheelbarrow after another.
Warren Quinn (1990) presents a fanciful Sorites cycle with a Downhill Slide narrative and ac-
knowledges that it has the same structures as Tversky (1969). In Quinn’s case, we strap a device on a
subject that delivers electroshocks. At level 0, the device is off. At level 1, the electroshock is so
minimal that the difference between 0 and 1 is imperceptible to the subject. The same holds for the
difference between 1 and 2, between 2 and 3, and so on. But at level 1000, the shocks are extremely
painful. Each week, the subject is offered $10K for turning up the device with one notch. This is a
straightforward downhill slide case: the subject will advance the settings from 0 all the way to 1000,
wishing that they could return to setting 0. This cycle is presented in Table 7.18.
A few years later, Larry Temkin (1996) presents a fanciful Sorites cycle with a No Way Up
narrative. He credits Rachels for the example, who had developed a similar example in unpublished
work. Rachels (1998) publishes a version that is close to Temkin (1996). He also includes an ex-
ample that has a Downhill Slide structure, which I will not discuss here.2
Table 7.17 Sorites Cycle Based on the Original Sorites Paradox
0 Loads 1 Load 2 Loads 1,000 Loads 0 Loads
View Sublime Sublime minus Sublime minus 2 … Obstructed Sublime

Cash earned $0 $1K $2K $1M $0
90
Table 7.18 Sorites Cycle of Quinn’s Self-Torturer with the Starting Point in Boldface
Level 0 1 2 3 … 1000 0
Remuneration $0 $10K $20K $30K $10,000K $0
Table 7.19 Sorites Cycle of Temkin-Rachels Torture-to-Hangnail with the Starting Point in Boldface
Suffering Hangnail Intense … Intense minus 2 Intense minus Intense Hangnail

Duration 2n yrs 2 yrs 8 yrs 4 yrs 2 yrs 2n yrs
Here is Temkin’s (1996) version. Our starting point is the option of 2 years of intense torture in
boldface in Table 7.19. We then introduce an option in which the intensity of the pain is slightly
reduced but lasts double the amount of time, to the left of two years of intense torture in the Table.
A subject would rather have the initial option of two years of intense torture. We continue this series
in the same manner: with each step, there is a slight reduction of the pain, but there is a doubling of
the time period. For each successive pair, the subject would opt for the greater but shorter-lasting
pain. At the end of the series, we have a trivial pain (say, a hangnail) lasting many, many years. This
option would be preferred to two years of intense torture, but it cannot be reached due to the way
the choices are structured.
We will end with an application to public policy that shares this Sorites structure. Tragedy of the
Commons problems are often analyzed as n-person Prisoner’s Dilemmas. To take Hardin’s example
(1968), each herder has a small benefit from bringing their sheep to the commons, but in doing so
imposes a cost on all other herders due to a decrease in the quality of the grasslands. If each herder
pursues their individual benefit, the lands will be massively overgrazed. Similarly, there is an older
example of lawn-crossing by Jonathan Harrison (1952–3): a person crossing the (presumably Oxford
College) lawns is pursuing a benefit for themselves due to timesaving but is imposing a cost on
others since lawn-crossing trails ruin the aesthetics of the grounds. The moral is that individual
rationality does not lead to collective rationality and that we need to coordinate our actions to secure
the collectively rational outcome. (See also Bovens (2015).)
However, Erik Carlson (1996) makes the apt observation that there is a one-person variant of
Harrison’s lawn-crossing example, which is an instance of Quinn’s Self-Torturer. In this variant, I
and I alone decide to cross the lawn on my estate, reap the benefit, and bear the cost. With every
crossing, I save time, but the lawn turns from sublime into horror. The structure is as laid out in
Table 7.20.
Chrisoula Andreou (2006) independently develops this point. When explaining how we arrived
at a situation of despoliation, the standard response is that the situation has a Prisoner’s Dilemma
structure in which individually rational actions lead to collectively irrational outcomes. Hence, if we
can put the decision in the hands of a collective agent, the tragedy of the commons will be averted.
But this is not quite right. We can just as well arrive at a situation of despoliation in single-person
Table 7.20 Sorites Cycle of the One-Person Variant of Harrison’s Lawn-Crossing Example
0 Crossings 1 Crossing 2 Crossings 100 0 Crossings

Crossings
Aesthetic experience Sublime Sublime Sublime … Horror Sublime

of lawn minus minus 2
Time Saved 0 min 5 min 10 min 100 min 0 min
91
Luc Bovens
decision-making. A single person may hold an intransitive ranking over the outcomes. If options are
offered in a stepwise sequence, choices will lead to an option they disprefer to the option that they
started from, as in the single-person lawn-crossing case. And this could also be the predicament of a
single collective agent. Hence, neutralizing the Prisoner’s Dilemma by moving decision-making
into the hand of a collective agent may not be sufficient to avert the despoliation of nature. For a
discussion of Andreou (2006), see Rebecca Livernois (2018).
6 Conclusion
I have taxonomized a half-century of examples of intransitive preferences into four structures: (i)
Negligible-Value-Differences and Missing-Values Cycles; (ii) Condorcet-Voting-Paradox Style
Cycle; (iii) Sen’s-Libertarian-Paradox Style Cycles; and (iv) Sorites Cycles.
Some recurring mechanisms are present that cut across these structures. First, the Condorcet-
Voting-Paradox style cycles and Sen’s-Libertarian-Paradox style cycles are transpositions of
intransitive social rankings as we encounter them in social choice. Second, both in Negligible-
Value-Difference cycles and in Sorites cycles, the intransitivity comes about because we ignore small
value differences on a particular feature in setting our preference, and these small value differences
will come to matter when they add up. And third, what is at the core of Missing-Value cycles and in
Sen’s-Libertarian-Paradox style cycles is that the relevant features have pull or at least have greater
pull for a subset of the options. These three mechanisms would be another way to categorize
intransitive preferences that crosscuts my taxonomy.
The grand philosophical questions remain. It may seem that people hold intransitive preferences,
but do they really? And if they do, is it reasonable for them to do so? What is the structure of value?
Might goodness be intransitive? We did not touch on these questions. As I said, my task has been
more modest. This has been an exercise in taxonomy. I categorized a wide variety of scenarios in the
literature into four structures and added some policy-relevant examples. This, I am sure, will already
prove sufficiently controversial. And when addressing the grand philosophical questions, it may
prove helpful that the cases under consideration are arranged in boxes so that like cases can be
addressed jointly.3
Notes
1 Sen himself doesn’t use the term “paradox” and introduces the principle of “Minimal Libertarianism” as the
principle of “Minimal Liberalism” in Sen (1970a, 1970b), but he shifts to “Minimal Libertarianism” in Sen
(1976).
2 Rachel’s Ecstasy example (1998) has a Downhill Slide structure. I leave this as an exercise for the reader.
3 I am grateful for comments from Ryan Doody, Chris Melenovsky, Daniel Muñoz, and Wlodek Rabinowicz.
References
Amartya Sen. (1970a) Collective Choice and Social Welfare. Volume 11 in Advanced Textbooks in Economics.
Amartya Sen. (1970b) The Impossibility of a Paretian Liberal. Journal of Political Economy, 78, 152–157.
Amartya Sen. (1976) “Liberty, Unanimity and Rights.” Economica, 43, 217–245.
Amos Tversky. (1969) “Intransitivity of Preferences.” Psychological Review, 76, 1, 31–48.
Chrisoula Andreou. (2006) “Environmental Damage and the Puzzle of the Self-Torturer.” Philosophy and Public
Affairs, 34, 1, 95–108.
Dorothy Edgington. (1992) “Validity, Uncertainty and Vagueness.” Analysis, 52, 4, 193–204.
Dorothy Edgington. (1996) “Vagueness by Degrees.” in Rosanna Keefe and Peter Smith eds. Vagueness: A
Reader. Cambridge, MA; MIT Press, pp. 294–315.
Erik Carlson. (1996) “Cyclical Preferences and Rational Choice.” Theoria, 62, 144–160.
Frances Myrna Kamm. (1985) “Supererogation and Obligation.” Journal of Philosophy, 82, 3, 118–138.
92
Garrett Hardin. (1968) “Tragedy of the Commons.” Science, 162, 3859, 1243–1248.
George F. Schumm. (1987) “Transitivity, Preference and Indifference.” Philosophical Studies: An International
Journal for Philosophy in the AnalyticTradition, 52, 3, 435–437.
John Broome. (1991) Weighing Goods. Oxford, UK; Blackwell.
Jonathan Harrison. (1952–3) “Utilitarianism, Universalisation, and Our Duty to Be Just.” Proceedings of the
Aristotelian Society—New Series, 53, 105–134.
Kenneth O. May. (1954) “Intransitivity, Utility, and the Aggregation of Preference Patterns.” Econometrica, 22,
1, 1–13.
Larry S. Temkin. (1996) “A Continuum Argument for Intransitivity.” Philosophy & Public Affairs, 25, 3,
175–210.
Luc Bovens. (1990) Reasons for Preferences. PhD. Dissertation. University of Minnesota.
Luc Bovens. (1993) “Contextual Pluralism and the Libertarian Paradox.” Archiv Für Rechts Und Sozialphilosophie,
79, 2, 188–197.
Luc Bovens. (1994) “Coherence Arguments and Cyclical Moral Rankings.” Philosophical Studies, 74, 3,
369–384.
Luc Bovens. (2015) “The Tragedy of the Commons as a Voting Game.” in Martin Peterson ed. The Prisoner’s
Dilemma. Classic philosophical arguments. Cambridge, MA; Cambridge University Press. pp. 156–176.
Maya Bar-Hillel and Avishai Margalit. (1988) “How Vicious Are Cycles of Intransitive Choice?” Theory of
Decision, 24, 119–145.
Paul Slovic and Douglas MacPhillamy. (1972 [1971]) “Dimensional Comparability and Cue Utilization in
Comparative Judgment.” Research Bulletin, 11, 14, https://scholarsbank.uoregon.edu/xmlui/handle/1794/
22309
Rebecca Livernois. (2018) “Regretful Decisions and Climate Change.” Philosophy of the Social Sciences, 48, 2,
168–191.
Stuart Rachels. (1998) “Counterexamples to the Transitivity of Better Than.” Australasian Journal of Philosophy,
76, 1, 71–83.
Warren Quinn. (1990) “The Puzzle of the Self-Torturer.” Philosophical Studies: An International Journal for
Philosophy in the Analytic Tradition, 59, 1, 79–90.
93
8
THEORIES OF CHOICE BEHAVIOR
Sudeep Bhatia
Understanding how people make choices is of crucial interest to researchers across many academic
disciplines, including economics, psychology, and neuroscience (Camerer et al., 2011; Glimcher and
Fehr, 2013; Oppenheimer and Kelso, 2015; Weber and Johnson, 2009). Choice behavior is also
central to the formulation and implementation of government policy (Halpern, 2016; Thaler and
Sunstein, 2009). Thus, unsurprisingly, scientists have developed numerous mathematical and
computational theories that can be used to describe and predict choice behavior. These theories take
as inputs the set of choice options available to the decision maker and produce as outputs de-
terministic or probabilistic predictions of choice. Different theories make different assumptions
about how the outcomes offered by choice options are evaluated, transformed, and aggregated to
determine choice. This chapter will summarize these assumptions and review prominent theories in
three key domains of decision-making: multiattribute choice (e.g., Keeney and Raiffa, 1993; Russo
and Dosher, 1983), intertemporal choice (e.g., Loewenstein and Prelec, 1992; Mazur, 1987;
Samuelson, 1937), and risky choice (e.g., Kahneman and Tversky, 1979; Von Neumann and
Morgenstern, 1944). Researchers have also developed theories for other domains, such as social and
strategic choice (e.g., Camerer et al., 2011; Fehr and Schmidt, 1999); however, the three domains
considered in this chapter have received by far the most attention and are the focus of the best
known and most influential theoretical work on choice behavior.
1 A Brief History
Although the study of choice behavior now falls within the purview of many different academic
disciplines, its roots are in an area of psychology known as behavioral decision theory (Edwards,
1954, 1961; Einhorn and Hogarth, 1981; Rapoport and Wallsten, 1972). Behavioral decision theory
started in the 1950s as a response to economic theories of individual decision-making, particularly
decision-making under risk and uncertainty. Economists offered sophisticated approaches to study
choice, such as expected utility theory (Von Neumann and Morgenstern, 1944), and economic
theories were both mathematically elegant and normatively compelling. These approaches came to
the attention of psychologists, who focused on developing experimental tools with which to test the
assumptions and recover the parameters associated with these theories.
At first, behavioral decision theory approached standard economic theories with cautious opti-
mism. In one of the earliest reviews in this field, Ward Edwards stated that “the psychologically
important question is: Can [expected utility theory] be used to account for simple experimental
94 DOI: 10.4324/9780367808983-10
Theories of Choice Behavior
examples of risky decisions?” (Edwards, 1954, 394). By the 1970s the answer to this question was
clear. Individual behavior displayed systematic, pervasive, and robust departures from expected
utility theory and its economic and rational counterparts in domains such as multiattribute and
intertemporal choice. This rejection of standard theories stimulated the development of behavioral
theories formulated to describe choice behavior as it occurred, with a focus on predicting empirical
violations of standard economic theories. The best known behavioral theory, prospect theory, was
published in 1979 (Kahneman and Tversky, 1979), and was soon accompanied by many rival
theories of risky choice (Bell, 1985; Loomes and Sugden, 1982, 1986; Rubinstein, 1988; Thorngate
1980; Yaari, 1987). Since then, the number of different behavioral decision theories has grown
rapidly and they have been an important factor in the development and popularization of new fields
such as behavioral economics, judgment and decision-making, and neuroeconomics (Camerer et al.,
2011; Glimcher and Fehr, 2013; Oppenheimer and Kelso, 2015; Weber and Johnson, 2009).
2 The Choice Problem

Empirically, theories of decision making are tested on their ability to describe people’s choices, often
observed in laboratory experiments or surveys involving real or hypothetical choice options. In such
experiments, decision makers are asked to select one or more options from a larger choice set
consisting of multiple options. In multiattribute choice, options are defined on distinct character-
istics, features, or attributes. For example, choices between laptops may involve attributes such as
memory, hard drive space, and battery life. In such choices, decision makers need to trade-off
relevant attributes in order to select the most preferred option. Decision theories, in turn, attempt to
describe the mathematical rules and mental processes involved in making these tradeoffs.
The outcomes in most multiattribute choice settings are typically obtained at the time of choice
and with certainty. By contrast, intertemporal choice settings involve choice outcomes occurring
with differing degrees of delay, and risky choice settings involve outcomes occurring with differing
degrees of likelihood. Examples of the former include choices between a larger monetary outcome
occurring in the future and a smaller outcome occurring in the present, whereas examples of the
latter include choices between a larger outcome occurring with a small probability and a smaller
outcome occurring with certainty. Again, the goal of the decision maker is to trade off the size of the
monetary outcome with the delay or risk involved in obtaining the outcome, and decision theories
try to describe how these tradeoffs are made by decision makers, and how outcomes, delays, and
probabilities are evaluated and aggregated to determine choice. Note that risky and intertemporal
choices are sometimes applied to outcomes other than monetary payoffs (such as health states) in
applied work, though such applications are less common in theoretical behavioral research.
Formally, the choice set offered to the decision maker can be understood as a collection of N
choice options, x1, x2, … xN. Each of these options is itself described as a vector or list of some M
outcomes, with xi= [xi1, xi2, … xiM]. In multiattribute choice, these outcomes usually correspond to
the amounts of different attributes, with xij capturing the amount of an attribute j in a choice option i
(e.g., 8GB RAM in an available laptop). In intertermporal choice, these outcomes usually corre-
spond to monetary payoffs obtained at different points in time (which could occur with some delay),
and these outcomes correspond with states of the world (which could occur with probability less
than one) with risky choice. Here, xij captures the jth outcome in option i, and is associated with
some time delay, tij, or probability, pij.
Of course, most real-world decisions involve a combination of multiple attributes, delay, and risk, as
well as richer information structures that are hard to describe using the above structure. For example,
deciding whether to eat a cheeseburger or a salad for lunch involves numerous taste and nutrition-
related tradeoffs, combined with risk and time delays in health outcomes such as heart disease, as
well as complex social and moral considerations, and various forms of temptation and self-control.
95
Sudeep Bhatia
Such decisions are currently too complex to be modeled using formal mathematical and computational
theories. Thus, to reduce complexity and increase tractability, the dominant approach in decision-
making research has been to develop separate theories of multiattribute, intertemporal, and risky
choice. This can be seen as a “divide and conquer” tactic, with different theories specifying each of the
different aspects of choice, and researchers and research programs focusing on individual theories and
domains.
3 Standard Theories
Common to multiattribute choice, intertemporal choice and risky choice are standard theories that
historically have been advanced by economists and have gained popularity due to their mathematical
elegance, analytical simplicity, and normative properties. These theories assign utility values to options
based on a weighted average over the options’ outcomes. For an option xi, this utility is given by u
(xi) = Σwij∙v(xij), with wij corresponding to a weight assigned to the outcome j in option i, and v(xij)
corresponding to the subjective value of outcome j in option i. For a given choice set, the option with
the highest utility (i.e., the highest weighted aggregate subjective value) is assumed to be chosen.
In the standard theory of multiattribute choice, known as the weighted additive model,
weights capture the relative importance of the attributes and values capture transformed mental
representations of attribute quantities (Keeney and Raiffa, 1993). For example, a decision maker that
cares a lot about memory would attach a high weight, wij, to the memory attribute in a laptop.
Additionally, a given laptop’s memory would be transformed into a numeric value with the function
v(). Thus wij∙v(8) for the memory attribute would represent the utility offered by xij = 8GB RAM in
an option. This utility would be added to the utilities of all the other attributes in option xi to
determine u(xi).
In intertemporal choice, the standard theory is discounted utility. In this theory, weights are
exponential functions of the time delays and values capture transformed representations of monetary
outcomes (Samuelson, 1937). Consider, for example, an option offering an immediate monetary
outcome of $50 as well as an outcome of $100 in 5 months. Here the weights associated with the
two time points would be given by a discounting function δt (with δ < 1 being a flexible discount
factor that determines how strongly the decision maker discounts future outcomes, and t being the
time delay for the outcome). This would result in weights of δ0 = 1 and δ5, respectively, and the
values would be determined by a function v() that transforms monetary outcomes, xij, into utilities.
Utility for option xi would thus be given by u(xi) = δ0∙v(50) + δ5∙v(100).
Finally, in risky choice, the standard theory is expected utility theory, where weights are equal
to state probabilities and values (as in intertemporal choice) capture transformed representations of
monetary outcomes (Von Neumann and Morgenstern, 1944). Consider, for example, an option
offering an outcome of $100 with a 20% chance and an outcome of $10 with an 80% chance. Here
the probabilities would be used simplistically as weights, and a value function v() would be used to
transform monetary outcomes into utilities. Utility for option xi would thus be given by u(xi) = 0.2∙v
(100) + 0.8∙v(10). Note that both discounted utility theories of intertemporal choice and expected
utility theories of risky choice are assumed to operate over final wealth states generated by choice
options and not the individual payoffs offered by the options.
As mentioned earlier, the weighted additive model, discounted utility theory, and expected
utility theory are sometimes considered to be normatively correct, as their functional forms can be
derived from seemingly self-evident principles of rational choice. For example, the weighted ad-
ditive model stems from simple assumptions of transitivity across options and independence of utility
across attributes (Keeney and Raiffa, 1993). Transitivity implies that decision makers who choose an
option x1 over an option x2 and option x2 over option x3 should not choose option x3 over option
x1. In other words, choice cycles should be impossible. Transitivity also often implies that choices
96
between options should not be influenced by irrelevant options (i.e., options that would never be
chosen by decision makers). Independence axioms propose that the quantities of attributes in a
choice option should not influence how the quantities of other attributes in that choice option are
evaluated. Discounted and expected utility theories further apply such independence and transitivity
axioms (along with additional axioms) to outcomes in different time periods and states of the world
(Samuelson, 1937; Von Neumann and Morgenstern, 1944). However, note that the normative
status of these theories is somewhat controversial, and the rationality of choice axioms can be cri-
ticized on numerous philosophical grounds (see Sugden, 1991, for a discussion).
Standard theories are also frequently criticized on descriptive grounds, since, as predictors of
experimentally observed choice patterns these theories are completely inadequate. In his now-classic
work, Tversky (1969) showed that decision makers can sometimes violate transitivity in carefully
constructed choice sets. Related work (Huber et al., 1982) has shown reversals in choice as nor-
matively irrelevant options (e.g., dominated options that are worse than others on all attributes) are
added or removed from the choice set. These findings cast doubt on the core assumptions of the
weighted additive model.
Independence axioms have also been shown to be violated in both intertemporal and risky
choice. For example, monetary outcomes in a given time period can influence how outcomes in
subsequent time periods are evaluated, and decision makers often display a preference for increasing
sequences of payoffs (Loewenstein and Sicherman, 1991). Decision makers also tend to display a
disproportionately strong preference for immediate payoffs, which cannot be captured by the ex-
ponential discounting function of discounted utility theory (Green et al., 2005). Likewise, monetary
outcomes in one state of the world can influence how outcomes in other states of the world are
evaluated, and decision makers often display a disproportionately strong preference for outcomes
that are obtained with certainty (Allais, 1953; Kahneman and Tversky, 1979). These patterns cannot
be captured by the functional form of expected utility theory. Relatedly, decision makers have been
shown to reverse choices between risky options as their underlying wealth state is changed, and to
display absurdly high degrees of risk aversion when offered small-scale gambles with both gains and
losses. Both these patterns are hard to accommodate within expected utility theory (Kahneman and
Tversky, 1979; Rabin and Thaler, 2001).
4 Prominent Behavioral Theories

These, and other descriptive limitations, have stimulated the development of behavioral theories of
choice whose primary goal is to describe choice behavior as it is observed in the laboratory. Perhaps
the best-known behavioral theory is prospect theory, which was proposed as an alternative to
expected utility theory in the domain of risky choice (Kahneman and Tversky, 1979; Tversky and
Kahneman, 1992). Prospect theory departs from the assumptions of expected utility theory in two
crucial ways. First, it assumes that probabilities are transformed before being used as decision
weights. The transformations overweigh small probabilities and underweigh large probabilities.
Thus, for an option offering an outcome of $100 with a 20% chance and an outcome of $10 with an
80% chance, prospect theory would assign a disproportionately high weight to the first outcome and
a disproportionately low weight to the second outcome. Utility for this option could subsequently
be u(xi) = 0.3∙v(100) + 0.7∙v(10) rather than u(xi) = 0.2∙v(100) + 0.8∙v(10).
Such assumptions allow prospect theory to explain many of the independence axiom violations in
risky choice, summarized in the previous section. Prospect theory also assumes loss aversion in its
value function, that is, the property that losses have a larger impact on utility than equivalently sized
gains. Thus, for example, a loss of $100 would have an absolute value that is higher than that of a
gain of $100, that is, |v(−100)| > |v(100)|. This assumption, when combined with additional
assumptions about the concavity of the value function in losses and gains, allows prospect theory to
97
Sudeep Bhatia
explain preference reversals caused by changes to underlying wealth states and to explain why people
appear to be extremely risk averse for small-scale gambles offering a combination of gains and losses.
Another prominent behavioral theory of decision-making is hyperbolic discounting
(Loewenstein and Prelec, 1992; Mazur, 1987). This theory applies to intertemporal choice and
assumes that the discount function used by individuals is not exponential, but rather hyperbolic.
Consider again an option offering an immediate monetary outcome of $50 as well as an outcome of
$100 in 5 months. According to Mazur’s (1987) discount function, which assumes that the weight
1
on an outcome is given by (1 + t ) / (with α and β being free parameters and t being the time delay),
1
the weight on the first outcome would be (1 + ) / but the weight on the second outcome would be
1
(1 + 5) / . Such hyperbolic functions can display a steeper dropoff than exponential functions for
smaller time delays, allowing the theory of hyperbolic discounting to explain people’s dis-
proportionately strong preference for immediate payoffs over delayed payoffs.
Finally, one well-known theory of multiattribute choice is the tallying heuristic (Payne et al.,
1988; Russo and Dosher, 1983). Heuristics are shortcuts for making choices that simplify the de-
cision process by ignoring information. According to the tallying heuristic, decision makers do not
weigh and aggregate the attribute values of choice options, as proposed by the weighted additive
model. Rather, they compare choice options on individual attributes and consider only whether one
option is sufficiently better than the other on that attribute. Choice, in this sense, involves taking the
option that is the best on the most number of attributes. For example, in the choice between a
laptop with 8GB RAM, 1TB hard drive, and 10 hours of battery life, and a laptop with 16GB
RAM, 1.5TB hard drive, and 5 hours of battery life, the tallying heuristic predicts that the decision
maker would choose the second option as it is superior on two out of the three attributes (by
contrast, a weighted additive model could attach a higher weight to battery life and pick the first
option). The types of operations assumed by heuristics like tallying can explain violations of tran-
sitivity that have been observed in multiattribute choice, as well as other empirical patterns in
multiattribute choice.
5 Key Behavioral Properties

Theories like prospect theory, hyperbolic discounting, and the tallying heuristic, accommodate
observed patterns in choice behavior by modifying one or more components of standard economic
theories. Typically, these modifications involve additional assumptions about how outcomes and
attributes in the choice options are compared and aggregated. In this section, we provide an
overview of the assumptions that are commonly applied in behavioral theories of decision-making.
5.1 Interactions Within Options

One of the most important assumptions of behavioral theories of choice is that the outcomes within
a single choice option can interact, leading to violations of independence axioms that are proble-
matic for standard theories of decision making. In risky choice, such interactions sometimes involve
the influence of payoff ranks within an option on the weights and values associated with those
payoffs, as in the transfer of exchange model (Birnbaum, 2008). Such interactions are also common
in variance-based theories (Markowitz, 1959; Weber et al., 2004), in which an option’s utility is
influenced by the overall variance of its monetary outcomes, as well as in disappointment-based
theories (Bell, 1985; Loomes and Sugden, 1986), in which a given outcome is evaluated relative to
the expected or average outcome in the option. Finally, within-option interactions appear in
heuristic theories of risky choice, such as Payne and Braunstein’s (1971) information processing
98
model and the priority heuristic (Brandstätter et al., 2006). These theories identify the largest or
smallest outcome in an option or the most likely or least likely outcome in an option, so that
changing one probability or payoff can affect how other probabilities and payoffs in the same option
are evaluated.
In contrast to risky choice, within-option interactions are relatively rare in behavioral theories of
multiattribute and intertemporal choice. However, configural weight theories of multiattribute
choice do allow for interactions between the outcomes in an option, such as through multiplication
or rank-based comparison (Birnbaum, 1974; Birnbaum and Zimmermann, 1998). Within-option
interactions are also central to neural network theories, such as the parallel constraint satisfaction
model of multiattribute choice (Glöckner and Betsch, 2008). Neural network theories allow for
attribute values to influence the activation of other attributes, and through this form of interaction,
bias choice. Last, conjunctive and disjunctive heuristics, which assume that decision makers focus on
the best or worst attributes within a choice option (Dawes, 1964) can display these interactions, as
can approximations of these heuristics that represent the decision process with various non-linear
utility functions (Einhorn, 1970).
In intertemporal choice, theories that allow for within-option interaction include the preferences
over sequences model (Loewenstein and Prelec, 1993), which proposes that increasing and decreasing
sequences of outcomes within an option are evaluated differently. Satiation and habit formation
theories (Baucells and Sarin, 2010) also allow for outcomes experienced at one point in time to
influence the desirability of outcomes at other points in time. Note, however, that much empirical
research has avoided the issue of intertemporal and multiattribute violations of independence by fo-
cusing on settings in which such interactions cannot occur (e.g., in choices involving a single delayed
outcome) or in which such interactions are less likely to occur (e.g., in choices involving attributes that
are largely independent of each other).
5.2 Interactions Between Options

Many behavioral theories also allow for outcomes and attributes to interact across options. This can
lead to violations of transitivity, as an option’s utility depends on the option it is being compared
against or on the presence or absence of irrelevant options in the choice set. Unlike interactions
within options, interactions between options are quite commonly studied in the multiattribute
choice domain. Multiattribute theories that assume between-option interactions include many
heuristics, such as the tallying heuristic (Russo and Dosher, 1983) discussed above, as well as variants
such as the lexicographic and lexicographic semiorder heuristics (e.g., Tversky, 1969). These the-
ories typically evaluate options based on the rank of attribute amounts across the choice set. Other
theories that utilize rank-based comparisons are the rank-weighted leaky accumulator (Tsetsos et al.,
2012), which aggregates an option’s utility with higher weights for attributes on which the option is
highly ranked, and the dominance model (Ariely and Wallsten, 1995), which gives options that
dominate others (i.e., have higher ranks for all attributes) disproportionately higher utilities.
Interactions between options are also present in many multiattribute choice theories that em-
phasize the role of option similarities, such as the feature matching model (Houston et al., 1989), the
focusing model (Kőszegi and Szeidl, 2013), the similarity contrast model (Mellers and Biagini,
1994), and multiattribute salience theory (Bordalo et al., 2013). All of these theories predict that
adding or removing options from the choice set, alters the perceived similarities of the remaining
options, which, in turn, may alter the choice between them. Typically, attributes on which options
are highly dissimilar to each other are given higher attention and thus higher weights.
We also observe between-option interactions in multiattribute theories that are based on loss
aversion, such as the componential context model (Tversky and Simonson, 1993) and the loss
aversion-based leaky competitive accumulator (Usher and McClelland, 2004). These theories
99
Sudeep Bhatia
assume that choice options are treated as gains or losses relative to other options in the choice set,
with a utility-penalty for an option that is perceived as a loss. Additionally, multiattribute theories
that utilize the distribution on a single attribute (e.g., the range of values for that attribute in the
choice set) to calculate the utilities of different options, display between-option interactions. These
include range-frequency theory (Parducci, 1974; Wedell and Pettibone, 1996) and the neuro-
computational range-normalization model (Soltani et al., 2012). In these theories, the distribution of
attributes in a choice set can be changed by adding or removing choice options, altering how each
individual attribute value is perceived and evaluated.
Many theories developed by cognitive scientists and neuroscientists assume that choice options
inhibit the activations of other options in the choice set, with highly preferred competitors reducing
a given option’s utility and choice frequency. Such theories include multi-alternative decision field
theory (Roe et al., 2001) and its many variants, as well as the attractor network model (Wang, 2002)
and the divisive normalization model (Louie et al., 2013). Between-option interactions can also be
observed in the associative accumulation model (Bhatia, 2013), and in other neural network theories
(Glöckner and Betsch, 2008). These form a related class of cognitive models, in which the attributes
in the choice options dynamically alter the activation of available options.
Some risky choice theories also allow for between-option interactions. The best-known example
of this is regret theory (Loomes and Sugden, 1982), which assumes that decision makers evaluate
outcomes in a given state of the world based on the outcomes offered by competing options.
Options that offer comparatively worse outcomes in a potential state of the world allow for the
possibility of regret. This type of assumption is also at play in many other prominent theories of risky
choice, including the perceived relative argument model (Loomes, 2010), salience theory (Bordalo
et al., 2012), the importance sampling model (Lieder et al., 2017), decision affect theory (Mellers
et al., 1999), and the similarity models of Leland (1994) and Rubinstein (1988).
Heuristic theories of risky choice also often involve between-option interactions. Such theories
include the minimax and maximax heuristics (Thorngate 1980). The former selects the option that
offers the highest minimum payoff across all states of the world, whereas the latter selects the option
that offers the highest maximum payoff across all states of the world. Variants of this assumption are
also found in other heuristic theories, such as the priority heuristic (Brandstätter et al., 2006), as well
as in sampling-based theories such BEAST (Erev et al., 2017) and the decision by sampling model
(Stewart et al., 2006).
In intertemporal choice, some theories assume that the differences between monetary outcomes
determine preference. This is the case with the trade-off model (Scholten and Read, 2010), the
absolute and relative differences dynamic models (Dai and Busemeyer, 2014), as well as similarity-
based models such as those of Leland (2002). Likewise, both the ITCH model (Ericson et al., 2015)
and the DRIFT model (Read et al., 2013) involve between option interactions, in that the utility for
an option is determined by the rate of return provided by that option compared to other options
in the choice set, as well as the differences between the monetary outcomes of different options.
5.3 Transformations of Values

Many behavioral theories transform the outcomes of choice options using a valuation function v()
not permitted by standard theories. The best known of these are theories of regret (Loomes and
Sugden, 1982), in which the values attached to monetary outcomes depend on the outcomes offered
by other options. For example, an outcome of $20 offered by a gamble in one state of the world
could be given low value if another gamble offers $30 in that state of the world, or be given a high
value if another gamble offers $10 in that state of the world. Similarly, theories of disappointment
(Bell, 1985; Loomes and Sugden, 1986) transform monetary outcomes relative to other outcomes in
the same option. For example, an outcome of $20 offered by a gamble in one state of the world
100
could be given low value if the same gamble offers $30 in a different state of the world, or be given a
high value if that same gamble offers $10 in a different state of the world. Decision affect theory
(Mellers et al., 1999), which allows for both regret and disappointment, also involves such trans-
formations. Likewise, theories that utilize the variance of an option to compute an option’s utility
(Markowitz, 1959; Weber et al., 2004) also entail value transformations. These theories penalize
outcomes that diverge strongly from the mean.
We also find value transformations in many theories of intertemporal choice. Prominent ex-
amples include the trade-off model (Scholten and Read, 2010), the DRIFT model (Read et al.,
2013), dynamic models (Dai and Busemeyer, 2014), and the ITCH model (Ericson et al., 2015).
These theories typically apply transformations to differences or ratios between monetary outcomes
across options. Value transformations are also a feature of sequence theories of intertemporal choice
(Loewenstein and Prelec, 1993), in which people have preferences for increasing, decreasing, and
dispersed sequences of consumption.
The componential context model (Tversky and Simonson, 1993) and the loss-averse leaky
competitive accumulator model (Usher and McClelland, 2004) are multiattribute theories that
feature value transformations. Here the attribute amounts for one option are evaluated based on
whether they are gains or losses relative to attribute amounts for other options. Many other mul-
tiattribute theories also involve transformations of attributes based on pairwise comparisons. These
include the additive difference rule (Tversky, 1969), the similarity contrast model (Mellers and
Biagini, 1994), the stochastic difference model (González-Vallejo, 2002), the nonlinear model
(Einhorn, 1970), the options as information model (Sher and McKenzie, 2014), and the decision by
sampling model (Stewart et al., 2006). Range-based multiattribute theories such as range-frequency
theory (Parducci, 1974; Wedell and Pettibone, 1996), and the range-normalization model (Soltani
et al., 2012), also often apply such transformations to the values of choice outcomes.
5.4 Transformations of Weights

Behavioral theories that assume weight-based transformations modify the weights, wij, put on each
outcome based on the other outcomes of the same option, or the outcomes of other options. These
modifications are not permitted in standard theories. Risky choice theories that involve weight-
based transformations include prospect theory and its variants (Schmidt et al., 2008; Tversky and
Kahneman, 1992; Yaari, 1987). Probability weighting in these theories often depends on the rank of
the outcomes that they correspond to, compared to the other outcomes in the same option. Many
such transformations do not involve the probability by itself, but rather differences in weighted
cumulative probabilities.
Other risky choice theories with weight-based transformations are salience theory (Bordalo et al.,
2012) and the importance sampling model (Lieder et al., 2017), in which probability weights depend
on the differences between outcome values in corresponding states of the world. These theories
assume that people pay more attention to (and thus place a higher weight on) states of the world that
yield highly diverging payoffs across options. Additionally, theories such as prospective reference
theory (Viscusi, 1989), the dual-systems model of risk (Mukherjee, 2010), the dual-systems model of
affect and deliberation (Loewenstein et al., 2015), the noisy retrieval model (Marchiori et al., 2015)
and distracted decision field theory (Bhatia, 2014), modify probabilities by combining them with a
uniform distribution. This assumption flattens the slope of the probability weights, leading to de-
viations from the expected utility theory.
Many multiattribute choice theories also feature weight transformations. For example, the as-
sociative accumulation model weighs attributes based on the presence or absence of these attributes
in other choice options (Bhatia, 2013). This is also a feature of neural network theories of multi-
attribute choice, in which this type of relationship is dynamic, and depends on the preferences for
101
Sudeep Bhatia
the options in consideration (Glöckner and Betsch, 2008). Other theories of multiattribute choice
that involve weight transformations include the feature matching model (Houston et al., 1989) and
multiattribute salience theory (Bordalo et al., 2013), which assume that attributes that involve large
differences are more salient and subsequently given higher weights. Finally, configural weight
theories (Birnbaum, 1974; Birnbaum and Zimmermann, 1998) assume interactions between out-
comes that take the form of biased attribute weights.
Intertemporal choice theories that utilize weight-based transformations include theories which
compare time delays of the options with each other, such as the trade-off model (Scholten and
Read, 2010), the proportional difference model (Cheng and González-Vallejo, 2016), the ITCH
model (Ericson et al., 2015), and the focusing model (Kőszegi and Szeidl, 2013). These theories all
transform delays based on the other delays (or outcomes) in the choice set.
5.5 Choice Processes

The standard theories introduced in the prior sections assume that choice is the product of utility
maximization. Such theories do not attempt to describe the mental steps (e.g., visual attention to
information or recall of experiences from memory) that determine the formation of utility, and
subsequently are unable to capture choice-relevant variables such as response time. This is also the
case with prominent behavioral theories such as prospect theory and hyperbolic discounting. Indeed,
nearly all behavioral theories proposed within economics abstract away from the specific psycho-
logical and cognitive processes involved in the calculation of utility and the generation of choice.
By contrast, psychologists and neuroscientists have focused extensively on these processes, and
have attempted to build into their theories, assumptions about the sequence of operations involved
in the calculation of utility. Heuristic theories (e.g., Brandstätter et al., 2006; Russo and Dosher,
1983; Payne and Braunstein, 1971; Tversky, 1969, 1972), for example, describe choice as the
outcome of a series of algorithmic steps, of the type that could be implemented in computer
programs. Such steps involve sequential attention to the available attributes or outcomes, com-
parison processes that binarize differences in attributes or outcomes across choice options, and simple
decision rules that specify when the algorithm should terminate and choice should be made (see
Gigerenzer and Gaissmaier, 2011; Payne et al., 1992, for an overview).
Another prominent approach in psychology and neuroscience involves theories of preference
accumulation (e.g., Bhatia, 2013; Busemeyer and Townsend, 1993; Roe et al., 2001). Such theories
assume that decision makers dynamically update a (potentially stochastic) utility variable, based on
attention to information about the choice options. Thus, as decision makers observe the various
attributes and outcomes offered in the choice set, their utilities for the available options rise or fall.
Choice is made when the utility of an option surpasses a threshold value. Such theories have had
great success in describing how response time varies as a function of the options offered to decision
makers, and how choice reflects attention and memory processes (Dai and Busemeyer, 2014;
Krajbich et al., 2010; Tsetsos et al., 2012; Usher and McClelland, 2004). Variants of these theories
have also been investigated from a neurobiological perspective, and have been shown to describe
activation patterns in neurons during choice (Gold and Shadlen, 2007; Ratcliff and McKoon, 2008).
6 Conclusion
The study of choice is central to our understanding of human behavior, and has shaped our un-
derstanding of markets and societies, of psychological processes and mental disorders, and of the
biological bases of behavior. Much of this work has relied on formal mathematical and computa-
tional theories to describe choice processes, predict choice outcomes, and interpret the relationship
between choices and various economic, psychological, and neurobiological variables.
102
In this chapter, we have provided an overview of theories in the domains of multiattribute, in-
tertemporal, and risky choice. We have started by examining three standard theories in these domains:
the weighted additive model (Keeney and Raiffa, 1993), discounted utility theory (Samuelson, 1937),
and expected utility theory (Von Neumann and Morgenstern, 1944). These theories are normatively
compelling and mathematically elegant and have been the focus of much empirical and theoretical
work over the past seven decades. This work has found that standard theories fail to describe the full
spectrum of behaviors observed in laboratory experiments and surveys.
In response, researchers have developed behavioral theories that attempt to describe observed
violations of standard theories by making additional assumptions about how choice outcomes are
evaluated and aggregated into utilities. We have examined three such theories in detail: prospect
theory (Kahneman and Tversky, 1979) in the domain of risky choice, hyperbolic discounting
(Loewenstein and Prelec, 1992; Mazur, 1987) in the domain of intertemporal choice, and the
tallying heuristic (Russo and Dosher, 1983) in the domain of multiattribute choice. We have also
provided a broad overview of additional assumptions that are made by various behavioral theories.
These include assumptions about interactions within and between options, assumptions about the
transformation of values and weights, and assumptions about the mental steps involved in delib-
eration and choice. Examples of theories that have modified these basic assumptions are listed in
Table 8.1. Most of these assumptions (and associated theories) involve violations of normatively
compelling axioms (such as transitivity and independence) that have motivated the standard theories
discussed above, though these violations are necessary in order to accurately predict human beha-
vior. It is this author’s belief that the use of descriptive behavioral theories should be prioritized over
normatively desirable standard theories, especially in policy applications and other applied settings
involving human behavior. Theories that specify choice processes, and describe the mental steps at
play in deliberation and decision making, may also be better suited for such applications. These
theories offer rigorous insights about how people perceive, evaluate, and aggregate the information
Table 8.1 Examples of modifications to theories of choice behavior
Multiattribute theory Risky choice theory Intertemporal choice theory
Prominent The Tallying Hueristic Prospect Theory Hyperbolic Discounting

Theories (§4) ( Payne et al., 1988) ( Kahneman and ( Loewenstein and
Tversky 1979) Prelec, 1992)
Interactions Within Rank-based Comparison Transfer of Exchange Preferences Over
Options (§5.1) ( Birnbaum and Model( Birnbaum, Sequences Model
Zimmermann, 1988) 2008) ( Loewenstein and
Prelec, 1993)
Interactions Between Lexicographic Semiorder Regret Theory ( Loomes The Tradeoff Model
Options (§5.2) Heuristics( Tversky, and Sugden, 1982) ( Scholten and Read,
1969) 2010)
Transformation of Componential Context Disappointment Theory The DRIFT Model ( Read
Values (§5.3) Model( Tversky and ( Bell, 1985) et al., 2013)
Simonson, 1993)
Transformation of The Associative The Importance Sampling The Proportional
Weights (§5.4) Accumulation Model Model ( Lieder et al., Difference Model
( Bhatia, 2013) 2017) ( Cheng and González-
Vallejo, 2016)
Choice Hueristic Theories (e.g., Brandstätter et al., 2006)
Processes (§5.5) Preference Accumulation Theories (e.g., Bhatia, 2013)
103
Sudeep Bhatia
involved in the decision, and can be used to understand how such choice processes can be influ-
enced and improved.
At each point in this chapter, we have attempted to give examples of well-known behavioral
theories reflecting common assumptions. Yet, there are far more theories in the literature than we
can cover in this chapter. Each existing theory involves a rich set of mathematical or computational
assumptions, and is studied in the context of a large set of empirical regularities. The parameters of
these theories are also often fit to choices, and the theories themselves are evaluated in terms of their
ability to quantitatively predict the choices. Interested readers could examine Starmer (2000) for an
overview of theories of risky choice, Frederick et al. (2002) for an overview of theories of inter-
temporal choice, Weber and Johnson (2009) for an overview for theories of choice process,
Gigerenzer and Gaissmaier (2011) for an overview of heuristic theories, Busemeyer et al. (2019) for
an overview of preference accumulation models, and He, Zhao, and Bhatia (2020) for metascientific
work attempting to integrate distinct theories within a single conceptual structure. Building and
testing formal theories of choice behavior has been an active area of academic inquiry for decades,
and with increased interest in theoretical decision making from fields such as cognitive science and
neuroscience, and growing applications to policy and practical problems, it is likely that this interest
will continue for many years to come.
References
Allais, M. (1953) “L’extension des théories de l’équilibre économique général et du rendement social au cas du
risque,” Econometrica, 21, 4, 269–290.
Ariely, D., & Wallsten, T. S. (1995) “Seeking subjective dominance in multidimensional space: An explanation
of the asymmetric dominance effect,” Organizational Behavior and Human Decision Processes, 63, 3, 223–232.
Baucells, M., & Sarin, R. K. (2010) “Predicting utility under satiation and habit formation,” Management Science,
56, 2, 286–301.
Bell, D. E. (1985) “Disappointment in decision making under uncertainty,” Operations Research, 33, 1, 1–27.
Bhatia, S. (2013) “Associations and the accumulation of preference,” Psychological Review,120, 3, 522.
Bhatia, S. (2014) “Sequential sampling and paradoxes of risky choice,” Psychonomic Bulletin and Review, 21, 5,
1095–1111.
Birnbaum, M. H. (1974) “The nonadditivity of personality impressions,” Journal of Experimental Psychology, 102,
543–561.
Birnbaum, M. H. (2008) “New paradoxes of risky decision making,” Psychological Review, 115, 2, 463.
Birnbaum, M. H., & Zimmermann, J. M. (1998) “Buying and selling prices of investments: Configural weight
model of interactions predicts violations of joint independence,” Organizational Behavior and Human Decision
Processes, 74, 2, 145–187.
Bordalo, P., Gennaioli, N., & Shleifer, A. (2012) “Salience theory of choice under risk,” The Quarterly Journal of
Economics, 127, 3, 1243–1285.
Bordalo, P., Gennaioli, N., & Shleifer, A. (2013) “Salience and consumer choice,” Journal of Political Economy,
121, 5, 803–843.
Brandstätter, E., Gigerenzer, G., & Hertwig, R. (2006) “The priority heuristic: Making choices without trade-
offs,” Psychological Review, 113, 2, 409.
Busemeyer, J. R., & Townsend, J. T. (1993) “Decision field theory: A dynamic-cognitive approach to decision
making in an uncertain environment,” Psychological Review, 100, 3, 32.
Busemeyer, J. R., Gluth, S., Rieskamp, J., & Turner, B. M. (2019) “Cognitive and neural bases of multi-
attribute, multi-alternative, value-based decisions,” Trends in Cognitive Sciences, 23, 3, 251–263.
Camerer, C. F., Loewenstein, G., & Rabin, M. (Eds.). (2011) Advances in behavioral economics. Princeton
University Press.
Cheng, J., & González-Vallejo, C. (2016) “Attribute-wise mechanism vs alternative-wise mechanism in in-
tertemporal choice: Testing the proportional difference model, trade-off model and hyperbolic model,”
Decision, 3, 3, 190–215.
Dai, J., & Busemeyer, J. R. (2014) “A probabilistic, dynamic, and attribute-wise model of intertemporal
choice,” Journal of Experimental Psychology: General, 143, 4, 1489.
104
Dawes, R. M. (1964) “Social selection based on multidimensional criteria,” The Journal of Abnormal and Social
Psychology, 68, 1, 104.
Edwards, W. (1954) “The theory of decision making,” Psychological Bulletin, 51, 4, 380.
Edwards, W. (1961) “Behavioral decision theory,” Annual Review of Psychology, 12, 1, 473–498.
Einhorn, H. J. (1970) “The use of nonlinear, noncompensatory models in decision making,” Psychological
Bulletin, 73, 3, 221.
Einhorn, H. J., & Hogarth, R. M. (1981) “Behavioral decision theory: Processes of judgement and choice,”
Annual Review of Psychology, 32, 1, 53–88.
Erev, I., Ert, E., Plonsky, O., Cohen, D., & Cohen, O. (2017) “From anomalies to forecasts: Toward a descriptive
model of decisions under risk, under ambiguity, and from experience,” Psychological Review, 124, 4, 369.
Ericson, K. M., White, J. M., Laibson, D., & Cohen, J. D. (2015) “Money earlier or later? Simple heuristics
explain intertemporal choices better than delay discounting does,” Psychological Science, 26, 6, 826–833.
Fehr, E., & Schmidt, K. M. (1999) “A theory of fairness, competition, and cooperation,” Quarterly Journal of
Economics, 817–868.
Frederick, S., Loewenstein, G., & O’donoghue, T. (2002) “Time discounting and time preference: A critical
review,” Journal of Economic Literature, 40, 2, 351–401.
Gigerenzer, G., & Gaissmaier, W. (2011) “Heuristic decision making,” Annual Review of Psychology, 62,
451–482.
Glimcher, P. W., & Fehr, E. (Eds.). (2013) Neuroeconomics: Decision making and the brain. Academic Press.
Glöckner, A., & Betsch, T. (2008) “Modeling option and strategy choices with connectionist networks:
Towards an integrative model of automatic and deliberate decision making,” Judgment and Decision Making,
3, 215–228.
Gold, J. I., & Shadlen, M. N. (2007) “The neural basis of decision making,” Annual Review of Neuroscience, 30,
535–574.
González-Vallejo, C. (2002) “Making trade-offs: A probabilistic and context-sensitive model of choice beha-
vior,” Psychological Review, 109, 1, 137.
Green, L., Myerson, J., & Macaux, E. W. (2005) “Temporal discounting when the choice is between two
delayed rewards,” Journal of Experimental Psychology: Learning, Memory, and Cognition, 31, 5, 1121.
Halpern, D. (2016) Inside the nudge unit: How small changes can make a big difference. Random House.
He, L., Zhao, W. J., & Bhatia, S. (2020) “An ontology of decision models,” Psychological Review, Forthcoming.
Houston, D. A., Sherman, S. J., & Baker, S. M. (1989) “The influence of unique features and direction of
comparison of preferences,” Journal of Experimental Social Psychology, 25, 2, 121–141.
Huber, J., Payne, J. W., & Puto, C. (1982) “Adding asymmetrically dominated alternatives: Violations of
regularity and the similarity hypothesis,” Journal of consumer research, 9, 1, 90–98.
Kahneman, D., & Tversky, A. (1979) “Prospect theory: An analysis of decision under risk,” Econometrica: Journal
of the Econometric Society, 47, 263–291.
Keeney, R. L., & Raiffa, H. (1993) Decisions with multiple objectives: Preferences and value trade-offs. Cambridge
University Press.
Kőszegi, B., & Szeidl, A. (2013) “A model of focusing in economic choice,” The Quarterly Journal of Economics,
128, 1, 53–104.
Krajbich, I., Armel, C., & Rangel, A. (2010) “Visual fixations and the computation and comparison of value in
simple choice,” Nature Neuroscience, 13, 10, 1292–1298.
Leland, J. W. (1994) “Generalized similarity judgments: An alternative explanation for choice anomalies,”
Journal of Risk and Uncertainty, 9, 2, 151–172.
Leland, J. W. (2002) “Similarity judgments and anomalies in intertemporal choice,” Economic Inquiry, 40, 4,
574–581.
Lieder, F., Griffiths, T. L., & Hsu, M. (2017) “Overrepresentation of extreme events in decision making reflects
rational use of cognitive resources,” Psychological Review, 125, 1–32.
Loewenstein, G. F., & Prelec, D. (1993) “Preferences for sequences of outcomes,” Psychological Review, 100, 1, 91.
Loewenstein, G., & Prelec, D. (1992) “Anomalies in intertemporal choice: Evidence and an interpretation,”
The Quarterly Journal of Economics, 107, 573–597.
Loewenstein, G. & Sicherman, N. (1991) “Do Workers Prefer Increasing Wage Profiles?,” Journal of Labor
Economics, 9, 1, 67– 84.
Loewenstein, G., O’Donoghue, T., & Bhatia, S. (2015) “Modeling the interplay between affect and delib-
eration,” Decision, 2, 2, 55–62.
Loomes, G. (2010) “Modeling choice and valuation in decision experiments,” Psychological Review, 117, 3, 902.
Loomes, G., & Sugden, R. (1982) “Regret theory: An alternative theory of rational choice under uncertainty,”
The Economic Journal, 92, 368, 805–824.
105
Sudeep Bhatia
Loomes, G., & Sugden, R. (1986) “Disappointment and dynamic consistency in choice under uncertainty,”
The Review of Economic Studies, 53, 2, 271–282.
Louie, K., Khaw, M. W., & Glimcher, P. W. (2013) “Normalization is a general neural mechanism for context-
dependent decision making,” Proceedings of the National Academy of Sciences, 110, 15, 6139–6144.
Marchiori, D., Di Guida, S., & Erev, I. (2015) “Noisy retrieval models of over-and undersensitivity to rare
events,” Decision, 2, 2, 82.
Markowitz, H. (1959) Portfolio selection, efficient diversification of investments. J. Wiley.
Mazur J. E. (1987) “An adjusting procedure for studying delayed reinforcement,” in Commons M. L., Mazur J.
E., Nevin J. A., Rachlin H., (Eds.). Quantitative analysis of behavior: Vol. 5. The effect of delay and of intervening
events of reinforcement value. Hillsdale, NJ; Erlbaum, pp. 55–73.
Mellers, B. A., & Biagini, K. (1994) “Similarity and choice,” Psychological Review, 101, 3, 505.
Mellers, B., Schwartz, A., & Ritov, I. (1999) “Emotion-based choice,” Journal of Experimental Psychology:
General, 128, 3, 332.
Mukherjee, K. (2010) “A dual system model of preferences under risk,” Psychological Review, 117, 1, 243.
Oppenheimer, D. M., & Kelso, E. (2015) “Information processing as a paradigm for decision making,” Annual
Review of Psychology, 66, 277–294.
Parducci, A. (1974) “Contextual effects: A range-frequency analysis,” Handbook of Perception, 2, 127–141.
Payne, J. W., Bettman, J. R., & Johnson, E. J. (1988) “Adaptive strategy selection in decision making,” Journal
of Experimental Psychology: Learning, Memory, and Cognition, 14, 3, 534–552
Payne, J. W., Bettman, J. R., & Johnson, E. J. (1992) “Behavioral decision research: A constructive processing
perspective,” Annual Review of Psychology, 43, 1, 87–131.
Payne, J. W., & Braunstein, M. L. (1971) “Preferences among gambles with equal underlying distributions,”
Journal of Experimental Psychology, 87, 1, 13.
Rabin, M., & Thaler, R. H. (2001) “Anomalies: Risk aversion,” Journal of Economic Perspectives, 15, 1, 219–232.
Ratcliff, R., & McKoon, G. (2008) “The diffusion decision model: theory and data for two-choice decision
tasks,” Neural Computation, 20, 4, 873–922.
Rapoport, A., & Wallsten, T. S. (1972) “Individual decision behavior,” Annual Review of Psychology, 23, 1,
131–176.
Read, D., Frederick, S., & Scholten, M. (2013) “DRIFT: An analysis of outcome framing in intertemporal
choice,” Journal of Experimental Psychology: Learning, Memory, and Cognition, 39, 2, 573.
Roe, R. M., Busemeyer, J. R., & Townsend, J. T. (2001) “Multialternative decision field theory: A dynamic
connectionst model of decision making,” Psychological Review, 108, 2, 370.
Rubinstein, A. (1988) “Similarity and decision-making under risk (Is there a utility theory resolution to the
Allais paradox?),” Journal of Economic Theory, 46, 1, 145–153.
Russo, J. E., & Dosher, B. A. (1983) “Strategies for multiattribute binary choice,” Journal of Experimental
Psychology: Learning, Memory, and Cognition, 9, 4, 676.
Samuelson, P. A. (1937) “A note on measurement of utility,” The Review of Economic Studies, 4, 2, 155–161.
Scholten, M., & Read, D. (2010) “The psychology of intertemporal tradeoffs,” Psychological Review, 117, 3, 925.
Sher, S., & McKenzie, C. R. (2014) “Options as information: Rational reversals of evaluation and preference,”
Journal of Experimental Psychology: General, 143, 3, 1127.
Schmidt, U., Starmer, C., & Sugden, R. (2008) “Third-generation prospect theory,” Journal of Risk and
Uncertainty, 36, 3, 203.
Soltani, A., De Martino, B., & Camerer, C. (2012) “A range-normalization model of context-dependent
choice: A new model and evidence,” PLoS Computational Biology, 8, 7, e1002607.
Starmer, C. (2000) “Developments in non-expected utility theory: The hunt for a descriptive theory of choice
under risk,” Journal of Economic Literature, 38, 2, 332–382.
Stewart, N., Chater, N., & Brown, G. D. (2006) “Decision by sampling,” Cognitive Psychology, 53, 1, 1–26.
Sugden, R. (1991) “Rational choice: A survey of contributions from economics and philosophy,” The Economic
Journal, 101, 407, 751–785.
Thaler, R. H., & Sunstein, C. R. (2009) Nudge: Improving decisions about health, wealth, and happiness. Penguin.
Thorngate, W. (1980) “Efficient decision heuristics,” Behavioral Science, 25, 3, 219–225.
Tsetsos, K., Chater, N., & Usher, M. (2012) “Salience driven value integration explains decision biases and
preference reversal,” Proceedings of the National Academy of Sciences, 109, 24, 9659–9664.
Tversky, A. (1969) “Intransitivity of preferences,” Psychological Review, 76, 1, 31–48.
Tversky, A. (1972) “Elimination by aspects: A theory of choice,” Psychological Review, 79, 4, 281.
Tversky, A., & Kahneman, D. (1992) “Advances in prospect theory: Cumulative representation of un-
certainty,” Journal of Risk and Uncertainty, 5, 4, 297–323.
Tversky, A., & Simonson, I. (1993) “Context-dependent preferences,” Management Science, 39, 10, 1179–1189.
106
Usher, M., & McClelland, J. L. (2004) “Loss aversion and inhibition in dynamical models of multialternative
choice,” Psychological Review, 111, 3, 757.
Viscusi, W. K. (1989) “Prospective reference theory: Toward an explanation of the paradoxes,” Journal of Risk
and Uncertainty, 2, 3, 235–263.
Von Neumann, J., & Morgenstern, O. (1944) Theory of Games and Economic Behavior. Princeton University Press.
Wang, X. J. (2002) “Probabilistic decision making by slow reverberation in cortical circuits,” Neuron, 36, 5,
955–968.
Weber, E. U., & Johnson, E. J. (2009) “Mindful judgment and decision making,” Annual Review of Psychology,
60, 53–85.
Weber, E. U., Shafir, S., & Blais, A. R. (2004) “Predicting risk sensitivity in humans and lower animals: Risk as
variance or coefficient of variation,” Psychological Review, 111, 2, 430.
Wedell, D. H., & Pettibone, J. C. (1996) “Using judgments to understand decoy effects in choice,”
Organizational Behavior and Human Decision Processes, 67, 3, 326–344.
Yaari, M. E. (1987) “The dual theory of choice under risk,” Econometrica: Journal of the Econometric Society, 55, 1,
95–115.
107
9
RULE-FOLLOWING
Man is as much a rule-following animal as he is a purpose-seeking one.

~ F.A. Hayek (1973)
1 Introduction
As Hayek’s memorable line reminds us, our social and economic lives are governed as much by rules
as by our tastes, our incomes, and market prices. People are rule-following animals – but why do
they follow rules? In this chapter, we provide a interdisciplinary perspective on rule-following,
sketching various answers to this question offered by social scientists. We start with a broad and
colloquial notion of rules that spans from concrete, formal regulations to inarticulate, informal
norms. Thus, broadly speaking, a rule generates a regularity of some sort in human behavior and,
perhaps most importantly, in human interactions. Rules direct, constrain, guide, and temper our
actions. The result of rule-following is a perceptible order, and the result of rule-breaking is a
disorder for observers (and perhaps punishment for the rule-breaker).
In economics, the answer to why we follow rules is that, when we do, it is because doing so is in
our interests. We follow rules when the benefits exceed the costs. Economic accounts of rule-
following differ in how they model the nature of these costs and benefits: for example, do the costs
of violating a rule arise within the context of an interaction or are they imposed from outside by a
third-party like the government? Do the benefits include only material benefits, or should we think
of people as intrinsically valuing rules and norms, or perhaps as valuing the reputational benefits of
being seen to follow the rules? Setting aside these subtleties, the economic viewpoint treats rule-
following the same way it treats any other choice, as the product of constrained optimization – that
is, choices reflect people doing the best they can for themselves (“maximizing utility”) given the
options available to them (“subject to constraints”).
Psychological accounts of rule-following bear some similarity to economic accounts in that they
recognize the fundamental role of individuals’ interests in shaping the decision to follow rules;
however, psychologists do not collapse interests into a singular notion of utility or assume that choices
are the product of optimization. The psychology of rule-following contends that norms are followed
for different reasons at different times (e.g., some rules and norms are followed just because someone
doesn’t know what to do or how to do something, and they assume that the rule or norm is a guide to
effective action; other rules and norms are followed because they help build and maintain relationships;
108 DOI: 10.4324/9780367808983-11

Rule-Following
others are followed because they make us look, or feel, like “good” people). Psychological accounts
also try to explain how we come to feel so strongly about a particular set of rules and norms (i.e., how
norms are internalized), how we know which rules and norms to follow in particular circumstances, and
how we might have come to be the kind of creatures that follow rules and norms in the first place.
Philosophers attempt to explain what a rule is. What definition of a norm fits the observed social
facts? What are the necessary and sufficient conditions for a normative principle to be a norm among
a group of people? Philosophers produce arguments about what constitutes differences in types of
norms and whether or not they can be reduced to facts and mental predicates. A common theme
across the disciplines, whether empirical or theoretical, is that they assume that rules and norms make
demands of people to act or not act – that is, rules and norms cause people to act or not act.
2 Three Economic Perspectives on Rule-Following

Within the rational choice framework of economics, rule-following is sometimes understood as an
individual choice and sometimes understood as an emergent outcome of an interaction between
multiple actors. While these perspectives turn out to be mutually compatible, they reveal different
ways of thinking about the nature of a rule. In this section, we introduce both ways of thinking
about the problem and then highlight the empirical successes of the economist’s view, while also
pointing out some empirical puzzles and highlighting how economists have attempted to amend
their theoretical framework to address them.
2.1 Rational Choice Rule-Following

In the standard rational choice framework of economics, the decision to follow a rule arises from a
consideration of the costs and benefits of doing so. If the benefits of following a rule exceed the
costs, economic theory predicts someone will follow the rule; if the benefits of violating a rule
exceed the costs, then they will violate the rule instead. In Becker’s (1968) classic work on the
economics of crime, he develops this perspective explicitly and works out its implications. Let vf be
the utility associated with rule-following and vb be the utility associated with rule-breaking. If vf >
vb the problem is trivial, as people will gladly follow the rule on their own. In the more interesting
case in which vb > vf, someone who wants to encourage compliance must find a way to impose
costs on rule-breakers. Let p be the probability that a rule-violation is detected and let c be the cost
imposed on violators when they are detected. The expected payoff from violating the rule can then
be rewritten as (1-p)vb + p(vb-c); thus, compliance would be incentivized if: vf > (1-p)vb + p(vb-c).
This simple model generates some very intuitive implications; namely, compliance can be en-
couraged both by increasing the magnitude of punishment and increasing the probability of de-
tection, for a given level of punishment. The model tidily explains why we punish grand larceny
more harshly than petty theft, why “treble damages” may be an appropriately chosen punishment for
anti-trust violations, and so on. As improving monitoring is presumably more difficult than in-
creasing the magnitude of punishment, a number of wits have half-joked about reducing crime and
increasing traffic safety by substantially raising the costs. For example, Becker has suggested dras-
tically reducing policing and instead making death the penalty for all crimes, and Landsburg (2012)
suggests that we could encourage safe driving by removing seatbelts from cars and placing a shar-
pened spike on the steering wheel pointed at the chest of the driver. Incentives matter after all.
While these are clearly intended as black humor, there is empirical evidence that incentives do
matter, at least for driver behavior – Peltzman (1975) famously showed that seat belt mandates had
the perverse effect of increasing the rate of accidents, because drivers who feel safer drive more
recklessly. Whether the death penalty actually serves as an effective deterrent remains a subject of
controversy (Donohue and Wolfers, 2005).
109
While the rational choice perspective on rule-following provides a lot of explanatory power, it
cannot be the whole story about why people follow rules. A famous experiment conducted at a
daycare in Israel provides a neat illustration of the complications for theories that reduce rule-following
to incentives alone. The daycare officially closed at 4 PM (Gneezy and Rustichini, 2000, 4), but some
parents had begun to show up late and the managers of the daycare decided to introduce a monetary
penalty for late pickups. The paper shows that, when the penalty was introduced, late pickups actually
increased, clearly contradicting the basic theory laid out above. Why? The title of the paper, “A Fine Is
A Price,” reveals the authors’ interpretation – there was an informal rule in place that was widely
followed and self-enforced. When this rule was replaced with a formal rule and enforced with a
penalty, it changed the nature of the interaction; support for the informal rule dissipated, and parents
began to think about late pickups as something that money can buy rather than as a violation of a
taboo. Strikingly, when the penalty was subsequently eliminated, the rate of late pickups did not return
to their initial level (see Gneezy and Rustichini, 2000).
2.2 Rule-Following in Equilibrium

The game-theoretic approach preserves the economist’s emphasis on optimization and interprets
rule-following as an equilibrium outcome of some game. In game theory, an equilibrium consists of
a set of strategies for all players that are best responses to one another, meaning that no player can
unilaterally improve their expected payoff by choosing a different strategy. There often exist
multiple equilibria in a game, but some equilibria may be preferable to others. For example, consider
a “game” in which two drivers going in opposite directions are choosing which side of the road to
drive on. If both drive on the same side of the road, they will crash, but if they drive on opposite
sides, they will reach their destinations unscathed. If the game is played once, there are three Nash
equilibria (1) each player drives on the right side of the road (from their point of view), (2) each
player drives on the left side of the road (from their point of view), and (3) each player flips a fair
coin and drives on the right side of the road if heads and the left of side of the road if tails. All three
of these combinations of strategies are Nash equilibria, but (1) and (2) Pareto dominate (3) since the
drivers will crash ½ of the time under (3) and will never crash under (1) or (2). There is an
“equilibrium selection” problem here, and establishing common expectations can resolve it. The
resolution involves players following a common rule or convention.
Young (1996) defines a convention as “an equilibrium that everyone expects in interactions that
have more than one equilibrium” and shows how even myopic players’ expectations can converge
through repeat interaction to establish a convention. While a process of trial and error can generate
conventional, rule-governed patterns of interaction, there are other ways of selecting (or creating)
equilibria that are welfare improving. One alternative comes from Aumann’s (1987) notion of a
correlated equilibrium. In a correlated equilibrium, players can condition their strategy on a publicly
observable signal (e.g., the location of signage and road markings). For example, a strategy that
conditions driver behavior on road signage results in people in the US always driving on the right
and people in the UK always driving on the left. Once established, such rules and conventions are
self-enforcing (Schelling, 1960) since deviations are harmful to those who deviate.
In repeated games, in which people expect to interact again in the future (or expect their re-
putations to follow them into interactions with others), a richer set of possibilities arises. Various
“folk theorems” show that people can employ strategies that punish deviations from some expected
pattern of behavior and thereby support almost any desired convention as an equilibrium (e.g.,
Kandori, 1992).
Notice that in the game-theoretic account, it remains the case that rule-following behavior arises
only when adherence is consistent with self-interest. Nevertheless, this needn’t mean that an equili-
brium rule is welfare-maximizing. For example, theorists have offered game-theoretic accounts of
110
Rule-Following
many rules and norms that most people would regard as harmful (e.g., Mackie, 1996, on female genital
mutilation and foot-binding and Thrasher and Handfield, 2018, on honor killings).
Despite the wide-ranging applicability of the game-theoretic account, there remain a number of
empirical findings that suggest that equilibrium arguments are insufficient to explain the scope of
rule-following behavior. One such piece of evidence from experimental economics is the will-
ingness of laboratory subjects to punish those who violate their expectations, even in one-shot
interactions in which punishment cannot be understood as future-oriented (e.g., Fehr et al., 2002).
2.3 “Behavioral” Rule-Following

In light of the evidence that people often follow rules and norms, even when doing so cannot be
explained by a straightforward analysis of costs and benefits or as part of the equilibrium of some
game, some economists have proposed that the standard model of preferences should be augmented
to include a utilitarian benefit from adhering to rules and norms (i.e., a “preference for norm-
following”; see Cappelen et al., 2007; López-Pérez, 2008; Kessler and Leider, 2012; Krupka and
Weber, 2013; Kimbrough and Vostroknutov, 2016). These models assume that people are moti-
vated both by material costs/benefits and by psychic costs/benefits of violating/following commonly
known and widely shared standards of behavior (or by concerns about social image and self-image,
which are themselves dependent on one’s adherence to rules and norms; see Bernheim, 1994;
Bénabou and Tirole 2006, 2011). They preserve the basic logic of the economic theory, but by
introducing an intrinsic benefit of following rules and norms and an intrinsic cost of violating them,
they are able to account for the fact that people often seem to follow the rules, despite the lack of
material incentive to do so. The intuition is simple: when the material costs of following a rule are
sufficiently low, the psychic benefit of rule-following exceeds the material cost, and thus people will
follow the rule.
One feature of the behavioral view is that it provides an account of differences in behavior across
people via differences in the extent to which they gain utility from following rules and suffer
disutility from violating them. Allowing for such heterogeneity comports with the intuition that
some people tend to be “conformists” and others tend to be “rebels.” In fact, experiments suggest
that a person’s willingness to incur a cost to follow a totally arbitrary rule reliably correlates with
their willingness to follow rules and norms in social settings. For example, those who incur more
costs following an experimenter-stated arbitrary rule are also more prone to follow egalitarian norms
in simple bargaining games (Kimbrough and Vostroknutov, 2016, 2018). This suggests that one can
learn a lot about someone by observing their willingness to follow rules, an idea that has also been
picked up in the literature on clubs and religious organizations where strict rules on diet, clothing,
and other behaviors are seen as “sacrificial costs” which help groups screen out insincere prospective
members (e.g., Iannaccone, 1992; Aimone et al., 2013).
This kind of model has been used to provide an account of seemingly anomalous behavior in a wide
variety of contexts, but there is reason to worry that the amendments to preferences grow almost as quickly
as the set of contexts that have been studied. For an extreme example, see Abeler et al. (2019), who design
experiments to help prune this rampant growth by distinguishing the best fitting model from among 22
models of “preferences for truth-telling.” Despite these concerns, this kind of model is also consistent with
the explanations for rule-following that emerge from evolutionary approaches. See section 3.4 below.
3 Psychological Accounts of Rule-following

In psychology, there is a longstanding debate about the extent to which rules and norms are arbitrary
impositions that are valued and reinforced by a logic internal to a particular cultural group and the
extent to which they are functional adaptations designed to solve some social problem. In reality,
111
both perspectives seem to contain elements of the truth; many practices governed by rules and
norms are highly culturally variable but nevertheless seem designed to solve social problems that are
human universals (e.g., signals of sexual (un)availability, greetings, dwellings, clothing styles, etc.; see
Cialdini and Trost, 2008). Note that this basic contention is consistent with the economic view,
which can help fill in the details about how the (perhaps socially imposed) costs and benefits of
following a particular rule or norm shape its adoption and persistence.1 However, psychologists also
offer accounts of peoples’ motivation to follow rules and norms that differ from and expand upon
the mechanistic logic offered by economists.
3.1 Effective Action, Relationship Building, and Self-Concept Management

Cialdini and Trost (2008) offer three distinct, but complementary, accounts of why people follow
rules and norms. First, they note that rules and norms are often reliable guides to effective action;
when we want to know what to do or how to do something, there is often a rule or norm we can
follow that provides a best-practice of some kind. A descriptive norm – what others typically do – is
often a guide to effective action, and thus we end up following a rule or norm just because we want
to make good choices. Second, Cialdini and Trost (2008) highlight that adherence to rules and
norms is often essential to building and maintaining a social relationship; we follow rules and norms
because doing so yields social approval, and we refrain from violating them to avoid disapproval. An
injunctive norm – what others believe we ought to do – informs us how our actions will be viewed by
others, and we follow injunctive norms in order to please those others. Finally, they argue that we
often follow rules and norms to make ourselves feel good about who we are. A personal norm – or,
more precisely, an internalized injunctive norm – allows us to monitor and evaluate our own
actions, and we follow such norms to ensure that we continue to see ourselves as “good, kind, and
helpful people” (p. 160). In this third account of why and how we follow rules and norms, the
notion of “internalization” or “interiorization” looms large because, though the ultimate cause of
rule- and norm-following may be some combination of the factors highlighted so far, the proximate
cause is often that following rules and norms just feels like the right thing to do. In the next section,
we examine one account of how such feelings come to develop.
3.2 Interiorization of Rules and Norms

Sherif (1936), whose work was influenced by contemporary psychoanalytic theory and who remains
essential reading, provides an account of rule- and norm-following that combines features of all
three motivations described by Cialdini and Trost (2008). According to his theory, a person’s sense
of self “is chiefly made up of social values derived from parents, teachers and others close to [them]”
(p. 179). Regularities in the behavior, approval, and disapproval of others affirm a set of values and
delineates a set of rules; exposure to those regularities over time thus inculcates those values and
teaches respect for those rules.2 Eventually, these values and rules become “interiorized” (or in-
ternalized) such that actual (or prospective) violations of them cause feelings of frustration, conflict,
or guilt because violating them conflicts with our sense of who we are and insults our perceived
status or our dignity. Similarly, adherence to them is a source of satisfaction and pleasure and an
important cause of action and striving. In this view, we first follow rules and norms because of
pressures external to us, but by learning and following those rules, they become part of us, and thus
we eventually follow them because they affirm to us who we are.
In a sense, the psychological perspective complements recent economic models of rule- and
norm-following by providing a developmental perspective on the process by which the positive and
negative emotional valences associated with following and violating rules and norms emerge.
112
Rule-Following
An important implication of this view is that what we internalize depends on with whom we
interact, and to the extent that social life involves navigating ties with multiple groups, we will often
find ourselves pulled in different directions by standards internalized from different groups. The idea
that our norms are derived from our group memberships has been captured in the models of
“identity economics” due to Akerlof and Kranton (2000), in which adherence to values and rules
both enter the utility function and are indexed to identity groups.
3.3 The “Role-Rule” Model of Social Behavior

In social psychology, Harré and Secord (1973) occupy one extreme of a spectrum ranging from what
might be called naïve behaviorism to what they call the “anthropomorphic model of man.” In their
view, behaviorist assumptions that describe human social interaction in simple stimulus-response
terms do not give justice to the capacity of humans to adjust their actions based on whether they satisfy
internalized evaluative criteria. They argue that all social behavior should be understood in terms of
rule-following: “human actions are generated by the conscious self-monitoring of […] performance
in accordance with certain sets of rules” (p. 93). Rules both specify the actions one ought to take and
provide the evaluative criteria by which humans create “anticipatory,” “monitoring,” and “retro-
spective” commentaries on their actions.
They introduce the “role-rule” model of social behavior. “Rules are propositions and they guide
action, through the actor being aware of the rule and what it prescribes […] Rules are future-
directed; they not only guide action, but also determine expectations regarding the actions of
others” (p. 12). There are many such rules and systems of rules, but one’s role in a particular setting is
defined by “the set of actions he is expected to take” (p. 13). That is, in a given situation, each
person recognizes and adopts a role and, because of the expectations associated with that role,
follows a set of rules governing their behavior. Both before and after acting, the rule provides a
measuring stick against which any action can be evaluated; thus, actors and observers can assess how
well an actor plays a given role.
An important implication of this view is that actors’ and observers’ ordinary language accounts of
an actor’s behavior (their “commentaries”) become relevant data for understanding and explaining
social behavior. In offering accounts of actors’ motivations and their intentions, both actors and
observers shed light on the meaning of social interaction and allow us to glimpse the role-rule
structure of the interaction. Importantly, it should be clear that there can be a mismatch between the
perception of the actor and that of the observer (or between that of different observers); this kind of
complication is regularly ignored in behaviorist accounts of social interaction with the result that
observers too readily impose their own view of the meaning of an interaction on the actors, without
considering whether the actors see it the same way.
Note that the strong emphasis on the rule-governed nature of all social behavior leaves little room
in the role-rule model for the explicit consideration of trade-offs that characterizes the economic view
of when and why people follow rules. That said, a weaker version of the claim – that people use
contextual information to determine their role and thereby the rules that one in such role ought to
follow – could be compatible with the economic view if it is taken to merely describe the process by
which people identify the rules that enter into their cost-benefit analysis.
3.4 Evolutionary Accounts of Rule-Following

In evolutionary psychology and anthropology, the propensity to follow rules and norms is under-
stood as a human universal that has evolved as an adaptation to our history as a social species.
Chudek and Henrich (2011) provide a succinct version of this view (see also the extensive references
therein). In their telling, humans have an evolved “norm psychology” that gives us an intrinsic desire
113
to acquire and follow social rules and norms as well as a willingness to teach such rules and punish
violations thereof. Our norm psychology emerged via a gene-culture coevolutionary process in
which, as the dependence of members of our species on culturally transmitted information about the
world grew in importance for survival and reproduction, those who were better able to acquire
cultural information were more likely to survive and reproduce. Key to this view is the premise that
culture is cumulative and that what any person is capable of depends on the quality and fidelity of
cultural information transmitted to them from parents, elders, peers, and other role models.
Members of those groups whose culturally transmitted rules and norms better enabled them to
exploit their environment, survive shocks and compete with other groups were also more likely to
survive and reproduce, and within those groups, individuals who were better equipped to learn such
useful information would themselves have an advantage over others who struggled to learn and
follow the rules and norms of their group.3
Evidence for this account highlights the facility and fidelity with which people learn rules and
norms, even as young children, the seeming adaptiveness of the means by which we choose models
from whom to learn, the early-emerging and persistent tendency to punish norm violators, and the
relative superiority of social learning over individual learning in generating solutions to complex
problems. One particularly striking piece of evidence for an intrinsic propensity to follow rules
comes from experiments comparing humans to non-human primates, in which only humans are
known to copy wholly irrelevant (perhaps “ritualistic”) steps in a process that yields some desirable
outcome (see Henrich, 2017).
The existence of an evolved “norm psychology” is consistent with some “behavioral” models of
rule-following in economics, especially those that posit the existence of preferences for norm-
following. The evolutionary account can be seen as providing an ultimate explanation for the
existence of such preferences, and the economic models are seen as working out how this aspect of
our psychology interacts with our other “purpose-seeking” tendencies in determining our behavior.
4 Rules and Norms in Philosophy

Whereas in economics rules and norms are either constraints to be maximized against or just-so
parameters in the utility maximand (Smith and Wilson, 2019), philosophers of the economics
persuasion ask, what composes a rule or norm, and what distinguishes the various kinds of rules
and norms?
4.1 Conditional Preferences

Bicchieri (1993, 2006, 2017) has had a longstanding interest in explaining what a norm is. In her
third book, she applies her theoretical model to the practical policy problem of changing the norms
that people follow. Bicchieri (2017) distinguishes three kinds of “collective patterns of behavior”:
collective customs/shared moral rules, descriptive norms, and social norms. A custom is “a pattern of
behavior such that individuals (unconditionally) prefer to conform to it because it meets their needs”
(p. 15). For Bicchieri, the distinguishing feature of customs is that people do not follow the moral
rule or collective custom because they expect other people to follow the moral rule or collective
custom. They follow it regardless of what other people do. A descriptive norm is “a pattern of
behavior such that individuals prefer to conform to it on condition that they believe that most
people in their reference network conform to it (empirical expectation)” (p. 19). Unlike a custom,
people follow a descriptive norm because they expect their peers to follow it. Moreover, they follow
the descriptive norm conditional upon observing or believing that people in their reference group
follow it. If people no longer believe that people in their reference group follow the descriptive
norm, then they will not follow it either. For Bicchieri, the descriptive norm causes people to act
114
Rule-Following
(p. 27). Descriptive norms are the kinds of patterns of behavior we observe in coordination games
(e.g., rules of traffic, as in section 2.2).
Social norms are much more complicated. A social norm is
a rule of behavior such that individuals prefer to conform to it on condition that they
believe that (a) most people in their reference network conform to it (empirical expectation),
and (b) that most people in their reference network believe they ought to conform to it
(normative expectation). (p. 35)
As with descriptive norms, people follow a social norm because they believe people in their re-
ference group follow the social norm, but people also follow a social norm because they believe that
people in their reference group believe they are obligated to follow the social norm. A nonmoral
prudential “ought” backs up a social norm. Social norms are patterns of behavior we observe in
games of cooperation, such as the prisoner’s dilemma and voluntary contribution mechanisms for
public goods.
Descriptive and social norms are both grounded on a “conditional preference,” as Bicchieri calls
it, a preference to act conditional on the belief that others act the same way. For a descriptive norm
to change, the belief that most people in their reference group conform to the descriptive norm must
simply change. But for a social norm to change, such a corresponding belief must change as well as
the belief that most people believe they ought to conform to the social norm. Bicchieri takes care to
explain that her model of conditional preferences, empirical expectations, and normative expecta-
tions does not literally describe the psychology of norm-following in everyday life. As with utility
maximization models in economics, her account of norms is an “as-if” account of human conduct.
4.2 A Non-reductive Explanation

Brennan, Eriksson, Goodin, and Southwood (2013) present a general analysis of norms, which they
distinguish based on two dimensions: formal versus informal and moral versus social. Norms, in
general, are “clusters of normative attitudes” accepted in a community “plus knowledge of these
attitudes” by a significant portion of the community (p. 41). The key function of norms, they say, is
that they “make us accountable to one another” (original italics, p. 36).
Formal norms, such as laws, are established by secondary procedures, say a legislature, that de-
lineate the mechanisms for explicitly enforcing the norms and imposing sanctions for breaching
them. For example, a legislature establishes which side of the road to drive on and the judicial
mechanisms for punishing those who do not follow the norm. Nonformal norms lack such sec-
ondary rules for enactment. For example, there is no established procedure for sanctioning a
mourner who shows up graveside in hot pink.
Formal and informal norms also entail different normative attitudes. Lacking a hierarchy of
adoption and application procedures, nonformal norms consist of de dicto normative attitudes, that is,
of the explicit attitudes the norm is said to cover. If the informal norm is to wear black at a funeral,
the normative attitude is about specifically the color of the clothes to be worn at a funeral. If there is
no normative attitude towards wearing black at a funeral, there is no norm of wearing black at a
funeral. Formal norms are not about what is said, but about the thing, de re. An American in London
may not have the attitude of driving on the left, but he may judge Parliament as having the le-
gitimate procedures for establishing and enforcing a driving on the left norm in England. The
secondary procedures make formal norms about the whole process of establishing and enforcing
norms and not just what the norm is said to be about.
Brennan et al. distinguish moral and social norms by the grounds upon which the norm is
justified by those who hold the normative attitudes. The social norm of wearing black at a funeral
115
rests on the practice itself of wearing black to a funeral. That a significant portion of the community
has such a normative attitude is the reason why people justify to themselves that they have such a
norm. The practice of the norm is the justification for the norm. A moral norm against rape is not
justified because it just so happens that rape is something that people in the community do not do.
The normative attitude against rape is justified by the dignity, respect, and justice that people in the
community afford to everyone.
The key difference between Bicchieri’s and Brennan et al.’s account of norms is that norms can
be reduced to clusters of preferences and social facts in Bicchieri’s, while norms cannot in Brennan
et al. The latter emphasize that their explanation of norms is “non-reductive” and that the normative
element of norms is “ineliminable” (p. 4). Nevertheless, they argue that their account of a norm can
be included as an additional argument in a standard rational choice model of utility maximization
(p. 204). Instead of representing A’s preferences over x1, …, xn as UA = u(x1, …, xn), Brennan et al.
say the utility function could do a better job at fitting social facts if it had a few more arguments
to maximize, for example, a function like this: UA = u(x1, …, xn; N1(x1, …, xn); N2(x1, …, xn); …),
in which utility depends not only on allocations but also on normative judgments thereof.
Of course, this raises issues about how one identifies N, how many norms ought to be included,
and so on.
5 Discussion and Conclusion

Modern social scientists and philosophers tend to think through conduct and rule-following with
the concepts of both rules and norms, though individual projects tend to favor one term or the
other. The astute reader will notice how quickly and easily we – authors and readers alike – moved
in this chapter from rules to norms and to “rules and norms.” H.L.A. Hart’s seminal text on jur-
isprudence relies exclusively on “social rules,” the existence of which involves “a combination of
regular conduct with a distinctive attitude toward that conduct as a standard” (1961, p. 85). Brennan
et al. (2013) treat norms as a direct synonym for Hart’s rules.4 In setting up their quotation of Hart,
Brennan et al. explicitly refer to “the example of the norms of chess” (p. 28) when Hart (1961)
himself refers to the “rules of any game” (p. 56). While Bicchieri’s project is explicitly about norms,
her definition of a social norm is in terms of a “rule of behavior” with specific empirical and
normative expectations. A descriptive norm, however, is a mere pattern of behavior because the
“behavioral rule” of a social norm often proscribes behavior from being observable as a pattern
(p. 35, fn. 14). The legal scholar and philosopher Stephen Perry (2006) “use[s] the terms ‘rule’ and
‘norm’ more or less interchangeably” with the fine and somewhat counterintuitive distinction that
“a rule… is a norm which is general at least in the sense that it has application to more than one case,
and which is also usually general in the sense of applying to more than one person” (p. 1176).
Hayek’s treatises on how abstract rules spontaneously create order rarely ever refer to norms. His
definition of a rule is “a propensity or disposition to act or not act in a certain manner, which will
manifest itself in what we will call a practice or custom” (1973, p. 75, original italics). Adam Smith
(1759) and David Hume (1740) always refer to rules, frequently “general rules,” for the use of the
words norm and norms do not begin to take off until the early 20th century.5 For Smith, “the rules of
justice” are “precise, accurate, and indispensable” (p. 175), while “the general rules of almost all the
[other] virtues” are “loose, vague, and indeterminate” (p. 175).
We are unaware of any work which attempts to distinguish norms from rules, but we believe the
distinction is a useful one. As we discuss below, the distinction we draw can help us see the dif-
ference between Smith’s precise rules of justice and the looser norms (as we would now call them)
“of prudence, of charity, of generosity, of gratitude, of friendship,” to which we must attempt to fit
our conduct (pp. 174, 175). Their respective Latin roots, norma and regula, are synonyms but not
exactly so. Both norma and norm connote the notion of a standard differently than regula and rule.
116
Rule-Following
Compare the following four sentences, the first and third from the Collins Cobuild Dictionary
for Learners:
1. A norm is an official standard or level that organizations are expected to reach.

∗
1′. A rule is an official standard or level that organizations are expected to reach.
2. Norms are ways of behaving that are considered normal in a particular society.
∗
2′. Rules are ways of behaving that are considered normal in a particular society.
Sentences (1′) and (2′) have an infelicitous ring. In (1′) “a rule” doesn’t serve as an official standard of
an organization like “a norm” does in (1). Respect, transparency, and accountability are all examples
of company norms that employees may be expected to reach (for). Each could be an official standard
of an organization that new hires learn about as part of their onboarding process. None would ever
be presented on the first day at work as “a rule” of the company. If anything, they would be
described as part of the company culture; “this is the kind of company we want to be.” Expense
reimbursements, on the other hand, require following a company rule. An employee does not reach
for the expense reimbursement policy. If they want to be reimbursed, they follow it to a T. To
follow a rule is to be exact in one’s actions. To follow a norm is to accurately conform one’s action
to fit the expectations of a community.6
While norms can be actions in (2), rules can never be actions themselves as (2′) seems to indicate.
Rules are statements or instructions about the actions that people ought to undertake or are allowed
or not allowed to undertake. Consider the example of “No Diving” as a pool rule versus a pool
norm. A posted sign of “Pool Rules” means that even if everyone else were at the shallow end, no
one ought to be diving at the deep end. Following the pool’s rule is a matter of doing exactly what
the sign says. The sign directly bears on the situation. If the posted sign instead read “Pool Norms,”
as it curiously never does, one ought not dive out of a consideration for the other swimmers in the
deep end. But if no one else is remotely in the vicinity, a quick dive could still conform to the
expectations of the other swimmers. Following the “pool’s norm” is a matter of accurately con-
forming to a standard of the swimming community.
Where the two words begin to overlap is in the description of something as definitely “the rule”
or “the norm” (also from the Collins Cobuild Dictionary):
3. If you say that a situation is the norm, you mean that it is usual and expected.
4. If something is the rule, it is the normal state of affairs.
But even here, there is slippage as an exact synonym:
∗
3′. If you say that a situation is the rule, you mean that it is usual and expected.
∗
4′. If something is the norm, it is the normal state of affairs.
A situation is not quite the rule in (3′) because the rule is something distinct from a situation. The rule
is about a situation; the rule has an external bearing on the situation. The rule is more felicitously a
statement that describes the way that things usually happen in a particular situation. Similarly, in (4′)
the norm is more than simple normalcy, as in:
5. If something is the norm, it is the standard state of affairs.
The norm (versus a norm) involves reference to a definite standard. The norm is about what accurate
behavior in a situation entails, that is, about the standard to which we are expected to fit our actions.
The normalcy of the norm is the standardness that the norm connotes.
117
A rule also involves a standard for conduct. A rule (regula) marks out a line for what we ought or
ought not to do, and as we track the line, we may fail to one side or the other on this occasion or
another. The question regarding our actions with respect to a rule is a matter of exactness. How
exact is our path in conforming to a rule that governs our conduct? For matters of justice, Smith says
we better track the line pretty closely. For rules of prudence, charity, generosity, gratitude, and
friendship, the line is a little more vague and our deviations from it judged a little more loosely, if
not indeterminately. The question regarding our actions with respect to a norm (norma) is how
accurately do our actions fit with a standard of the community? A norm is about the attention we
expend to conform our actions to fit with the expectations of the community.
Being charitable to your critics is a rule, for we may not exactly follow it on this occasion or that.
Cleaning up after your dog is a norm, for we walk the dog daily, and removing the dog’s waste, even
when time is short, shows respect for our neighbors. We may call being charitable to your critics the
norm of an academic community and cleaning up after your dog the rule in your neighborhood, but
the meaning of the principle has slightly changed. Being charitable to your critics as a norm is about a
certain academic community reaching for a standard of showing respect towards its commenters, and
cleaning up after your dog as the rule is a matter of exactly removing the waste or not. The exactness
of rules is about an instance in time and space. The care bestowed upon or the attention expended in
fitting conduct to norms is of many instances.
Identifying such distinctions is not mere pedantry, for it highlights the one-sidedness with which
social scientists and philosophers answer the question of why humankind is a rule-following animal. The
direction of causation in economics, psychology, and economics-inspired philosophy generally flows
from the rule or norm to the individual decision to act (or to not act). The incentives, the preferences,
the empirical and normative expectations, the normative principles are the cause for someone to conform
with the rule or norm. Rules and norms do indeed make us accountable to others. But human conduct
is bidirectional (Smith and Wilson, 2019; Wilson, 2020). What social scientific and philosophical ac-
counts tend to ignore is the reverse direction of rule- and norm-following, from the individual to the
community. Why do we care to be accountable to others? Why do we care that we ourselves conform to a
norm, or that most people believe we ought to conform to a norm? It is not an answer to simply say we
prefer to, or as Frank Knight put it in an essay on ethics and economics interpretation a century ago, “it
is not enlightening to be told that conduct consists in choosing between possible alternatives” (1922,
p. 467). One does not simply prefer death to tyranny. The value of doing a1 and not a2 is a matter of
virtue and ethics, even for such mundane matters as cleaning up after your dog.
We follow neighborly norms because we care about our neighbors, because we care about
showing respect for neighbors, and because we care about being the kind of person who shows
respect for their neighbors. Rules and norms make external demands upon us to act, but that we
simultaneously supply such actions is also a matter of the internal feelings and thoughts, cultivated as
virtues and an ethical code, which prompt us to act in harmony with our community. The solution
to the problem of why people follow rules and norms is analogously the simultaneous solution of
two equations, of the demand and supply of conduct. Rules make regulative demands of us and
norms normative demands, and, simultaneously, we conform (or fail to conform) with either be-
cause we internally care (or fail to care) about tracking the line of the rule exactly or fitting the norm
of a community accurately. The open questions for research are how to correct for the bias and
inconsistency that come from the assumption of one-way, external causation in following rules and
norms. Humankind is as much a virtue-ethical animal as it is a rule- and norm-following one.
Notes
1 Economists’ accounts of various rules and norms sometimes emphasize their potential arbitrariness and
sometimes their functional adaptiveness (Elster, 1989).
118
Rule-Following
2 This needn’t be conceived as a passive process, since each person’s own actions, approval, and disapproval
are inputs into the same process for others. “Norms thus arising are not imposed from without, but are
products of situations of which the individual is a part” (p. 181). See also Piaget (1932) on how children’s
games reveal the process by which norms emerge.
3 Note that excessive conformity could lead to too little innovation, while excessive innovation could create
instability and cause useful traditions to be lost. If we think of conformity as a heritable trait, perhaps there is
balancing selection which results in a mix of conformist rule-followers and rebellious rule-breakers.
4 Hart himself never uses norm, except on the second page in quoting Hans Kelsen’s General Theory of Law and
State (1949).
5 See Google Ngram Viewer (https://books.google.com/ngrams).
6 “Her answer was accurate” does not mean the same thing as “Her answer was exact” or “Her answer was
precise.”
References
Abeler, J., Nosenzo, D. and Raymond, C. (2019) “Preferences for truth‐telling,” Econometrica, 87, 4,
1115–1153.
Aimone, J. A., Iannaccone, L. R., Makowsky, M. D. and Rubin, J. (2013) “Endogenous group formation via
unproductive costs,” Review of Economic Studies, 80, 4, 1215–1236.
Akerlof, G. A. and Kranton, R. E. (2000) “Economics and identity,” The Quarterly Journal of Economics, 115, 3,
715–753.
Aumann, R. J. (1987) “Correlated equilibrium as an expression of Bayesian rationality,” Econometrica, pp. 1–18.
Becker, G. S. (1968) “Crime and punishment: An economic approach,” in G. S. Becker and W. M. Landes eds.
The economic dimensions of crime, London, UK; Palgrave Macmillan, pp. 13–68.
Bénabou, R. and Tirole, J. (2006) “Incentives and prosocial behavior,” American Economic Review, 96, 5,
1652–1678.
Bénabou, R. and Tirole, J. (2011) “Identity, morals, and taboos: Beliefs as assets,” The Quarterly Journal of
Economics, 126, 2, 805–855.
Bernheim, B. D. (1994) “A theory of conformity,” Journal of Political Economy, 102, 5, 841–877.
Bicchieri, C. (1993) Rationality and coordination. New York, NY; Cambridge University Press.
Bicchieri, C. (2006) The grammar of society: The nature and dynamics of social norms. New York, NY; Cambridge
University Press.
Bicchieri, C. (2017) Norms in the wild: How to diagnose, measure, and change social norms. New York, NY; Oxford
University Press.
Brennan, G., Eriksson, L., Goodin, R. E. and Southwood, N. (2013) Explaining norms. New York, NY; Oxford
University Press.
Cappelen, A. W., Hole, A. D., Sørensen, E. Ø. and Tungodden, B. (2007) “The pluralism of fairness ideals: An
experimental approach,” American Economic Review, 97, 3, 818–827.
Chudek, M. and Henrich, J. (2011) “Culture–gene coevolution, norm-psychology and the emergence of
human prosociality,” Trends in Cognitive Sciences, 15, 5, 218–226.
Cialdini, R. B. and Trost, M. R. (2008) “Social influence: Social norms, conformity and compliance,” In
Gilbert, D. T., Fiske, S. T. and Lindzey, G. eds. The Handbook of Social Psychology, Vol. 1–2, pp. 151–192,
McGraw‐Hill.
Donohue, J. J. and Wolfers, J. (2005) “Uses and Abuses of Empirical Evidence in the Death Penalty Debate,”
Stanfod Law Review, 58, 791–846.
Elster, J. (1989) “Social norms and economic theory,” Journal of Economic Perspectives, 3, 4, 99–117.
Fehr, E., Fischbacher, U. and Gächter, S. (2002) “Strong reciprocity, human cooperation, and the enforcement
of social norms,” Human Nature, 13, 1, 1–25.
Gneezy, U. and Rustichini, A. (2000) “A fine is a price,” The Journal of Legal Studies, 29, 1, 1–17.
Harré H. and Secord, P. F. (1973) The explanation of social behaviour. Totowa, NJ; Littlefield & Adams.
Hart, H. L. A. (1961) The concept of law. New York, NY; Oxford University Press.
Hayek, F. A. (1973) Law, legislation and liberty, vol. 1: Rules and order. Chicago, IL; University of Chicago Press.
Henrich, J. (2017) The secret of our success: How culture is driving human evolution, domesticating our species, and making
us smarter. Princeton, NJ; Princeton University Press.
Hume, D. (1740/2000) A treatise on human nature, Norton, D. F. and Norton, M. J. eds. New York, NY;
Oxford University Press.
119
Iannaccone, L. R. (1992) “Sacrifice and stigma: Reducing free-riding in cults, communes, and other collec-
tives,” Journal of Political Economy, 100, 2, 271–291.
Kandori, M. (1992) “Social norms and community enforcement,” The Review of Economic Studies, 59, 1, 63–80.
Kessler, J. B. and Leider, S. (2012) “Norms and contracting,” Management Science, 58, 1, 62–77.
Kimbrough, E. O. and Vostroknutov, A. (2016) “Norms make preferences social,” Journal of the European
Economic Association, 14, 3, 608–638.
Kimbrough, E. O. and Vostroknutov, A. (2018) “A portable method of eliciting respect for social norms,”
Economics Letters, 168, 147–150.
Knight, F. (1922) “Ethics and the economics interpretation,” Quarterly Journal of Economics, 36, 3, 454–481.
Krupka, E. L. and Weber, R. A. (2013) “Identifying social norms using coordination games: Why does dictator
game sharing vary?,” Journal of the European Economic Association, 11, 3, 495–524.
Landsburg, S. E. (2012) The armchair economist: Economics and everyday life. Simon and Schuster.
López-Pérez, R. (2008) “Aversion to norm-breaking: A model,” Games and Economic Behavior, 64, 1, 237–267.
Mackie, G. (1996) “Ending footbinding and infibulation: A convention account,” American Sociological Review,
61, 999–1017.
Peltzman, S. (1975) “The effects of automobile safety regulation,” Journal of Political Economy, 83, 4, 677–725.
Perry, S. (2006) “Hart on social rules and the foundations of law: Liberating the internal point of view,”
Fordham Law Review, 75, 3, 1171–1209.
Piaget, J. (1932) The moral development of the child. London, UK; Kegan Paul.
Schelling, T. C. (1960) The strategy of conflict. Cambridge, MA; Harvard University Press.
Sherif, M. (1936) The psychology of social norms. New York, NY; Harper & Brothers.
Smith, A., (1759/1982) The theory of moral sentiments. Indianapolis, IN; Liberty Fund.
Smith, V. L. and Wilson, B. J. (2019) Humanomics: Moral sentiments and the wealth of nations for the twenty-first
century. New York, NY; Cambridge University Press.
Thrasher, J. and Handfield, T. (2018) “Honor and violence,” Human Nature, 29, 4, 371–389.
Wilson, B. J. (2020) The property species: Mine, yours, and the human mind. New York, NY; Oxford University
Press.
Young, H. P. (1996) “The economics of convention,” Journal of Economic Perspectives, 10, 2, 105–122.
120
10
IMPLICIT BIAS AND
DECISION-MAKING
Lacey J. Davidson
1 Bias
When we make decisions, we might understand ourselves as collecting all the available reasons,
weighing them against each other, and subsequently selecting the best course of action given the
reasons we have considered. However, empirical research shows that human decision-making rarely
happens this way. First, not all our decision-making happens through conscious reflection, like with
a pros and cons list. This is because conscious reflection is costly – it takes a lot of cognitive energy
for us to think about the millions of small decisions we make each day. It is only when we take a
moment, pause, and bring our decisions to mind that we use conscious decision-making
(Kahneman, 2013). So, lots of the time, it is a good thing that we do not have to reason about
every single decision we make – it would be too costly to do so. Second, we are subject to a host of
biases and heuristics that affect our reasoning and decision-making processes (Gilovich, Griffin, and
Kahneman, 2002). We might think that these biases affect us most when we are not consciously
reflecting on our reasons; however, some biases affect us even when we are thinking hard.
Take, for example, the anchoring bias (Tversky and Kahneman, 1974; Chapman and Johnson,
2002). Imagine that you are shopping for a car. Although you have a particular budget in mind, you
are somewhat flexible in how much you are willing to spend. First, the salesperson shows you an
expensive car – one at a price that you are certainly unwilling to pay. Second, the salesperson shows
you another car, one far less expensive than the first car (but perhaps still more expensive than you
were originally thinking). The anchoring bias causes you to compare the price of the second car
with the price of the first; the price of the first car serves as an “anchor” for your comparison. In
comparison to the anchor, the second car seems very reasonably priced! Even though you are
consciously weighing your options for the big decision of purchasing a car, the anchoring bias may
still affect your reasoning without you knowing. This anchoring bias occurs even when the anchor is
generated randomly, for example, through the spin of a fortune wheel (Tversky and Kahneman,
1974) or a role of the dice (Englich, Mussweiler, and Strack, 2006). The anchoring bias is one of
many that affect our reasoning, judgment, and decision-making.
While the anchoring bias affects how we think about numbers, the main topic of this chapter will
be biases that affect how we think about people. Implicit social bias is one of the many biases that
influence our reasoning, but it is significant because of its social consequences. The chapter is
structured as follows: First, I will give an overview of implicit bias. I will identify what researchers
mean when they use the term implicit bias and explore some of its general features. After we have a
DOI: 10.4324/9780367808983-12 121

Lacey J. Davidson
basic understanding of what is captured by the term, I will explain several relationships that implicit
bias has with decision-making, showing the ways our implicit biases can influence our decisions
without our conscious awareness and the ways that this influence cannot be easily prevented
through deliberate reflection. After we understand the relationship between decision-making and
implicit bias, I will review a critique of the implicit bias literature. This critique claims that the focus
on implicit bias overemphasizes the role of individuals’ biased decision-making in explaining the
persistence of unjust material conditions along lines of social identity (racism, sexism, etc.)
(Haslanger, 2015, 2017). To finish the chapter, I will review several interventions that may reduce
the influence of implicit bias on decision-making (Doris, 2015).
2 What Is Implicit Bias?

Although many of our biases, like the anchoring bias discussed above, are implicit in the sense that
they don’t come about as a result of explicit, conscious reasoning, the term implicit bias most often
refers to biases that target members of socially marginalized groups via associations or stereotypes.
This will be the sense that I will focus on in this chapter. Although interest in implicit cognition,
prejudice, or stereotyping isn’t new in philosophy, interest in implicit bias specifically within phi-
losophy grew rapidly starting around 2008 (Kelly and Roedder; Gendler 2008b), and the two-
volume Implicit Bias and Philosophy (Brownstein and Saul, 2016a, 2016b) was published in 2016. The
primary motivation for the growth in curiosity about implicit bias is like the interest in all sorts of
biases; namely, there are cognitive, affective, motivational, and evaluative processes that influence
our reasoning, judgment, and decision-making that are difficult to access through conscious re-
flection. This is in tension with the way that philosophers have historically thought about reasoning
and decision-making. Even more curious is that implicit bias may influence decision-making in ways
that contradict a person’s avowed beliefs, that is, those beliefs they would endorse upon reflection or
when asked. For example, a person may have an avowed commitment to egalitarianism or merit-
based rewards but may also reason or make decisions that privilege a dominant group without
realizing they are doing so.
Researchers have developed a host of indirect measures for tracking implicit biases. The most
common measure is the Implicit Association Test (IAT; Greenwald, McGhee, and Schwartz,
1998).1 The IAT asks participants to sort words or images that are representative of different social
identities or groups into categories. In the stereotype-consistent cases, participants are asked to sort
the words or pictures into categories that are consistent with the social stereotype of that group. For
example, they are asked to sort names associated with women into the category associated with
family or staying at home. In the stereotype-inconsistent cases, participants are asked to do the
opposite, for example, to sort names associated with women into the category associated with
having a career outside of the home. The IAT is calculated by measuring reaction time and errors in
sorting. On average, participants are faster and make fewer errors in the stereotype-consistent case
than in the stereotype-inconsistent case, demonstrating an implicit preference for one social group
over the other (Nosek, Banaji, and Greenwald, 2002). Although there is disagreement about exactly
what implicit bias “picks out” in the mind, we can see here that there is a relationship between
stereotypes about social identities and implicit biases. Some examples of Implicit Association Tests
currently available on the Project Implicit website are Race, Gender, Sexuality, Skin-Tone, Age,
Weight, Religion, and Ethnicity (Project Implicit 2021). Although the IAT is the most popular
implicit measure, there are others such as semantic priming, the Affect Misattribution Procedure,
and the Go/No-go Association Task (Brownstein, 2019). These measures allow researchers to study
the ways that social biases function and to measure the strength and prevalence of these biases. This is
crucial because we care about identifying our biases and the effects these biases have, especially when
these biases produce patterns in decision-making across individuals, places, and times.
122
Implicit Bias and Decision-Making
Let’s make the effects of implicit bias on decision-making more concrete with an example. One
might hold the egalitarian view that a person’s gender or race should not negatively influence
whether they are offered a job. This is also reflected institutionally in anti-discrimination laws that
make it illegal to discriminate based on race, gender, or sexuality. Given both the endorsement of
the egalitarian view and a general desire to obey the law, we might think that prospective employers
are just as likely to hire equally qualified women and candidates of color as they are to hire white
men. However, resume field studies (Bertrand and Mullainathan, 2004; Rooth, 2010; Agerström
and Rooth, 2011) suggest that implicit biases2 affect the evaluation of resumes and call-back rates of
the candidates. Kessler, Low, and Sullivan (2019) found that employers that are focused on Science,
Technology, Engineering, and Mathematics (STEM) candidates are more interested in hiring white
male candidates than either white female candidates or non-white male candidates after reviewing their
resumes (24). Further, the expected call-back rates for white females and non-white candidates were
up to 10% lower than for white men in STEM-focused job searches (24). As noted by Kessler et al.,
we should expect to see no differences between gender and racial identities because the companies
included in the study had explicit egalitarian beliefs and employers had “attempted to override any
discriminatory preferences” (24). The authors attribute the inconsistency between the employer’s
explicitly-held beliefs and their behaviors to implicit biases that affect the resume evaluation process,
especially when fatigued. Researchers found that subjects spent less time with resumes as they
became fatigued within the study and use fatigue as evidence of the influence of implicit biases
(24–25, 82). From this example, we can see the ways that implicit biases can influence decision-making
and contribute to the maintenance of social inequity. Implicit bias is a concern for those with
commitments to egalitarianism, anti-racism, or feminism.
3 Implicit Bias and Decision-Making

Now that we have described implicit bias, we can explore some further contexts and situations in
which implicit bias influences decision-making. Like our other biases, our implicit biases are with us
as we walk through the world. They may influence whether we interpret someone as passionate
rather than aggressive or angry, whether we cross the road when we pass a stranger on the sidewalk,
how we like or dislike a character on an evening sitcom, and whether we listen or interrupt a person
when they are talking. These everyday contexts are certainly very serious, worthy of our attention,
and shape our experiences with the world.3 However, there are contexts where the influence of
implicit bias on decision-making is especially salient given the potential for compounding effects or
because of the life-or-death stakes involved. In what follows, I will discuss the role of implicit bias in
education, employment, and the medical and legal systems.
3.1 Education
In the United States, it is illegal to discriminate on the basis of age, disability, gender, national origin,
race, or religion for educational opportunities, and we hope that educators have avowed com-
mitments to educating all of their students equitably. However, there are differences in educational
attainment across social identities that map onto or align with other social and economic disparities
(Espinosa et al., 2017). Implicit bias is likely one among the many, complexly interrelated factors
contributing to differences in educational outcomes. Jacoby-Senghor, Sinclair, and Shelton (2016)
found that implicit racial biases of white instructors (measured by the subliminal priming task)
predicted lower assessment outcomes for Black, but not white students. In addition, they found that
coders rated the instructors as more anxious and their lessons as lower in quality when they were
teaching to Black rather than white students. In fact, even non-Black participants who watched
recordings of white instructors who had presented material to Black students learned less, as
123
Lacey J. Davidson
indicated by a test measuring how well they absorbed the material. As the researcher’s note, one
limitation of their study is the simulated nature of the study (the participants weren’t teachers in
actual classrooms); however, they remain confident that their findings are applicable given that
cross-race teaching dynamics are common in the United States and that new teachers are more likely
to teach in schools with higher populations of minoritized students. In line with this study, van den
Bergh et al. (2010) found that IAT scores explained the size of ethnic achievement gaps within
classrooms. Research in Italy also shows the influence of implicit bias on education. Carlana (2019)
found that the gender gap in math performance increases as the implicit bias of the instructor
increases (as measured by the gender-science IAT) and Alesina et al. (2018) found that math teachers
with stronger implicit biases gave lower scores to immigrant children than to non-immigrant
children even when the performance was identical.
Alongside the quality of instruction, there are also documented differences in disciplinary
choices. Intuitively, students need to be in the classroom and at school in order to learn effectively.
This means that decisions to use disciplinary practices that exclude students from the learning en-
vironment or make students feel like they do not belong can have compounding effects for student
learning. Okonofua and Eberhardt (2015) found that teachers are more likely to increase the severity
of the punishment after only two infractions for Black students than for white students and more
likely to view Black students as troublemakers. Further, teachers perceived Black students’ infrac-
tions as indicative of a pattern rather than an isolated incident, and they were more likely to predict
that they would suspend the child in the future. Gilliam et al. (2016) found that early childhood
educators spent significantly more time watching Black boys in expectation of poor behavior than
other students. Researchers explain the difference in dwelling times with implicit bias findings that
measure associations between race and criminality. In other words, the educators may have more
strongly associated the Black boys with poor or delinquent behavior, which causes them to spend
more time dwelling on the Black boys in the study. In these cases, implicit biases of educators
influence the decisions they make about attention and disciplinary practices, which influences the
overall experience of students at school and their ability to achieve academically.
3.2 Employment
As discussed above, implicit biases affect the evaluation of resumes and the likelihood that applicants
receive a callback for an interview. If we have egalitarian commitments, we should be interested in
and concerned about the ways that implicit biases shape hiring practices and company culture. A
further concern is that implicit biases shape our self-conception and beliefs about others that embody
the same identities. For example, when primed with images of men occupying traditionally male-
associated roles (medical doctor) and women occupying traditionally female-associated roles (nurse),
women showed greater implicit biases on a power-warmth IAT (Rudman and Phelan, 2010). In
other words, the stereotype-consistent prime increased their work-related implicit biases in com-
parison to a control condition. However, women who were shown the atypical condition (women
in traditionally male roles and men in traditionally female roles), scored lower on a self-concept IAT
meant to test their association between themselves and leadership characteristics. The authors suggest
that the atypical prime represents an “upward social comparison threat” that reduces women’s ability
to identify with the role and have similar career aspirations.4 Further, in comparison to controls,
both the traditional and non-traditional primes led to reduced interest in masculine-typed jobs.
Rudman and Phelan conclude that these findings demonstrate that “modifying implicit gender
stereotypes can have downstream consequences that impact gender equality” (198). With both
examples, we see the role that implicit bias may be playing in the production and maintenance of
employment inequity and the wage gap.
124
3.3 Medical System

Implicit biases are also present and play a role in the provision of medical care. This is particularly
critical given the stakes involved and the avowed commitments to equal treatment by medical
doctors and other healthcare providers. We think that a general principle guiding healthcare pro-
vider decision-making is that everyone is deserving of quality medical regardless of their social
identity. In a meta-analysis of thirty-seven studies, Maina et al. (2018) found that twenty-six out of
the thirty-one studies that reported Race IAT data found evidence of pro-white/anti-Black implicit
biases in healthcare providers. Studies within the meta-analysis also found evidence for slight to
strong pro-white/anti-Hispanic implicit biases, moderate pro-white/anti-American Indian implicit
biases, and slight to modern pro-light-skinned/anti-dark-skinned bias. Although the results are
varied, some vignette-based studies (but not a majority) showed an association between implicit bias
and healthcare outcomes. Seven out of nine studies demonstrated an association between implicit
bias and actual real-world patient-provider communication. For example, Black patients ranked
providers with high pro-white implicit biases lower on multiple measures and reported lower overall
satisfaction. In other words, implicit biases seem to influence the decisions that primary care pro-
viders make as they communicate with and treat their patients.
In addition, medical doctors, alongside the general population of IAT test-takers, show strong
implicit anti-fat biases (Sabin, Marini, and Nosek, 2012). The biases of medical doctors have the po-
tential to influence the medical care that fat people receive. Medical students who harbor anti-fat
implicit biases spend less time with fat patients and were more likely to be exposed to discriminatory
behavior against fat people within their medical training (Phelan et al., 2015). These findings de-
monstrate some of the ways that implicit biases can influence the practice of medicine and the decisions
made by healthcare providers. However, unlike some other biases, medical doctors are often explicitly
biased against their fat patients; Foster and colleagues found that over 50% of surveyed doctors found
that their fat patients were “awkward, unattractive, ugly, and noncompliant” (1168). Given that
medical care is a matter of life and death, implicit biases within medical practice are very concerning.
Implicit bias can certainly help us understand the experiences of patients who embody targeted
identities, and the structural features of access to healthcare – such as the affordability, proximity, and
availability of preventative care and treatments – as well as other experiences caused by the oppression
that negatively influence wellbeing, are also critical in explaining disparities in healthcare outcomes.
3.4 The Criminal System

As with the previous contexts, there are stark and well-documented differences along social identity
lines, especially race, within the legal system. These disparities have many causes, and most of them
cannot be boiled down to what’s in the heads or bodies of individuals. Nevertheless, implicit biases
play a documented role in decision-making within the legal system.
As demonstrated by the worldwide uprising against racism in policing and police murders, a
critical area of investigation is the role that implicit racial bias might play in a police officer’s decision
to use excessive force that leads to the assault or murder of an unarmed citizen. One explanation is
that implicit biases alter our perception. In a police officer’s dilemma study, Correll et al. (2015)
found that the race of the subject influenced the participant’s perception of the object that the
subject was holding in a simulation. For example, participants were quicker to identify a gun in a
Black man’s hand than in a white man’s hand and slower to identify the object in the reverse case.
These findings are consistent with previous studies focused on race and crime-related objects
(Correll et al., 2007; Sadler et al., 2012), and Sadler et al. (2012) found that this negative bias is also
demonstrated for Latinos, but not Asians. These studies suggest that implicit biases and racial ste-
reotyping play a role in perception in high-stakes scenarios like a police-civilian interaction. The
125
Lacey J. Davidson
differences in perceptual processing speeds may then influence the decisions that are made by police
officers. When taken together, the implicit biases of individuals play a role in producing differential
treatment that leads to higher incarceration rates and police brutality.
Implicit biases play a role in decision-making in police-civilian interactions, but the influence of
implicit bias in the legal system doesn’t end there. An implicit bias may cause a person to encode,
store, or retrieve a memory in a biased way, which can influence decisions made by jury members.
In an empirical study using legal stories, Levinson (2007) demonstrated that the race of an individual
involved in a case can implicitly affect the memory of mock jurors. For example, when contrasting
the ability of the mock jurors to remember details about a legal story with “Tyronne” versus
“William,” the mock jurors were more likely to “remember” Tyronne as aggressive, even when
Tyrone and William’s behaviors were described in the same way. As is consistent with other implicit
bias measures, those who were most likely to misremember details of the legal story did not always
harbor explicit racial biases and sometimes even demonstrated less explicit racial bias. The purpose of
Levinson’s study is to demonstrate the connection between implicitly biased memory and legal
decision-making. Although Levinson’s study doesn’t show that implicit bias is affecting a juror at the
time a decision is made (say in the moment of deciding between an “innocent” or “guilty” verdict),
it does show that implicit biases are influencing their memory of the details of the case itself, which
subsequently influences their deliberations and decision-making. As with the medical system,
decision-making within the legal system can be life or death and certainly have long-term con-
sequences for individuals, and can contribute to the maintenance of unjust social conditions.
I started this section by emphasizing that implicit biases influence the ways that we interact with
the world in small and large ways, and I have reviewed some of the domains in which implicit biases
have critical and severe consequences. This is because of the compounding effects and stakes within
these domains. These examples demonstrate the ways in which implicit biases affect decision-
making, even when individuals hold avowed commitments to equity. In addition, it often does not
seem to the person that implicit bias is influencing their decisions, so the influence of implicit bias is
hidden or covered over. As mentioned above, implicit bias is not the only thing that matters within
these domains, and as will be discussed below, it is likely not the most important feature in ex-
plaining the persistence of social oppressions. However, it is clear that implicit biases influence our
decision-making in ways that are inconsistent with our goals.
4 The Importance of Implicit Bias

One of the primary areas of debate within philosophy connected to implicit bias is the relative
importance of implicit bias in explaining inequitable outcomes or conditions of oppression. Some
have argued that the swift growth in interest in implicit bias overemphasized the importance of an
individual psychological measure in explaining continued conditions of social inequity, injustice,
and oppression (Haslanger, 2015). For example, Banks and Ford (2011) write, “While we do not
doubt the existence of unconscious bias, we doubt that contemporary racial bias accounts for all, or
even most, of the racial injustice that bedevils our society. The racial injustices that are most sub-
stantive … stem from a complex interplay of economic, historical, political and social influences”
(1). This debate is sometimes understood as the individualist versus structuralist debate (Madva,
2016, Davidson, 2019). The primary question within this debate is what kinds of things – individuals
or structures – are more important when analyzing the social world and the stark differences in
situations and outcomes along the lines of race, gender, sexuality, or other social identities? Within
this debate, implicit bias is largely understood as an individualist entity (inside the heads of in-
dividuals), where things like policies, laws, and institutions are understood as structures (parts of
social and physical reality outside the head that incentivize and constrain what individuals do).5
126
Although this debate certainly is not settled, there are a growing number of philosophers who argue
for the priority of structures over individuals with some arguing that individuals do not matter at all for
explaining oppression (Martín 2020). Some people who are attributed with the structural view are
Anderson (2010) in her discussion of segregation and the moral imperative of integration, Banks and
Ford (2011) in their discussion of the relative non-importance of implicit bias, Haslanger (2015) in her
analysis of cases that have racist or sexist outcomes without individual implicit biases, and Ayala-López
(2018) in their analysis of the norms that produce discursive injustices. In her argument against in-
dividualistic explanations, Martín argues that we should re-conceptualize implicit biases as structural
entities; in other words, the individualist versus structuralist debate is resolved by arguing that there is
only one category of things to be concerned about with respect to oppression.
In a move in the other direction, Kelly and I (Davidson and Kelly, 2018) argue that social norms
are the soft structures that connect individuals up to formal structures. We argue that there are (at
least) three levels to attend to rather than the binary choice between individuals and structures. The
framework of social norms allows us to explain why biases are internalized by individuals, protected
through unconscious social processes, and are sustained group-level behavioral regularities.
Although the addition of this paper does not resolve the individualist versus structuralist debate, it
provides us with more tools for understanding the relation between individuals and structures.
Madva (2016, 2020) has also emphasized the importance of individual attitudes and beliefs in making
structural changes; the main idea here is that it takes individuals who are committed to justice and
liberation to develop, propose, and accept the kinds of widespread structural reforms that are ne-
cessary for transforming our world.
No matter where you land in the debate, there is widespread agreement that implicit biases are
not the only thing that matters when it comes to describing or explaining unequal, unjust, or
oppressive conditions and most are unwilling to give priority to implicit biases over structures,
though some argue for equal priority (Soon, 2020). My own preferred view is that we should be
pluralists about the kinds of things we think matter depending on our goals and the context that we
are in.6 For example, if we are trying to understand our own or another’s experiences of being
harmed by a person (maybe us!) with explicitly held commitments toward a liberated future, then
appealing to implicit bias and more particularly the implicit biases of that person will allow us to
understand our experiences. We might even have an opportunity to call on that person to reflect on
their biases, which might change our or other’s experiences with them in the future. But it certainly
is not the case that the implicit biases of that particular person will account for the patterns of
experiences that people with shared identities will have in common. This does not make implicit
bias irrelevant to our discussions of injustice or oppression, but it does demonstrate that the context
and goals of our discussion will influence whether or not implicit bias is central.
5 Interventions on Implicit Bias

The debate about the relative importance of implicit bias in comparison to structural features in
explaining unjust or oppressive conditions naturally leads to a discussion of interventions. We might
think that if we view individuals or structures as more relevant to our explanations of oppression,
then one or the other will be more important for addressing and mitigating the racist, sexist, classist
conditions we wish to alter. There is certainly a healthy debate about this. Alongside Madva (2016,
2020), I prefer a both/and or pluralistic approach that focuses on individuals and structures. For
example, a legislative change (a structural change) might be an important step in achieving justice,
but it takes very many dedicated individuals with particular attitudes, beliefs, commitments, and
skillsets to produce the kind of movement that results in landmark legislative decisions. However,
the evidence is clear that focusing on individuals alone is not going to be enough to change the
world.
127
Lacey J. Davidson
One of the most troubling findings in the implicit bias literature is that just learning more about
implicit bias does not by itself do much to reduce implicit bias. For example, research shows that
addressing bias directly through diversity trainings for management can do very little or even harm
the odds for increasing diversity in management, especially for Black women (Kalev, Dobbin, and
Kelly, 2006). In addition, researchers found that some interventions, like exposure to vivid counter
stereotypic scenarios or using implementation intentions (e.g., If I see a Black face, I will respond by
thinking “good.”), are effective in reducing implicit biases immediately but are not effective after a
delay. In other words, although interventions have a short-term effect on implicit biases, there is no
evidence of lasting changes to implicit biases (Lai et al., 2016).
However, there are some strategies for reducing bias that show some promise. For example,
Gehlbach et al. (2016) found that identifying similarities or common ground within groups of high
school students and their teachers led to better-perceived relationships and higher course grades.
Similarly, Simonovits, Kézdi, and Kardos (2018) found that a perspective-taking game reduced
prejudice against those whose perspective was taken and that the reduction in prejudice lasted for at
least a month. Levontin, Halperin, and Dweck (2013) were also able to reduce prejudice and ne-
gative attributions by introducing information like “traits are malleable” to participants. In other
words, when the participants were prompted to think of decisions or behaviors as contextual or
circumstantial rather than stable character traits of groups, their negative attributions were reduced
(for a review of the fundamental attribution error, see Knight, 2008). Finally, there is evidence that
using an approach mindset (focusing on what to do rather than what not to do) in anticipation of an
interracial interaction reduces the negative impact on an executive function that would otherwise
occur after an interracial interaction (Trawalter and Richeson, 2006). This is related to implicit bias
given that one thing we might think as we navigate the world is “don’t be biased,” but this research
shows that identifying things you want to do rather than what you don’t is more effective. Madva
(2020) discusses these interventions in more detail.
Another potential intervention is mindfulness as an integrated practice. Leboeuf (2018) argues
that mindfulness practices like meditation have the potential to allow people to shift their perception
and change their affective responses, specifically their “racializing bodily habits,” in ways that reduce
the influence of implicit biases. This is because rather than conceptualizing implicit bias as “all in the
head” as they are often conceptualized, Leboeuf reconceptualizes implicit bias as an embodied
phenomenon shaped by the social world (Leboeuf, 2020). In addition, Leboeuf emphasizes the ways
that our bodies are also shaping the social world as we enact our habits. Focusing on the embo-
diment of implicit bias shifts the conversation from bad attitudes to bodily habits7 that can be
transformed or shifted through somatic practices. Strategies of this type are explored in Menakem’s
(2017) My Grandmother’s Hands.
Though there are some individual strategies that seem to either reduce biases or reduce their
influence, structural interventions are also essential. As discussed above, implicit bias plays a role in
hiring and promotion practices, but requiring anti-bias trainings may have little or no effect.
However, the same study that shows the limited efficacy of bias trainings demonstrated that both
diversity committees and affirmative action plans, strategies that address implicit bias indirectly, are
effective in increasing the diversity of management over time (Kalev et al., 2006).
In an interview focused on reducing implicit biases, Greenwald (Mason, 2020) suggests that
discretion elimination is a promising structural strategy for reducing the influence of implicit biases.
The primary strategy is to remove identity markers in evaluation when possible; this strategy is also
referred to as anonymization. The famous example that Greenwald discusses are the results of
“blind” auditions for major symphonies. The practice of auditioning behind a screen increased the
gender diversity, even though the audition panels did not take themselves to be gender-biased.
And there is more evidence to support this strategy. In an analysis of four actual medical cases,
researchers identified the elimination of discretion as a possible way to increase equitable outcomes
128
(Capers, Bond, and Nori, 2020). In one of the four cases, a Black woman was not placed on the
transplant list while a white man was placed on the list, even though they both displayed similar
concerning behaviors (participation in illegal activity, potential non-compliance, and being “opi-
nionated”). The authors argue that discretion elimination may have prevented the committee’s
implicit preferences for white people from influencing their decision about the kidney transplant. In
this case, discretion is eliminated by pre-determining the factors that can be considered when
making a transplant list decision. For example, if engaging in illegal activity was not an approved
consideration for not being placed on the transplant list, then illegal activity could not be used as
justification for denial in either case. In this case, this would mean that both the white man and the
Black woman would have been placed on the transplant list. Cases like this reveal how the reduction
of discretion can sometimes be a matter of life and death.
The discussion of interventions is intertwined with the individualist versus structural debate. The
kinds of interventions we think are most salient will likely map onto the features of our social world
that we think explain the existence and persistence of oppression in virtue of social identity. Of
course, we might think that the solution to a problem might be different than the cause (Madva,
2016); in other words, we might think that the kinds of processes that led to a particular social
inequity is not what should be targeted to address it. Given limited resources, including time, as well
as the potential for our interventions to have unintended consequences, our decisions about how to
go about dismantling systems of oppression are complex and vital. What is clear, however, is that
whatever intervention or solution we take on, it won’t be a one-and-done fix, but a lifelong project.
6 Conclusion
The empirical and philosophical literature on implicit bias is large and the debates are lively. In this
chapter, I have situated implicit bias as a family of biases with similar effects among many various
biases, explained the primary features of implicit bias, identified situations and contexts in which
implicit biases influence individual decision-making, highlighted a lively debate with respect to
implicit bias, and reviewed some of the potential interventions on implicit bias. Other questions and
disagreements remain about what exactly implicit biases are measuring or the cognitive structure of
implicit bias (e.g., what mental structures or processes do the implicit measures “pick out”) (for a
start, see Gendler, 2008a, 2008b; Schwitzgebel, 2010; Brownstein, 2018; Johnson, 2020) and re-
sponsibility for implicit bias (e.g., can we be responsible for something that influences our decisions
and behaviors without our knowledge or control?) (for an overview, see Holroyd, Scaife, and
Stafford, 2017; for further examples, see chapters in Brownstein and Saul, 2016a).
Although the evidence regarding the influence of implicit bias on our decisions and behaviors is
vast, a large and growing body of evidence also highlights the promises and challenges of using this
knowledge to change our world. While implicit bias is not the primary obstacle against transforming
the material conditions of our world, implicit biases continue to harm individuals who embody
already marginalized identities within the current social and political conditions. Anyone interested
in changing the world must navigate these tensions within themselves and as they make decisions in
their interactions with family and friends and as they raise their children, buy and consume products,
engage in their vocations and as a part of their communities, and vote in local, state, and national
elections. Our commitments to egalitarianism or anti-oppression demand us to stay in the struggle
against intersecting and interlocking systems of oppression.
Notes
1 For an overview of the methods, design, and methodological strengths and challenges of the IAT, see
Nosek, Greenwald, and Banaji (2007). For a very accessible overview and introduction, see Banaji and
Greenwald (2013).
129
Lacey J. Davidson
2 Bertrand and Mullainathan (2004) used names to represent the race of the candidates and did not collect
IAT data.
3 For example, the weathering hypothesis claims that consistent exposure to social and economic dis-
advantages, microaggressions, and racial trauma negatively influences well-being and causes ill-health (Forde
et al., 2019; Allahand, 2021).
4 This replicates a finding reported in Davies, Spencer, and Steele (2005).
5 As noted by Ayala-López and Beeghly (2020), Madva (2020), and Brownstein (2020), there is not one
consistent way in which theorists deploy “individuals” and “structures” within this debate. Haslanger (2015)
uses the term “social structures,” which she understands as a “network of practices” that “consist of behavior
that conforms to cultural schemas in response to resources” (3–4). On this broad interpretation of social
structures, structure includes everything but the attitudes and behaviors of individual people. Structures then
include patterns of behavior that are made to be patterns by many individuals, but not the behaviors of the
individuals themselves.
6 And perhaps lots of sturcturalists will also be open to this. As explored briefly above, Martín (2020) focuses on
explaining oppression and argues that implicit biases as individual features cannot explain oppression. It could be
then on Martín’s view that implicit bias could be relevant to other kinds of non-explanatory projects.
7 Brownstein (2018) also argues for a habits-based understanding of implicit attitudes, which he calls the “the
habit stance.”
References
Alesina, Alberto, Michela Carlana, Eliana La Ferrara, and Paolo Pinotti. (2018) “Revealing Stereotypes:
Evidence from Immigrants in Schools” (Working Paper #25333). National Bureau of Economic Research.
10.3386/w25333.
Allahand, Ashley. (2021) “The Trauma of Racial Microaggressions: The Psychological Impact of
Discriminatory Actions Affecting the Minority Population.” Journal of Advances in Social Science and
Humanities 7, 5, 1779–1787. 10.15520/jassh.v7i5.615.
Anderson, Elizabeth. (2010) The Imperative of Integration. Princeton, NJ; Princeton University Press.
Agerström, Jens and Dan-Olof Rooth. (2011) “The Role of Automatic Obesity Stereotypes in Real Hiring
Discrimination.” The Journal of Applied Psychology 96, 4, 790–805. 10.1037/a0021594.
Ayala-López, Saray. (2018) “A Structural Explanation of Injustice in Conversations: It’s About Norms.” Pacific
Philosophical Quarterly 99, 4, 726–748. 10.1111/papq.12244.
Ayala-López, Saray, and Erin Beeghly. (2020) “Explaining Injustice: Structural Analysis, Bias, and Individuals.”
in Erin Beeghly and Alex Madva edited by. An Introduction to Implicit Bias: Knowledge, Justice, and the Social
Mind, 211–232. New York, NY; Routledge.
Banaji, Mahzarin R., and Anthony G. Greenwald. (2013) Blindspot: Hidden Biases of Good People. Delacourte
Press.
Banks, Ralph Richard, and Richard Thompson Ford. (2011) “Does Unconscious Bias Matter?” Poverty and
Race 20, 5, 1–2.
Bertrand, Marianne, and Sendhil Mullainathan. (2004) “Are Emily and Greg More Employable than Lakisha
and Jamal?” American Economic Review 94, 991–1012. 10.3386/w9873.
Brownstein, Michael. (2018) The Implicit Mind: Cognitive Architecture, the Self, and Ethics. New York, NY;
Brownstein, Michael. (2019) “Implicit Bias.” in Edward Zalta edited by. The Stanford Encyclopedia of Philosophy,
Metaphysics Research Lab, Stanford University. https://plato.stanford.edu/archives/fall2019/entries/
implicit-bias/.
Brownstein, Michael. (2020) “Skepticism about Bias.” in Erin Beeghly and Alex Madva edited by. An
Introduction to Implicit Bias: Knowledge, Justice, and the Social Mind, 57–76. New York, NY; Routledge.
Brownstein, Michael, and Jennifer Saul, eds. (2016a) Implicit Bias and Philosophy, Volume 2: Moral Responsibility,
Structural Injustice, and Ethics. Oxford, United Kingdom; Oxford University Press. 10.1093/acprof:oso/
9780198766179.001.0001.
Brownstein, Michael, and Jennifer Saul, eds. (2016b) Implicit Bias and Philosophy, Volume 1: Metaphysics and
Epistemology. Oxford, United Kingdom; Oxford University Press. 10.1093/acprof:oso/9780198713241.001.
0001.
Capers, Quinn, David A. Bond, and Uday S. Nori. (2020) “Bias and Racism Teaching Rounds at an Academic
Medical Center.” Chest 158, 6, 2688–2694. 10.1016/j.chest.2020.08.2073.
130
Chapman, Gretchen B., and Eric J. Johnson. (2002) “Incorporating the Irrelevant: Anchors in Judgments of
Belief and Value.” in Thomas Gilovich, Dale Griffin, and Daniel Kahneman edited by. Heuristics and Biases,
1st ed., 120–138. Cambridge; Cambridge University Press. 10.1017/CBO9780511808098.008.
Carlana, Michela. (2019) “Implicit Stereotypes: Evidence from Teachers’ Gender Bias.” The Quarterly Journal of
Economics 134, 3, 1163–1224.
Correll, Joshua, Bernadette Park, Charles M. Judd, and Bernd Wittenbrink. (2007) “The Influence of Stereotypes
on Decisions to Shoot.” European Journal of Social Psychology 37, 6, 1102–1117. 10.1002/ejsp.450.
Correll, Joshua, Bernd Wittenbrink, Matthew T. Crawford, and Melody S. Sadler. (2015) “Stereotypic Vision:
How Stereotypes Disambiguate Visual Stimuli.” Journal of Personality and Social Psychology 108, 2, 219–233.
10.1037/pspa0000015.
Davidson, Lacey J. (2019) “That’s (Also) Racist! Entity Type Pluralism, Responsibility, and Liberatory Norms.”
Doctoral Dissertation, Purdue University.
Davidson, Lacey J., and Daniel Kelly. (2018) “Minding the Gap: Bias, Soft Structures, and the Double Life of
Social Norms.” Journal of Applied Philosophy, December. 10.1111/japp.12351.
Davies, Paul G., Steven J. Spencer, and Claude M. Steele. (2005) “Clearing the Air: Identity Safety Moderates
the Effects of Stereotype Threat on Women’s Leadership Aspirations.” Journal of Personality and Social
Psychology 88, 2, 276–287. 10.1037/0022-3514.88.2.276.
Doris, John. (2015) Talking to Ourselves: Reflection, Ignorance, and Agency. Oxford, UK; Oxford University Press.
Englich, Birte, Thomas Mussweiler, and Fritz Strack. (2006) “Playing Dice With Criminal Sentences: The
Influence of Irrelevant Anchors on Experts’ Judicial Decision Making.” Personality and Social Psychology
Bulletin 32, 2, 188–200. 10.1177/0146167205282152.
Espinosa, Lorelle L, Jonathan M Turk, Morgan Taylor, and Hollie M Chessman. (2017) “Race and Ethnicity in
Higher Education: A Status Report.” American Council on Education. https://www.equityinhighered.org/
indicators/u-s-population-trends-and-educational-attainment/educational-attainment-by-race-and-ethnicity/.
Forde, Allana T., Danielle M. Crookes, Shakira F. Suglia, and Ryan T. Demmer. (2019) “The Weathering
Hypothesis as an Explanation for Racial Disparities in Health: A Systematic Review.” Annals of Epidemiology
33, May, 1–18.e3. 10.1016/j.annepidem.2019.02.011.
Gehlbach, Hunter, Maureen E. Brinkworth, Aaron M. King, Laura M. Hsu, Joseph McIntyre, and Todd
Rogers. (2016) “Creating Birds of Similar Feathers: Leveraging Similarity to Improve Teacher–Student
Relationships and Academic Achievement.” Journal of Educational Psychology 108, 3, 342–352. 10.1037/
edu0000042.
Gendler, Tamar Szabó. (2008a) “Alief and Belief.” The Journal of Philosophy 105, 10, 634–663.
Gendler, Tamar Szabó. (2008b) “Alief in Action (and Reaction).” Mind & Language 23, 5, 552–585. 10.1111/
j.1468-0017.2008.00352.x.
Gilliam, Walter S, Angela N Maupin, Chin R Reyes, Maria Accavitti, and Frederick Shic. (2016) “Do Early
Educators’ Implicit Biases Regarding Sex and Race Relate to Behavior Expectations and Recommendations
of Preschool Expulsions and Suspensions?,” 18.
Gilovich, Thomas, Dale Griffin, and Daniel Kahneman, eds. (2002) Heuristics and Biases: The Psychology of
Intuitive Judgment. 1st ed. Cambridge; Cambridge University Press. 10.1017/CBO9780511808098.
Greenwald, Anthony G., Debbie E. McGhee, and Jordan L. K. Schwartz. (1998) “Measuring Individual
Differences in Implicit Cognition: The Implicit Association Test.” Journal of Personality and Social Psychology
74, 6, 1464–1480. 10.1037/0022-3514.74.6.1464.
Haslanger, Sally. (2015) “Distinguished Lecture: Social Structure, Narrative and Explanation.” Canadian Journal
of Philosophy 45, 1, 1–15. 10.1080/00455091.2015.1019176.
Haslanger, Sally. (2017) “Racism, Ideology, and Social Movements.” MIT Web Domain, January. https://
dspace.mit.edu/handle/1721.1/116159.
Holroyd, Jules, Robin Scaife, and Tom Stafford. (2017) “Responsibility for Implicit Bias.” Philosophy Compass
12, 3, e12410. 10.1111/phc3.12410.
Jacoby-Senghor, Drew S., Stacey Sinclair, and J. Nicole Shelton. (2016) “A Lesson in Bias: The Relationship
between Implicit Racial Bias and Performance in Pedagogical Contexts.” Journal of Experimental Social
Psychology 63, March, 50–55. 10.1016/j.jesp.2015.10.010.
Johnson, Gabbrielle M. (2020) “The Structure of Bias.” Mind 129, 516, 1193–1236. 10.1093/mind/fzaa011.
Kahneman, Daniel. (2013) Thinking, Fast and Slow. 1st pbk. ed. New York; Farrar, Straus and Giroux.
Kalev, Alexandra, Frank Dobbin, and Erin Kelly. (2006) “Best Practices or Best Guesses? Assessing the Efficacy
of Corporate Affirmative Action and Diversity Policies.” American Sociological Review 71, 4, 589–617. 10.1177/
000312240607100404.
131
Lacey J. Davidson
Kelly, Daniel, and Erica Roedder. (2008) “Racial Cognition and the Ethics of Implicit Bias.” Philosophy
Compass 3/3, 522–540. 10.1111/j.1747-9991.2008.00138.x.
Kessler, Judd, Corinne Low, and Colin Sullivan. (2019) “Incentivized Resume Rating: Eliciting Employer
Preferences without Deception.” w25800. Cambridge, MA; National Bureau of Economic Research. 10.3386/
w25800.
Knight, Kathleen. (2008) “In Their Own Words: Citizens’ Explanations of Inequality Between the Races.”
inJon Hurwitz and Mark Peffley edited by. Perception and Prejudice: Race and Politics in the United States,
202–232. Yale University Press.
Lai, Calvin K., Allison L. Skinner, Erin Cooley, Sohad Murrar, Markus Brauer, Thierry Devos, Jimmy
Calanchini, et al. (2016) “Reducing Implicit Racial Preferences: II Intervention Effectiveness across Time.”
Journal of Experimental Psychology: General 145, 8, 1001–1016. 10.1037/xge0000179.
Leboeuf, Céline. (2018) “Reforming Racializing Bodily Habits: Affective Environment and Mindfulness
Meditation.” Critical Philosophy of Race 6, 2, 164–179.
Leboeuf, Céline. (2020) “The Embodied Biased Mind.” In Erin Beeghly and Alex Madva edited by. An
Levinson, Justin D. (2007) “Forgotten Racial Equality: Implicit Bias, Decision-Making and Misremembering.”
Duke Law Journal 57. https://papers.ssrn.com/abstract=927547.
Levontin, Liat, Eran Halperin, and Carol S. Dweck. (2013) “Implicit Theories Block Negative Attributions
about a Longstanding Adversary: The Case of Israelis and Arabs.” Journal of Experimental Social Psychology 49,
4, 670–675. 10.1016/j.jesp.2013.02.002.
Madva, Alex. (2016) “A Plea for Anti-Anti-Individualism: How Oversimple Psychology Misleads Social
Policy.” Ergo, an Open Access Journal of Philosophy 3, 20180912. 10.3998/ergo.12405314.0003.027.
Madva, Alex. (2020) “Individual and Structural Interventions.” in Erin Beeghly and Alex Madva edited by. An
Maina, Ivy W., Tanisha D. Belton, Sara Ginzberg, Ajit Singh, and Tiffani J. Johnson. (2018) “A Decade of
Studying Implicit Racial/Ethnic Bias in Healthcare Providers Using the Implicit Association Test.” Social
Science & Medicine 199, February, 219–229. 10.1016/j.socscimed.2017.05.009.
Martín, Annette. (2020) “Explaining Oppression: An Argument Against Individualism.” American Philosophical
Association. https://www.apaonline.org/page/2021C_Accepted.
Mason, Betsy. (2020) “Curbing Implicit Bias: What Works and What Doesn’t.” Knowable Magazine | Annual
Reviews, June. 10.1146/knowable-060320-1.
Menakem, Resmaa. (2017) My Grandmother’s Hands: Racialized Trauma and the Pathway to Mending Our Hearts
and Bodies. Central Recovery Press.
Nosek, Brian A., Anthony G. Greenwald, and Mahzarin R. Banaji. (2007) “The Implicit Association Test at
Age 7: A Methodological and Conceptual Review.” In Automatic Processes in Social Thinking and Behavior,
edited by J. A. Bargh, 265–292. New York; Psychology Press.
Nosek, Brian A., Mahzarin Banaji, and Anthony G. Greenwald. (2002) “Harvesting Implicit Group Attitudes
and Beliefs from a Demonstration Web Site.” Group Dynamics: Theory, Research, and Practice 6, 1, 101–115.
10.1037//1089-2699.6.1.101.
Okonofua, Jason A., and Jennifer L. Eberhardt. (2015) “Two Strikes: Race and the Disciplining of Young
Students.” Psychological Science 26, 5, 617–624. 10.1177/0956797615570365.
Phelan, Sean M., Rebecca M. Puhl, Sara E. Burke, Rachel Hardeman, John F. Dovidio, David B. Nelson, Julia
Przedworski, et al. (2015) “The Mixed Impact of Medical School on Medical Students’ Implicit and Explicit
Weight Bias.” Medical Education 49, 10, 983–992. 10.1111/medu.12770.
Rooth, Dan-Olof. (2010) “Automatic associations and discrimination in hiring: Real world evidence.” Labour
Economics 17, 3, 523–534. 10.1016/j.labeco.2009.04.005
Rudman, Laurie A., and Julie E. Phelan. (2010) “The Effect of Priming Gender Roles on Women’s Implicit
Gender Beliefs and Career Aspirations.” Social Psychology 41, 3, 192–202. 10.1027/1864-9335/a000027.
Sabin, Janice A., Maddalena Marini, and Brian A. Nosek. (2012) “Implicit and Explicit Anti-Fat Bias among a
Large Sample of Medical Doctors by BMI, Race/Ethnicity and Gender.” Edited by Richard Fielding. PLoS
ONE 7, 11, e48448. 10.1371/journal.pone.0048448.
Sadler, Melody S., Joshua Correll, Bernadette Park, and Charles M. Judd. (2012) “The World Is Not Black and
White: Racial Bias in the Decision to Shoot in a Multiethnic Context.” Journal of Social Issues 68, 2,
286–313. 10.1111/j.1540-4560.2012.01749.x.
Schwitzgebel, Eric. (2010) “Acting Contrary to Our Professed Beliefs or The Gulf Between Occurrent
Judgment and Dispositional Belief.” Pacific Philosophical Quarterly 91, 531–553.
132
Simonovits, Gábor, Gábor Kézdi, and Péter Kardos. (2018) “Seeing the World Through the Other’s Eye: An
Online Intervention Reducing Ethnic Prejudice.” American Political Science Review 112, 1, 186–193. 10.1017/
S0003055417000478.
Soon, Valerie. (2020) “Implicit Bias and Social Schema: A Transactive Memory Approach.” Philosophical Studies
177, 7, 1857–1877. 10.1007/s11098-019-01288-y.
Trawalter, Sophie, and Jennifer A. Richeson. (2006) “Regulatory Focus and Executive Function after
Interracial Interactions.” Journal of Experimental Social Psychology 42, 3, 406–412. 10.1016/j.jesp.2005.05.008.
Tversky, A., and D. Kahneman. (1974) “Judgment under Uncertainty: Heuristics and Biases.” Science 185,
4157, 1124–1131. 10.1126/science.185.4157.1124.
van den Bergh, Linda, Eddie Denessen, Lisette Hornstra, Marinus Voeten, and Rob W. Holland. (2010) “The
Implicit Prejudiced Attitudes of Teachers: Relations to Teacher Expectations and the Ethnic Achievement
Gap.” American Educational Research Journal 47, 2, 497–527.
133
PART III
Social Structures
Introduction to Part III

PPE provides an opportunity for social science to say something general about social structures. In
Political Science the emphasis is often – but not always – on legal structure and political con-
stitutions. These aspects of society are linked to and supported by (ostensibly) legitimate coercion. In
Economics, the emphasis is often – but not always – on the spontaneous order that emerges as
people respond to existing incentives. If we want to say something general about society, we need
both perspectives. Society has legislated structures, emergent structures, and structures that are in
between. PPE is in a privileged position to leverage the insights from both disciplines to build a
more comprehensive view. Part III gives a sense of how PPE scholars have tried to better understand
the patterns and norms that organize social life.
To do this, the first chapter defends an account of social norms. Such norms are difficult to
explain because they seem to have aspects that are in the world and in the mind. They are not mere
attitudes or beliefs because they are linked with how people generally act in the world. Yet, they are
not simply patterns of individual action because people respond to them in distinct ways. In “Social
Norms,” Ryan Muldoon explains and expands on the theory of social norms originally offered by
Cristina Bicchieri. The key to this account is understanding social norms as rules of behavior that
people follow under the condition that (a) other members of a relevant community also follow the
rule and (b) people believe that other members of a relevant community believe they should follow
the rule. This approach recognizes that social norms are not simply in the mind because people
choose to follow a rule based on what other people do. They are not simply patterns of behavior,
because they depend on conditional preferences to act in rule-guided ways. After introducing this
approach, Muldoon shows how it can differentiate social norms from other forms of collective
behavior, how norms relate to learning processes, and how norms can be changed.
The second chapter reviews how “institutions” are understood across the three PPE disciplines.
In each discipline, there has been a rise in the number of “institutionalists” since the 1980s. In an
attempt to build a shared conception of institutions, Chris Melenovsky explains how institutions are
discussed in each discipline. Overall, each approach puts a greater explanatory emphasis on the
background social structures that contextualize choice, but they define institutions in different ways.
Melenovsky explains the choices that are made in defining institutions and briefly defends his own
preferred definition of “variable systems of social rules that are deemed to be socially significant.” He
DOI: 10.4324/9780367808983-13
Social Structures
ends by contrasting the different kinds of institutions including four different ways of distinguishing
formal and informal institutions.
In the third chapter, Bas van der Vossen explores a basic conflict in how we should understand
the institution of property. On the one side, any form of property seems to necessarily involve
immunity from expropriation. It is hard to think of what a functional property system would be if
what we “owned” could be taken from us without our consent, and such an immunity seems
necessary to incentivize care of one’s property. On the other side, a strict immunity from ex-
propriation would make taxation – and the government services funded by it – unjustified. To
motivate this conflict, van der Vossen surveys attitudes about what property is and why it is im-
portant. He recognizes that the specific features of property rights change from one society to
another, but there is an important core for property rights to be effective. At the end, he challenges
common ways that people have tried to justify taxation. In between, he shows why the strong
protection of property rights can be economically beneficial and why the creation of externalities
and inequality can be a negative side effect of such protection.
One of the most significant contemporary institutions is the corporation. Fundamentally, this
form of organization allows investors to pool their money in a way that shields them from personal
liability. Without this legal structure, it would be more difficult to gather the funds needed for large-
scale projects. However, the exact terms of the corporation form have changed over time. Instead of
being set up for specific government-issued tasks, for example, the modern corporation has open-
ended purview. What the legal structure of the corporation should be is itself a complex PPE
question. In “Corporations in Our Polity,” Amy Sepinwall identifies three ways in which cor-
porations exercise unexpected levels of control over US citizens’ daily lives. First, she discusses the
influence that corporations have over elections, especially following the Citizen’s United v. FEC
Supreme Court decision. Second, she explains the expansion of corporate religious freedom that is
exemplified by the Burwell v. Hobby Lobby decision. Last, she discusses the control that corporations
have in social causes. A corporate boycott can have more impact than boycotts from mere citizens,
and many social causes draw funding from corporate donors. She argues that these cases show that
corporations have used government powers to their advantage rather than the government putting
limits on corporate power.
The final chapter of Part III explains an overlooked form of social structure. Often, we make a
hard contrast between central authority or no authority – between the state or anarchy. For ex-
ample, Hobbes saw government as the solution to the dangers from the state of nature. On the other
side, freedom from authority is viewed as the solution to oppression. Even a view towards the
economy as an “emergent order” frames such order as arising outside conditions of authority. In
“Polycentricity,” Vlad Tarko explains a form of social organization that resists this contrast. In a
polycentric system, decision-units make their choices against a background of rules that they
themselves establish and maintain. The rules structure interactions and incentivize cooperation, and
no single centralized body has the capacity to change or enforce those rules. It is neither the state nor
anarchy. After introducing the concept, Tarko discusses its development with emphasis on the work
of Michael Polyani, Vincent Ostrom, and Elinor Ostrom. Because of their work, we have a better
understanding of both the theory and practice of polycentric systems. We even have a sense of when
such systems are more likely to be successful. Tarko ends the chapter by applying the idea to one of
the world’s leading problems. Perhaps, a polycentric system is the best hope for fighting global
climate change.
11
SOCIAL NORMS
Ryan Muldoon
1 Introduction
Very often, when we think about why people do what they do, we consider their personal pre-
ferences, beliefs, desires, or commitments. These are meant to give an explanation for their in-
dividual actions. Bob mowed his lawn because he likes to keep his yard looking nice. Carol shows
up to work a bit before 9 am every day because she values punctuality. And so on. When we think
about larger-scale issues, we often assume that the way to change people’s behaviors is to change the
laws. If you want to reduce jaywalking, make it illegal and punish jaywalkers with a hefty fine. If you
want to promote education, legally mandate that boys and girls must go to school. On this sort of
view, we can see people’s actions as some combination of their individual wants and desires, as
constrained by whatever is legal. To complicate this picture further, we could add some kind of
budget constraint. People do as much to satisfy their preferences as they can while constrained by
their resources and what’s legally permissible.
Thinking this way has some virtues: we have the state and other formal institutions that specify
what is allowed and what is required. We also have a wide range of individual choices to explain
what we do within those legal boundaries. This is all fairly easy to describe or measure. However, it
is obviously incomplete. For instance, if you ask Bob why he mows his lawn, it might be the case
that he really doesn’t enjoy it at all – he just feels obligated to do it, fearing his judging neighbors.
Maybe Carol shows up to work when she does because that’s the latest that she can arrive without
her coworkers rolling their eyes at her as she walks in the door. Likewise, even though most places
have made jaywalking illegal, it’s still incredibly common. Plenty of people jaywalk in front of police
officers, knowing full well that police officers by and large just don’t care about jaywalking.
Very often, the reason for these mismatches between laws, personal attitudes, and the behaviors
that we see is that people are responding to social norms. Bob might dislike mowing his lawn, but
every Saturday his neighbors all mow their lawns and give him a funny look when he doesn’t do the
same. So, Bob dutifully mows each week. Jaywalking is a crime that is on the books, but most
people do it anyway because it is convenient, and most police don’t think that it’s worth pursuing,
so it isn’t taken seriously as a crime.
Social norms are incredibly pervasive in our normal lives. How we dress, which words we use in
which situation, how we greet each other, and all kinds of other behaviors are governed by us and
our neighbors, often without noticing it. Social norms can be about trivialities like which hand you
hold your fork in or more serious matters like who you should treat with respect. As we will see,
DOI: 10.4324/9780367808983-14 137

Ryan Muldoon
social norms are a powerful but complex social tool. Our first challenge is to define social norms, and
then distinguish them from other social phenomena, like customs and conventions. We will then
look at the dynamics of norms, and how they are used.
2 Defining Social Norms

Given the importance of the concept of social norms, there have been a variety of accounts of norms
in the literature. This literature is quite extensive, so I will describe the most prominent approaches
here.1 Perhaps the most common conception of a social norm is that it’s a set of common values in a
population. These common values can take on an injunctive property, where we feel the need to act
on the basis of these shared values. On this account, we are socialized into a set of values and then act
on the basis of these internalized ideas. Parsons and Shils (1951) presents an early view of this sort.
One recent philosophical treatment of norms is by Brennan et al. (2013) and is part of what we
might call the functionalist school of social norms. This functionalist program argues that social norms
have some particular purpose – norms arose to solve a problem of some kind. An important in-
tellectual contribution of the Brennan et al. account is the idea that social norms, unlike other sorts
of social behaviors, provide a mechanism for accountability to others. Social norms provide a system
of praise and blame. On this account, not only do people have a first-order set of values, but know
that others in the population also share these values. This provides the basis for praise and blame.
Another sort of account of social norms focuses on norms as solving a problem for us but lacks the
praise and blame element. Game theoretic approaches to social norms focus on norms as solutions
for coordination problems in games. That is, they focus on norms as Nash Equilibria, which focuses
attention on the epistemic and rational conditions that can support those norms. Ullman-Margalit
(1977) is a prominent example of this sort of approach.
While there have been several candidate accounts of social norms in the literature, it is worthwhile
to consider for a moment what makes an account of social norms useful. First, we want an account to at
least have the potential to be universal – that is, we want an account that can describe the structure of
all social norms, not just some. On this ground, many functionalist accounts of norms may come up
short for the reason that not all norms solve a problem for us. Some norms cause problems! (Thrasher
and Handfield, 2018). And indeed, whether a particular social norm is a “problem” or a “solution”
requires some values deliberation. A second consideration is whether a proposed account of norms
accounts for empirical evidence. Here, Parsons-style accounts seem to come up short – your personal
normative considerations are frequently not decisive for determining your behavior. Last, we would
like an account of norms that allows us to measure and predict real behaviors. That is, we want an
account that can be operationalized into, say, experimental work, or surveys and other measurement
instruments. This last consideration is often under-appreciated. Theories can be elegantly constructed
and useful to help us understand behaviors, but they risk being inert if we can’t actually use them to
help diagnose problems, measure behavior, or try and design new institutions.
On these three grounds, the most useful prominent model of social norms is the Bicchieri account.
It offers a picture of a universal structure of norms, it has empirical support (Bicchieri, 2016), and is
amenable to operationalization. On her approach, a social norm is a rule of behavior R which applies
in a particular context C in a reference network P. An agent in P prefers to follow R in C when:
1. Empirical expectations: Agent A believes that most people in P follow R in C

2. Normative expectations: Agent A believes that most other people in P expect A to follow R in C,
and may sanction A for failure to comply (Bicchieri, 2006, p. 11)
Let’s unpack this definition and get a better sense of how everything fits together. First and foremost,
a social norm is about a behavioral rule – that is, social norms are about behaviors and not mere beliefs or
138
Social Norms
attitudes. Social norms are about things we must or must not do in some context. So, for instance, we
might have a social norm to form a line at a ticket window instead of just mobbing around it.
Likewise, we might have a social norm against cutting in line. This is distinct from disliking cutting
in line or thinking that crowding a ticket window is inefficient. Social norms are always grounded in
a particular behavioral rule – something that we can socially monitor – not just what’s in our heads.
The next component is that a social norm applies only in some particular context. We have
context-sensitive rules for what kinds of clothes we wear – it’s different between going to work,
going to a funeral, going to the beach, and going bowling. Likewise, we have different behavioral
rules for which words we can use between adult friends at a party, an office with coworkers, or
around small children. In some contexts, we have a great many social norms that structure our
behaviors, and, in others, we have far fewer norms.
Relatedly, social norms apply to particular reference networks. That is, particular norms are active
amongst particular sets of people, who hold each other accountable based on some sort of shared set
of relationships (Bicchieri, 2016, Ch 1, Bicchieri and Fukui, 1999, pp. 131). For instance, a group of
friends may have strong reciprocity norms amongst themselves – if Alice gives Bob a ride to the
airport, Bob owes Alice a favor in return. Likewise, teachers or other professionals that have unions
or professional societies may well have social norms that govern their behavior as professionals. This
will be similar for families, members of a religious group, club members, and other kinds of social
networks. Individuals will generally be in multiple different reference groups: Carol may have one
set of social norms that govern her behavior as a friend with her friends, another set of social norms
that govern her professional behavior, and still another set that informs how she behaves with her
family. Indeed, it’s entirely possible that she occasionally finds herself in a situation in which fol-
lowing one social norm for one reference group may involve violating another for a different
reference group (Bicchieri, 2016, pp. 76). For instance, perhaps Carol is getting married, and de-
ciding whether to keep her last name or take her husband’s. She may come from a conservative
family that has the social norm of women taking their husband’s last name. But she may also be in a
more progressive industry, where it is unusual for women to change their last names. These are two
opposing social norms that have some grips on Carol, but it is because she has multiple reference
groups that she is responsive to. What clothes you wear, or words you use, or your manner of
greeting someone is easy enough to switch between contexts. But some things are difficult to switch
between contexts. Hair coloring, piercings, and tattoos are some examples of behavioral choices that
are going to be context-invariant, just because it’s somewhat impractical to alter these across
contexts. Choices such as these may be areas where individuals find themselves trying to be re-
sponsive to conflicting social norms.
Now that we have a sense of how social norms are scoped, we can turn to how they are
maintained in those contexts for a given reference network. To do that, we need to look more
closely at social expectations. There are two kinds of social expectations: empirical expectations and
normative expectations (Bicchieri, 2016, p. 11). Empirical expectations are the simplest. Alice has
empirical expectations about how many people in her reference network follow some rules. For
instance, Alice has an estimate of how many (or what proportion) of her neighbors mow their lawn
every Saturday morning. She also would have a belief about how many of her friends attend church
each week, and how many of her coworkers show up for work before 9 am. In each case, Alice has
identified a relevant behavioral rule, a context, and a reference network. Her beliefs are about how
many people in the reference network actually follow the rule in the relevant context. She might not
care, for instance, about how many of her coworkers mow their lawn on Saturdays. She doesn’t live
near her coworkers, and so doesn’t participate in a shared norm about lawn mowing with them. She
likewise doesn’t need to care about what time her neighbors go to work in the morning. She has
empirical expectations for how many people in her reference network follow the rule in the relevant context.
Information about what other kinds of people do is just irrelevant. In general, empirical expectations
139
Ryan Muldoon
can be determined by observation or social monitoring – it’s easy to see who is at work by 9 am, or
who has mowed their lawn, or who is in church. Anytime we have rules that govern public be-
haviors, it is generally quite easy for people to have accurate empirical expectations. They can see
what others do. This gets trickier when we have rules that govern private behaviors. We have limited
ability to monitor whether people wash their hands after they use the restroom. We have no real
capacity to monitor what people do in the privacy of their own homes or in their bedrooms. And
yet, we often have rules that govern those behaviors. In those cases, we typically rely on people’s
reporting on what they do, or some proxies for the behavior that we can observe. So, especially with
rules that govern private behaviors, our empirical expectations could quite easily be incorrect.
The more complex kind of social expectation is a normative expectation. Whereas empirical
expectations are just beliefs about how many people are following the rule, a normative expectation
is a belief about other people’s beliefs. A normative expectation is a belief about how other people
think that you should follow the rule (Bicchieri, 2006, pp. 11). These second-order beliefs – beliefs
about beliefs – are less straightforward to reliably learn, because you’re trying to learn about what is
in other people’s heads. So, like with empirical expectations of private behaviors, we are left with
what others tell us, and behavioral proxies for their beliefs. People can (and do!) tell us what they
think we should do all the time. Just as importantly, people can punish or praise people for their
behavior (Chavez and Bicchieri, 2013). So, if Bob doesn’t mow his lawn one weekend, do Alice and
Carol show their disapproval somehow? Perhaps they loudly complain to each other about
neighbors who don’t do their part to keep the neighborhood looking good. Perhaps they leave Bob
off of the guestlist for a backyard barbecue. Maybe they shake their heads at him. Bob, and other
members of the reference network, can adjust normative expectations based on these behaviors. We
can surely pick up on patterns of praise and blame in this way, and come to form beliefs about what
others think we should do. These instances of praise and blame may be coordinated, as with an
organized campaign, or more organic responses that come from normal interactions. Humans seem
to be very good at identifying praise and blame and learning from it. Of course, as with trying to
learn from proxies for private behaviors, we could still be mistaken or even misled. For instance,
people could publicly support a rule that they privately dislike, thus making the rule appear to have
more support than it actually does. This can happen in instances where the perceived cost of dis-
agreeing with a social norm is sufficiently high such that people will follow the social norm even if
they disagree with it. Indeed, in a situation called pluralistic ignorance, most people privately dislike a
social norm but publicly support it (Bicchieri and Fukui, 1999, Katz et al., 1931, Miller and
McFarland, 1987, O’Gorman 1975). So, the norm persists even if most would like it to go away
because it is too costly for people to reveal their true beliefs.
A classic example of pluralistic ignorance is from the children’s story, “The Emperor’s New
Clothes.” A clever-but-deceitful tailor convinces a vain Emperor that an empty box contains the
most beautiful clothes constructed by the finest fabrics, but only those with excellent taste and
refinement can see their beauty. The Emperor fears that since he can’t see the clothes, that means he
lacks taste and refinement. So, he lies – he praises how beautiful the clothes are, and has the tailor put
the clothes on him. Word spreads quickly that the Emperor has a beautiful new set of clothes, but
only those with taste and refinement can see their beauty. As the Emperor parades through his Court
and Kingdom, everyone loudly praises his new clothes, as they fear that they are the only ones who
can’t see the clothes. Eventually, a child – who has no real concern for their social standing – shouts
that the Emperor is naked. This lets others realize that everyone else is seeing a nude Emperor. The
praise for the clothes quickly shifts to jeers and laughter. This story illustrates how pluralistic ig-
norance can be maintained – people all fear the cost of revealing their private beliefs. This public
support makes it even more costly for the next person to reveal their private belief. So, we can be in
equilibrium for public support of something counter to everyone’s personal beliefs. However, this
equilibrium is fragile – once there is a signal that others don’t support the public statements (revealed
140
Social Norms
by the child in the story), things can collapse rapidly. This story is frivolous, but the same structure
can appear in support for oppression, authoritarian governments, and other dangerous social si-
tuations. Timur Kuran’s Private Truths and Public Lies (1995) is a detailed look at this phenomenon.
So, let’s consider again what a social norm is. It’s a rule of behavior for members of a particular
reference network that applies to some particular context. A member of the reference network will
follow the rule in that context when she believes that most other members of the reference network
are also following the rule, and she believes that most other members of the reference network think
she should follow the rule as well. So, her preference for following the rule in that context is
conditional on her social expectations. Her choice crucially depends on her beliefs about others. If she
believes that other people are not following the rule, or don’t expect her to, and she still follows the
rule, then this is not a social norm for her – that behavior would just be what she unconditionally
prefers to do. Social norms are interdependent behaviors where Alice’s choice about what to do is
conditional on what she thinks Bob and Carol will do, and what she thinks they want her to do.
3 Other Kinds of Collective Behavior

To get a richer sense of what social norms are, we can now turn to what they are not. In this section,
we will consider other forms of collective behavior and see how they are different from social
norms. This matters not merely as a matter of labeling, but because if we draw useful distinctions, we
can have better analytic grips on social phenomena. We can then understand our social lives in a
deeper way and also use this knowledge to make better policy choices.
The most important distinction to make is between independent and interdependent behaviors.
Social norms, as we have seen, are fundamentally about behavioral interdependence. Alice follows a
neighborhood social norm of mowing her lawn on Saturday morning if she thinks Bob and Carol
will, and she thinks Bob and Carol expect her to as well. This is very different from a situation where
Alice just loves the smell of freshly cut grass, and so she excitedly wakes up each Saturday morning
eager to mow. In the first instance, Alice is following a social norm. In the second, Alice is doing
what she wants to do. If no one else mowed her lawns on Saturday, Alice would still get pleasure
from mowing and smelling the freshly cut grass. For Alice, mowing her lawn is just something she
enjoys, and so she has plenty of independent reasons to choose to do it. It is then irrelevant to her if
her neighbors have a social norm around mowing or not.
Many collective behaviors can occur without interdependence. For instance, on a rainy day, if
you were to look at how people were behaving, you’d likely observe that most of them had an
umbrella or a raincoat. If social norms were on your mind, you might think that this is yet another
example of the power of social norms – everyone’s doing the same thing! But if you instead step
back and think a little bit about people’s reasons (or even better: ask them!), you’d realize that for
most people, getting wet in one’s clothes is pretty unpleasant. It is uncomfortable, it might make you
colder than you’d like to be, and clothes can take a while to dry. So, it’s just a good idea to take steps
to avoid the thing you don’t like. Umbrellas and raincoats are great technologies for avoiding getting
wet.2 That we see a lot of people using umbrellas or raincoats doesn’t provide evidence for social
expectations or even for the presence of a behavioral rule. That pattern of behavior is more likely
explained by the idea that umbrellas and raincoats are good solutions to a common problem.
Independent behaviors are just that – independent. Bob uses an umbrella because it’s a good idea,
not because other people are doing it. It’s the same with wearing a hat or sunglasses on a sunny day,
wearing water shoes on a rocky beach, or any number of other behaviors. Plenty of independent
behaviors are individualized – Bob likes to drink orange juice alongside a peanut butter sandwich
because he thinks they taste good together. Carol likes to bike to work because she thinks it’s a good
way to get exercise. But other behaviors, like using umbrellas or wearing sunglasses, are both in-
dependent and form a collective pattern of behavior. Bicchieri calls this class of behaviors a custom
141
Ryan Muldoon
(2016, p.15). We might see variation between these collective patterns of behavior across cultures, but
that variation is better explained by social learning and path dependence rather than social norms.
Customary behavior is commonly seen, but not something that’s socially monitored. Bob might have
the empirical expectation that most people in his reference network engage in some behavior in a
particular context, but his choice to act is not conditional on those expectations. He will do what he
wants to do, and his choice does not depend on other people’s behaviors or expectations.
A variant of a custom is shared moral norms. While a custom is something that’s a good solution to
a common problem, a moral norm is something that an individual views as a moral obligation
(Bicchieri, 2006, pp. 20–21). When Carol thinks that she should give 10% of her salary to the poor,
she is responsive only to what her personal normative beliefs are. When she is deciding to give, she
does not consider whether others are doing it or whether she thinks they expect her to do it. Instead,
she acts out of her sense of moral obligation. When she has these kinds of reasons for her actions, she
is following a moral norm. It is often the case that many people have similar normative beliefs, but
even if everyone in Carol’s community gives 10% of their money to the poor, this is not a social
norm if they are doing so out of their personal convictions about what the right thing to do would
be. If Carol was in a community whose members never did so, she still would. That is, her pre-
ference for giving to the poor is unconditional on social expectations. She is doing it because she
believes she ought to do it – she thinks it is the right thing to do. It may well be the case that many
people in some community feel similarly – they may have a common set of values or a common
religion – but the key issue is whether they are acting from their personal normative commitments
or if they are acting conditionally from their social expectations. When they are acting based on their
personal commitments, they are following a moral norm and not a social norm.
So, we have seen that it is possible, and even likely, to find common patterns of behavior that
arise from common but fully independent behaviors. Now, we will turn to interdependent behaviors
that are not social norms. Here, we will look at two cases: descriptive norms and conventions.
Descriptive norms rely on empirical expectations, but no normative expectations. Conventions are
solutions to coordination problems. Let’s first consider descriptive norms.
A descriptive norm is a behavioral rule R that applies in a context C for some reference network P
(Bicchieri, 2006, pp. 31–32). An agent in P prefers to follow R in C if she believes that most other
people in P follow R in C (Cialdini, 1991, pp. 203). Notice that this is almost identical to the
definition of a social norm, just with normative expectations removed. For instance, Alice shows up to
a party “fashionably late” by getting there about 15 minutes late. She thinks most people will show up
about 15 minutes late, but she doesn’t think that anyone would think any less of her for arriving on
time or a bit later. Bob wears black socks rather than white socks with pants, because he noticed that
most of his coworkers and friends do the same. He doesn’t think anyone has very strong views about
what color socks he wears, but he figures he may as well just go along with the majority. Notably,
descriptive norms can be one-sided, in the sense that Bob can seek to copy the behavior of people who
don’t likewise pay attention to what Bob is doing. We take cues from “tastemakers” or “influencers”
all the time, never thinking that we’re engaged in some kind of reciprocal relationship with them.
A convention is a descriptive norm where the behavioral rule R is also a solution to a co-
ordination game present in context C (Bicchieri, 2006, pp. 38). Where a descriptive norm can be
one-sided, conventions are always two-sided. Unlike social norms, where the social norm might be
counter to one’s interests, and so may require punishment to enforce, a convention is self-enforcing.
Failing to coordinate is the punishment. The classic example of a convention is driving rules. Carol
drives on the right-hand side of the road because everyone else does. If she drove on the left-hand
side, she’d crash her car. We don’t strictly need to add any punishments on top of this at all. People
have plenty of motivation to not get into car crashes. In this traffic example, even before we think
about social norms or conventions or anything else, there’s still an underlying coordination problem:
how do we all share the roads? There are plenty of other “natural” coordination problems that we
142
Social Norms
regularly face: where and when we meet up with friends after work, who brings what to a potluck
dinner, how roommates divide up housework, and so on. Conventions are a tool to solve those
coordination problems. It is important to note, however, that the solutions are in some sense ar-
bitrary. That we drive on the right-hand side instead of the left-hand side is more of a fluke than an
insight into the superiority of right-hand-side driving. Bob and Carol meeting at 6:00 pm at the
brewpub is a perfectly good solution to the coordination problem of meeting after work, but
probably no better than meeting at 6:15 pm at the cocktail bar across the street. The important part
is that we’ve coordinated. Insofar as coordination is the important part, once we’ve settled in on a
way to do it, there’s no reason to try something different.
This is indeed a common feature of interdependent behaviors. Once we’ve settled on a behavioral
rule, whether it is a social norm, a descriptive norm, or a convention, it becomes difficult to move away
from that rule. In a social norm, deviation usually results in punishment. In a convention, there will be
mis-coordination, which has costs for everyone involved. Descriptive norms are low-key enough that
simply not following the norm is ok, but shifting to a new norm is not easy once one is established.
Interdependent behaviors are “sticky” in ways that independent behaviors are not. Independent be-
haviors like customs or moral rules can more readily change if people have reason to.
While we can make sharp conceptual distinctions between all of these categories, when we
consider real empirical cases of social behaviors, we are likely to see some diversity in people’s
motivations. Sometimes, we have both strong personal normative beliefs and strong empirical ex-
pectations, all pointing us in the same direction. Sometimes, what’s a social norm for Alice might be
a moral norm for Bob and a descriptive norm for Carol. We may still treat the social phenomenon as
a social norm if most people are like Alice, but it may be useful to know that there are personalities
like Bob’s and Carol’s around.
4 Social Norms and Social Learning

As we just saw, sometimes a collective pattern of behavior is a custom rather than a social norm.
Customs are a common solution to a common problem. But we may wish to pause for a moment to
think about how they come into existence. Customs are a collection of independent behaviors, but
they may well arise from a social process. It is useful to see how this differs from a social norm.
While it is possible for a custom to arise purely from lots of individuals making their own choices
in isolation, it is far more likely that customs come about from social learning. Social learning, as you
might expect, just means that people learn from observing what others do or talking to them about
it. Bob the farmer might experiment with different kinds of fertilizers for his crops, but a much
lower-cost way of learning is asking his friends and neighbors what has worked for them. Bob can
even just look around and see that Alice’s crops look the nicest or that Carol’s are the most plentiful,
and he could ask Carol or Alice what they did and try to copy them next season. In this case, Bob
isn’t doing something because Alice or Carol did it. He’s doing it because he has evidence that the
methods they used were better than Bob’s. Bob just cares about doing better himself, not being like
someone else. If Carol or Alice switch to some other method and it doesn’t work as well for Bob,
Bob wouldn’t simply copy them. He just wants to identify the best techniques, and he looks around
him for new ideas and better solutions. Social learning may facilitate the spread of new ideas or new
behaviors, but what people most care about in those contexts is picking the better option, not
following the crowd.
While classic examples of social learning involve technology uptake, this is also the basic model
for extending moral norms. Moral norms spread by Alice convincing Bob that R is a good rule in C.
Perhaps it’s just that Bob observes Alice living her life in accordance to certain principles, and Bob
thinks that she is living a better life than him, and so he takes those principles on himself. Carol could
evangelize to Bob, and Bob might find that ultimately convincing. But crucially, Bob adopts those
143
Ryan Muldoon
moral principles because he thinks they are the right principles not because Alice and Carol are following
them. If Alice or Carol change their moral principles, that does not give Bob a reason to change his.
He might learn more from them as they try something new, but no matter what, he relies on his
own judgments. Alice and Carol might just be easier to learn from than Kant or Bentham.
The important thing to notice is that processes of social learning are social but not conditional. Bob
may learn about better tools or better rules from Alice and Carol, but once he learns, he doesn’t need
to worry about what Alice or Carol are doing. Bob could become a hermit, never again concerned
with what others do, and continue to rely on what he learned from Alice and Carol. He could have
learned what he learned from them by some other means. That he learned socially might tell us
something about how we can effectively spread useful information, but it does not tell us anything
about what reasons Bob has for doing what he’s doing. Bob may have decided to become an ethical
vegetarian after learning about it from Alice, but he’s not presently taking his eating cues from Alice.
A social norm, on the other hand, would rely on Alice, Bob, and Carol all monitoring each other’s
behavior, ready to punish any transgressions. Their behavior would be conditional on each other’s
behavior and expectations. Then, if Bob moved away from Alice and Carol, he may well change
his behavior if he entered into a new reference network that has a different rule of behavior. Social
norms are only conditionally followed. Social learning, on the other hand, doesn’t involve social
reasons to persist in some behavior.
5 Norm Dynamics: How Do Norms Come and Go?

Now that we have a more solid grasp of what a social norm is and how it is distinct from other social
patterns of behavior, we can turn to the questions of how a social norm comes into existence, and how
they go away. There isn’t a single answer to these questions, in part because many – likely most –
norms come and go “naturally.” That is, no one decides to make a social norm governing some
behavior. It just develops over time. Likewise, some norms just go away without anyone deciding that
they should. There is a great deal of interest in deliberate norm creation and elimination, whether led by
a state effort or from civil society. There is a mixed record of success with these efforts.
Let’s first turn to natural norm formation. The simplest case is the emergence of a descriptive
norm. Muldoon et al. (2012, 2014) consider different mechanisms for how descriptive norms can
emerge without any kind of collective intention. What matters most is that agents observe what
others do, and they use that as a cue for how they should behave in the future. Put another way,
people in some reference network are taking social information about what others are doing, and
people are using it as evidence of the presence or absence of particular behavioral rules instead of
using it to just assess for what good ideas might be. When we see a number of people doing the same
thing in the same context, we imagine that this is an important regularity – we assume that this is for
some rule-driven reason. Of course, we could be entirely mistaken. It’s possible that it was a co-
incidence that a number of people, for their own reasons, adopted the same behavior. But this belief
can quickly become self-fulfilling: if we think others are following a rule, we’ll follow it too, which
can make others believe there’s a rule. Soon enough we’re all following a rule that didn’t used to
exist. One can imagine this as a more general version of a standing ovation after a concert. At first,
no one is standing. A few people stand simply because they loved the performance and want to
celebrate it. Now, a few others see those standing and think that might be the thing to do, so they
stand up as well. Now that more people are standing, the next group of people who were a bit more
reticent will join in. And so it goes until the last people stand. What’s nice about this ovation kind of
model is that we can see that people can differ in their social sensitivities, and that will shape whether
they take on the new norm. A range of social sensitivities in the population will lead to what looks
like a cascade of support – something that starts small, but grows until everyone is involved.
144
Social Norms
One can see an easy path to extending this basic story to how some social norms emerge. Once
there is a descriptive norm present, some people can start to place some value on it, or believe that
others do. Once this happens, some number of them will begin punishing transgressors. Those
punishers will feel justified in doing so, as they are punishing in support of the shared rule. Seeing
others get punished for failure to comply will increase everyone’s normative expectations for this
behavioral rule. This will induce both more compliance and more punishment behavior. We should
expect more compliance just because punishment increases the cost of deviance. The reasons for
more punishment may be more varied. Some people may choose to engage in punishment behavior
because they find value in the rule and want it to be successfully maintained, and punishment is just
an effective means to do so. Some others might just enjoy opportunities to punish. Gossiping
behaviors are often examples of this kind of pleasure in (relatively mild) punishing behavior.
Social norms can emerge naturally through our interactions, but we can also (at least try to) create
them by agreements. This is discussed in some detail in Bicchieri (2016), but the core idea is that to
establish a new social norm by social fiat, it is necessary to credibly coordinate the relevant reference
network’s social expectations. That is, people need to believe that others will follow the rule, and
think that others expect them to follow the rule as well. People likewise need to believe that there
are sanctions attached to these expectations, to help make their preference for following the rule
conditional on social expectations.
It is important to note that the intentional creation of a new social norm might follow after an
extensive persuasion effort – people should have reasons to want to endorse the social norm that they
are creating, and have reasons to believe that others endorse the norm as well. But just individual
persuasion is not sufficient – social expectations do need to be coordinated. Public proclamations,
pledges, or other such efforts to socially commit one another to a new behavior are needed to help
set people’s social expectations. The details of how these steps are taken need to be sensitive both to
what would work for setting expectations and what makes some sense in the appropriate cultural
context. Public pledges are sometimes seen as valuable as there are pre-existing norms around promise-
keeping that help support a new norm. Bicchieri (2016) examines a number of real-world efforts with
these tools.
The flip side of norm emergence is norm elimination. Let’s start by thinking about intentional
efforts to eliminate norms. One approach is to replace a pre-existing norm with a novel social norm.
In this case, the major focus is once again on setting expectations to support the new practice. If we
instead want to just eliminate a norm with nothing to replace it, it’s a question of trying to eliminate
(or at least reduce) the relevant social expectations that people are attentive to. This can either be by
directly intervening on those expectations, or by trying to change the reference network.
A nice case of norm elimination by means of changing social expectations is in the case of
pluralistic ignorance. For example, commonly on college campuses, students would like to have one
or two beers at a party, but see others drinking four or five beers, and believe that others expect
them to binge drink as well. So they do. A successful intervention (Haines and Spear, 1996) has been
to just privately survey students about their actual drinking preferences, and then publicly reveal the
results with informational posters. When everyone discovers that most everyone else also wants to
have fewer drinks, behavior shifts. People who wanted to drink fewer drinks now feel comfortable
in doing so. Lawrence Lessig (1995) discusses the way in with the Civil Rights Act freed business
owners up stop discriminating without fear of retribution. Passing a law that required that businesses
not discriminate changed the social meaning of a business integrating. Rather than being a social
statement against the norms of segregation, it could also be interpreted as a desire to comply with the
law. This allowed for a large shift in behavior that likely cannot be explained by the state’s capacity
to monitor and enforce the law.
While these are intentional efforts to eliminate norms, many norms can simply fall away over
time. Again, this stems from a dwindling of social expectations. In general, dwindling empirical
145
Ryan Muldoon
expectations lead to social norm collapse. As Kogelmann, Muldoon, and Vallier (2022) argue in the
context of democratic norms, this can happen if the reference network is split, so sub-groups stop
following a norm, which leads to its general collapse. Norms can also collapse due to new members
of a reference group who are not sufficiently sensitive to a norm and therefore don’t follow it . This
can lead to a cascade against the norm, as more and more people stop following the rule.
While the body of knowledge on norm dynamics is growing, it is far from complete. There are
likely important interaction effects between the social norm in question, habits, mental models, and
the other social norms that are active for that population. These interactions remain under-studied,
but may ultimately help explain why we see differences in norm dynamics across contexts.
6 Conclusion
As the literature on norms has grown, there has been greater recognition of their importance in our
social lives and growing account of how norms can mediate a variety of social situations and relations.
What Bicchieri has focused on is the explanatory and predictive role that her theory of norms can
have when we seek to better understand social behaviors in a wide variety of contexts, and how social
norms can be used as a tool to encourage pro-social behaviors. Of particular focus for Bicchieri and
others has been on public health measures like promoting toilet use and handwashing in the devel-
opment context (Chavez and Bicchieri, 2013; Curtis, 2013; Sanan and Moulik, 2007). In similar de-
velopment contexts, social norms have been used as parts of attempts to eliminate female genital cutting
(Hadi, 2006), to understand sources of support for bride kidnapping (Muldoon and Casabonne, 2017),
and explain the rise and collapse of footbinding (Mackie, 1996; Blake, 1994), amongst other issues.
Undoubtedly, this use of norms as a potential tool for “Norm entrepreneurs,” whether they be
policymakers or citizens, will only increase, in part because norms look to be a more appealing tool
for encouraging particular social behaviors compared to the coercive apparatus of the state
(Bicchieri, 2016, Ch 4, Sunstein, 1996). However, as Muldoon has raised (2017, 2018), social norms
remain coercive, and their coercion can under certain circumstances be more difficult to keep under
democratic control. Social norms can “solve” problems with extremely detrimental methods
(Thrasher and Handfield 2018). States have a clear mechanism for eliminating or modifying laws,
and they take immediate effect. While we have a general understanding of the mechanisms of norm
change, we do not have nearly the same capacity to reliably change norms as we do to change laws.
As discussed, there are important interactions between social norms and laws, however. We have
already briefly seen how laws that deviate too far from what social norms require or allow are largely
impotent: jaywalking is a common practice despite being illegal. Social norms can lead police or
prosecutors to just not bother with certain kinds of crimes, or be overzealous with others (Campbell
et al., 2015; Corrigan, 2013; Irving, 2008). Most interestingly, as discussed earlier, the law can alter
the meaning of our behaviors, as they can obscure our motivations to act. Social norms and the law
interact in fascinating ways to enable or inhibit each other. In the same way that our personal
preferences don’t necessarily predict our behaviors, without taking into account social norms,
merely looking at what laws have been passed will give us a partial understanding of what behaviors
are actually forbidden and what behaviors are encouraged.
Social norms are an incredibly potent social tool that we frequently use without realizing it. The
Bicchieri framework gives us an analytic grip on a wide variety of complex social behaviors and
some insights into why people do what they do.
Notes
1 For a more detailed review of the literature, see Bicchieri, Muldoon, and Santuouso (2018).
2 The author pledges that he has not accepted any money from Big Raincoat to make these statements.
146
Social Norms
References
Blake, C. (1994) “Foot-Binding in Neo-Confucian China and the Appropriation of Female Labor,” Signs, 19,
3, 676–712.
Bicchieri, Cristina. (2006) The Grammar of Society: The Nature and Dynamics of Social Norms, Cambridge;
Cambridge University Press. doi: 10.1017/CBO9780511616037
Bicchieri, Cristina. (2016) Norms in the Wild: How to Diagnose, Measure, and Change Social Norms, New York;
Oxford University Press. doi: 10.1093/acprof:oso/9780190622046.001.0001
Bicchieri, Cristina, and Yoshitaka Fukui. (1999) “The Great Illusion: Ignorance, Informational Cascades and
the Persistence of Unpopular Norms,” Business Ethics Quarterly, 9, 1, 127–155. doi: 10.2307/3857639
Bicchieri, Cristina, and Alessandro Sontuoso. (2015) “I Cannot Cheat on You after We Talk”, in Martin
Peterson ed. The Prisoner’s Dilemma, Cambridge; Cambridge University Press, pp. 101–114. doi: 10.1017/
CBO9781107360174.007
Bicchieri, Cristina, and Alessandro Sontuoso, forthcoming. “Game-Theoretic Accounts of Social Norms,” in
M. Capra, R. Croson, T. Rosenblatt, and M. Rigdon eds. The Handbook of Experimental Game Theory,
Cheltenham; Edward Elgar Publishing.
Brennan, Geoffrey, Lina Eriksson, Robert E. Goodin, and Nicholas Southwood. (2013) Explaining Norms,
New York; Oxford University Press. doi: 10.1093/acprof:oso/9780199654680.001.0001
Campbell, R., Feeney, H., Fehler-Cabral, G., Shaw, J., & Horsford, S. (2015) “The National Problem of
Untested Sexual Assault Kits (SAKs): Scope, Causes, and Future Directions for Research, Policy, and
Practice,” Trauma, Violence, & Abuse, 18, 4, 363–376. doi: 10.1177/1524838015622436
Chavez, Alex K. and Cristina Bicchieri. (2013) “Third-Party Sanctioning and Compensation Behavior: Findings
from the Ultimatum Game,” Journal of Economic Psychology, 39, 268–277. doi: 10.1016/j.joep.2013.09.004
Cialdini, Robert B., Carl A. Kallgren, and Raymond R. Reno. (1991) “A Focus Theory of Normative
Conduct,” in Mark P. Zanna ed., Advances in Experimental Social Psychology, volume 24, New York;
Academic Press, pp. 201–234. doi: 10.1016/S0065-2601(08)60330-5
Corrigan, R. (2013) “The New Trial by Ordeal: Rape Kits, Police Practices, and the Unintended Effects of
Policy Innovation,” Law & Social Inquiry, 38, 4, 920–949.
Curtis Valerie (2013) Don’t Look, Don’t Touch, Don’t Eat: The Science Behind Revulsion Chicago. UC Press.
Hadi, A. A. (2006) “A Community of Women Empowered: The Story of Deir El Barsha,” in R. M. Abusharaf
edited by. Female Circumcision: Multicultural Perspectives, 104–124. Philadelphia; University of Pennsylvania Press.
Haines M., Spear S. F. (1996) “Changing the perception of the norm: a strategy to decrease heavy-episodic
drinking among college students”. J Am Coll Health. 45, 134–140.
Irving, T. (2008) “Decoding Black Women: Policing Practices and Rape Prosecution on the Streets of
Philadelphia,” NWSA Journal 20, 2, 100–120. https://www.muse.jhu.edu/article/247356.
Katz, D., Allport, F. H., & Jenness, M. B. (1931) Students’ attitudes; a report of the Syracuse University reaction study.
Craftsman Press.
Kogelmann, Brian, Muldoon, Ryan, and Vallier, Kevin. “Democratic Norms” manuscript.
Kuran, Tim. (1995) Private Truths, and Public Lies: The Social Consequences of Preference Falsification, Cambridge,
Massachusetts; London, England; Harvard University Press. doi: 10.2307/j.ctvt1sgqt
Lessig, Lawrence. (1995) “The Regulation of Social Meaning,” University of Chicago Law Review, 62, 3, Article
1. https://chicagounbound.uchicago.edu/uclrev/vol62/iss3/1
Mackie, G. (1996) “Ending Footbinding and Infibulation: A Convention Account,” American Sociological
Review, 61, 6, 999–1017. doi: 10.2307/2096305
Miller, D. T., and McFarland, C. (1987) “Pluralistic ignorance: When similarity is interpreted as dissimilarity,”
Journal of Personality and Social Psychology, 53, 6, 298–305.
Muldoon, Ryan. (2017) “Perspectives, norms, and agency,” Social Philosophy and Policy 34, 1, 260–276.
Muldoon, Ryan. (2018), “Understanding norms and changing them,” Social Philosophy and Policy 35, 1, 128–148.
Muldoon, Ryan, and Casabonne, Ursula. (2017) “Gender Norms in Flux: Bride Kidnapping and Women’s
Civic Participation in the Kyrgyz Republic:. World Bank, Washington, DC. © World Bank. https://
openknowledge.worldbank.org/handle/10986/28989 License: CC BY 3.0 IGO.”
Muldoon, R., Lisciandra, C., Bicchieri, C., Hartmann, S., & Sprenger, J. (2012) “On the emergence of de-
scriptive norms,” Politics, Philosophy & Economics, 13, 1, 3–22. doi: 10.1177/1470594X12447791
Muldoon, R., Lisciandra, C. & Hartmann, S. (2014) “Why are there descriptive norms? Because we looked for
them,” Synthese 191, 4409–4429 (2014). doi: 10.1007/s11229-014-0534-y
O’Gorman, Hubert J. (1975) “Pluralistic ignorance and white estimates of white support for racial segregation”.
Public Opinion Quarterly, 39(3), 313–330.
147
Ryan Muldoon
Parsons, T., & Shils, E. A. (Eds.). (1951) Toward a general theory of action. Harvard University Press. doi: 10.4159/
harvard.9780674863507
Sanan, D., Moulik, S. G. (2007) “Community-Led Total Sanitation in Rural Areas - An approach that works,”
Water and Sanitation Program (WSP) [available online: https://documents1.worldbank.org/curated/en/
672891468324551045/pdf/396690Total0sanitation01PUBLIC1.pdf. Accessed: January 4th, 2022].
Sunstein, C. (1996) “Social Norms and Social Roles,” Columbia Law Review, 96, 4, 903–968. doi: 10.2307/
1123430
Thrasher, John and Handfield, Toby. (2018) Honor and Violence: An Account of Feuds, Duels, and Honor Killings
(March 20, (2018)). Available at SSRN: https://ssrn.com/abstract=3057147 or 10.2139/ssrn.3057147
Ullman-Margalit, Edna. (1977) The Emergence of Norms. Oxford University Press.
148
12
INSTITUTIONS AND
INSTITUTIONALISM
C. M. Melenovsky
In common usage, many different things are called institutions: private property, marriage, the
common law, the US Dollar, the UK Constitution, The African Union, the KMT political party of
Taiwan, UNICEF, local-funding of primary schools, the corporation, the “Blue Line of Silence”
among police, the New York Times, and – of course – Turtle Swamp Brewery in Jamaica Plain.
Any attempt to characterize institutions tries to find a single unity shared by examples like these.
Most often, an institution is characterized as a set of behavioral patterns that establishes background
conditions for decision-making in some set context. By establishing these background conditions, an
institution can change incentives, generate expectations, establish roles, anchor meaning, and narrow
down the information needed to make a decision. The institution of contract law, for example, sets
expectations about what is needed to make a mutually recognized and binding agreement. By
entering such an agreement, the parties have greater trust in one another because there is a shared
understanding and because the state acts as a third-party enforcer. Because institutions set the
background conditions for our interactions, institutions are often said to set the “rules of the game.”
However, what game is being played and what counts as a rule differs from one analysis to the next.
For many purposes, a vague characterization of institutions is sufficient. So long as we know that
institutions are behavioral patterns that contextualize choice, we can see their importance. However,
a vague characterization does not adequately explain what distinguishes institutions from other
features of a social structure. How do we differentiate what institutions are from, say, what culture
is? The fact that many undergraduates study abroad during their junior year might impact an un-
dergraduate’s choice of whether to study abroad, but that does not make patterns of travel an
institution. Similarly, we would not think that prices are institutions even if, for example, landscape
engineers make certain designs based on expectations about hydrangea prices. All sorts of social
conditions set expectations and encourage behavior, so all sorts of social conditions contextualize
choice. To isolate what institutions do, we need a clear conception about what institutions are that
differentiates them from culture, prices, social norms, patterns of belief, and habits of behavior.
However, the way that any given theorist distinguishes “institutions” from other background
social conditions will depend on their aims, the traditions they work within, and the rival theories
that act as contrasts. As such, there is no single shared conception of institutions across the dis-
ciplines. So, to understand institutions in PPE, §1 will briefly review some of the major versions of
institutionalism. This will give some sense of how institutions are thought about in Economics,
Political Science, and Philosophy. From there, §2 identifies some of the major divisions in how
theorists understand institutions. I’ll distinguish expansive from narrow conceptions of institutions
DOI: 10.4324/9780367808983-15 149

C. M. Melenovsky
and why some theorists include elements that others do not. §3 concludes with a brief explanation of
my own preferred conception of institutions as “variable systems of social rules that are deemed to be
socially important.” I offer it as a potential conception that could be shared amongst those looking to
find an interdisciplinary conception of institutions.
1 Institutionalism
A concern with institutions is now common across the social sciences. Different theorists identify
themselves as “institutionalists” or with some subgenre of institutionalism. As different disciplines
have different concerns, methods, and traditions, it should not be a surprise that “institutionalism”
means different things in different disciplines. As such, there is no clear agreement about (a) what an
institution is or (b) what an institutionalist cares about. The most unifying thing one can say is that
an “institutionalist” puts greater emphasis on background social structures to explain some ex-
planandum. If one wants to explain economic growth, for example, an institutionalist would em-
phasize how the background social structure explains different growth rates. An institutionalist that
explains the change in the ideology of political parties would emphasize the importance of back-
ground structures more than, say, specific personalities or relationships. An institutionalist ex-
planation of moral obligations would emphasize the importance of variable social structures over
transcendent moral principles. Beyond this basic unity, we can only understand institutionalism by
looking at the different institutionalisms.
1.1 Institutional Economics

Classical Institutional Economics is often associated with Thorsten Veblen, John Commons, and
Clarence Ayers. For Veblen, the dynamics of an economy must be partly explained by the norms in a
society. So, Veblen took an anthropological view towards economics. The patterns of behavior in a
particular class or culture have an affect on the larger distribution of resources. Rather than focusing on
the rules that constituted markets, property, and governance, Veblen was concerned with the back-
ground norms that structure conduct. While he would not have identified his concern as “institutions,”
he has been identified as one of the leading “institutionalists” because his work inspired many of the
institutionalists that followed. Foremost among these was Clarence Ayres. Ayres was a student of both
Veblen and John Dewey. He took much of Veblan’s analysis but redirected his concern from “cere-
monial behavior” towards “institutions.” For Ayres, the institutions of society had a restraining effect,
and we need conscious adjustment of these institutions in order to adjust to technological advancement
and social change (Ayres, 1944). The version of institutions that Ayres advanced downplayed the
explanatory power of individual preferences and desires, which he thought were themselves condi-
tioned by surrounding institutions (Ayres, 1961, Ch. 4). John Commons, by contrast, had a different
emphasis. He took Institutional Economics as the study of how interest groups rigidify processes and
empower people through institutional rules (Commons, 1934, 1–9). Instead of bringing greater at-
tention to norms and culture, Commons was focused on the purposeful action of collective agents.
Classical Institutional Economics had a number of successful research programs, including Wesley
Mitchell’s study of business cycles (Rutherford, 2003, 369) and Adolf Berle’s influence on New Deal
policy (Berle, 1933). However, its influence flagged in Economics following the 1940s. Part of the
reason for this stems from the link between Institutional Economics and Psychology. Rather than
building economic understanding from a formal model of rational decision-making, Institutional
Economics wanted to ground their views in psychology as an empirical science. Partly because of
Dewey’s influence, there was the belief that psychology would vindicate the importance of social
context for explaining individual behavior. Hence, institutions would provide a more foundational
explanation of economic outcomes than individual preferences. However, psychology turned
150
Institutions and Institutionalism
towards behaviorism (Rutherford, 2003, 371). With that turn, it became more difficult to study the
long-term effects of social context on decision-making. You cannot, for example, easily use a be-
havioral lab to prove that institutions instill members of society with certain dispositions. Without
the empirical psychological backing that initial institutionalists expected, they did not have enough
academic support to rival Neoclassical Economics and its microeconomic foundation.
However, a new form of Institutional Economics emerged in the 1970s. Instead of challenging the
microeconomic foundation of Neoclassical Economics, this new form of Institutional Economics
addressed perceived failures in the Neoclassical model. This new form of Institutional Economics is
most associated with four Noble Prize-winning economists; Ronald Coase, Douglas North, Oliver
Williamson, and Elinor Ostrom. For Coase, North, and Williamson, the motivating problem was that
Neoclassical Economics ignored transaction costs. In “The Problem of Social Cost,” Coase famously
argued that – in the absence of transaction costs – free agreements between agents would reach an
efficient equilibrium (1944). However, he did not mean for this argument to vindicate Neoclassical
economics. Instead, he was trying to emphasize that the real world does not reach efficient equilibrium
because transactions costs do exist (Ronald Coase Institute, 1997). Wallis and North famously argued
that 45% of national income was dedicated to transaction costs (North, 1990, 28). For a society to
achieve greater efficiency, we must find ways of lowering these transaction costs. Institutions do that.
Douglas North identified two kinds of transaction costs; measurement costs and enforcement
costs (1990, 30–33). Measurement costs arise because transactions are done with incomplete in-
formation, and finding the desired information about a product is costly. Enforcement costs arise to
maintain honesty across transactions and incentivize agents to keep agreements. For North, in-
stitutions develop in repeated interactions to change the costs of transaction (1990, 61). Admittedly,
he recognizes that these institutions can have deep effects on individual values and motivations
beyond the more immediate effects on transaction costs. However, the basic cause of institutions is
their adjustment to transaction costs.
Williamson looks at transaction costs from a different perspective. He is best known for a focus
on the organization of the modern firm. For Neoclassical economics, the existence of larger
companies can seem odd. Why have different productive activities organized together under a single
firm rather than having independent agents at different steps of a production process contract with
one another? For Williamson – like Coase (1937) – the modern firm exists because unifying these
activities together reduces transaction costs. Yet, of course, one should not suppose that unification
is always conducive to efficiency. Unification also increases the level of bureaucracy, and the effi-
cient size of a firm must balance between reducing transaction costs and reducing bureaucratic costs
(Williamson, 1999, 19). From such an analysis, we can understand the modern firm as one specific
institution structured to decrease transaction costs.
Elinor Ostrom would not deny the importance of reducing transaction costs for explaining the
emergence of institutions. Yet, her work is most associated with another problem of Neoclassical
Economics; collective action. Neoclassical models suggest that either the state would enforce rules
strictly enough to ensure compliance or people would suffer from collective action problems. In
reality, people regularly find ways of overcoming collective action problems without state in-
volvement (Ostrom, 1990). For her, institutions are a mechanism for regulating conduct in a way
that does this. By surveying the myriad forms of institutions that were developed across different
societies, she sought to identify general principles that were associated with the long-term success
and stability of such institutions (Ostrom, 2005, 259).
1.2 Institutional Political Science

In Economics, Classical Institutionalism was considered a heterodox approach (Rutherford, 2003,
360), so it is easy to identify the work of those distinguishing themselves by this idea. While New
151
C. M. Melenovsky
Institutionalism in Economics is less heterodox, it still represents a distinct research program with its
own intellectual lineage. In Political Science, a focus on institutions is more mainstream, so theorists
distinguish themselves as institutionalists less often. In 2002, Pierson and Skocpol said of political
scientists, “we are all institutionalists now” (706). However, we can still get a sense of the in-
stitutionalist perspective in Political Science by tracing its historical development.
A “neo-institutionalist” method in Political Science emerged in the 1980s. As is our theme, the
point was to turn analysis away from the particular choices that political agents make and towards the
structures that give those choices context. This approach is associated with James March and Johan
Olsen, whose influential article, “The New Institutionalism: Organizational Factors in Political Life”
set an agenda for the tradition (1984). Their goal was to distinguish the institution as a unit of
analysis. As such, they (a) denied that institutions were merely another reflection of the larger
society, (b) denied that institutions could be reduced to the aggregate consequences of individual
decision-making, and (c) denied that institutions were merely vehicles that brought about efficiency
(March and Olsen, 2008). Research should focus on the development, change, and effects of these
distinct elements of society.
Theorists took this neo-institutionalist method in three different directions (Hall and Taylor,
1996). First, the Rational Choice Institutionalists explained political behavior with the rational actor
model familiar from Economics. One side of this approach holds institutional rules fixed and
analyzes patterns of choice as the rational pursuit of goals within those rules. As an example, one
might analyze how political parties are unified by the legislative rules that must be followed to pass a
bill. The other side of Rational Choice Institutionalism makes institutional rules variable and ex-
plains the emergence and change of these rules as the result of rational choices. For example,
Buchanan and Tullock’s famous argument for the emergence of simple majority rule in Calculus of
Consent(1962) could be considered a version of this second side of Rational Choice Institutionalism.
Sociological Institutionalism offers a substantively distinct approach. Instead of modeling the
behavior of individuals as the instrumentally rational pursuit of ends, they analyze behavior as
emerging from a given culture. The existing set of norms and beliefs gives meaning to certain
actions, and people are motivated to act because of the way they understand the world around them
(Hall and Taylor, 1996, 947). As such, we often cannot make sense of the actions of individuals
unless they are viewed from within a social and institutional context. The dominant approach for
understanding the ways institutions develop and change is to understand what institutional forms
would be viewed as legitimate in a given social context. As an example, Neil Fligstein argues that the
spread of American firms resulted from a shared understanding among professions that it was a
valuable form of organization (1990). Yasemin Soysal argued that immigration policy emerged from
a shared understanding of human rights (1994).
Finally, Historical Institutionalism eschewed a more general analysis of institutions in favor of
understanding the particular contexts that institutions arise in. They focus on the way that in-
stitutional structure is path-dependent, as developed institutions establish a background for the
formation of new institutions (Sanders 2008). As an example, Stephen Skowronek emphasized the
ways that regulations of the railroads in 19th century United States established a government
structure that influenced how future government regulations were carried through (1982). Ken
Kersch explained the transitions in Supreme Court decisions based on the political context of
elections and specific party ideologies (2004).
Today, there are further sub-genres of institutionalism in Political Science. We can, for example,
distinguish “Constructivist Institutionalism” (Hay, 2008), “Discursive Institutionalism” (Schmidt,
2008), and “Network Institutionalism” (Ansell, 2008). Each offers different ways of studying the
emergence and change of political institutions. However, I will leave discussion of these methods
aside lest this chapter turn to a review of all the different methods in political science. After all,
political scientists are “all institutionalists now.”
152
1.3 Institutionalism in Political and Moral Philosophy

In Philosophy, there are two trends that could be identified as “institutionalist.” The first – and most
dominant – develops from Rawlsian political philosophy. In A Theory of Justice, Rawls identified the
primary subject of justice as the “basic structure of society” which consists of the “major social in-
stitutions” (1971, 7). His two principles of justice apply specifically to these institutions and do not have
any (direct) implication for individual behavior. This is a radical departure from consequentialist moral
theory, which claims that both individuals and institutions should further certain aims. Liam Murphy
argues against the Rawlsian form of institutionalism because it focuses our efforts on institutional
reform and therefore causes us to miss opportunities for making the world better sooner (1998). G. A.
Cohen argued that one cannot rationally support the idea that institutions should enhance equality
without also supporting the idea that individual choices should enhance equality (1997). When pressed
on this argument, Cohen made the further claim that justice is a basic moral ideal, and all basic moral
ideas must apply to both institutions and individual choice (Cohen, 2008, Ch. 6).
In response to these criticisms, a range of defenses emerged for a Rawlsian focus on institutions.
Thomas Nagel argued for a division of labor wherein we best promote what is valuable by dis-
tinguishing principles for institutions from principles for people (1991, 53–62). Samuel Scheffler
argued for a different division of labor wherein certain values are within the appropriate purview of
institutions and other values are in the purview of individual choice (2010, 107–128). Kok-Chor
Tan argued that institutionalism is necessary for people to have the space to pursue personal projects,
which is crucial for any liberal political theory (2004). Melenovsky argued that the distinct principles
for institutions come from Rawls’s dual commitment to conventionalism and constructivism (2014).
Each approach – and the many others that have been offered – gives a slightly different defense for
Rawlsian institutionalism.
A second strand of institutionalism in political philosophy is associated with a communitarian re-
sponse to liberalism. Here, the political allegiance of “institutionalism” is switched. Rawls was an
institutionalist liberal, but communitarians often cast liberals as individualists that treat social structures
as if they were simply emergent from the choices of free and independent atoms (Kymlicka, 2002).
Against this view, Communitarians recognize that individuals are developed within a social context
and act within particular roles established by their society. As such, the communitarian kind of “in-
stitutionalism” is a response to a form of liberalism. Communitarian institutionalism seeks to shift the
focus of value theory away from the individual and towards the nexus of norms, practices and social
values that situate people.
Different as they are, both forms of institutionalism in political philosophy emphasize how
distinctive the evaluation of institutions is from the assessment of individual choices. This has a major
impact on theories of political philosophy. Moreover, both recognize that choice occurs within
a social context partly established by institutions, which has a greater impact on the subfield of moral
philosophy.
1.4 Legal Institutionalism

Because of the work of H.L.A. Hart, there is a close link between legal institutionalism and
positivism. Hart argued that a legal systems is constituted by a set of interacting social rules, and he
(originally) believed that appeal to such rules did not require moral commitments. Since the pub-
lication of The Concept of Law, this belief has been the object of significant scrutiny. Hart char-
acterized social rules as involving an “internal aspect” of approval by members of a social group
(2012, 56), and it was argued that we can only make sense of this internal aspect as expressing moral
commitments. In that case, the law would be constituted by rules that express moral commitments,
and Hart’s positivism would be significantly softened.
153
C. M. Melenovsky
Much of contemporary work in jurisprudence has followed Hart and tried to analyze the legal
system as a social practice constituted by rules (Postema, 2011). This is often – but not always –
linked with an attempt to explain how legal normativity is consistent with positivism. For example,
Coleman (1982) and Postema (1982) both suggest that Hart saw legal practices as solutions to
coordination problems. Such an analysis followed David Lewis’s (2002) own game-theoretic
account of “conventions” as solutions to coordination problems. For Coleman and Postema, Hart’s
“rule of recognition” was a solution to a coordination problem among officials that seek to identify
and apply the law. Later, Coleman (2001) modified and added to his account by arguing that the
conventions were grounded on a joint commitment to cooperative activity. This commitment
better explains the normativity of law because being committed to this cooperative activity would
obligate a person to follow what is recognized as the law even when the law does not align with
their other goals and values. Andrew Marmor offered a third way to understand the law as a
convention. Using a concept associated with Rawls (1955) and Searle (1995), Marmor analyzed the
law as a “constitutive” convention. In his analysis, just as a system of rules constitutes the game of
chess, a system of rules constitutes the legal system. Our reason to follow either rules will depend on
our reasons to engage in the activity, but our reason to follow the rules is not fully reducible to our
reason to engage in the activity. Because it is a constitutive practice, we have reason to respect a legal
system and thereby follow its rules, even if every act of rule-following does not satisfy our reasons to
respect the legal system.
While these versions of “legal conventionalism” can be interpreted as a form of legal in-
stitutionalism, the phrase “institutionalism” itself is most associated with the work of Neil
McCormick. For McCormick, the legal system is a “normative order.” People act in ways according
to what they believe is a shared conception of the right thing to do. The rules of a legal system are
simply expressions of that shared conception. When McCormick refers to the legal system as an
“institutional order” he uses “institution” in a way that is shared with John Searle. For Searle and
McCormick, An “institutional fact” is something that is made true in virtue of our shared re-
cognition of it is as such (MacCormick, 2007, 12; Searle, 1995). A piece of metal in my pocket is a
“coin,” for example, because we both recognize it as a coin. The normativity of law is explained by
the fact that we have shared recognition that it expresses a way that it is right to act. As such, the
normativity of law is viewed, by this tradition, as an institutional fact.
For most institutionalists, the legal system and its constituent parts are paradigm institutions. As
such, many accounts of institutions can be tested by whether they adequately describe a legal system.
In this way, debates in legal theory have been an arena where different views about institutions come
into direct conflict. The legal theory literature is also especially interesting to those concerned with
the normativity of institutions because so much of the focus in the philosophy of law is on the
normativity of legal systems. As such, it is interesting to see how each of the above approaches draws
from more general accounts of conventions and institutions in the philosophical literature. The 1982
views of Coleman and Postema associate law with David Lewis’s analysis of conventions. Coleman’s
later view (2001) relies more strongly on views about collective intention offered by Margaret
Gilbert (1989) and Michael Bratman (1987). Both Marmor and McCormick draw on Searle’s work,
with Marmor developing the idea of law as a deep constitutive convention and McCormick
identifying law as normative order composed of institutional facts.
2 Defining Institutions
As this brief review shows us, each of these different forms of “institutionalism” have distinct aims,
make different contrasts, and work within unique traditions. So, it should be no surprise that these
institutionalists have developed different definitions of what an institution is. To demonstrate these
differences, we can look at a few of the most often quoted definitions.
154
We may define an institution as Collective Action in Control of Individual Action.

Collective action ranges all the way from unorganized Custom to the many organized
Going Concerns, such as the family, the corporate, the holding company, the trade as-
sociation, the trade union, the Federal Research System, the group of affiliated interests,
the State. The principle common to all of them is more or less control of individual action
by collective action. (Commons 1934, 69–70)
Institutions are the humanly devised constraints that structure political, economic, and
social interaction. (North, 1990, 3)
a relatively enduring collection of rules and organized practices, embedded in structures of
meaning and resources that are relatively invariant in the face of turnover of individuals and
relatively resilient to the idiosyncratic preference and expectations of individuals and
changing external circumstances. (March and Olsen, 2008, 3)
a social institution is, in its most general characterization, nothing more than a “stable,
valued, recurring pattern of behavior.” That characterization might be a little too general
to be terribly helpful… Still, a relatively general characterization is precisely what is needed
to capture the diverse range of social activities that we would want to deem to be in-
stitutions and to theorize alongside one another as such. (Goodin, 1996, 21; quoting
Huntington, 1968, 12)
institutions are the prescriptions that humans use to organize all forms of repetitive and
structured interactions. (Ostrom, 2005, 3)
a public system of rules which defines offices and positions with their rights and duties,
powers and immunities, and the like. These rules specify certain forms of action as per-
missible, others as forbidden; and they provide for certain penalties and defenses, and so on,
when violations occur. As examples of institutions, or more generally social practices, we
may think of games and rituals, trials and parliaments, markets and systems of property.
(Rawls, 1971, 47–48)
While each definition is developed for a particular project, there are some clear similarities. Most
importantly, each is concerned with the social conditions that set a structure for individual choice.
However, I don’t expect that we will find a consensus definition of institutions. Instead, we can
explain some of the common differences in this section. Doing this will, I hope, accomplish two
goals. First, it will make sense of individual conceptions of institutions by contrasting them with
rivals. Second, it will contextualize the definition of institutions that this author defends in §3.
2.1 Patterns of Behavior vs. Rules

One of the bigger dividing lines in how theorists understand institutions is whether to think of them as
primarily consisting of rules or as consisting of stable patterns of behavior. When there is a social
practice that consists of rules, we will often see stable patterns of behavior that align with those rules.
However, there are also certain stable patterns that are independent of any rules. To use an example
from the previous chapter, we will see people carrying umbrellas in the city when it rains. This is not
because rules require umbrella use, it is simply because umbrellas are a generally available tool for
staying dry. Some will argue that institutions are partly constituted by such stable patterns (like umbrella
use) while others will instead focus on rules that sometimes (but not always) underlie stable patterns.
For those concerned with the ways that institutions set expectations – and thereby make certain
choices rational or sensible – there is strong reason to define institutions as consisting of stable
155
C. M. Melenovsky
patterns that are not structured by rules. After all, such stable patterns set expectations regardless of
whether rules apply. By contrast, if people are concerned with the way that institutions put social
pressure on people to act in certain ways, they have reason to think of institutions as rule-structured.
It is the recognition of social rules, after all, that often explains the social pressure. Moreover, those
interested in the ethics of institutions will have additional reasons to focus on rules. We might have
reason to follow rules that go beyond any reason to align our actions with stable patterns.
2.2 Common Beliefs and Values

The most expansive conceptions of institutions will say that – beyond patterns of action and rules –
institutions are also composed of common beliefs and values. For example, we might suppose that it
is common among Americans to think that earned wealth is a sign of intelligence. Or, members of a
Parliament might think it is beneath them to make calls to donors to raise funds for elections. Some
might regard these common beliefs as institutions in themselves. After all, such common beliefs do
form a background against which people make their choices. So, there is little reason to exclude
common beliefs and values from their list of institutions for those concerned with characterizing
institutions as setting the background conditions for individual choice
On the other side, including common beliefs and values threatens to make the class of in-
stitutions too expansive. We lose the ability to analyze a feature of society when it takes so many
different forms and works in different ways. For those that focus on patterns of behavior, for ex-
ample, the unifying idea is that institutions represent instances of interdependent choices. People act
in accordance with an institution because they expect others to act according to that institutions.
Such mass-scale interdependent choices are unique from other aspects of society, and we can focus
on that kind of social structure by isolating it. A narrower conception of institutions allows us to do
that. Similarly, those that definite institutions as consisting of social rules can focus even more
precisely. Not only are they concerned with mass-scale interdependent choices but they are also
concerned with those choices for which there are social rules. While these approaches might exclude
some parts of the society that have influence similar to – what they call – institutions, they have the
capacity to isolate causes and effects that are specific to a more delineated phenomenon.
However, one contentious issue makes this distinction more difficult: some allege that it is
impossible to distinguish a conception of institutions based on rules from one based on common
beliefs. Those that define institutions as consisting of (observable) patterns of behavior do not need
to say anything about mental states. However, it is not obvious what the rules are. It might be that
rules are best understood as commonly held beliefs about how people should act. In that case, the
distinction between (a) institutions as systems of rules and (b) institutions as commonly held beliefs is
called into question.
2.3. Different Conceptions of Rules

Even those that think of institutions as systems of rules often have very different views on what rules
are. There are at least five different ways of understanding rules. First, one can think of rules simply
as predictions of positive and negative responses. A rule exists, for example, when we think that
people will punish us for deviating from some expected behavior. Such a conception of rules might
collapse a distinction between (a) institutions as constituted by patterns of behavior and (b) in-
stitutions as constituted by rules.
Second – as mentioned above – a social rule might simply be a shared belief about how people
should act. In this case, a social rule would be similar to a personal rule (e.g., “I’ll only eat 3 oreos at a
single sitting”) except that (i) it applies to more than just you (e.g., “Everyone should only eat 3 oreos
at a single sitting”) and (ii) many people in a social group share the belief (e.g., most people in this
156
social group think that “Everyone should only eat 3 oreos at a single sitting”). This threatens to
collapse the distinction between (a) institutions as constituted by rules and (b) institutions as con-
stituted by shared beliefs.
A third view – most associated with Margaret Gilbert (1989) – sees rules as expressions of joint
intentions. As the primary examples goes, for us to go on a walk together requires more than two
separate intentions to walk. Instead, it requires that we each have the intention to walk together. We
have an “we-intention” such that each of us has the intention, “we will go for a walk.” To have this
intention is to commit oneself to a joint activity with the other person that obligates them to the
strictures of walking together. From there, Gilbert says that a social rule exists “if and only if the members
of some population P are jointly committed to accepting as a body a requirement of the following form:
members of P are to do A in C” (1999, 163). The joint commitment to such a rule obligates individuals
to follow that rule, so the normativity of rules is explained by the commitments made to others.
A fourth view – most associated with Michael Bratman (2021) – sees rules as expressions of
shared individual intentions, which are constructed from our plans. When I form a plan to paint a
house with you, such an intention rationally requires me to be sensitive to the ways that you are
acting. Similarly, your intention to paint a house with me rationally forces you to act in ways that are
sensitive to my ways of acting. Together, we share an intention to paint a house together, and our
choices depend on one another in interlocking ways. Social rules are the result of a shared intention
to endorse a public pattern of action (58). The normativity of such rules is thus explained by the
requirements of rational planning. Because such an intention is shared with others, I am rationally
compelled to follow and apply that standard only when there is general conformity with that public
pattern of action. In this way, social rules are constructed from our planning capacity, which
produce shared intentions.
A fifth view sees rules simply as irreducible ways that we understand the social world. To say that
they are “irreducible” does not mean that they have their own mass, force, or energy. Rules do not
have any existence independent of people. Rules are irreducible because we understand social
life in terms of rules much as we see the world in terms of colors. While both color-perception and
rule-understanding occur through explicable mechanisms, that does not mean we fully capture
color-perception or rule-understanding when we describe it in other terms. For example, think of
the practice of waiting in line. We learn from an early age to wait in line. If any of us were to explain
the practice to someone unfamiliar with it, we could easily explain it in terms of rules. We might say
something like, “well, when you wait for service, you are supposed to wait until after a person that
arrived before you has gotten service. Often, this order of service is communicated by people
waiting behind those that arrived before them.” This phrase, “you are supposed to” is simply the
expression of a rule. It is as if we imagine a social group to be represented by the judgment of a single
person, and we interpret the rules of interaction as what such a singular person believes we should
and should not do. We don’t associate this single judgment with the edict of any one actual person
and there is no definitive articulation written down somewhere. Regardless, it is how we interpret a
given social context. So, it is an irreducible feature of our understanding.
Anyone that understands institutions in terms of rules invites these complications. While we have
an intuitive understanding of what rules are, these different conceptions show that it’s not easy to
explicate this intuition. However, we understand what social rules would have a significant impact
on what institutions are.
2.4 Formal vs. Informal Institutions

In the literature on institutions, there is a common distinction made between “formal” and “informal”
institutions. For some, only “formal” institutions are truly institutions. “Informal” institutions would
then be more appropriately identified as “social practices.” The formal/informal distinction is often
157
C. M. Melenovsky
used as if it means the same thing to all discussants, but there are actually four different ways of making
the distinction. Any of them, therefore, could be used to distinguish institutions from social practices.
The first, and most common, way of distinguishing formal vs. informal institutions is based on
whether they are legal vs. non-legal. Put simply, does the institution emerge from the legal system or
not? The rules can “emerge” from a legal system in three ways. First, the institutions can be the
result of direct legislative action. If, for example, the federal government establishes an
Environmental Protection Agency, that might be called a legal institution. Second, the institution
might emerge because of the way rules are interpreted by a judicial authority. When the US
Supreme Court, for example, gives an authoritative judgment about the law of contracts, the newly
clarified rules have emerged from the legal system. Third, an institution can emerge from a legal
system if an agency has the legal authority to establish new rules. For example, if the Securities and
Exchange Commision changes the reporting requirements for publicly traded companies, the new
rules emerge from a legal authority even though they did not pass directly through a legislative
process or by court interpretation.
A second way of distinguishing formal and informal institutions is by whether the institution
directs government activities. The rules apply to those who occupy (or seek to occupy) roles in a
government. A prime example might be the rules of legal interpretation. Importantly, not all
governmental institutions are legal institutions. For example, suppose that those running for office
are expected to seek the endorsement of local paper, and local papers are expected to responsibly vet
each candidate. Someone that fails to do this might be viewed as an unserious candidate, and hence
candidates might reliably follow the rule to seek these endorsements even when they do not think
they will receive them. This would not have any legal status, but it would nonetheless be a gov-
ernmental institution. Such institutions show that not all “governmental” institutions are “legal”
institutions. So, someone that discusses “formal” institutions must choose between them.
There is still a third way that someone might identify institutions as “formal.” Institutions might
be planned and purposefully implemented. Most of the time, this will coincide with a legal in-
stitution, but it does not need to. For example, a (smaller) social group might come together to
decide to establish a set of rules. After such a group decision, people might start following the rules
and it could be stable over time. Such an institution would be purposefully established, but it would
not necessarily involve the law. An example of such an institution would be a social group that aims
to eliminate the practice of foot-binding by promising to not bind their daughters feet and refusing
to marry their sons to women with bound feet (Mackie, 1996). The new rules established by this
group would create a purposive institution, but it would not involve government. Elinor Ostrom
offers many examples of institutions established by communities without government involvement.
These would be “formal” institutions according to this third distinction, but they don’t necessarily
involve government. By contrast, some institutions emerge over time without any purposive es-
tablishment. Perhaps there was simply a common way of doing things that became institutionalized.
Perhaps people would eliminate the institutions if they could. This would be an “emergent” in-
stitution. The third distinction identifies these emergent institutions as “informal” institutions.
A fourth, and final, kind of “formal” institution is one that establishes a jointly recognized au-
thority. For example, suppose that pro tennis players get together and establish a group to negotiate
payments to players from tournaments. The negotiations of this group would be respected by the
players, so it would have a kind of authority. The Organization of the Oil Producing Countries
(OPEC) would be an authoritative institution of this type. It sets production standards for oil-
producing countries that individual countries have incentive to deviate from. When countries
respect the decision of this body, they grant it a kind of authority. Most obviously, the establishment
of a government represents the creation of an authoritative institution. By contrast, the institution of
the filibuster in the US Senate does not establish any authority. It is set of rules that organize the
actions of senators, but it does not have the power to make new decisions that are respected by
158
participants. Throughout time, there have been certain rules that structured war between countries.
Such rules could be considered institutions, but they are not the institutions that establish authorities.
When agreements establish the International Court of Justice, by contrast, they establish an au-
thoritative institution.
For some cases, the various distinctions will align. The Food and Drug Administration, for
example, seems to be a formal institution in all four senses. It is a legally established, government-
directed, purposefully created, and authority-creating institution. Yet, many institutions will not be
so unified. The Blue Line of Silence among policy is not legally established, is government-directed,
may have been purposefully created, and does not establish an authority. In cases like this, we need
to be more discerning to identify it as either “formal” or “informal.”
2.5 Choosing What an Institution Is

In drawing all these contrasts, I hope to have better shown what is at stake in a given definition of an
institution. As I understand it, to name a feature of social life an “institution” is to identify it as a kind
different from other features of social life. Anyone who organizes their understanding of the social
world in a particular way does so for reasons. There will not be any obvious definition of institutions
that is independent from a theorist’s aims and tradition. Foremost among these choices are whether
institutions are constituted by patterns of choice, rules, common beliefs, or a combination of these
elements. Those who include rules will need an understanding of what rules are. Those that see
institutions as “formal” will need to identify what kind of formality they have. As the traditions of
institutionalism stand, there is no consensus on these issues, but could there be?
3 A Reasonable Consensus?
I want to end by offering one view of how to differentiate institutions from other social conditions
Specifically, I offer my view. I take this to be an appropriate conclusion for two reasons. First, it puts
my own cards on the table. I don’t claim to be an impartial observer of the literature. I am a
disputant, and you might want to interpret my previous remarks in the light of my avowed position.
Second, I want to show that interdisciplinary complexity does not mean that we cannot say
something general about how to think about institutions.
As I’ve argued throughout, how one understands what institutions are will depend on one’s aims
and the tradition one works in. So, let me be explicit about my aim. I want to find ways of bringing
the disciplines together in the study of social phenomena. As such, I seek a definition of institutions
that economists, political scientists, philosophers, and legal theorists are all comfortable with. Yet, I
do not think such a consensus position exists, and I do not think that one could construct a position
that all would accept. However, I do think that we can identify a position that everyone looking for a
consensus could accept. Said differently, I am not looking for agreement amongst all social theorists. I’m
looking for a definition that a sub-group of social theorists could accept. I’m looking for agreement
amongst those willing to accept a consensus definition. I think that there is value in a shared de-
finition of institutions that can coordinate research disciplines with different aims. Others likely
agree. The definition that I put forward is meant to accomplish the aim of better coordinating the
study of institutions across disciplines.
The conception of institutions that I support is “variable systems of social rules that are deemed to
be socially significant.” This conception has four features that are worth explaining in more detail.
First, institutions are “variable” in the sense that they differ from one society to the next, and they
could be different in our current society. Often, people refer to conventions as “arbitrary” to capture
that they could have been different, but “arbitrary” has the unfortunate connotation that nothing is
at stake in organizing an institution one way rather than another. Second, institutions are constituted
159
C. M. Melenovsky
by social rules. This is a clear choice that differentiates this approach from those that focus on shared
beliefs or patterns of action. I defend this choice more below. Third, this favored conception
identifies institutions as “systems” of social rules. There are different kinds of rules that interact with
one another in complex ways. Rather than single rules, institutions have interlocking rules that can
constitute roles, objects, and activities.
The fourth feature of my preferred conception is that institutions are “deemed to be socially sig-
nificant.” This is, perhaps, the strangest aspect of this conception because calling a social structure an
institution would then communicate a personal judgment. However, this feature is crucial for cap-
turing the way in which institutions differ from games. “Freeze tag,” for example, is a variable system
of rules, but it is not an institution. We do not call it an institution because we do not view it as socially
significant enough. A property system, by contrast, would be an institution because it is a variable
system of rules that are deemed to be socially important. This also captures a feature of common
linguistic usage. You might hear someone say, “The Boston Globe isn’t just a paper, it is an institu-
tion.” When they say that, they seem to be indicating that it has a certain standing in the community.
Part of it is that it has a certain place that is respected by others. Perhaps, they indicate that politicians
treat it a certain way, other journalists treat it a certain way, or members of the community rely on it as
a source of honest information. It has importance in the community, and it should be regarded as such.
The foremost reason to define institutions in this way is because we can thereby isolate a unique
feature of our social structure that is conducive to focused study. In the name of precision and
accuracy, it is worth distinguishing socially significant systems of rules from other social conditions.
My reason to focus on systems of rules – rather than including shared beliefs or all patterns of
conduct – is because social rules have unique motivational power. They motivate action for four
distinct kinds of reasons. First, one often best accomplishes their own aims by coordinating with
others. Systems of social rules establish the expectations of others that make this possible. In her
analysis of social norms, Cristina Bicchieri calls the expectations of what others will do “empirical
expectations” (Bicchieri, 2005, 13). So, when I have reason to coordinate with others and the rules
of an institution set my empirical expectations, I am best able to coordinate by following the rules of
the institution. Second, a system of rules expresses what a member of a social group believes other
members of the social group think people should do. Bicchieri calls these “normative expectations”
(2005, 14). I might worry that others will punish me if I deviate from these normative expectations.
I might simply fear being judged badly. I might even just want to act as others want me to act. Either
way, such normative expectations are a second motivation to follow social rules. Third, one might
have a latent disposition to be a rule-follower. Perhaps, I have internalized the rules of a given
institution and thereby follow the institution from my own commitments. Perhaps, one is un-
thinkingly predisposed to follow all social rules unless given reason to reject them. Regardless, such
dispositions do motivate people to act in accordance with institutions. Fourth and finally, I might
have a moral commitment to following rules that I deem to be part of a worthwhile system of rules.
Even when this commitment goes against my predispositions, I might follow the institution because
I think it best to do so.
Alternative proposals might focus exclusively on one of these motivations, but I see little reason
to do so. People are complex, and different people will be motivated for different reasons at different
times. A system of rules is especially powerful because it can incite all these motivations, thereby
capturing the widest conformity amongst the broadest group. Some will follow an institution be-
cause they thereby better coordinate with others, some because they think others think they should,
some because they are disposed to be rule-followers generally, some because they have internalized
the particular rules of an institution, and some because of a moral commitment to the institution.
We do not need to pick one motivation to understand an institution as a system of rules.
As an objection, one might say that the aim of narrowing our focus to a social structure with
unique causes and effects would lead us to ever more fine-grained concerns. After all, we could see
160
even more unique causes and effects if we defined institutions as systems of rules relating to the
handling of human waste. This reduction seeks to show that we should not be concerned with the
isolation of phenomena in defining institutions. Yet, narrowing our conceptions of institutions even
further would fail to capture how the term is generally used. In the literature, “institution” is used to
refer to a wide range of activities across different contexts. My conception strikes that balance
between typical use and unique social phenomenon.
Now, the point of this chapter has not been to defend my preferred conception of institutions.
Much more would need to be filled in to do that. Instead, the point of the chapter has been to give a
brief tour of institutionalism that can be used to better contextualize any conception of institutions.
We thereby understand what is at stake in choosing a particular conception of institutions. While I
do not think that one definition is capable of capturing what every theorist is concerned with, I do
believe it is possible to identify a consensus position amongst those that seek a consensus position.
We can identify a definition of institutions that better coordinates the research of those working in
the different traditions. The idea of institutions as “variable systems of social rules that are deemed to
be socially important” is offered as one possibility.
References
Ansell, Chris. (2008) “Network Institutionalism” in Binder, Rhodes, and Rockman ed. The Oxford Handbook of
Political Institutions. Oxford; Oxford University Press.
Ayres, Clarence. (1944) The Theory of Economic Progress. Chapel Hill, NC; University of North Carolina Press.
Ayres, Clarence. (1961) Towards a Reasonable Society. Austin, TX; University of Texas Press.
Berle, Adolf. (1933) “The Social Economics of the New Deal.” New York Times, October 29, 1933.
Bicchieri, Cristina. (2005) The Grammar of Society. Cambridge, UK; Cambridge University Press.
Bratman, Michael. (1987) Intention, Plans, and Practical Reason. Cambridge, MA; Harvard University Press.
Bratman, Michael. (2021) “Shared Intention, Organized Institutions” in David Shoemaker ed. Oxford Studies in
Agency and Responsibility, Vol. 7, Oxford, UK; Oxford University Press.
Buchanan, James M. and Gordon Tullock . (1962) The Calculus of Consent: Logical Foundations of Constitutional
Democracy. University of Michigan Press.
Coase, Ronald. (1937) “The Nature of the Firm.” Economica, 4, 16, 386–405.
Coase, Ronald. (1944) “The Problem of Social Cost.” Journal of Law and Economics, 3, 1–44.
Cohen, G. A. (1997) “Where the Action Is: On the Site of Distributive Justice” Philosophy & Public Affairs,
26, 1, 3–30.
Cohen, G. A. (2008) Rescuing Justice and Equality. Cambridge, MA; Harvard University Press.
Coleman, Jules. (1982) “Negative and Positive Positivism.” Journal of Legal Studies, 11, 1, 139–164.
Coleman, Jules. (2001) The Practice of Principle. Oxford, UK; Oxford University Press.
Commons, John R. (1934) Institutional Economics: It’s Place in Political Economy. Macmillan Company.
Fligstein, Neil. (1990) The Transformation of Corporate Culture. Cambridge, MA; Harvard University Press.
Goodin, Robert. (1996) “Institutions and their Design” in Robert E. Goodin ed. The Theory of Institutional
Design. Cambridge, UK; Cambridge University Press.
Gilbert, Margaret. (1989) On Social Facts. Princeton, NJ; Princeton University Press.
Gilbert, Margaret. (1999) “Social Rules: Some Problems for Hart’s Account, and an Alternative Proposal.” Law
and Philosophy, 18, 2, 141–171
Hall, Peter A. and Rosemary C. R. Taylor. (1996) “Political Science and the Three New Institutionalism.”
Political Studies, 44, 936–957.
Hart, H. L. A. (2012) The Concept of Law. Oxford, UK; Oxford University Press.
Hay, Colin. (2008) “Constructivist Institutionalism” in Binder, Rhodes, and Rockman ed. The Oxford
Handbook of Political Institutions. Oxford; Oxford University Press.
Huntington, Samuel. (1968) Political Order in Changing Societies. New Haven, CT; Yale University Press.
Kersch, Ken. (2004) Constructing Civil Liberties: Discontinuities in the Development of American Constitutional Law.
Cambridge, UK; Cambridge University Press.
Kymlicka, Will. (2002) Contemporary Political Philosophy. Oxford, UK; Oxford University Press.
Lewis, David. (2002) Convention: A Philosophical Study. Malden, MA; Blackwell Publishers.
161
C. M. Melenovsky
MacCormick, Neil. (2007) Institutions of Law: An Essay in Legal Theory. Oxford, UK; Oxford University Press.
Mackie, Gerry. (1996) “Endding Foot-Binding and Infibulation: A Conventional Account.” American
Sociological Review, 61, 6, 999–1017.
March, James and Johan Olsen. (1984) “The New Institutionalism: Organizational Factors in Political Life.”
The American Political Science Review, 78, 3, 734–749.
March, James and Johan Olsen. (2008) “Elaborating the ’New Institutionalism’” in Binder, Rhodes, and
Rockman ed. The Oxford Handbook of Political Institutions. Oxford; Oxford University Press.
Melenovsky, C. M. (2014) “Incentives, Conventionalism, and Constructivism.” Ethics, 126, 3, 549–574.
Murphy, Liam. (1998) “Institutions and the Demand of Justice.” Philosophy and Public Affairs, 27, 4, 251–291.
Nagel, Thomas. (1991) Equlity and Partiality. New York, NY; Oxford University Press.
North, Douglas. (1990) Institutions, Institutional Change, and Economic Performance. Cambridge, UK; Cambridge
University Press.
Ostrom, Elinor. (1990) Governing the Commons. New York, NY; Cambridge University Press.
Ostrom, Elinor. (2005) Understanding Institutional Diversity. Princeton, NJ; Princeton University Press.
Postema, Gerald. (1982) “Coordination and Convention at the Foundations of Law.” The Journal of Legal
Studies, 11, 1, 165–203.
Postema, Gerald. (2011) Legal Philosophy Is the Twentieth Century: The Common Law World. Volume 11 of A
Treatise of Legal Philosophy and General Jurisprudence. New York, NY; Springer Publishing.
Rawls, John. (1955) “Two Concepts of Rules.” The Philosophical Review, 64, 1, 3–32.
Rawls, John. (1971) A Theory of Justice. Cambridge, MA; Harvard University Press.
Ronald Coase Institute. (1997) “Interview With Ronald Coase” accessed October 6, 2021, available at:
https://www.coase.org/coaseinterview.htm
Rutherford, Malcolm. (2003) “American Institutional Economics in the Interwar Period” in Samuel, Biddle,
and Davis ed. A Companion to the History of Economic Thought. Malden, MA; Blackwell Publishing.
Sanders, Elizabeth. (2008) “Historical Institutionalism” in The Oxford Handbook of Political Institutions.
(Edited by Sarah A. Binder, R. A. W. Rhodes, and Bert A. Rockman). Oxford University Press.
Scheffler, Samuel. (2010) Equality and Tradition. New York, NY; Oxford University Press.
Schmidt, Vivian. (2008) “Discursive Institutionalism: The Explanatory Power of Ideas and Discourse.” Annual
Review of Political Science, 11, 303–326.
Searle, John. (1995) The Construction of Social Reality. New York, NY; Simon and Schuster Press.
Skowronek, Stephen. (1982) Building a New American State. Cambridge, UK; Cambridge University Press.
Soysal, Yasemin. (1994) Limits of Citizenship: Migrants and Postnational Membership in Europe. Chicago, IL; The
University of Chicago Press.
Tan, Kok-Chor. (2004) “Justice and Personal Pursuits.” The Journal of Philosophy, 101, 7, 331–362.
Williamson, Oliver. (1999) The Mechanisms of Governance. New York, NY; Oxford University Press.
162
13
PROPERTY
Bas van der Vossen
Property rights are central to every society on earth. They structure economies, both developed and
developing, and determine how valuable resources are allocated. Of course, there are other ways of
allocating resources, and not all societies use property rights to the same degree to allocate resources.
But no society exists in which property rights do not play a very important role.
The system of property with which most of us are most familiar is a system of private property. It’s
distinctive of private property rights systems, compared to other ways of allocating resources, that
they decentralize economic decision-making. Private property rights can belong to individuals or
collectives of individuals, and they give these individuals and collectives significant authority with
regards to their possessions. When things are yours, others cannot use or take those things without
your consent. And so, part of the economy is under your control or authority, with many other parts
under the control of many others.
Many find the decentralized nature of private property a major attraction. Others find it proble-
matic. What’s nice about decentralizing economic decision-making is that it empowers individuals to
act on the reasons and information that they can best perceive in the part of the economy where they
live and act. But many find it worrisome that this means that the allocation of resources is sensitive
to just these facts. The more private property structures the allocation of resources, the less room
there will be for considerations of the overall distribution, income inequality, and so on.
This chapter discusses the nature and value of property rights. It will explain (1) what property
rights are, (2) the relationship between private property and economic development, and (3) some
objections to structuring societies around such rights. This discussion throughout focuses on the
decentralizing nature of private property rights, asking what implications it has from philosophical,
social, and political points of view.
1 What Are Property Rights?

At their most abstract, property rights give people claims over resources. Such claims can be complex –
with at least as many different dimensions as there are different possible claims one might have.
Typically, when one is the owner of some object O, one gets to decide who gets to use O in certain
ways. If I own this car, I get to decide whether you may drive it. To be sure, one need not have
absolute authority over how O gets to be used. Recognizing I cannot run you over with the car in no
way detracts from my owning the car. But, typically, those who lack authority over some O are not the
owners of O.
DOI: 10.4324/9780367808983-16 163

Bas van der Vossen
Typically, but not necessarily. There are other claims one might have over the car. Suppose I
bought the car using a loan from my bank. In that case, I own the car in the sense of being able to
decide whether or not you may drive it. But I am not the owner in the sense of having a right to the
full economic value of the car. If I sell the car, I owe a sum to the bank. Should I stop making
payments, the bank will repossess the car. Correspondingly, my property rights to my car can be
violated in many ways, too. My car might be stolen or damaged without my consent. I might be
coercively prevented from using it by others. I could be prohibited from selling it or giving it away,
wrongfully expropriated by the government, and so on. Each of these actions we could describe as a
violation of my property right. But they are meaningfully different.
We seem to have a rather complex kind of property right in this example, therefore. I own the
car in one respect, the bank owns it in another. And in the grand scheme of things, car ownership is
still a fairly straightforward matter. Property rights get far more complex. (Consider, for example,
financial derivatives.) In a famous study, Tony Honoré (1961) claimed there exist no less than eleven
“incidents” of ownership. These include the right to exclusive physical control of a thing, the right
to use the thing, the right to decide who is allowed to use the thing and how, the right to the
income earned with the thing, the right to consume, waste, or destroy the capital of the thing, the
right to transfer the ownership to another person, and more.
There is some debate about whether any of these incidents are essential to property rights. The
question here is whether one or several of these incidents must be present for one to have a genuine
property right. Looking at the complexity of property rights, some have argued that there is no such
“core” or essential incident. We can find examples of property rights that lack any one of these
incidents. Consider, for example, the incident of exclusion. You might own a piece of land but lack
the right to keep out hikers. Or consider the incident of use. You might own an environmentally
protected piece of land, meaning you cannot build certain structures on it. And so on.
Those who think there are no essential incidents are fond of the following metaphor. We should
think of property rights like a bundle of sticks, it is argued. Each stick represents one of the various
incidents, and it is the bundle that represents the property right. As no particular stick is necessary to
make a bundle, we should really talk about what matters about these bundles as a whole, not
whether or not they contain this or that stick. There are no essential incidents, therefore (Honoré,
1961; Grey, 1980).
But this view is not without its problems. For example, it seems that having some sticks in the
bundle makes it very difficult to leave out certain others. So, while no particular incident is essential,
at least the presence of certain incidents requires the presence of others (Attas, 2006). And it does
seem odd to imagine a world in which no one had the right to exclude anyone else, and then claim
we’re still imagining a world containing property. The right to exclude may be practically un-
avoidable if a system of property rights is to function as it should (Schmidtz, 2011).
This last point suggests that we might usefully distinguish two separate questions. On the one
hand, we might wonder whether the presence of any particular rights-element is necessary for any
given right to qualify as a property right. On the other hand, we might wonder whether the presence
of any particular rights-element in a certain proportion of rights is necessary for a system, consisting
of those rights, to qualify as a genuine system of property rights.
Perhaps the best answer to the first question is that no particular element is, strictly speaking,
necessary. That is, for any given element, we can come up with plausible examples of a property
right that lacks them. That would seem to include the element of exclusion, as the example of
environmentally protected land above suggests. You might lack the right to exclude hikers from
trails that traverse your land yet still be the clear and sole owner of that land.
It’s worth noting here that it’s not just because of government regulation that particular elements
of property rights can be taken away. Another possible source is private contract. Suppose you take
out a mortgage with a bank to buy a house. The bank can transform the mortgage loan into
164
Property
collateralized debt obligations, making a number of entities the shared (conditional) owner of your
house in the event of your nonpayment. All of these effectively unpack what was once a thick
bundle of elements into several separate sticks (or: smaller bundles of sticks that together once made
up the thick bundle).
But this answer to our first question does not entail a similar answer to our second question. Even
if any particular property right may contain or lack any particular rights-element, it’s a further
question whether any just or recognizable system of property can omit any element to a significant
degree. It’s possible – and I think, plausible – that a well-functioning or just system of property must
have certain general or recurrent features, even if the rights within that system need not all share
those features.
Perhaps the best candidate for an essential or necessary element to individual property rights (and,
by extension, property systems) is proposed by John Locke (1988 [1689]) in his famous discussion of
property rights. Locke wrote that “I truly have no Property in that, which another can by right take
from me, when he pleases, against my consent” (ibid., 138). The point is that, whatever incidents of
property we might possess, if what we have is a property right it must at least be the case that it’s
genuinely ours. And for it to be genuinely ours it cannot be the case that others can remove said right
from our possession without our consent (Cf. Attas, 2006).
We might call this element immunity to expropriation.1 It’s probably too strong to say that this is an
absolutely essential element to property rights. After all, many societies allow for legal doctrines like
(what in the American context is called) eminent domain. Such doctrines allow governments in
exceptional circumstances to expropriate people, provided just compensation is paid. Nevertheless,
we naturally think of people in those societies as owning private property. Still, to the extent that the
immunity to expropriation becomes weakened, we also naturally think of people’s property rights
diminishing. And so here we do seem to have one element that at least must be widely and robustly
(if not absolutely) respected if a system is to fully count as a system of property rights.
2 The Importance of Private Property

The above points might be summarized as follows. Property rights are (a) robustly immune to
expropriation and (b) entitle owners (at least as a rule) to a measure of exclusive control over things.
Rights to property become rights to private property when the owners of property are individuals.
These are mere analytical observations, observations about what property rights are. They do not
mean that property rights are inevitable, a good idea, or specify who owns what. That said, almost
no one today disputes that private property rights of some kind are highly important. This leaves
room for plenty of disagreement, of course, but the case that private property is at least an important
element of just and prosperous societies has few serious detractors left.
One reason for this is straightforward: there are simply no examples of societies with sustained high
living standards that have not given private property rights pride of place. There exist compelling
empirical correlations between respect for private property (and related economic freedoms, such as
freedom of contract, stable and reasonable taxation levels, etc.) and things like per capita Gross
Domestic Product (GDP) and life expectancy (Gwartney et al., 2020). Here’s how economist Daron
Acemoglu and political scientist James Robinson put it in their influential book, Why Nations Fail:
[E]conomic institutions foster economic activity, productivity growth, and economic

prosperity. Secure property rights are central, since only those with such rights will be
willing to invest and increase productivity. A businessman who expects his output to be
stolen, expropriated, or entirely taxed away will have little incentive to work, let alone any
incentive to undertake investments and innovations. (2012, 75)
165
Bas van der Vossen
Note the role of the elements above: the rights to exclusion and immunity to expropriation enjoyed
by individuals. Without secure and exclusionary property rights, people often can’t afford to invest
and increase their productivity. Workers, entrepreneurs, or traders who cannot expect control over
their output have little incentive to work, let alone invest or innovate. And these rights are the
engines of prosperity and growth. It’s precisely because they put economic decision-making au-
thority in the hands of individuals that those societies become richer (Acemoglu, Johnson, and
Robinson 2005, 403). Those individuals are the ones who do the work, know most about what
they’re good at, what other people want, and so on.
Of course, countries may be rich but have several other problems. Most people who are
or remain skeptical of private property worry about the inequality that such wealth seems to
inevitably bring along. Below we will return to the topic of inequality. But for now, one thing that
is worth noting is that these benefits do not just redound to the rich. Poor people in these societies
also benefit from living in societies with private property, even disproportionally so (Dollar and
Kraay, 2002; Bergh and Nilsson, 2014). At least in practice, then, there seems to be a difference
between fighting poverty and fighting inequality. Countries that do well in terms of fighting poverty
may nevertheless end up with considerable inequality.
That said, countries that offer strong protections for private property rights include Norway,
Germany, and Spain. These are countries with extensive redistributive government programs,
funded through taxation, and taxation constitutes at least a prima facie breaching of people’s property
rights over their income and wealth. But while these facts may not demonstrate that all forms of
taxation ought to be regarded as wrongful on balance, they do show that any responsible view of
taxation ought to be compatible with a strong veneration of private property. In our world, there is
simply no good alternative to respecting such rights.
3 Property Rights and Externalities

People, we saw Acemoglu and Robinson write, have little incentive to work, invest or innovate if
they expect their output to be stolen, expropriated, or taxed away. The friend of a private property
agrees. Private property rights secure for people the benefits of their labor, decisions, and possessions.
They reward productive activities – the kinds of activities that benefit everyone – and thereby
encourage people to do more of these. And they prohibit harmful or extractive activities – the kinds
of activities that involve taking from others what they produced. These are the benefits of de-
centralizing economic decision-making, essential to a prosperous society.
But there are downsides to such decentralized economic systems as well. Private property rights
discourage certain kinds of harmful behavior because it would constitute a violation of those rights.
But systems of private property allow and even encourage behavior that does not violate these rights,
and that can include other kinds of harmful behavior. The story about prosperity is at least more
complicated than just the one about private property summarized above, then.
One challenge is that our individually productive activities, the ones that private property rights are
supposed to encourage, also often produce bad outcomes for others. In particular, our productive
activities often create waste, pollution, noise, and so on. These are of course real problems. We can
harm other people’s ability to live their own lives, we can harm the environment, we can harm future
generations. Acemoglu and Robinson point out that the benefits of productivity should redound to
their producers. How should we treat the harms (the pollution, noise, etc.) that they also produce?
In economics, the term externalities refers to those outcomes of people’s actions that do not
redound to themselves. Positive externalities are beneficial outcomes, negative externalities harmful
ones. As a rule, property rights systems operate better to the extent that they produce fewer ex-
ternalities – or better: when people internalize externalities. People internalize positive externalities
when they profit from the beneficial things they do, for example by being allowed to keep the
166
Property
rewards of their investments, labor, and so on. This gives them the incentive to do socially useful
things. People internalize negative externalities when they must bear the costs of their activities by,
for example, paying for the pollution they create. This incentivizes them to avoid doing destructive
things. When all externalities are internalized, people are maximally incentivized to create value and
not create disvalue.
Friends of property rights will thus insist that it’s a problem when producers are allowed to, for
example, pollute the environment without having to bear the costs for this. In these cases, not all
externalities are internalized by producers. As a result, the incentives producers face are not well-
aligned with socially beneficial outcomes, and we should expect them to produce too much pol-
lution. Or, more precisely, we should expect them to produce pollution beyond the point where the
value of their products does not outweigh the disvalue of the pollution. After all, from the point of
the producer, every additional unit of production is worth it – for they do not need to bear the cost
of pollution. But this means they will continue to create pollution even beyond the point where the
badness of pollution is no longer outweighed by the goodness of further production.2
There are several ways in which negative externalities might be internalized. If my pollution
somehow harms your property, I might be made to pay you a sum of money to offset the harm. In
cases where such individual payments are not possible or feasible, a tax might be imposed on
producers representing the negative value of the pollution they produce. Other possibilities exist,
too. Governments might auction (tradeable) permits to pollute up to a certain point, such as in cap-
and-trade systems for carbon emissions. All such measures aim at the same thing: making sure that
the negative effects of productive activities do not outweigh their benefits.
4 What Do We Own?
The observations above assume one important thing: that we can always tell whose activities create
which results. And often this is indeed easy: when I build a factory and the stack begins to emit
smoke, the smoke is “mine” in the relevant sense. If anyone has to pay for producing the smoke,
that is, it’s me.
The same is true for many positive results of people’s activities. Consider John Locke’s famous
argument that people own the fruits of their labor:
Whatsoever then he removes out of the State that Nature hath provided, and left it in, he
hath mixed his Labour with, and joyned to it something that is his own, and thereby makes
it his Property.
(Locke, 1988 [1689], II, 27).
To Locke, the point is the same one. The benefits I produce with my labor belong to me. So, this
establishes my ownership, my having a private property right, over the fruits of my labor. Thus, the
soil I till, the crops I plant, all belong to me. Similarly, the book you write, the painting you create,
the company you build, all belong to you. These were your actions, so the decision about what
happens to their effects belongs to you as well.
Many commentators have focused on Locke’s claim about mixing in the above passage, won-
dering if it implies something like the following formula: “if you mix something you own with
something that is unowned, you thereby come to own the (previously) unowned thing.” But that
formula is false. As Robert Nozick points out, if I own a can of tomato juice and pour it into the
ocean, I do not acquire an ocean. I lose my juice (Nozick, 1974, 174–175).
Locke’s larger point does not depend on this particular formula, however. His larger point is
simply that we own what we produce through our labor. It’s because of this that, quite clearly, we
come to own fruits when we pick them, we come to own land if we till and develop the field, and
167
Bas van der Vossen
so on. (II, 32) When someone tries to take these things from us, Locke argues, “it is plain he desired
the benefit of another’s pains, which he had no right to” (Locke, 1988 [1689], II, 34).
Still, we might press on. Why is it that taking the fruits I have picked even counts as “the benefits
of my pains” in the first place? One possible answer would be “because you mixed your labor with
those fruits.” But that answer won’t do, as we’ve just seen. A better answer might go as follows:
these benefits would not exist but for the labor that you chose to perform. And when benefits would
not have existed (or at least not existed in this particular form) but for your labor, you will normally
have a claim to them. This idea, then, is what the Lockean theory sees as the basis of a property right.
We might wonder whether this principle is true. Are there counter-examples, examples like
Nozick’s tomato juice example, to it as well? The matter is up for debate, but even if this principle is
true, it still hides a lot more complexity. For we can imagine cases where this principle, too, does not
generate clear answers.
Even some of the cases Locke thought were clear cases raise questions. Consider the following:
“the Hare that any one is hunting, is thought his who pursues her during the Chase. For
being a Beast that is still looked upon as common, and no Man’s private Possession;
whoever has imploy’d so much labour about any of that kind, as to find and pursue her, has
thereby removed her from the state of Nature, wherein she was common, and hath begun a
Property. (Locke, 1988 [1689], II, 30)
Locke thinks the case is an easy one. And at the time, the common law considered the matter clear.
Yet, a little over a century later, the case would become controversial. In the famous case Pierson v
Post (3 Cai. R. 175, 2 Am. Dec. 264, 1805), Post was hunting a fox. Post chased the fox and, as he
was about to capture the animal, an interloper (Pierson) appeared, killed the fox, and ran off with the
carcass. Post sued on the argument Locke seems to endorse, namely that his pursuit established his
property right to the fox. That was, we might say, the point at which the labor was “mixed” with
the fox, so to speak.
The local court ruled in favor of Post, following the common law tradition (which Locke is
invoking in the passage above). Pierson, however, appealed and subsequently got the ruling re-
versed. In the end, the court’s majority decided that the one who killed, captured, or wounded the
animal mortally was the owner. The court argued that this was because such acts bring the animal
within “certain control.” And such control is the beginning of ownership, the argument went.
Perhaps you agree with one side or the other. But it’s at the very least not clear that Post or Pierson
was in the right. Locke thought that pursuit is the point of labor. But it’s not ridiculous to think that
“certain control” is the point where value is created, or labor is mixed with the animal. The theory,
at least in this application, just seems not that clear.
It’s notable here that the dissenting Judge Livingston (in the appeal) pointed out that the “certain
control” rule would discourage hunting and encourage the behavior of a “saucy intruder” like
Pierson. And this seems a good point. But note that this is not a point about who really mixed their
labor, at least if we understand that phrase in the literal sense invoked by Nozick’s example above.
Judge Livingston’s point was that a good rule of property should encourage productive activities
(like, in his view, hunting), and that is the way we figure out who gets ownership. This is not a
question about who’s act really is more connected to the fox or what really counts as mixing one’s
labor. This is a more complicated social or legal question – one that asks which rules might help us
live together more safely and prosperously.
Perhaps the most plausible interpretation of Locke’s position, then, is that he’s making a point
about how we should treat people’s labor if we really want to live together well. He is arguing that
when people engage in actions that create value for themselves and others, we better consider them
good candidates for holding property rights. And when people lay around to grab what others
168
Property
create, we should not reward them in that manner. Only such a rule would actually encourage
people to continue doing beneficial things.3
Such a rule also comports well with the way economists typically answer the question of when
property rights appear in the first place. After all, societies do not apply private property to every
resource around. It’s physically possible, for example, to privately own fresh air by putting it in tanks
and selling them. Yet we don’t have a market in fresh air. We do have markets in land, and many
other things, however. What might explain the difference?
In a seminal article, Harold Demsetz (1967) argued that the difference can be understood by
comparing the costs that property systems pose for communities with the costs that their absence or
alternatives pose. Obviously, creating and maintaining a system of property rights is costly. It takes
time, effort, institutions, and more. Property rights are costly also in our day-to-day behavior. They
impose transaction costs – if I want to buy a piece of land, say, I must first find out who owns it,
negotiate, draw up a contract, and so on. Because of these costs, communities won’t create property
systems unless the costs are worth it, argued Demsetz.
The most obvious way these costs can be worth it is if the alternatives are worse. And the absence of
private property can lead to terrible outcomes. The insight goes back to at least David Hume (1978
[1739]). A world without property, Hume argued, has three problems. First, resources are scarce,
meaning nature doesn’t provide enough for everyone to be happy. Second, we’re not naturally
motivated to work for others quite as hard as we’re willing to work for ourselves. So, while we could
overcome this scarcity by working and producing enough for everyone else, we don’t want to do this if
we do not stand to also gain from it ourselves. And third, stuff can be moved around. If we do produce
enough to overcome scarcity, others might take it from us for their own benefit.
These three elements combine to create a problem: because we all want more than is naturally
available, but we can all serve our needs by taking what others have produced, no one will feel safe
to produce things. After all, as soon as your work is done, someone else might come and take the
proceeds. Under these conditions, human beings are practically unable to overcome the scarcity that
they’re facing.
Demsetz’s point, then, is that communities begin creating property rights when the costs of
maintaining and enforcing such rights (as well as deal with the transactions costs) are outweighed by
the benefits of ending circumstances like the ones Hume described.4 Those costs can include several
things in addition to the monetary costs of maintaining an institutional apparatus of courts, registers,
and so on. But they can also include the perceived costs of parceling land, say, into private lots
(Ellickson, 1993). Demsetz is thus inviting us to think of property as an institution, created to address
social problems.
This way of viewing property rights helps us understand why property systems may end up
creating externalities. The externalities we create become internalized when we own the benefits
and costs of our activities. This happens, in other words, by extending our property rights to include
all the effects of our activities. But since extending property rights in this way is itself costly, it’s an
open question whether these costs are worth it from the point of view of society as a whole. When
the costs are not worth it, the effects of our activities can remain unowned in the same way that
objects can remain unowned.
5 The Question of Inequality

Property rights are important. They are also controversial, especially in politics. While many think
that property is extremely important, for reasons such as those outlined in this piece, others worry
about life in a society organized around private ownership. Or at least they might worry about life in
a society organized around very strong rights of private ownership.
169
Bas van der Vossen
The most common worry, of course, concerns the material inequality that tends to characterize
such societies. When individual people are allowed to make consequential economic decisions in a
decentralized manner, the outcome distributions of material goods will not always look the way we
might like. In fact, as Nozick pointed out in another of his famous arguments, societies in which
individuals are allowed to make such decisions will end up disturbing whatever overall distribution
(or, in Nozick’s terms, “pattern”) of resources you might find desirable. People don’t make their
decisions with those outcomes in mind. And even if they did, they might not be able to bring about
an overall social pattern all by themselves. (Nozick, 1974, 163)
In other words, even egalitarian-minded societies will, if private ownership is allowed, end up
disturbing equality. The overall distributional results will not fit whatever your preferred view of
justice entails – and that is true irrespective of what your preferred view of justice-in-patterns might
be. The source of this disruption is private property and the freedom of economic choice it gives to
individual persons. To end such inequality, this decentralized system of property needs to be limited,
curtailed (or, at the limit, abolished altogether) in favor of redistribution. Such redistribution in-
variably occurs through taxation.
Earlier, however, we said that property rights are robustly immune to expropriation. As Locke
put it, “I truly have no Property in that, which another can by right take from me, when he pleases,
against my consent” (Locke, 1988 [1689], II, 138) Private property rights protect one from having
things taken away without one’s consent. And taxation does just that.
It seems, therefore, that we are facing a crossroads here. Tax-and-transfer schemes aimed at
fighting inequality straightforwardly conflict with robust property rights. They take from some to
give to others, and they do so without asking our permission. On the other hand, robust property
rights conflict with a desire for material equality. Protecting people in their possessions will lead to
outcomes that many find problematic. So, what is to be done?
Some people think this seeming conflict between property and equality is not a genuine one. As
we’ve seen, many countries that score very high on indexes of property rights protections also have
strongly redistributive governmental policies. Philosophically, too, people debate whether taxation
poses a conflict with property rights. There are several ways in which theorists have tried to square
respect for property with an endorsement of redistributive taxation.
One attempt is to argue that taxation does indeed conflict with property rights, but that the
conflict is not something we should be worried about. The importance of taxation is here said to
justify the overriding of what (ordinarily would be inviolable) property rights. Rights are not ab-
solute, this argument goes, and when they conflict with other important goals, rights can be justly
curtailed. Taxation then is said to be such a goal.
To be sure, we can imagine situations in which people’s property rights might be justly over-
ridden. For example, if I’m seriously wounded and need to cross your property to get to the hospital
in time, I can permissibly cross it even without your consent. Such a crossing would ordinarily
constitute wrongful trespass. But in this situation, the circumstances mean that my trespass is not
wrongful. Similarly, there might be cases in which the money you own might be justly taken from
you. Perhaps if the only way I can save my life is to hop in a cab, and I have no money on me to pay,
I may snatch the $20 bill from your hand and go.
But this is a problematic argument, as this kind of justification is difficult to extend to taxation in
general. The cases above involve life and death. And while perhaps other cases, not involving life and
death, might be conceivable as well, it’s plain that something morally very important is needed to
override people’s rights. But many (perhaps most) things that governments do with tax revenue simply
aren’t important enough. I cannot take your money just because without it I would have to take on
debt to pay my kid’s university tuition. I cannot take your money just because without it I couldn’t
afford to run an art museum. I cannot take your money to pursue foreign wars. And it’s not at all
170
Property
obvious that achieving some overall more desirable pattern of resource distribution can justify in-
fringing what ought to be other people’s rights.
One might respond here that the costs of being taxed to owners aren’t all that great, and so we
shouldn’t worry so much about taxation. Maybe it’s permissible to tax people because the harms it
imposes just aren’t very serious? But rights simply don’t work like that. When I see a beautiful
person standing in front of me while waiting in line, I might feel an extremely urgent need to softly
stroke their hair. And I might be able to do this without them even noticing. In other words, such
an act would impose zero cost or harm. Nevertheless, I am plainly not permitted to touch them
without their consent. Their rights block my doing so, even if not much harm is involved.
This example involves bodily rights. But the same is true for property rights. Suppose I know you
are going on a vacation. When you’re gone, I break into your house and remodel your kitchen to
your exact preferences (you showed me your plans right before you left). Again, I will have imposed
no significant costs on you. I probably even benefited you, adding tremendous value to your home.
Nevertheless, my actions are wrong. And they are wrong for a simple reason: i violated your rights.
Here is another way of seeing the same point. Cases like these (i.e., justified rights violations) are
typically characterized by a kind of moral residue that remains after the taking occurs. When I take
your money to get to the hospital, I might not wrong you in the same way a robber would. But after
I’m cured, I should probably pay you back. Or at the very least I owe you an apology. None of this
applies to taxation. Governments don’t think they need to pay us back, nor are they in the habit of
apologizing for taking our money.
Moreover, if this really were the justification of taxation, there should be no reason why the
government would be the only one with the right to tax. After all, the justification here depends on
the moral importance of that for the sake of which property rights are violated. But that means that,
at least in principle, anyone else could do the same with equal justice. And very few people think
that’s the truth about taxation.
If the argument that taxes are justified violations, but still violations, of people’s property rights
doesn’t work, we might consider a second approach. Perhaps the seeming conflict between re-
specting property and caring about material equality can be overcome in the following manner:
perhaps taxes really are not infringements of people’s rights at all?
This second option might seem surprising at first. After all, this would be just to say that taxes
don’t really take your money at all. And at least at first sight, that just seems plainly false. The money
surely was mine until it got taxed away. My employer paid me, not somebody else. And the money
was paid in return for the labor I, and not someone else, performed. At the very least, it seems clear
therefore that the money was mine.
Nevertheless, there are several philosophical arguments that aim to establish that taxes don’t really
take your money. I will here review two of these. The first of these has gotten some traction in
recent years. In an influential book, Murphy and Nagel (2004) have argued that taxes really do not
violate people’s property rights. Stronger, their view holds that taxes simply cannot violate property
rights. To think otherwise, they claim, involves a conceptual mistake. It is to confuse property rights
for something that they are not.
This view of course seems initially pretty implausible. But Murphy and Nagel think we tend to
make a simple mistake here. When we receive payment for the goods we sell or salaries for the
services we provide, we think the money we receive thereby becomes ours, something we thereby
have a moral right to in some straightforward sense. It’s true that we have a legal right to our money
in the sense that other people cannot forcibly take it from us. But this legal right of course does not
extend to the government.
Nor does it follow that we have a moral right to our money that is claimable against the gov-
ernment. It cannot be said, they claim, that it’s morally wrong for the government to tax us because
“the money is ours.” Here’s the argument for that claim: the activities by which we make our
171
Bas van der Vossen
money all take place in an intricate institutional context, including laws, courts, and of course
property rights. This context is the product of government activity. And for this reason, Murphy and
Nagel claim, the property rights that are the product of the government cannot be claimed against
the government.
The upshot is that taxation does not really involve the taking of something that belongs to you.
It’s merely the rearranging of allocations by a government agency responsible for that task. And since
all our property is only such an allocation, we also cannot claim that the reallocation via taxation
violates our rights. When the IRS sends us a bill, this is more like a public accounting exercise – the
balancing of society’s books – than taking of what’s ours.
It’s not quite clear why we should believe Murphy and Nagel when they assert that we cannot
claim property rights against the government because property rights are part of a wider institutional
context. Of course, they are part of a wider context. But so too are our other rights. And those other
rights apparently can be claimed against the government. So it’s not clear why we should move from
this observation about context to a conclusion about taxation.
One possible argument in defense of this move relies on the following principle: if the current
scheme of property rights would not exist in their current form without the government and the
taxes that support it, then those rights cannot be claimable against the government. But that
principle is simply unacceptable. Again, all our rights would not exist in their current form without
government. And so this principle entails that we cannot any rights against the government. And
this, quite clearly, is false. (It’s also not the position Murphy and Nagel themselves defend.)
Consider applying Murphy and Nagel’s principle to our rights over our bodies. Why can’t the
government set up a corvée-style scheme – a tax paid by labor, not money? Applying the principle
above, it seems we own our pre-corvée time no more than we own our pre-tax income. After all,
without a government, we wouldn’t nearly have the same kinds of lives (or quality of life) as we
have with it. And the rights over our bodies are no less defined through institutional mechanisms of
the state than the rights over our (other) possessions. It follows, then, that we have no rights over
precorvée time. (Similar arguments could be made for other core liberal rights.)
Murphy and Nagel’s argument thus seems to justify not only, say, a 35% tax rate on our earnings.
It would also justify a regime that would force us to labor for 35% of our time for the government’s
or society’s benefit. But we really shouldn’t accept that this argument is a good reason to think that
governments can force us to labor like that. To accept that would be to deny the basic protections
that are the building blocks of free and prosperous societies. And it’s unclear why things should be
different for property rights.
Murphy and Nagel try to draw a distinction between property rights and rights over our bodies.
They write: “Egalitarian liberals simply see no moral similarity between the right to speak one’s mind,
to practice one’s religion … and the right to enter into a labor contract … unencumbered by a tax
bite” (Murphy and Nagel, 2004, 65). They say that some liberties are at “the core of the self ” and must
be protected against the state; others are not at the core. But this simply gives up on the argument
about the nature of property rights. It’s another way of simply saying that property rights are not
morally important enough to block taxation. And that argument we’ve already discussed above. (For
more detailed discussion, see Brennan and Van der Vossen, 2019.)
A different defense of Murphy and Nagel’s claim asserts that property rights are conventional,
where this is taken to mean that such rights depend on government legislation. And this dependence
is said to make these rights incapable of being claimed against the government on which they
depend. The same might not be true of bodily rights, and thus the objections above could be
avoided. But such a move is no less troublesome.
For one, it’s not clear why (a) rights being conventional would entail (b) that rights cannot be
claimed against the government. To label rights conventional is presumably to say something about
their genesis. It is to say, roughly, that these are something humanly created, not something naturally
172
Property
part of our moral fabric. But even if this is true of property rights, that says nothing about their being
or not being claimable against the government. Humanly created rights can be perfectly claimable
against the government. (Consider, for example, rights of due process.)
Nor does labeling rights as conventional entail that they depend on government institutions or
legislation. Conventions are social systems of norms, but such norms need not involve the law. And
as the history of the common law suggests, it is perfectly possible for social conventions about rights
to constrain governments and the law. Common law rights are part of social conventions of rules
and norms that developed independently of direct government legislation, and courts have re-
cognized these as applying to citizens and governments alike.
The idea that property rights are conventional is supposed to justify Murphy and Nagel’s con-
clusions in the following way. Conventional rights bind only those who are party to the convention,
or members of the community within which the convention exists (cf. Freeman, 1991, 45). And the
government stands outside of (above?) the convention. As a result, governments enjoy the freedom
to not respect its subjects’ property rights, even though those subjects themselves lack that freedom.
But this claim is no less problematic. Even if we grant that it allows governments to ignore
people’s rights over their possessions, this thesis also allows others who are not party to the con-
vention to ignore such rights. This includes foreign individuals and other governments. But the idea
that foreign individuals and governments are not morally required to respect people’s property rights
is highly troubling. For example, it makes it difficult to condemn as unjust colonial forces ex-
propriating native people’s possessions. If this thesis were true, those colonial forces could simply
point out that they, as outsiders, are no more bound by conventionalist property rights than the
native governments and proceed to expropriate the native people with impunity. But plainly such
expropriation constitutes a massive and severe injustice. And the injustice lies in the violation of the
native people’s property rights (among other things).
The thought that property rights have a kind of (moral) validity that applies to those within and
without whatever convention they might be a part of is therefore simply unavoidable. But this
thought just contradicts the conventionalist argument for Murphy and Nagel’s position above.
Yet more arguments for taxation might be conceived along these lines, of course. Perhaps rights
in general, including all other liberal rights, ought to be understood as depending on their relative
costs and benefits for society, and this allows for taxation in the case of property but not, say, in the
case of rights over bodies. Some theorists, such as F. A. Hayek (1960) for example, suggest that it can
be permissible for people to be forced to pay for basic police services, but not other things. Perhaps
such arguments can be sustained, but they would need a detailed defense of their own. And they
would need to avoid problems like the ones we have just seen plague the influential proposal from
Murphy and Nagel. They would need, in other words, to plausibly weaken our property rights
against taxation without at the same time weakening other important rights as well. Such arguments
are difficult to imagine.
A different view arguing that taxation does not really violate property rights is more philoso-
phically promising. This view is often called left-libertarianism. It does not hold that taxes cannot
violate property rights. Rather, it holds that some taxes take from people possessions that aren’t
legitimately theirs. Different left-libertarians get to this position in different ways. But one influential
line of argument makes a distinction between two factors that go into the things we make or earn.
One of these is the labor that goes into producing things. The other consists of the natural resources
that went into production. Most, if not all, products contain both. A table has to be made by
workers, but it also consists of wood, metal nails, and so on. Left-libertarians insist that all of us have
an equal right to the world’s natural resources (or their equivalent value). But we don’t all use equal
parts of the total set of natural resources. Some people produce a lot, and thus use a lot of resources,
others much less. In those cases, it’s fair to tax the former, and redistribute to the latter, the value of
173
Bas van der Vossen
the natural resources they used in excess of their equal share. (The best version of this argument
remains Steiner, 1994. A famous, much earlier statement is George, 1879. For a different, related
argument see Otsuka, 2003.)
Much here depends, of course, on the claim that people have an equal right to the world’s natural
resources. It’s common here for left-libertarians (e.g., Steiner, 1994) to claim that this is an implication
of Locke’s famous assertion that property rights must leave “enough and as good” for others in
common (Locke, 1988 [1689], II, 27). As an interpretative matter, this is not very plausible, however,
as Locke is simply not concerned with arguing for equal distributive shares (Van der Vossen, 2021).
Others claim intuitive support for the egalitarian claim. Otsuka (2018), for example, claims that if
two persons are stranded together on an island, equal division would be clearly the intuitively fair
solution. But such egalitarian intuitions typically apply only when we ignore relevant circumstances.
For instance, suppose that one person had arrived earlier to the island, and already cultivated, say,
two-thirds of the island, while leaving more than enough for the second person to independently
make a living, is willing to cooperate, trade, and so on. In that case, a latecomer insisting that she has
a right to half the island is not only counter-intuitive, but probably just wrong. (The intuition of
equal division becomes even less appealing if we imagine more than two parties, capable of pro-
duction, trade, and cooperation, arriving at different times.)
In any case, even if the left-libertarian argument succeeds, it justifies only some taxes. Left-
libertarians hold that taxes are just if they reflect excessive takings of natural resources. But they see
taxes on people’s labor as violations of our rights, indeed violations of our property rights. Such taxes
are, therefore, and for that reason, unjust. This is a virtue of the view, as it maintains the intuitively
very plausible judgment that governments can violate our rights. However, it does not license taxing
people merely to achieve a more equal distribution of resources. What the best distribution looks
like depends on people’s productive activities. To that extent, the left-libertarian position will share
some of the worries that come with decentralizing economic decisions in general.
In the end, there seems to be no way out of the tension with which we began. On the one hand,
property rights ought to protect our possessions from expropriation. That protection applies to ex-
propriation by other people. And there is no reason to think that it should not also apply to ex-
propriation by governments. Arguments designed to show that the latter kind of expropriation is
somehow not as troublesome as the former are difficult to sustain. And at least one influential version of
this plainly fails.
On the other hand, claiming that redistributive taxation violates property rights comes at a high
price. If redistributive taxation is impermissible because it constitutes a violation of individual
property rights, then this may well be true for all kinds of taxation. In that case, governments should
not only stop trying to mitigate inequality in our societies but effectively stop doing anything at all.
That, to put it mildly, would be an implausible conclusion as well.
What we’re left with, then, is simply a conflict that needs further thought. We lack a good
solution to this problem. Property rights need not be fully immune to expropriation. So, taxation
might be possible without necessarily denying people’s rights, but rights mean little if they’re not at
least robustly immune to such expropriation. And we lack a good way of understanding why and
when expropriation might be acceptable that does not also threaten other important rights at the
same time. The conclusion is an unsatisfying one: there is still more work to be done.
Notes
1 Immunity of expropriation is not uniquely important to property rights, however. It’s a key feature of many
rights. After all, any right is not securely held unless it’s protected by such immunity. In the absence of
immunity, others could simply remove our rights unilaterally – leaving us effectively as vulnerable as we
would be without rights. This, too, fits Locke’s view, who often uses the term “property” to refer to the
total set of rights a person holds. (For discussion, see Van der Vossen, 2021.)
174
Property
2 It’s worth noting that this need not be mere selfish behavior. Absent some mechanism for imposing the cost
of pollution on producers, they might not even know how much disvalue their pollution creates.
3 This principle accords well with the arguments we’ve discussed above. Letting people own the fruits of their
labors is a good way of incentivizing productivity. And Locke indeed makes much of the beneficial and
productivity-enhancing nature of labor in general. He argues at length that it’s the engine of prosperity and
wealth (II, 40–43).
4 It’s an implication of this analysis that, should air become so polluted as to become worth storing and selling
in tanks, people will start privately appropriating air.
References
Acemoglu, Daron, Johnson, Simon, and Robinson, James A. (2005) “Institutions as a fundamental cause of long‐run
growth.” Handbook of Economic Growth, vol. 1, Part A, (Edited by Philippe Aghion, Steven N. Durlauf),
pp. 385–472.
Attas, Daniel. (2006) “Fragmenting Property.” Law and Philosophy 25, 1, 119–149.
Bergh, Andreas and Nilsson, Therese. (2014) “Is Globalization Reducing Absolute Poverty?” World
Development 62, 42–61.
Brennan, Jason and Van der Vossen, Bas. (2019) “The Myths of the Self-Ownership Thesis.” In Jason Brennan, Bas
van der Vossen, David Schmidtz eds. The Routledge Handbook of Libertarianism. New York; Routledge Press.
Demsetz, Harold. (1967) “Toward a Theory of Property Rights.” The American Economic Review 57, 347–359.
Dollar, David and Kraay, Aart. (2002) “Growth Is Good for the Poor.” Journal of Economic Growth 7, 195–225
Ellickson, Robert C. (1993) “Property in Land.” The Yale Law Journal 102, 1315–1400.
Freeman, Sam. (1991) “Property as an Institutional Convention in Hume’s Account of Justice.” Archiv für
Geschichte der Philosophie 73, 1991, 20–49.
George, Henry. (1879) Progress and Poverty. 5th edition. New York; D. Appleton and Company; reprinted by
Robert Schalkenbach Foundation, 1966.
Grey, Thomas. (1980). “The disintegration of property.” Nomos, 22, 69–8.
Gwartney, James, Robert Lawson, Joshua Hall, Niclas Berggren, Fred McMahon, Ryan Murphy, and Therese
Nilsson. (2020) Economic Freedom of the World. Fraser Institute. Available at: https://www.fraserinstitute.org/
sites/default/files/economic-freedom-of-the-world-2020.pdf
Honoré, Anthony. (1961) “Ownership.” in A. Guest ed. Oxford Essays in Jurisprudence. London; Oxford
University Press, pp. 107–147.
Hume, David. (1978 [1739]) A Treatise of Human Nature. L. A. Selby–Bigge and P. H. Nidditch eds. Oxford;
Clarendon Press.
Hayek, F. A. (1960) The Constitution of Liberty. Chicago, IL; University of Chicago Press.
Locke, John. (1988) [1689]. Two Treatises of Government. Laslett, P. ed. Cambridge; Cambridge University Press.
Murphy, Liam, and Nagel, Thomas. (2004) The Myth of Ownership. Oxford; Oxford University Press.
Nozick, Robert. (1974) Anarchy, State and Utopia. New York; Basic Books.
Otsuka, M. (2003) Libertarianism without Inequality. Oxford; Clarendon Press.
Otuska, M. (2018) “Appropriating Lockean Appropriation on Behalf of Equality.” In James Penner and
Michael Otsuka eds. Property Theory: Legal and Political Perspectives. Cambridge; Cambridge University Press,
pp. 121–137.
Schmidtz, David. (2011) “Property and justice.” Social Philosophy and Policy, 27, 1.
Steiner, H. (1994) An Essay on Rights. Cambridge, MA; Blackwell Publishers.
Van der Vossen, Bas. (2021) “As Good As ‘Enough and As Good’.” The Philosophical Quarterly 71, 1, 183–203.
175
14
CORPORATIONS IN OUR POLITY
Amy J. Sepinwall
When the Supreme Court issued its decision in Citizens United v. Federal Election Commission in
January 2010, political commentators and concerned citizens cried foul. Citizens United is, of course,
the landmark case holding that corporations enjoy the same rights as individuals to spend as much
money as they would like on ads supporting or opposing candidates for political office. Prior to the
decision, corporations were not permitted to spend any money on political speech. Their newfound
entitlement threatened to make the problem of money in politics magnitudes worse. After all, not
only is the average publicly traded corporation far wealthier than the average individual,1 cor-
porations also have the added benefit of immortality, which could allow them to accumulate wealth
over many human lifespans (Beard, 1936, 345; Torres-Spelliscy, 2018).
Much of the public outcry targeted the insidious effects of this new fount of political spending.
But much attention was also directed to the larger significance of the decision: in the public ima-
gination at least, the Court had anointed corporations as people (Bowie 2019, 2010; Chemerinsky,
2010, 54; Ripken, 2011, 212–13). The internet exploded with memes ridiculing the supposed
conflation of human individuals with inanimate, single-mindedly profit-driven corporations. A New
Yorker cartoon featured a corporate lawyer beseeching a panel of judges, “If you prick a corporation,
does it not bleed?” (Sipress, 2011) One woman offered an internet post entitled, “Ten Rules for
Dating a Corporation” and another staged a press conference so that she could announce her search
for a corporate husband (Halper, 2010; Bainbridge, 2011). The comedian Stephen Colbert formed a
Super PAC, which “asked whether voters would be comfortable letting Mitt Romney date their
daughters’ corporations” (Totenberg, 2014). More seriously, the decision launched multiple in-
itiatives seeking statutory declarations that corporations were not people, or even a constitutional
amendment to that effect.2
In point of fact, Citizens United nowhere declared that corporations were persons; nor did the
decision turn on a belief that corporations are persons (more on this below). Nonetheless, the
anxiety about the corporation’s ascendancy was not entirely misplaced. In the years since Citizens
United, corporations have gained rights to exercise religion (Burwell v. Hobby Lobby, 2014); violate
anti-discrimination laws on conscientious grounds (Masterpiece Cakeshop, Ltd. v. Colorado Civil Rights
Commission, 2018); and even claim recognition for having a discrete race.3 Corporations have also
played an enormously influential political role outside of election cycles. For example, North
Carolina jettisoned its infamous “bathroom bill,” which would have required individuals to use the
bathroom corresponding to the sex they were assigned at birth, in response to a corporate boycott
( Jenkins and Trotta, 2017). And social media corporations exercise ever-increasing power over the
176 DOI: 10.4324/9780367808983-17

Corporations in Our Polity
content of our political discourse, as demonstrated powerfully by Twitter’s permanent expulsion of

Donald Trump (Twitter 2021).
How should we think about corporations’ expanding legal rights and growing political influence?
This chapter aims to answer that question across three different contexts where corporate rights or
power threaten central democratic values: corporations and electoral politics; corporations and re-
ligious freedom; and corporations and political protest.
1 Corporations and Electoral Politics

When Citizens United struck down the ban on corporate political spending, it repudiated a com-
mitment to keeping corporations out of politics that had reigned for over 100 years. In 1907,
Congress passed the Tillman Act, which made it a criminal offense to use corporate funds for
campaign contributions (US Congress). Some 40 years later, Congress extended the prohibition to
independent political expenditures – that is, money spent on one’s own speech supporting or op-
posing candidates for office, without coordination with any political campaign (Bowie, 2019).
When, in 1971, Congress undertook what at the time was its most extensive effort to reform
campaign finance, passing the Federal Election Campaign Act, it left in place the bans on corporate
campaign contributions and corporate independent expenditures. So too did the 2003 Bipartisan
Campaign Reform Act (BCRA, aka McCain-Feingold) (Citizens United, 434–41).
As of the 1970s, however, the Supreme Court began to weaken the ban on corporate in-
dependent expenditures. In 1978, the Court ruled that a Massachusetts law criminalizing corporate
spending in support of or opposition to a state ballot referendum violated corporations’ First
Amendment rights. It did so because it rejected the proposition that “speech that otherwise would
be within the protection of the First Amendment loses that protection simply because its source is a
corporation” (First National Bank of Boston v. Bellotti 1978).
Still, the Court’s path to Citizens United was not a straight line. In 1990, in Austin v. Michigan
Chamber of Commerce, the Court upheld a Michigan law that banned corporations from using their
own money on speech supporting or opposing candidates for office. There, the Court reasoned that
the Michigan legislature had a compelling justification for the ban: it aimed to prevent “the cor-
rosive and distorting effects of immense aggregations of wealth that are accumulated with the help of
the corporate form and that have little or no correlation to the public’s support for the corporation’s
political ideas” (Austin v. Michigan Chamber of Commerce, 1990, 660). The Court’s holding was
reaffirmed in 2003 in McConnell v. FEC, when it sustained a similar ban in the BCRA.
By the time of Citizens United, the Court took itself to be subject to two conflicting lines of
precedent – the Austin line, which held that the bans on corporate political spending were con-
stitutional, and the Bellotti line, which held that they were unconstitutional (Citizens United, 348).
One might have thought that the first of these two lines was straightforwardly more authoritative.
For one thing, the Austin line was more recent and had been reaffirmed, whereas no subsequent case
had explicitly reaffirmed Bellotti. Moreover, there was an easy way to reconcile the two: the question
in Bellotti was whether corporations could speak to particular referendum issues; the question in
Austin and McConnell was whether they could influence general candidate elections. So the two lines
of precedent were not in fact in direct conflict.
Nonetheless, in a striking turn, the Supreme Court in Citizens United overruled Austin and
McConnell to hold that, thereafter, corporations would enjoy the same rights as individuals to spend
unlimited amounts of money on independent political speech. The Court, in an opinion that five
Justices joined, based its holding on two rationales. First, speaker identity is immaterial under the
First Amendment: individuals should enjoy the same free speech protections whether they speak in
their individual voices or through the corporate form. Second, because listeners have a right to hear
177
Amy J. Sepinwall
as much speech as might be on offer, there too the state could not justify treating corporate speakers
differently from individual speakers.
As already suggested, the decision unleashed a torrent of criticism. Curiously, though, much of it
was aimed at a false target: critics lambasted the Court’s recognition of corporate personhood. They
launched the “Movement to Amend,” which sought to add a provision to the Constitution denying
that corporations were persons (Kavoussi, 2012). Voters in cities across the country passed resolu-
tions declaring that “corporations are not human beings” (Bowie, 2019). Memes attacking the
Court’s (supposed) conflation of corporations and persons abounded. Yet the Court’s opinion
nowhere declared that corporations were persons; nor did it rely on that proposition.4 And at any
rate, any suggestion in the opinion that corporations were persons would hardly have been re-
volutionary: the Supreme Court had summarily declared that they were over 120 years earlier, in
Santa Clara County v. Southern Pacific R. Co (1886).
Santa Clara involved a railroad challenge to corporate tax rules that prohibited corporations from
deducting mortgage payments on their tax returns even while individuals could make these de-
ductions. The railroad argued that the differential treatment violated the Equal Protection and Due
Process clauses of the Constitution. But that argument could succeed only if corporations were
persons, as the two constitutional clauses applied only to persons. As such, the railroad had devoted
much energy to establishing the truth of corporate personhood in its arguments in the courts below
(Winkler, 2018, 144–45). Before oral argument in the Supreme Court even began, Chief Justice
Waite made what turned out to be a momentous announcement:
The Court does not wish to hear argument on the question whether the provision in the
Fourteenth Amendment to the Constitution which forbids a state to deny to any person
within its jurisdiction the equal protection of the laws applies to these corporations. We are
all of opinion that it does. (Santa Clara, 397)
Ironically, the announcement was meant to dismiss the question of corporate personhood, not to
resolve it (see Winkler, 148). Waite believed that the Court could decide the case on narrow
grounds that would allow the Court to avoid weighing in on whether corporations were persons.
Sure enough, the Court determined that the tax assessment had been miscalculated, and it in-
validated it on that ground. But Waite’s off-handed remark ended up in a headnote for the decision,
and it has been taken to stand for the Court’s view of the matter forever after. As a result, corporate
personhood was settled not through argument but by fiat. And so Citizens United did not need to
establish that corporations were persons; Supreme Court jurisprudence already (improvidently)
granted that they were.
None of this is to suggest that the consternation around Citizens United has been groundless; it is
instead to suggest that it has been misdirected, and that is so for two reasons. First, personhood would not
be sufficient to establish that corporations ought to enjoy the same political speech rights as individuals.
Foreign individuals are undoubtedly persons. Yet we might well question whether they ought to have
the same ability to influence political matters through their speech, or speech expenditures, as individual
citizens do. As it stands, under the Federal Election Campaign Act (FECA), foreign nationals are
prohibited from “making contributions or expenditures in U.S. elections” (Garrett, 2019). And we can
see that the threat of foreign influence is real in light of its effect on the 2016 election.5
Our polity might privilege the speech of citizens – and accordingly impose more spending re-
strictions on the speech of non-citizens – because citizens bear unique associative obligations.
Perhaps most notably, citizens enjoy both the right, but also the duty, to vote. Further, citizens – and
only citizens – can be called upon to serve in the military (and perhaps sacrifice their lives as a result).
One could then argue that the obligations of citizenship go hand-in-hand with robust political
participation, and they provide reason to have citizens’ voices largely predominate in political
178
discourse. In short, corporate personhood might be a red herring, for what really matters when
determining the scope and strength of rights of political participation might not be personhood but
civic status instead.
A second reason to question whether the reaction to Citizens United aims at the right target arises
because of the immense influence corporations exert directly over politicians through their lobbying
activities. In 2018, lobbying spending reached $3.4 billion (Evers-Hillstorm, 2019). This type of
spending yields results. For example, one study found that when corporate lobbying expenditures
increase by 1%, a 0.5–1.6% lower effective tax rate follows for the average firm that lobbies (Richter
et al., 2009).
In sum, the focus on corporate spending in election cycles distracts us from far more insidious
dynamics in the interplay between money and political power: the real problem with current
campaign finance law is that it allows wealth – whether corporate or individual – to exert undue
influence over election outcomes. To the extent that corporate wealth, in particular, is problematic,
we ought to worry about corporate lobbying at least as much as corporate support for political
candidates. And at the same time, the idea of corporate political or constitutional rights might well
have merit – at least where recognizing these rights is necessary to ensuring the full exercise of the
rights of the individuals constituting the corporation.
2 Corporations and Religious Freedom

Once corporations won the right to speak freely (or expensively!) on political matters, it was perhaps
but a short stretch for them to claim – and eventually win – rights of religious freedom. The
landmark case, Burwell v. Hobby Lobby, involved a for-profit corporation owned by a single five-
member family – the Greens – but employing 13,000 people at the time of the decision (2014, 702).
The case sought to challenge a key provision in the Affordable Care Act (ACA) that required
employers to subsidize all twenty FDA-approved methods of contraception through their employee
health insurance plan. The Greens, practicing Christians, objected to four of these forms of con-
traception, all of which, they believed, operated in ways that posed a risk of embryo destruction
(703). Their objection prompted them to believe that they would individually be complicit in a
wrong were the corporation they owned to subsidize these forms of contraception for their employees
and the employees’ dependents.6
Federal law provides protection for individuals’ religious freedom through two modalities – the
First Amendment’s Free Exercise clause and the Religious Freedom Restoration Act (RFRA). In
1990, the Supreme Court ruled in Employment Division v. Smith that the First Amendment would no
longer allow for religious exemptions from a general law of neutral application (i.e., one that did not
single out religion by design or effect). The decision prompted such outcry that Congress, in an
unusual show of unity, passed RFRA in order to restore through statute the constitutional pro-
tections that Smith had withdrawn.7 Under RFRA, a religious adherent can seek an exemption from
any federal law that substantially burdens their religious freedom unless the government can establish
that the law is necessary to serve a compelling state interest (US Congress, 1993). The Green family
challenged the ACA’s contraceptive mandate on just this ground. Because the Obama
Administration had already exempted religious non-profit organizations from the mandate, it could
not establish that the mandate was necessary to serve a compelling state interest. Hobby Lobby
became the first for-profit corporation to enjoy rights to religious freedom.
Just as in Citizens United, the decision unleashed criticism about the Court’s having conflated the
rights of individuals and corporations. In Hobby Lobby, however, the now-protected capacity of the
corporation seemed even more fantastic. It is not unusual for corporations to speak – they hold press
conferences, mount advertising campaigns, testify before Congress, etc. But can a corporation be-
lieve in a deity? Engage in religious ritual? Care for the purity of its “soul” (whatever that might be)?
179
Amy J. Sepinwall
While a minority of corporate law scholars argued that a for-profit corporation could be religious
(Meese and Oman, 2014), the view of religious exercise this involved was flawed.8 The Court
rightly rested its holding not on the corporation’s capacities but instead on the rights of its owners,
which, the Court insisted, warranted protection through the corporate form.9 In so doing, the
Court declined to adopt the arguments that some of the nation’s most eminent corporate law
scholars had pressed upon it. These scholars argued directly against the claim that the owners were
entitled to have their own religious freedom rights grafted onto the corporation. Their principal
rationale? Limited liability (Katz and Dyer, 2014, 14).
The key legal innovation that the corporation affords is a separation between the business and its
owners: the owners are not liable for corporate wrongdoing because the corporation is an entity
separate from them. But if that is so in cases of wrongs, why should it not also be so in cases of rights?
The problem, as these scholars saw it, was that allowing the corporation to enjoy the rights of its
members, while allowing its members to disclaim the wrongs of the corporation, would be to allow
them to “have their corporate cake and eat it, too” (Winkler, 2014; Corbin, 2014).
In response, one could reasonably worry that these scholars sought to make too much of the
separation between the corporation and its owners.10 The fact that some of the corporation’s ob-
ligations, powers, and privileges are different from those of its owners need not entail that all must
be. There is nothing in the nature of a corporation, or in the nature of limited liability, that renders it
incompatible with treating the corporation as if it enjoyed rights of free exercise as a way of re-
specting the free exercise rights of its members. Corporations enjoy many rights that are grounded in
the rights of their members. Rights of free association, for example, can be ascribed to the cor-
poration in the first instance, but their purpose is not to allow the corporation to associate with
whomever (or whatever) it wishes but instead to respect the rights of the corporation’s members to
associate with one another (NAACP v. Alabama, 1958).
The idea that owners cannot have it both ways seems to presuppose that the nature of the
corporation is fixed, and that once the corporate veil is in place, it erects a complete separation
between the corporation and its owners. But just like the corporation, the veil is, to paraphrase John
Marshall, the first Chief Justice of the Supreme Court, “an artificial” construct, “existing only in
contemplation of law” (Trustees of Dartmouth College v. Woodward, 1819, 636). As such, the extent of
the veil’s coverage or porosity is a matter for the law to decide. If the owners cannot both limit their
liability and assert their constitutional rights through the corporate form, this cannot be because of
what the concept of the veil means or what the veil or a corporation is. In other words, there is no
conceptual or metaphysical truth about the veil or the corporation upon which opponents of
corporate religious exemptions can rely.
The foregoing is not meant to establish that Hobby Lobby was rightly decided. It is meant instead
to establish that, just as the “corporations are not people!” battle cry was a red herring in Citizens
United so too the “corporations are not religious!” battle cry is a red herring here. Whether cor-
porations should enjoy rights of religious freedom is not a matter that can be decided by appeal to
the “true” nature or transcendental purpose of the corporation; the corporation is an artifact whose
nature and purpose are up to us to determine. And we should make these determinations on the
basis of moral and political considerations – in particular, what are the individual interests at stake
and how can they be best protected (whether through law, social practice, etc.)? Looking to what
the corporation is, or can do, in its own right is a distraction.
What then of Hobby Lobby? Providing people with affordable access to healthcare is obviously a
preeminent obligation in a liberal state (Daniels, 1985). Instituting a system where most people
receive their health insurance through their places of work imposes numerous complications. Most
relevantly here, it conscripts individuals with moral or religious objections to some medical treat-
ments into subsidizing those treatments – morally, if not financially. The state could avoid conflicts
between the employer’s conscience and the employee’s health needs if only it offered a public
180
option; better still, if we moved to a single-payer system, à la Medicare for all. This is not the place
to defend such an arrangement. It is merely to notice that there would be no occasion for corporate
conscientious objections if the obligation to provide healthcare coverage were fulfilled by the state
in the first instance, rather than having it devolve onto private employers.
3 Corporations and Political Protest

Corporations have helped secure some of the landmark civil rights advances of the last decade.
Google’s “Legalize Love” campaign may well have contributed to Obergefell v. Hodges, the mo-
mentous Supreme Court decision finding that same-sex marriage was a constitutional right (Waddell
and National Journal, 2015). Paypal and other businesses launched a boycott to protest North
Carolina’s infamous “bathroom bill,” which prompted the state to roll back the law (Levin, 2019).
And corporations resoundingly promoted the Black Lives Matter movement in 2020, often lending
financial support to the cause too (Wellemeyer, 2020; Sepinwall, 2021). These are undoubtedly
admirable initiatives. But they pose a threat similar to the one Citizens United produced – perhaps to
an even graver degree.
Whereas Citizens United made it possible for corporations to influence the outcomes of candidate
elections, that influence would not directly translate into policy. The elected politicians would still
have to legislate in favor of corporate interests and win the support of a majority of their colleagues.
But the tactics just described have the corporation wield its voice and resources directly in support of
political change – oftentimes successfully, as we have just seen. Should corporations be permitted to
play this role in our polity? A few considerations should give us pause.
There is the obvious concern that the causes supported by corporations will undermine de-
mocratic equality. This is a concern with any form of collective action. We might celebrate every
campaign to boycott stores selling clothing made in sweatshops, but what then should we say of a
campaign to boycott companies because they support LGBTQ+ rights? (Reynolds, 2020) Because
the Supreme Court considers boycotts in support of a political cause to be speech of the “highest
value,” the state may not restrain these boycotts even if they express animus toward protected groups
or otherwise threaten the equal standing of all citizens (NAACP v. Claiborne Hardware, 1982, 913).
Yet boycotts are not only speech – they are also a form of market activity; in particular, they involve
refusals to deal. And so they can inflict significant economic harm, including harms that interfere
with settled democratic policy. For example, some commentators condemn the Movement to
Boycott, Divest and Sanction Israel (BDS) for Israel’s treatment of Palestinians in the West Bank
because the United States is committed to friendly relations with Israel. Those who support BDS,
the argument goes, threaten to interfere with that relationship, and thereby counteract or undermine
American foreign policy (Greendorfer, 2016, 118).
Whatever the risk that individuals engaging in a concerted boycott will quash the political ar-
rangements our elected officials have chosen in their best wisdom to pursue, it is greater still where the
boycotters are corporations. Whereas individual consumers can affect policy only if a significant number
of them join forces, the lone corporation might enjoy sufficient market share to effect policy change on
its own. And the direction of that change might be precisely opposite to the policies on which our duly
elected representatives have settled. This is just a version of a worry that Tom Christiano has articulated
about corporations’ making business decisions that undermine democratic policy (Christiano, 2010).
We should not think that the worry is any less serious when the corporation is motivated by political
rather than economic interests. The threat to democracy is the same.
A salient example of the threat emerged in the final days of President Trump’s administration,
where he arguably baited an insurrection using his social media accounts. In response, these sites
placed him “on mute” (Needleman and Wells, 2021). Most dramatically, Twitter instituted a
permanent ban, denying Trump the opportunity to communicate at all through any existing or
181
Amy J. Sepinwall
proxy accounts (Conger and Isaac, 2021). That these sites ought to have been permitted to take
down posts that would incite violence is not in question. But a permanent expulsion – a perpetual
boycott of sorts – raises questions about the role of for-profit corporations in deciding what speech
warrants a place in public discourse, and what speech deserves to be censored. The response might
be that other social media sites will emerge to provide a forum for the content (or the people)
Twitter, etc., condemns. Yet in the days following Twitter’s decision, the internet giants seemed to
arrive at a chilling convergence to silence Trump – with Facebook placing a two-week moratorium
on his account (ibid.), and Google refusing to make Parler, which was to have been the Trump-
friendly alternative to Twitter, available through its app store (Leswing, 2021). Given the de facto
scarcity of active social media sites, and the fact that the people who run them have not been chosen
through any kind of representative democratic procedure, it is especially concerning to have these
sites function as gatekeepers to the modern-day equivalent of the public square or town hall.
There is an additional reason that counsels against corporate political activism, which harkens
back to Citizens United and Hobby Lobby. For-profit corporations are not organized democratically.
They act in the world in ways that represent all their constituents – shareholders, managers, and
employees – even while they do not afford all these constituents a voice (let alone an equal voice)
(Bowie 2019, 2014). It was this feature that bred the concern in Citizens United that some share-
holders (to say nothing of employees, e.g., Harvard Law Review, 2014) might oppose the candidate
whom the corporation supports (Citizens United, 349). A similar concern was voiced in Hobby Lobby:
why should the five owners get to dictate the nature of the healthcare coverage tens of thousands of
employees and their family members receive?11 And the concern emerges too where corporations
wield their power to protest in support of policy change. Corporations recruit their members into
complicity with the political stance the corporation decides to take. It is not at all clear that this
arrangement is just given the hierarchical, undemocratic structure that most for-profit corporations
adopt (Anderson, 2017).
In short, corporate political protests threaten democracy in two ways. First, the causes cor-
porations pursue may run counter to settled governmental policies and, with their sometimes
outsized power, corporate protests might then undermine these policies. Second, most corporations
are not organized democratically, so they risk recruiting non-consenting workers into supporting
causes the workers oppose.
4 Conclusion
Many pro-business theorists champion market freedom because they imagine a well-functioning
state that will establish and enforce ground rules that will constrain business activity in ways con-
sonant with, or even supportive of, foundational democratic values (Friedman, 2002; Scherer and
Palazzo, 2007; Isaacs, 2014). This chapter has aimed to reveal that it may be naïve to rely on the state
to curb corporate overreach. This is because corporations have infiltrated the very elements of the
basic structure that were supposed to rein them in: They exert undue influence over candidate
elections as a result of Citizens United, thereby helping to select the people who will go on to
“regulate” them. They have leveraged the responsibilities the government has devolved onto them,
like healthcare coverage, to enlarge their control over workers’ private lives. And, under the guise of
constitutionally protected speech, they are permitted to wield their market power to achieve policy
outcomes that counter the ones our elected representatives have sought to enact. Progressive
commentators seek to critique these developments by likening the corporation to a “private gov-
ernment” (Anderson, 2017). But we might just as readily ask whether these developments under-
mine the public/private distinction. Perhaps corporations have amassed so much political power that
we ought to refer to them as “governments” tout court.
182
Notes
1 As of Dec. 31, 2020, the median corporation in the S&P 500 had a market capitalization of $26 billion
(Financial Knowledge and Information Portal, 2020. There are roughly 3530 publicly traded corporations
on U.S. markets; those in the S&P 500 rank in the top 14% (Krantz, 2020). To qualify for the “1%” in 2020,
a household need have earned no more than $531,000 (PK, 2021).
2 See, e.g., Sepinwall (2012, 581 n. 15) (referencing a group of attorneys and public servants advocating a
constitutional amendment for the purpose of invalidating Citizens United).
3 As the Court of Appeals for the Fourth Circuit wrote in Carnell Construction Corporation v. Danville
Redevelopment and Housing Authority, “We agree … that a minority-owned corporation may establish an
‘imputed racial identity’ for purposes of demonstrating standing to bring a claim of race discrimination under
federal law” (2014, 716).
4 If anything, the Court in several places explicitly distinguishes between “persons” and corporations It cites
the text of FECA imposing spending limits on “any person or group” (Citizens United, 339 italics added) and
recognizes that corporations are not “natural persons” (343). It also states that “All the provisions of the Bill
of Rights set forth the rights of individual men and women …. But the individual person’s right to speak
includes the right to speak in association with other individual persons” (391–92). It is precisely because the
Court never discusses whether corporations are persons, let alone anoints them with the banner of per-
sonhood, that Justice Stevens could bemoan that, while “[t]he fact that corporations are different from
human beings might seem to need no elaboration, [] that the majority opinion almost completely elides
it” (465).
5 Russia aimed to influence the election by posting fake news stories on social media, buying political ads, and
staging political rallies (Masters, 2018). Citizens United has facilitated this interference by providing an end-
run around ban on foreign political expenditures where American corporations are controlled by foreigners
(Vanderwalker and Lawrence, 2018).
6 The other family joining the Greens in challenging the contraceptive mandate described their concern in
similar terms: it is “against [their] moral conviction to be involved in the termination of human life after
conception, which they believe is a sin against God to which they are held accountable” (701, internal
quotation marks omitted).
7 The Senate vote in favor of RFRA’s passage was 97–3. (US Senate, 1993).
8 Briefly, because corporations have no soul, and no capacity to revere, feel guilt, or fear divine retribution,
they cannot be religious in any meaningful sense. The most they can do is to act in ways that line up with
religious observance; they cannot be observant, or spiritually guided by religious dictates. I argue as much at
greater length in Sepinwall, 2015, 173, 186–89.
9 This is not an implausible claim. For example, in NAACP v. Alabama, the Court upheld the NAACP’s
refusal to release its membership rolls to Alabama state officials by recognizing the NAACP’s rights of
freedom of association – not so that the group could freely associate with other groups but in order to
protect the free association rights of its members (1958).
10 This paragraph and the next are taken from Sepinwall, 2015.
11 Justice Ginsburg expressed this concern pointedly when she wrote, “The exemption sought by Hobby
Lobby … would override significant interests of the corporations’ employees and covered dependents. It
would deny legions of women who do not hold their employers’ beliefs access to contraceptive coverage
that the ACA would otherwise secure” (745–46).
References
Anderson, Elizabeth. (2017) Private Government: How Employers Rule Our Lives (and Why We Don’t Talk about It).
Princeton, NJ; Princeton University Press.
Austin v. Michigan Chamber of Commerce. (1990) 494 U.S. 652. Supreme Court.
Bainbridge, Stephen. (2011) “Citizens United and the Constitutional Rights of Corporations,”
Professorbainbridge.com (January 18, 2011).
Beard, Charles A. (1936) “Corporations and Natural Rights,” Virginia Quarterly Review 12.
Bowie, Nikolas. (2019) “Corporate Personhood v. Corporate Statehood,” Harvard Law Review 132.
Burwell v. Hobby Lobby. (2014) 573 U.S. 682. Supreme Court.
Carnell Construction Corporation v. Danville Redevelopment and Housing Authority. (2014) 745 F.3d. 703,
716 (4th Circuit).
Chemerinsky, Erwin. (2010) “The Most Important Decision of the Term.” Trial, May 2010.
183
Amy J. Sepinwall
Christiano, Tom. (2010) “The Uneasy Relationship Between Democracy and Capitalism.” Social Philosophy and
Policy 27.
Citizens United v. Fed. Election Commission. (2010) 558 U.S. 310. Supreme Court.
Conger, Kate and Mike Isaac. (2021) “Twitter Permanently Bans Trump, Capping Online Revolt.” New York
Times, January 12, 2021 https://www.nytimes.com/2021/01/08/technology/twitter-trump-suspended.
html (accessed 4 August 2021).
Corbin, Caroline Mala. (2014) “Hobby Lobby: Religious Liberty for the Elite.” American Constitution Society, July
11, 2014. https://www.acslaw.org/acsblog/hobby-lobby-religious-liberty-for-the-elite (accessed 4 August
2021).
Daniel, Norman. (1985) Just Health Care. New York; Cambridge University Press.
Employment Division v. Smith. (1983) 494 U.S. 872. Supreme Court.
Evers-Hillstorm, Karl. (2019) “Lobbying spending reaches $3.4 billion in 2018, highest in 8 years.”
OpenSecrets.org, January 25, 2019. https://www.opensecrets.org/news/2019/01/lobbying-spending-
reaches-3–4-billion-in-18/ (accessed 4 August 2021).
Financial Knowledge and Information Portal. (2020) “Market Cap of S&P 500 Companies.” FKNOL,
December 31, 2020. https://fknol.com/list/market-cap-sp-500-index-companies.php (accessed 4 August
2021).
First National Bank of Boston v. Bellotti. (1978) 435 US 765. Massachusetts Supreme Judicial Court.
Friedman, Milton. (2002) Capitalism and Freedom. Chicago; Chicago University Press.
Garrett, R. Sam. (2019) “Foreign Money and U.S. Campaign Finance Reform.” Congressional Research Service,
March 25, 2019. https://fas.org/sgp/crs/misc/IF10697.pdf (accessed 4 August 2021).
Greendorfer, Marc A. (2016) “The Inapplicability of First Amendment Protections to BDS Movement
Boycotts.” Cardozo Law Review 112.
Halper, Katie. (2010) “How to Date a Corporation: Dating Rules for a Post-Citizens United World,”
HuffPost.com. https://www.huffpost.com/entry/how-to-date-a-corporation_b_437110 (accessed 4 August
2021).
Harvard Law Review. (2014) “Note, Citizens United at Work: How the Landmark Decision Legalized
Political Coercion in the Workplace.” Harvard Law Review 128.
Isaacs, Daniel M. (2014) “An Unexpected Friend: Liberalism’s Response to Corporate Political Spending.”
Suffolk University Law Review 14.
Jenkins, Colleen and Daniel Trotta. (2017) “Seeking end to boycott, North Carolina rescinds transgender
bathroom law.” Reuters, March 30, 2017. https://www.reuters.com/article/us-north-carolina-lgbt/seeking-end-
to-boycott-north-carolina-rescinds-transgender-bathroom-law-idUSKBN1711V4 (accessed 4 August 2021).
Katz, Aaron M. and Susanna G. Dyer. (2014) Amicus Curiae Brief of Corporate and Criminal Law Professors in
Support of Petitioners, Hobby Lobby, 134 S. Ct. 2751 (Nos. 13-354, 13-356), 2014 WL 333889.
Kavoussi, Bonnie. (2012) “Move to Amend Movement Plans To Rubber Stamp Dollar Bills To Protest
Citizens United.” Reuters, June 7, 2012. https://www.huffpost.com/entry/move-to-amend-movement_n_
1578586 (accessed 4 August 2021).
Krantz, Matt. (2020) “Chasing Right Stocks to Buy Is Critical with Fewer Choices But Big Winners,”
Investors.com, November 27, 2020.
Leswing, Kif. (2021) “Google Removes Parler, Popular with Trump Supporters, from App Store for Android.”
CNBC, January 8, 2021. https://www.cnbc.com/2021/01/08/google-removes-parler-app-from-google-
play-store.html (accessed 4 August 2021).
Levin, Dan. (2019) “North Carolina Reaches Settlement on ‘Bathroom Bill.”’ New York Times, July 23, 2019,
https://www.nytimes.com/2019/07/23/us/north-carolina-transgender-bathrooms.html (accessed 4 August
2021).
Masterpiece Cakeshop, Ltd. v. Colorado Civil Rights Commission. (2018) 584 U.S. Supreme Court.
Masters, Jonathan. (2018) “Russia, Trump, and the 2016 U.S. Election.” Council on Foreign Relations. February 26,
2018. https://www.cfr.org/backgrounder/russia-trump-and-2016-us-election (accessed 4 August 2021).
McConnell v. FEC. (2003) 540 U.S 93. Supreme Court.
Meese, Alan J. and Nathan B. Oman. (2014) “Hobby Lobby, Corporate Law, and the Theory of the Firm:
Why For-Profit Corporations Are RFRA Persons.” Harvard Law Review 127.
NAACP v. Claiborne Hardware. (1982) 458 U.S. 883. Supreme Court.
NAACP v. Alabama. (1958) 357 U.S. 449. Supreme Court.
Needleman, Sarah E. and Georgia Wells. (2021) “Twitter, Facebook and Others Silenced Trump. Now They
Learn What’s Next.” Wall Street Journal, January 10, 2021. https://www.wsj.com/articles/twitter-facebook-
and-others-silenced-trump-now-they-learn-whats-next-11610320064 (accessed 4 August 2021).
184
PK. (2021). “Who Are the One Percent in the United States by Income and Net Worth?” Don’t Quit Your Day
Job. https://dqydj.com/top-one-percent-united-states/ (accessed 4 August 2021).
Reynolds, Daniel. (2020) “One Million Moms Threatens Boycott Oreo Over LGBTQ+ Campaign.” Advocate,
October 19, 2020. https://www.advocate.com/business/2020/10/19/one-million-moms-threatens-
boycott-oreo-over-lgbtq-campaign (accessed 4 August 2021).
Richter, Brian Kelleher, Krislert Samphanktharak, and Jeffery F. Timmons. (2009) “Lobbying and Taxes.”
American Journal of Political Science 53, 4.
Ripken, Susanna Kim. (2011) “Corporate First Amendment Rights After Citizens United: An Analysis of the
Popular Movement to End the Constitutional Personhood of Corporations.” University of Pennsylvania
Journal of Business Law 14.
Santa Clara County v. Southern Pacific R. Co. (1886) 118 U.S. 394. Supreme Court.
Sepinwall, Amy J. (2012) “Citizens United and the Ineluctable Question of Corporate Personhood,” Connecticut
Law Review 44.
Sepinwall, Amy J. (2015) “Corporate Piety and Impropriety: Hobby Lobby’s Extension of RFRA Rights to the
for-Profit Corporation” Harvard Business Law Review 5, 2.
Sepinwall, Amy J. (2021) “The State of U.S. Democracy: Corporations.” Penn Today, January 19, 2021.
https://penntoday.upenn.edu/news/state-us-democracy (accessed 4 August 2021).
Scherer, Andreas Georg and Guido Palazzo. (2007) “Toward a Political Conception of Corporate Responsibility:
Business and Society Seen from a Habermasian Perspective.” Academic Management Review 32.
Sipress, David. (2011) New Yorker Cartoon, New Yorker, March 14, 2011.
Torres-Spelliscy, Ciara. (2018) “Does ‘We the People’ Include Corporations?” ABA Human Rights Magazine
43, January 1, 2018. https://www.americanbar.org/groups/crsj/publications/human_rights_magazine_
home/we-the-people/we-the-people-corporations/ (accessed 14 February 2021).
Totenberg, Nina. (2014) “When Did Companies Become People? Excavating The Legal Evolution.” NPR,
July 28, 2014. https://www.npr.org/2014/07/28/335288388/when-did-companies-become-people-
excavating-the-legal-evolution (accessed 4 August 2021).
Trustees of Dartmouth College v. Woodward. (1819), 17 U.S. 518. Supreme Court.
Twitter Inc. (2021) “Permanent suspension of @realDonaldTrump.” Twitter Blog, January 8, 2021. https://
blog.twitter.com/en_us/topics/company/2020/suspension.html (accessed 4 August 2021).
US Congress, Act of Jan. 26, 1907, ch. 420, 34 Stat. 864 (1907).
US Congress, 42 U.S.C. § 2000bb-1(b).
US Senate. “Roll Call Vote 103rd Congress 1st Session.” https://www.senate.gov/legislative/LIS/roll_call_
lists/roll_call_vote_cfm.cfm?congress=103&session=1&vote=00331 (last visited 22 February 2021).
Vanderwalker, Ian and Lawrence Norden. (2018) “Getting Foreign Funds out of America’s Elections.” Brennan
Center, April 6, 2018. https://www.brennancenter.org/our-work/policy-solutions/getting-foreign-funds-
out-americas-elections (accessed 4 August 2021).
Waddell, Kaveh and National Journal. (2015) “How Tech Giants Became a Leading Civil-Rights Voice.” The
Atlantic, April 1, 2015. https://www.theatlantic.com/politics/archive/2015/04/how-tech-giants-became-
a-leading-civil-rights-voice/458547/ (accessed 4 August 2021).
Wellemeyer, James. (2020) “Want to know where all those corporate donations for #BLM are going? Here’s
the list.” NBC News, June 5, 2020. https://www.nbcnews.com/business/consumer/want-know-where-all-
those-corporate-donations-blm-are-going-n1225371 (accessed 4 August 2021).
Winkler, Adam. (2014) “Yes, Corporations Are People.” Slate, March 17, 2014. http://www.slate.com/
articles/news_and_politics/jurisprudence/2014/03/corporations_are_people_and_that_s_why_hobby_lobby_
should_lose_at_the_supreme.html (accessed 4 August 2021).
Winkler, Adam. (2018) We the Corporations: How American Business Won Their Civil Rights. New York, NY;
Liveright Publishing.
185
15
POLYCENTRICITY
Vlad Tarko
1 Introduction
Polycentricity is a system of many interdependent but autonomous decision centers operating under
an over-arching set of rules and norms that they themselves set up, monitor, and enforce (V.
Ostrom, 1972, 1991a; E. Ostrom, 1999, 2005, Ch. 9; Aligica and Tarko, 2012). A decision center is
part of the polycentric system (rather than an outsider that may be affected by the system’s actions) if
they are not only subjected to those over-arching rules and norms, but also have some say in their
creation, monitoring, enforcement, and change. Polycentricity is an alternative both to hierarchical
governance, where lower-level decision-makers have to obey commands from higher-level centers
and have little say about the rules imposed upon them from the top-down, and to anarchic (frag-
mented) governance, where no commonly accepted over-arching rules and norms exist. In the case
of fragmented governance, we cannot really speak of a “system” as there is no clear delineation
between insiders and outsiders of the system.
Markets are the most well-studied example of a decentralized “invisible hand” phenomenon, and
the concept of polycentricity was initially developed by drawing an analogy between markets and
other forms of emergent orders. However, coordination in markets occurs thanks to the operation of
the price system – prices generated by supply and demand acting as “signals wrapped in an incentive”
(Cowen and Tabarrok, 2014). By contrast, most of the other examples of polycentric systems lack a
coordination device akin to prices. Such examples include democracy as “polycentricity in the se-
lection of political leadership and the organization of political coalitions” (V. Ostrom, 1972), as well as
competitive public economies [like federalism and governance in metropolitan areas],

scientific inquiry, law and adjudicatory arrangements [especially in common law and
customary law systems], systems of governance with a separation of powers and checks and
balances, and patterns of international order. (V. Ostrom, 1991a)
What all these examples have in common are the following features: (1) a multiplicity of in-
dependent decision-makers (as opposed to a command-and-control hierarchy), (2) an over-arching
system of rules and norms constraining these decision-makers, and (3) a complex emergent order
resulting from the interactions of these decision-makers under the over-arching rules and norms,
what Adam Ferguson referred to as “the result of human action but not human design.” These three
features are the defining characteristics of polycentricity, and any system that has these characteristics is a
186 DOI: 10.4324/9780367808983-18

Polycentricity
polycentric system. In practice, these features are instantiated in a wide variety of ways (Aligica and
Tarko, 2012). For example, the decision centers may or may not have shared common goals, their
jurisdictions may or may not be territorially defined, the over-arching rules may be decided by
means of a variety of collective choice arrangements, entry, and exit into the polycentric system (i.e.,
being subjected to the rules and a participant in their creation) may be free or constrained in various
ways, and various forms of information may be private or public.
What guarantees that an emergent social-institutional-political order will be efficient, that is, that
it will have desirable characteristics from the point of view of the participants? Furthermore, how
can an emergent order be efficient in the absence of something like the price system to facilitate
coordination? In some cases, economists have proposed price-like mechanisms for coordination, for
example, Tiebout (1956) competition in a federal system is partially guided by differential tax rates
acting as a type of quasi-price for the local public services, or competition within the scientific
community is driven by reputation and prestige acting like a quasi-price (Tarko, 2015). But, while
in the case of markets we have a theory of market failure built on analyses of situations in which
prices might fail to fully capture the opportunity costs of the resources being exchanged, the
question of whether the emergent non-market orders are to be expected to be productive or ef-
ficient is far more difficult to address. This is so even in cases where some theory of quasi-prices is
proposed, but it is of exceeding difficulty in the other.
The concept of polycentricity is supposed to help us address this question in its most general
form, and, in Elinor Ostrom’s account, the failure or success of an emergent order is the con-
sequence of the nature of the over-arching rules and norms of the underlining polycentric system.
Part of her and her colleagues’ work has been to try to find criteria, which she called “core design
principles,” under which a community is more likely to succeed in creating effective operational
rules (i.e., rules that guide everyday activities and manage scarce resources) (E. Ostrom, 1990, 2005).
The Ostroms connect the idea of a productive emergent order to the concept of self-governance,
arguing that communities are more likely to gradually zero-in on good rules, in an evolutionary
fashion (Wilson, Ostrom, and Cox, 2013; E. Ostrom, 2014b), when they are governed in an in-
clusive and fair manner, such that relevant information is discovered and decision-makers are held
responsible to their constituents. One can think of this as an application of the idea of Pareto
efficiency to the problem of institutional design, along the lines proposed by the Calculus of Consent
(Buchanan and Tullock, 1962; E. Ostrom, 2011).
Polycentricity is, hence, one of the key concepts in Elinor Ostrom’s work on self-governance.
Indeed, her Nobel Prize address is titled, “Beyond Markets and States: Polycentric Governance of
Complex Economic Systems” (2010a). As argued by Elinor Ostrom, we need to move beyond the
strict state-market dichotomy, according to which markets should provide private and club goods,
while governments should provide public goods and establish the institutions of contract and
property necessary for markets to operate efficiently. This simplified picture is deceptive not only
because modern states do a lot more than just provide public goods, but also because (1) states are
not monolithic entities with coherent goals but need to be understood on the basis of their complex
internal structure (states themselves are polycentric), and (2) citizens in a democratic system are not
merely passive consumers of the goods and services provided to them by governments and firms, and
mere subjects of the rules imposed upon them, but are involved shareholders of the state and
coproducers of the institutions (Parks et al., 1981; Percy, 1984; Aligica and Tarko, 2013; Boettke,
Lemke, and Palagashvili, 2014).
This idea becomes particularly important for development economics, where we need to un-
derstand how institutions are set up in the first place (Rajan, 2004), out of the interactions of
different groups responding to past failures and new perceived opportunities. Understanding how
institutions emerge out of the interactions of numerous quasi-independent groups is essential for
187
Vlad Tarko
Table 15.1 Types of governance systems
Centralized Decentralized
Coordinated Top-down hierarchical (coordination as Polycentricity (coordination as

command-and-control) emergent order)
Not coordinated Rent-seeking (decentralized lobbying to a Fragmented (anarchic)
central authority)
discovering the possible reasons (and remedies) for the failure to create effective institutions
(E. Ostrom, 1990, 2005; McGinnis, 2016; Thiel, Blomquist, and Garrick, 2019).
Polycentricity plays an important role in emphasizing that centralized coordination is not the
only possible form of coordination (Table 15.1). A top-down hierarchical system can achieve co-
ordination by means of command-and-control mechanisms. Indeed, a key argument in favor of
hierarchy is that the absence of hierarchy leads to chaos and disorder because the system will lack
common goals and a unified direction. However, not only can order exist without top-down design
and control, but self-governing bottom-up emergent orders, facilitated by polycentric governance,
are often more productive, more equitable, and more resilient. As such, understanding poly-
centricity helps strengthening the arguments against hierarchical control and in favor of self-
governance in a wide variety of contexts. Hierarchy is not necessary for coordination.
It is also worth noting that many centralized systems in fact lack coordination, and, instead, set the
stage for wasteful rent-seeking contests. The typical example of this is mercantilism, in which the
central government auctions monopoly rights (Ekelund and Tollison, 1982). In such a system, would-
be monopolists compete for obtaining the monopoly privilege (which only one will get), wasting in
the process large amounts of resources that could have, in principle, been used for productive purposes
(Krueger, 1974). Real-world socialism also appears to have instituted a similar de facto rent-seeking
system (Anderson and Boettke, 1997), and a common argument is that instituting any form of top-
down hierarchical control automatically creates rent-seeking incentives that are difficult to curtail. The
presence of hierarchy should, thus, not be assumed to be a sufficient condition for coordination.
This being said, not all decentralized systems generate productive emergent orders. For co-
ordination to exist under decentralization, some over-arching rules and/or norms that effectively
constrain the decision-centers are necessary. Otherwise, conflict, zero-sum games, and tragedy of
the commons are pervasive. This is why it is important to distinguish polycentric governance from
mere fragmentation.
2 A Brief History of the Concept
2.1 Decentralized Production of Knowledge in Markets and Science

Both the word “polycentricity” and the basic idea were first introduced by the polymath Michael
Polanyi as a useful concept in the mid-20th century debate in Britain about the proper organization
of the scientific community (Polanyi, 1951, 1962). Polanyi was responding to highly prominent calls
at the time, although essentially forgotten today, to fully centralize and plan scientific research (e.g.,
see Caldwell, 2020). According to the centralization argument, scientists were wasting resources
overlapping their efforts and competing with one another. Instead, scientific research was supposed
to be organized as a fully comprehensive plan, where priorities would be established in a top-down
fashion, and each scientist and research center would be designated a specific non-overlapping
research role. This way, it was hoped, science could advance a lot faster.
188
Polycentricity
Polanyi created the concept of polycentricity as a way of explaining how the scientific community
was operating in a decentralized fashion but under an over-arching set of norms about the scientific
method, which were themselves created and evolved in a decentralized fashion. Contrary to what the
would-be central planners asserted, the scientific community was not wastefully chaotic and needlessly
anarchic – it was exhibiting a type of emergent order. Furthermore, Polanyi argued that this emergent
order was a lot more efficient than the proposed centralized planning, in terms of generating scientific
progress. On one hand, no one knows for sure what the most promising line of research is. The central
planners were delusional in thinking they could know any better than the trial-and-error process of
existing polycentric science. Key scientific developments often arrive out of unexpected directions,
which the central plan would not be able to guess. Instead, under the centralized system, these un-
expected directions would be drained of resources and shut down. On the other hand, the incentive
structure created by the proposed centralization would also be highly problematic – it would in effect
re-establish an inflexible authority structure that would stifle scientific innovation.
These ideas brought Michael Polanyi in close intellectual proximity with Friedrich Hayek, who
had made similar arguments about the efficiency of markets over socialist planning (Hayek, 1945).
Indeed, Polanyi later used the concept of polycentricity, acknowledging that markets are also an
example of a polycentric system, to engage in the socialist calculation debate (for more details on
this, see Aligica and Tarko, 2012). Markets are a polycentric system of consumers and producers
operating under the overarching set of rules about property and contract. Nevertheless, a crucial
difference existed between Polanyi’s argument and Hayek’s argument.
Hayek’s argument about markets is that their emergent order occurs as a side-effect of the op-
eration of prices. Prices aggregate the relevant information about the opportunity costs of various
resources and create the incentives that everyone takes this information into account. In Hayek
(1945)’s example, when, for some unknown reason, a resource becomes scarcer, its price rises,
which, in turn, determines consumers to economize it and creates a profit opportunity for producers
who find substitutes or new sources. In other words, the price change affects everyone’s behavior
nudging them in the direction of greater efficiency – that is, prices create large-scale coordination,
even if few have any idea about what the actual cause of the initial shortage was.
By contrast, there is no single “signal wrapped in an incentive” in Polanyi’s account of the
scientific community. Reputation might serve a somewhat similar purpose to prices, as a catalyst of
the emergent order, but the analogy is imperfect (Tarko, 2015). Instead, the key work here for
generating a productive and efficient emergent order is done by the set of over-arching rules and
norms. If those over-arching rules and norms are flawed, the emergent order would become dys-
functional, that is, the scientific community will fail at its truth-seeking goal. This idea about the role
of rules and norms in enabling a productive emergent order becomes the crucial guiding principle
for Vincent and Elinor Ostrom, who were also trying to make sense of emergent orders in the
absence of any obvious simple “signal wrapped in an incentive.”
2.2 Governance of Metropolitan Areas

In the late 1950s and early 1960s, Vincent Ostrom – the husband of Elinor, and a key figure in his
own right in the fields of public administration and political science, and one of the founders of
public choice – came up with the word “polycentricity” to mean virtually the same thing as Polanyi.
He quickly learned about Polanyi’s earlier work and embedded Polanyi’s insights into his own
analysis (e.g., see V. Ostrom, 1991a). But Vincent’s focus, and the reason he had for thinking about
polycentricity in the first place, was yet another example of a polycentric system – the metropolitan
urban governance (Tarko, 2017, Ch. 1).
By the 1950s, the growing urbanization in the United States had led to a curious phenomenon.
Small towns had expanded so much that they came in contact with one another, and merged into
189
Vlad Tarko
single metropolitan areas. But these de facto unified cities did not have unified governments. They
were still governed by the multitude of legacy local governments, that is, their organization was
polycentric reflecting the “multiplicity of political jurisdictions in a metropolitan area” (V. Ostrom,
Tiebout, and Warren, 1961). Should these small legacy governments be abolished in favor of single
city governments? This became a particularly salient issue with respect to police departments. The
common view at the time was that
[f]ragmentation of police services is extreme … Wasted energies are lost motion due to
overlapping, duplication, and noncooperation are not the worst consequences of this
fragmentation. Large areas of the United States – particularly rural communities and the
small jurisdictions in or near metropolitan areas – lack anything resembling modern,
professional police protection. (Committee for Economic Development, cited by E.
Ostrom, Parks, and Whitaker, 1974)
This argument bore a striking resemblance to the arguments that Polanyi had addressed earlier. The
centralization of the scientific community was also advocated on the grounds that such centralization
and top-down coordination would eliminate the overlapping efforts of competing scientists, and, it
was argued, would create greater efficiency by reducing redundant use of resources. Similarly, one of
the key arguments in favor of socialism, to which Mises and Hayek had responded in the 1920s and
1930s (Hayek, 1935, 1945), was also that private enterprises in the capitalist system were redundant,
producing duplicate products, while state enterprises in a socialist system would take better ad-
vantages of economies of scale and eliminate the “waste” of the “anarchic” competitive system. In
both these cases, the mistake was to ignore the possibility that a system can generate coordination in
a decentralized fashion and assume instead that the lack of top-down control implied disorder.
From a scientific point of view, the challenge was to actually build a theory of how such de-
centralized coordination occurs, and the concept of polycentricity was proposed in the attempt to
understand how large-scale coordination can occur even when market prices are not available, in
particular within social-political orders (V. Ostrom, 1972, 1991a; Tarko, 2017, Ch. 2). The concept
is proposed as a way of “[p]enetrating an illusion of chaos and discerning regularities that appear to
be created by an ‘invisible hand’ … Patterns and regularities which occur under an illusion of chaos
may involve an order of complexity that is counterintuitive” (V. Ostrom, 1972). Can this coun-
terintuitive order be better understood?
Building on Tiebout (1956)’s initial model of competitive local governments, Vincent Ostrom,
Tiebout, and Warren (1961) made a strong theoretical case that the polycentric organization of
public services in metropolitan areas might actually be beneficial. Their arguments mirrored Hayek’s
arguments against central planning and in favor of markets, and Polanyi’s arguments against the
centralization of the scientific community. Later on, Elinor Ostrom and her students engaged in an
extensive empirical study of metropolitan areas in the United States, and of police departments in
particular.1 These empirical results dealt a serious blow to the case for consolidation:
The presumption that economies of scale were prevalent was wrong; the presumption that
you needed a single police department was wrong; and the presumption that individual
departments wouldn’t be smart enough to work out ways of coordinating is wrong. … For
patrolling, if you don’t know the neighborhood, you can’t spot the early signs of problems,
and if you have five or six layers of supervision, the police chief doesn’t know what’s
occurring on the street. (Elinor Ostrom, interviewed by Zagorski, 2006)
The metropolitan police studies compared the performance of police in similar neighborhoods, and
in close vicinity to each other, but with different organizations, either consolidated or polycentric.
190
Polycentricity
The consistent finding from this series of studies is that small and medium-sized police
departments perform more effectively than large police departments serving similar
neighborhoods, and frequently at lower costs … Victimization rates tend to be lower,
police response tends to be faster, citizens tend to be more willing to call on police, citizens
tend to more positively evaluate specific contacts with the police, and citizens tend to rate
police higher across a series of other evaluative questions. Further, citizens living in small
communities tend to be more informed about how to change local policies, tend to know
more policemen serving their neighborhoods, and call the police more frequently to obtain
general information than do citizens living in large cities. Citizens served by small de-
partments tend to receive better services at lower costs than their neighbors living in the
center city. Instead of being a “problem” for the metropolitan area, small departments frequently
contribute to the improvement of police services in the area. (E. Ostrom, 2000, emphasis added)
In a nutshell, Elinor Ostrom and her students discovered that the advocates of consolidation not only
proposed a highly dubious policy “reform” that was likely to make things worse, but they also had
very little accurate knowledge of how police departments actually operated (e.g., see E. Ostrom,
Parks, and Whitaker, 1974). Many of the claims were based on imagination and theory, rather than
on actual observations. Contrary to the quote from the Committee for Economic Development
given earlier, Ostrom discovered that small-scale police departments actively cooperated in order to
take advantage of economies of scale, rather than inefficiently duplicating their efforts. For example,
jails or crime labs were in fact set up to service multiple police departments, and decentralization was
adopted only when it made the most sense (e.g., in the case of street patrols).
Furthermore, Elinor Ostrom and her student, Gordon Whitaker, were among the first to provide an
in-depth empirical political economy account of institutional racism within police departments. They
noted that “[p]olice seem to be failing to serve residents of many black neighborhoods in US cities” and
a key reason for this is that “[p]olice effectiveness depends, in part, on police understanding the nature
of the community being served and police openness to suggestions, criticism, and complaints”
(E. Ostrom and Whitaker, 1974). They analyzed a range of proposed reforms and concluded that
[c]ommunity control of police may … provide an institutional framework for the effective
expression of black citizen demands for impartial police service. … Professionalism alone
does not appear to provide sufficient controls so that police will be responsive to their
needs for protection and respect. Community control places that responsibility on the
people themselves and provides them with the mechanisms by which to exercise it. (E.
Ostrom and Whitaker, 1974)
At first glance, it may seem that the decentralized nature of polycentric governance might enhance
the problem of racism. This is why the idea of “community control” is critical. Unfortunately,
“community policing” has been adopted largely as an Orwellian slogan to describe police gathering
intelligence from the local population in the attempt to fight a top-down federally mandated War of
Drugs (Boettke, Palagashvili, and Lemke, 2013, 2016; Coyne and Hall-Blanco, 2016). By contrast,
community policing … properly understood involves not just a decentralized mechanism

for knowledge aggregation, but also … allowing the goals of the local government to be
determined by the local citizens. A decentralized police agency that takes orders from the
federal government, and it is dependent on revenues and equipment on the central gov-
ernment, is not going to be responsive to citizens’ needs, and, hence, will not perform well
in terms of most criteria of citizens’ satisfaction. (Tarko, 2017, 37)
191
Vlad Tarko
The basic understanding of this, including the potential trap of top-down funding, as well as the
worry that the idea of “community policing” will be misappropriated as a slogan, is already present
in the original papers. E. Ostrom and Whitaker (1974) defined it as such:
Local control of the police would involve the establishment of formal structures of ac-
countability to the public being served as well as indirect internal supervision of patrolmen
on the job. An effective means of establishing local control of the police in large cities
might be to set up neighborhood districts to handle a variety of locally confined public
problems. Such units would require some means of public selection of officials and the
authority to levy local taxes and establish local ordinances.
This was the first example among many in which Elinor Ostrom was driven to the conclusion that
collectively beneficial outcomes are the result of social-political institutions that facilitate and enable self-
governance. As such, the concept of polycentricity takes a deeper political and normative significance. It
is no longer only a descriptive theory designed to help us, in the words of Vincent Ostrom, discern the
“[p]atterns and regularities which occur under an illusion of chaos” (V. Ostrom, 1972). It is also a policy
tool for helping us enhance self-governance, which can be seen to have both instrumental value (in
promoting economic efficiency) and an inherent political value (e.g., Dahl, 1989).
2.3 Federalism
After these initial studies, Vincent and Elinor Ostrom went in two different but related directions.
Vincent used the perspective of polycentricity and self-governance to develop a theory of federalism
(V. Ostrom, 1987, 1991b; Bish, 1999, 2014; Wagner, 2005). Most theories of federalism are
concerned with describing interactions between different levels of government. As such, most of
these theories tend to theorize federalism as a type of hierarchical system, in which local-level
governments operate within the constraints created by upper-level governments. For example,
according to a well-known theory proposed by Weingast (1995), in order to preserve the market
order, the federal government needs to implement a variety of constraints upon lower-level gov-
ernments, such as denying them the ability to impose tariffs upon one another.
In contrast to such theories, Vincent Ostrom noted that, particularly in the case of American
federalism, it is relatively common for lower-level governments to implement laws that are at odds
with higher-level laws. For example, various US states have legalized marijuana although it remains
a restricted drug as far as federal law is concerned. Similarly, in the European Union, courts in
individual states have nullified EU-level laws on multiple occasions. As such, federalism is better
understood in terms of the checks-and-balances that different governments impose on each other –
local-on-local governments, thanks to Tiebout competition, as well as between different levels, for
example, state-on-federal and federal-on-state.
Not only is this checks-and-balances view a more accurate representation of the relations between
governments, but even individual governments are best understood as systems populated by agents
with conflicting interests, rather than homogeneous agents with clearly identifiable goals. Although
some systems can be described, for all practical purposes, as idealized representative agents – that is,
as rational agents with as-if beliefs and goals – according to Vincent Ostrom, governments are not
among such systems. The errors introduced by such homogenizing idealizations far exceed the
benefits one gets from the simplified picture.
Furthermore, in line with the self-governance perspective provided by the idea of polycentricity,
Vincent Ostrom argued that federalism should be understood not from the point of view of gov-
ernments, but from the perspective of individual citizens, that is, in terms of the choices and op-
portunities it provides (Bish, 2014; see also Inman and Rubinfeld, 1997). As he put it, the “principles
192
Polycentricity
of federalism permit people to function through self-governing institutions among local, regional,
and national communities of interest in organizing collective endeavors” (V. Ostrom, 1987, 173).
Once again, Elinor brought the empirical guns at a fight between political theorists, showing how
the polycentricity perspective on federalism enlightens the debate about police consolidation (E.
Ostrom, 1976).
3 A Practical Theory of Self-Governance

Following the police studies, Elinor returned to the topic of her dissertation – the study of how
communities or groups of communities manage to solve the tragedy of the commons problems. Elinor
Ostrom is most well known for studying this case of polycentric governance and received the Nobel
Prize for her efforts. This area of study was indeed particularly fruitful, as it allowed her and her
collaborators to identify various core institutional principles that enabled communities to prosper.
The statement that the efficiency of an emergent social-institutional-political order depends on
the overarching rules and norms of the underlining polycentric system remains rather vague and
open-ended, unless we specify more clearly the principles that these over-arching rules and norms
are supposed to comply with. In other words, the claim that the key for understanding the success or
failures of emergent orders rests in the study of the overarching rules is only methodology – it tells us
where to look for a solution, but it doesn’t offer any specific guidelines with respect to what the
solution might be. Is it possible to do more than just methodology? Indeed, it appears so.2
The first, and highly important, discovery that Elinor Ostrom made after she analyzed hundreds
of case studies of various communities from around the world dealing (some successfully and some
unsuccessfully) with different cases of tragedies of the commons was that there was no unique set of
practical rules that guaranteed success or avoided failure. As she kept saying: “No panacea!”
Economists have used game theory to explain the nature of the problem – the common incentive
problem that plagues all cases of tragedies of the commons. This is indeed part of the beauty of game
theory – allowing us to see the hidden structure behind an otherwise bewildering complexity of
different cases. Over-fishing in the ocean or in rivers, over-fishing lobsters in Maine, over-grazing in
mountain areas around the world, deforestation, depleting underground water supplies, workers
shirking at their job, workers failing to unionize because of free-riding, pollution, mass famine under
farm collectivization in China’s Great Leap Forward, global warming due to carbon emissions, etc. –
are all the same underlining Prisoners’ Dilemma-type problem in which shirking and free-riding is
the individually most beneficial choice although all would prefer that we behaved more responsibly.
However, the fact that the cause of the problem is the same does not imply that the remedy is the
same. This is, of course, unfortunate, but it is, nonetheless, true. The situation is even more un-
fortunate than most people, and most economists, assume. What Elinor Ostrom found was that the
remedy is substantially different even in cases that seem rather similar. Learning how to fix over-
fishing in Maine will not necessarily help in fixing over-fishing in Turkey. Deforestation in two
different places around the world will not have the same kind of institutional solution, and trying to
“transplant” an institution from a successful place to another will often inexplicably fail. Why does it
fail? Why should it be so hard to learn from other people’s experiences? The problem is less with
learning and more with the expert imposition of solutions. Rules need to be monitored and en-
forced; people who don’t create the rules themselves often don’t understand them or don’t see them
as legitimate. Moreover, rules work in the context of other rules and norms. The same rule will
often work differently in two different places because of the context provided by the other existing
rules and norms. These other rules and norms create different pathways for avoidance and shirking.
The Ostroms distinguish between different levels of rules (E. Ostrom and Ostrom, 2004):
(i) operational rules are rules about how to actually manage resources and interact with people about
specific matters; (ii) collective choice rules are rules about how to decide what the operational rules are;
193
Vlad Tarko
(iii) constitutional rules are rules about how to choose the collective choice rules; and (iv) meta-
constitutional rules/norms are general criteria about which constitutional rules are perceived to be
legitimate or not.
Because of the diversity problem mentioned above, there are relatively few, but nonetheless
important, direct guidelines, that is, operational rules, that we can identify:
1. Excludability principle: clearly defined group boundaries and membership.

2. Graduated sanctions principle: graduated sanctions for breaking the rules.
3. Local fit principle: correspondence of appropriation rules to local conditions.
4. Fairness principle: proportionality between the benefits and costs of various actors.
The first one is a very basic public choice principle and the key focus of Mancur Olson’s Logic of
Collective Action (1965). The second one is a basic law and economics principle that helps distinguish
between mere errors and deliberate offenses, and also makes sure that offenders are not incentivized
to compound their offenses (e.g., a robber who breaks into your house should not have the in-
centive to kill you if you discover him). The third one is a federalism principle according to which
the scale of administration should match the scale of the problem the administration is trying to solve
(V. Ostrom, Tiebout, and Warren, 1961). The fourth one is a social psychology principle about the
nature of fairness and the kind of inequality that people are willing to tolerate (Starmans, Sheskin,
and Bloom, 2017).
The collective choice principles are also relatively straightforward, and, yet, in practice, they are
often missing:
5. Accountability principle: monitors and enforcers of rules are accountable for their actions.
6. Political representation principle: most individuals affected by the rules are included in the collective
choice group that can modify these rules.
7. Conflict resolution principle: access to low-cost local arenas for conflict resolution providing de-
cisions perceived as fair.
These are all political economy rules about checks-and-balances, and about preventing abuses of
power and limiting rent-seeking. The last one is a basic idea in the economic theory of conflict. As
noted by Tullock (2005, 5), situations of conflict are inherently inefficient as they “lead to in-
vestment of resources by A to get B’s property and by B to defend it. Regardless of the outcome of the
conflict, the use of resources for this purpose is offsetting and therefore inherently wasteful.” [em-
phasis added] Consequently, “[s]ocial contrivances for reducing such investment are, on the whole,
desirable, although there may be cases where it is more efficient to place no institutional restrictions
on such conflict” (Ibid.). Similarly, V. Ostrom (1997, 136), argued that “[t]he development of order
out of chaos requires that each human being establish a basis for anticipating how others will behave,
so that each person can act with an expectation that other persons will act with constraint.”
Establishing a peaceful society requires “common knowledge and shared communities of under-
standing” (Ibid.).
Finally, at the constitutional level:
8. Subsidiarity principle: external governmental authorities recognize, at least to some extent, the
right to self-organize.
9. Nested enterprises principle: “Appropriation, provision, monitoring, enforcement, conflict re-
solution, and governance activities are organized in multiple layers of nested enterprise”
(E. Ostrom, 2005, 259).
194
Polycentricity
These principles are in line with Vincent Ostrom’s theory of federalism discussed earlier. The first of
these recognizes the fact that higher-level governments often have the power to subvert local
governance. As such, some explicit recognition of the right to self-govern is usually needed. The
second one recognizes that the scale of different collective problems differs a lot from problem to
problem, and, as such, insisting that a single apparatus, with a given scale, is responsible for solving
all problems (or a large number of problems) will inevitably lead to many inefficiencies due to
scale-mismatches.
As mentioned earlier, multiple phenomena can be understood as examples of polycentric systems.
A provocative conjecture put forth by Vincent Ostrom holds that several of those different poly-
centric systems – markets, democratic politics, federalism, common law, science, and peaceful in-
ternational relations – rely upon each other and reinforce each other. As Vincent Ostrom (1991b,
237) put it, “[p]olycentricity in each unit of government … is essential to the maintenance of
polycentricity in ‘the whole system of human affairs.’” Similarly, Aligica (2013, 50–51) argues that
[p]olycentricity is a complex system of powers, incentives, rules, values, and individual

attitudes, all combined in a complex system of relationships at different levels. One may
detect very interesting dynamics at work. Market polycentrism seems to entail judicial
polycentrism. Judicial polycentrism entails political polycentrism, and in its turn political
polycentrism entails constitutional polycentrism. Accepting the existence of such a sys-
temic logic, one may visualize the entire social system as defined by underlining currents
originating in pulsating polycentric domains. Polycentric order in one area entails and
produces polycentrism in other areas.
The extent to which this conjecture holds is still an open question.
4 Global Climate Change

These core principles have been tested primarily in cases of relatively small-scale communities (e.g.,
see E. Ostrom, 1990). There is a reasonable worry that the operation of large-scale communities
requires additional ideas or maybe different principles altogether. As a practical example of how
polycentricity has been used to provide a novel perspective on a large-scale problem, consider the
issue of global warming (E. Ostrom, 2010b, 2014a; Cole, 2015).
Global warming is a typical tragedy of the commons problem in which most people would be
better off if carbon emissions were lower, but everyone also has the incentive to shirk. Moral self-
restraint can only get us so far (not very far).
The typical point of view is to believe that because the scale of the problem is global, the scale of
the solution should also be global. This is not entirely unreasonable, and it is in fact in line with the
“local fit principle” mentioned above. Indeed, V. Ostrom, Tiebout, and Warren (1961) use air
pollution as an example of the importance of local fit, explaining that air pollution in California
could not be addressed while the policy was too localized – the source of the pollution was often
outside of the jurisdiction that was affected by pollution and which tried to fix the problem (this is
also in line with the “political representation principle”). Air pollution was addressed more suc-
cessfully when the task of environmental policy was moved at the state level.
This being said, Elinor Ostrom argued that, in the case of global warming, the above assumption,
that the scale of the problem implies the scale of the jurisdiction, was actually dangerously mis-
guided. The scale of the problem is global, but the transaction costs of building a global policy
consensus are so large that we risk doing almost nothing. Everyone is focused on the wrong scale,
and mitigation efforts are wasted lobbying for unfeasible global policies when these efforts would be
better spent on more realistic and more local policies.
195
Vlad Tarko
The problem with local policies for addressing global warming is that none of them, by them-
selves, can do much of a dent in addressing the problem. If California or Germany adopt some
“climate policy,” they are easy to ridicule as engaging in mere “virtue signaling” because such
policies obviously cannot have much of an effect globally. But this is where the emergent order
perspective of polycentricity comes into play. Elinor Ostrom’s point is that the result that we want –
a sufficiently large reduction in global emissions – can be the outcome of an emergent order,
emerging out of the interactions of many localized policies, rather than being the outcome of a
single top-down global policy. A large diversity of localized policies is more politically feasible than a
single meaningful global policy, and, as such, the polycentric approach could have a bigger effect in
terms of actual emissions reductions.
The problem of global warming brings to the forefront the distinction between polycentricity
and anarchic fragmentation. As mentioned earlier, decentralization without over-arching rules/
norms does not necessarily lead to good results. There are no formal enforceable global over-arching
rules with respect to climate policy, which leads to the reasonable suspicion that we are dealing with
merely an anarchic system which, given free riding problems, cannot be trusted to generate a good
emergent climate policy. However, one lesson from the study of the scientific community is that
large-scale polycentric systems can be coordinated by informal norms. To what extent can such a
norm-based mechanism apply here?
We can think of norms as a combination between an expectation about how others will in fact
behave and a normative element about how people should behave (Crawford and Ostrom, 1995). A
normative claim that no one obeys is empty, while an expected behavior without the normative
element is just a behavioral regularity.
One can think of international climate treaties as important less in terms of global rules, but more
in terms of creating global norms. But, as noted by Cole (2015), there is another, even more
powerful, mechanism for creating global coordination, without global rules:
numerous climate policies that have been, and are being, implemented at local, state, regional
and national governments, and even among private business associations. These polycentric
policies multiply opportunities for communication, trust-building, policy experimentation
and learning. Local-level governments have been experimenting with GHG mitigation
policies for many years now, and paying close attention to what others have been doing.
The key element here is that the behavior of some actors creates trust among other actors, and this
mutual trust can be expanding, engendering greater coordination. This is similar to the payment
mechanism instituted by websites like Kickstarter. Funding a project via numerous small donations
faces an apparently insurmountable trust and free riding difficulty: would-be small investors cannot
trust that sufficient funds will be gathered from others (similar to how different actors cannot trust
that others would also implement climate mitigation policies), and everyone is tempted to wait and
see, free-riding on other contributors (again similar to climate policy). The way websites like
Kickstarter fix the issue is by guaranteeing that the funds are collected only if the necessary aggregate
is pledged. A similar full guarantee does not exist for climate policies, but the Ostrom-Cole argu-
ment is that the process of expanding mutual trust acts as an imperfect substitute.
As such, the different decision-centers are not, in fact, acting in a completely disjointed and
uncoordinated fashion because the actions of one increases the likelihood that others will also act:
“Trust is not the same as blind faith, where parties simply sit down and ‘do the right things’ ac-
cording to someone’s moral compass. Rather, trust is earned by mutual commitments that are not
overly costly to monitor” (Cole, 2015).
So, is this polycentric account of a global problem at odds with the “core design principles”
described above, given that it breaks the fit principle? It does seem so, strictly speaking. However, a
196
Polycentricity
more insightful perspective is to say that transaction costs need to be considered more carefully.
Small scale communities have small transaction costs in terms of creating, changing, monitoring, and
enforcing rules. As such, the problem of transaction costs is in the background. But when we are
analyzing problems with severe transaction costs, for example, due to much larger populations or
larger geographical or temporal scales, the principles need to be re-assessed. Elinor Ostrom has, in
fact, always insisted that these principles should not be used mechanically, as a mere checklist. But
why not use them as a checklist? To put it in economic terms, there are costs and benefits to sticking
to those principles, and the transaction costs may become so high that the costs go beyond the
benefits. In the example of global warming, sticking to the local fit principle becomes positively
harmful and probably hampers our ability to solve the problem. Another answer, that may com-
plicate matters even more, is that significant transaction costs can also affect how the different
principles interact with one another.
We can also use this example to amend my earlier description of the relationship between the
concept of polycentricity and that of emergent order. It is often the case that the over-arching rules
are the object of collective choice, and the nature of the emergent order is determined by those
rules. This is indeed the bulk of the examples studied by Elinor Ostrom. But it may also be the case
that the over-arching rules are fixed – or changing them in a timely fashion is politically unfeasible –
as in the case of climate policy at a global level. If the over-arching rules are not the object of
collective choice, one needs to refocus on the fact that maybe different emergent orders are possible
under the same over-arching rules. In other words, maybe the emergent order is not determined by
those rules, but merely constrained by them. Phenomena like path-dependence or collective blind-
spots may be highly relevant. In the case of global warming, Ostrom is arguing that one important
collective blind-spot we have is that global problems require global solutions.
5 Conclusion
To summarize, the concept of polycentricity has been developed with two (related) goals in mind:
(1) as a descriptive tool for understanding emergent social-institutional-political orders, especially
emergent orders that lack a price system as a coordination device and (2) as a normative policy tool
for enabling greater levels of self-governance.
This being said, not any imaginable emergent order is productive, equitable, and resilient. Only a
subset of them are, and a key advance made by Elinor Ostrom and her collaborators was to identify a
set of “core principles” describing which types of social choice institutions are more likely to
generate good practical results. These core principles form, in effect, a practical theory of self-
governance. However, as emphasized by Elinor herself, as well as by some of the researchers fol-
lowing in her footsteps (e.g., see especially Cox, Arnold, and Villamayor Tomás, 2010; and the
discussion in Tarko, 2017, Ch. 4) we are still probably far from done. A key active research area
involves the attempt to develop a framework of analysis that applies to larger-scale systems (e.g., see
McGinnis and Ostrom, 2014; Cox, 2014), under the assumption that Ostrom’s core principles,
discovered in the context of analyzing relatively small-scale societies, are incomplete or maybe even
unhelpful for large-scale cases.
Notes
1 See E. Ostrom (1972) for a typical example. Many of the key Bloomington School papers on polycentricity
are collected in McGinnis (1999a) and McGinnis (1999b). See Tarko (2017, Ch. 1), for an overview of this
literature.
2 For a more in-depth discussion of the core design principles see Cox, Arnold, and Villamayor Tomás (2010)
and Tarko (2017, Ch. 4). The account here follows the latter.
197
Vlad Tarko
References
Aligica, Paul Dragos. (2013) Institutional Diversity and Political Economy: The Ostroms and Beyond. Oxford and
New York; Oxford University Press.
Aligica, Paul Dragos, and Vlad Tarko. (2012) “Polycentricity: From Polanyi to Ostrom, and Beyond.”
Governance-an International Journal of Policy and Administration 25, 2, 237–262.
Aligica, Paul Dragos, and Vlad Tarko. (2013) “Co-Production, Polycentricity, and Value Heterogeneity: The
Ostroms’ Public Choice Institutionalism Revisited.” American Political Science Review 107, 04, 726–741.
Anderson, Gary M., and Peter J. Boettke. (1997) “Soviet Venality: A Rent-Seeking Model of the Communist
State.” Public Choice 93, 1–2, 37–53.
Bish, Robert L. (1999) “Federalist Theory and Polycentricity: Learning from Local Governments.” in Donald
P. Racheter and Richard E. Wagner edited by. Limiting Leviathan. Cheltenham, UK; Northampton, MA;
Edward Elgar.
Bish, Robert L. (2014) “Vincent Ostrom’s Contributions to Political Economy.” Publius: The Journal of
Federalism 44, 2, 227–248. http://publius.oxfordjournals.org/content/44/2/227.
Boettke, Peter J., Liya Palagashvili, and Jayme Lemke. (2013) “Riding in Cars with Boys: Elinor Ostrom’s
Adventures with the Police.” Journal of Institutional Economics 9, 4, 407–425.
Boettke, Peter J., Jayme S. Lemke, and Liya Palagashvili. (2014) “Polycentricity, Self-Governance, and the Art &
Science of Association.” The Review of Austrian Economics 28, 3, 311–335.
Boettke, Peter J., Jayme S. Lemke, and Liya Palagashvili. (2016) “Re-Evaluating Community Policing in a
Polycentric System.” Journal of Institutional Economics 12, 2, 305–325.
Buchanan, James M., and Gordon Tullock. (1962) The Calculus of Consent: Logical Foundations of Constitutional
Democracy. Indianapolis; Liberty Fund [1999].
Caldwell, Bruce. (2020) “The Road to Serfdom After 75 Years.” Journal of Economic Literature 58, 3, 720–748.
10.1257/jel.20191542.
Cole, Daniel H. (2015) “Advantages of a Polycentric Approach to Climate Change Policy.” Nature Climate
Change 5, 2, 114–118. 10.1038/nclimate2490.
Cowen, Tyler, and Alex Tabarrok. (2014) Modern Principles of Economics. Third edition. New York; Worth
Publishers.
Coyne, Christopher J., and Abigail R. Hall-Blanco. (2016) “Foreign Intervention, Police Militarization, and
Minorities.” Peace Review 28, 2, 165–170.
Cox, Michael E. (2014) “Understanding Large Social-Ecological Systems: Introducing the SESMAD Project.”
International Journal of the Commons 8, 2, 265–276. 10.18352/ijc.406.
Cox, Michael E., Gwen Arnold, and Sergio Villamayor Tomás. (2010) “A Review of Design Principles for
Community-Based Natural Resource Management.” Ecology and Society 15, 4, 38. http://www.
ecologyandsociety. org/vol15/iss4/art38/.
Crawford, Sue E. S., and Elinor Ostrom. (1995) “A Grammar of Institutions.” American Political Science Review
89, 3, 582.
Dahl, Robert Alan. (1989) Democracy and Its Critics. New Haven; Yale University Press.
Ekelund, Robert B., and Robert Tollison. (1982) Mercantilism As a Rent-Seeking Society: Economic Regulation in
Historical Perspective. College Station, TX; Texas A & M University Press.
Hayek, Friedrich A., ed. (1935) Collectivist Economic Planning: Critical Studies on the Possibilities of Socialism.
London; Routledge.
Hayek, Friedrich A., ed. (1945) “The Use of Knowledge in Society.” The American Economic Review 35, 4,
519–530.
Inman, Robert P, and Daniel L. Rubinfeld. (1997) “Rethinking Federalism.” Journal of Economic Perspectives 11,
4, 43–64.
Krueger, Anne O. (1974) “The Political Economy of the Rent-Seeking Society.” American Economic Review 64,
3, 291–303.
McGinnis, Michael D., ed. (1999a) Polycentric Governance and Development: Readings from the Workshop in Political
Theory and Policy Analysis. Michigan; University of Michigan Press.
McGinnis, Michael D., ed. (1999b) Polycentricity and Local Public Economies: Readings from the Workshop in Political
Theory and Policy Analysis. Michigan; University of Michigan Press.
McGinnis, Michael D., ed. (2016) Polycentric Governance in Theory and Practice: Dimensions of Aspirations and
Practical Limitations Indiana University, Bloomington.
McGinnis, Michael D., and Elinor Ostrom. (2014) “Social-Ecological System Framework: Initial Changes and
Continuing Challenges.” Ecology and Society 19, 2, http://www.ecologyandsociety.org/vol19/iss2/art30/.
198
Polycentricity
Olson, Mancur. (1965) Logic of Collective Action: Public Goods and the Theory of Groups. Cambridge, MA; Harvard
University Press.
Ostrom, Elinor. (1972) “Metropolitan Reform: Propositions Derived from Two Traditions.” Social Science
Quarterly 53, 3, 474–493.
Ostrom, Elinor. (1976) “Size and Performance in a Federal System.” Publius 6, 2, 33–73. 10.2307/3329539.
Ostrom, Elinor. (1990) Governing the Commons: The Evolution of Institutions for Collective Action. Cambridge and
New York; Cambridge University Press.
Ostrom, Elinor. (1999) “Polycentricity, Complexity, and the Commons.” The Good Society 9, 2, 37–41.
Ostrom, Elinor. (2000) “The Danger of Self-Evident Truths.” PS: Political Science & Politics 33 (March), 33–46.
10.2307/420774.
Ostrom, Elinor. (2005) Understanding Institutional Diversity. Princeton; Princeton University Press.
Ostrom, Elinor. (2010a) “Beyond Markets and States: Polycentric Governance of Complex Economic Systems.”
American Economic Review 100, 3, 641–672. http://pubs.aeaweb.org/doi/abs/10.1257/aer.100.3.641.
Ostrom, Elinor. (2010b) “Polycentric Systems for Coping with Collective Action and Global Environmental
Change.” Global Environmental Change, 20th Anniversary Special Issue, 20, 4, 550–557.
Ostrom, Elinor. (2011) “Honoring James Buchanan.” Journal of Economic Behavior & Organization, Special issue
on the work of James Buchanan, 80, 2, 370–373. 10.1016/j.jebo.2011.04.002.
Ostrom, Elinor. (2014a) “A Polycentric Approach for Coping with Climate Change.” Annals of Economics and
Finance 15, 1, 97–134.
Ostrom, Elinor. (2014b) “Collective Action and the Evolution of Social Norms.” Journal of Natural Resources
Policy Research 6, 4, 235–252. 10.1080/19390459.2014.935173.
Ostrom, Elinor, and Vincent Ostrom. (2004) “The Quest for Meaning in Public Choice.” American Journal of
Economics and Sociology 63, 1, 105–147.
Ostrom, Elinor, Roger B. Parks, and Gordon P. Whitaker. (1974) “Defining and Measuring Structural
Variations in Interorganizational Arrangements.” Publius 4, 4, 87–108. 10.2307/3329480.
Ostrom, Elinor, and Gordon P. Whitaker. (1974) “Community Control and Governmental Responsiveness:
The Case of Police in Black Neighborhoods.” in Michael D. McGinnis edited by. Polycentricity and Local
Public Economies: Readings from the Workshop in Political Theory and Policy Analysis. Michigan; University of
Michigan Press.
Ostrom, Vincent. (1972) “Polycentricity (Part 1 and 2).” in Michael D. McGinnis edited by. Polycentricity and
Local Public Economies, 52–74 and 119–138. Michigan; University of Michigan Press.
Ostrom, Vincent. (1987) The Political Theory of a Compound Republic: Designing the American Experiment. Third,
revised edition. Lanham, MD; Lexington Books.
Ostrom, Vincent. (1991a) “Polycentricity: The Structural Basis of Self-Governing Systems.” In The Meaning of
American Federalism, 223–248. San Francisco, CA; ICS Press.
Ostrom, Vincent. (1991b) The Meaning of American Federalism. San Francisco, CA; ICS Press.
Ostrom, Vincent. (1997) The Meaning of Democracy and the Vulnerabilities of Democracies: A Response to Tocqueville’s
Challenge. Ann Arbor; University of Michigan Press.
Ostrom, Vincent, Charles M. Tiebout, and Robert Warren. (1961) “The Organization of Government in
Metropolitan Areas: A Theoretical Inquiry.” American Political Science Review 55, 4, 831–842.
Parks, Roger B., Paula C. Baker, Larry Kiser, Ronald Oakerson, Elinor Ostrom, Vincent Ostrom, Stephen L.
Percy, Martha B. Vandivort, Gordon P. Whitaker, and Rick Wilson. (1981) “Consumers as Coproducers of
Public Services: Some Economic and Institutional Considerations.” Policy Studies Journal 9, 7, 1001–1011.
10.1111/j.1541-0072.1981.tb01208.x.
Percy, Stephen L. (1984) “Citizen Participation in the Coproduction of Urban Services.” Urban Affairs Review
19, 4, 431–446. 10.1177/004208168401900403.
Polanyi, Michael. (1951) The Logic of Liberty. London and New York; Routledge.
Polanyi, Michael. (1962) “The Republic of Science: Its Political and Economic Theory.” Minerva 1, 1, 54–73.
10.1023/A:1026591624255.
Rajan, Raghuram. (2004) “Assume Anarchy.” Finance and Development 41, 3, 56–57.
Starmans, Christina, Mark Sheskin, and Paul Bloom. (2017) “Why People Prefer Unequal Societies.” Nature
Human Behaviour 1, 4, 1–7.
Tarko, Vlad. (2015) “Polycentric Structure and Informal Norms: Competition and Coordination Within the
Scientific Community.” Innovation: The European Journal of Social Science Research 28, 1, 63–80. 10.1
080/13511610.2014.985194.
Tarko, Vlad. (2017) Elinor Ostrom: An Intellectual Biography. London; Rowman & Littlefield.
Thiel, Andreas, William A. Blomquist, and Dustin E. Garrick, eds. (2019) Governing Complexity: Analyzing and
Applying Polycentricity. Cambridge, UK; New York, NY; Cambridge University Press.
199
Vlad Tarko
Tiebout, Charles M. (1956) “A Pure Theory of Local Expenditures.” Journal of Political Economy 64, 5, 416–424.
Tullock, Gordon. (2005) The Social Dilemma: Of Autocracy, Revolution, Coup d’Etat, and War. Indianapolis;
Liberty Fund.
Wagner, Richard E. (2005) “Self-Governance, Polycentrism, and Federalism: Recurring Themes in Vincent
Ostrom’s Scholarly Oeuvre.” Journal of Economic Behavior & Organization 57, 2, 173–188. 10.1016/j.jebo.
2004.06.015.
Weingast, Barry R. (1995) “The Economic Role of Political Institutions: Market-Preserving Federalism and
Economic Development.” Journal of Law, Economics, & Organization 11, 1, 1–31. http://www.jstor.org/
stable/765068.
Wilson, David Sloan, Elinor Ostrom, and Michael E. Cox. (2013) “Generalizing the Core Design Principles for
the Efficacy of Groups.” Journal of Economic Behavior & Organization 90 (June), S21–S32. 10.1016/j.jebo.
2012.12.010.
Zagorski, Nick. (2006) “Profile of Elinor Ostrom.” Proceedings of the National Academy of Sciences 103, 51,
19221–19223. 10.1073/pnas.0609919103.
200
PART IV
Markets
Introduction to Part IV
When one asks about the ethics of markets, there are certain well-recognized issues. Discussion often
turns towards the justifiability of markets for kidneys, markets for drugs, markets for sex, and markets
for military service. The value of interdisciplinary work is obvious for topics like these. We need to
understand how markets work, which values are at stake, and how market regulations would work.
However, the interdisciplinary study of markets goes beyond this. First, we should not assume that
those goods currently sold in markets should be. Healthcare is a clear case, but education, housing,
and labor are others. Raising questions about goods like these invites a more radical assessment of
economic structures. Second, there are the ever-present questions about how to measure the effects
of marketization and how to weigh the costs and benefits against each other. Third, we should not
assume that markets take only a single form. Because any market is a complex structure of prohi-
bitions, powers, and allowances, there are many ways to structure a market. As such, there is work to
be done on questions of market engineering.
The chapters in this Part IV introduce the interdisciplinary study of markets in two ways. The
first three chapters discuss general concerns with markets, and the latter two go into more detail
about the labor market and housing market. In this way, Part IV itself acts as a bridge between the
parts that precede and follow it. Markets are a particular social structure, and how markets are
organized partly determines a society’s economic structure. As such, it is worth looking at markets in
isolation from these other PPE concerns.
Part IV begins with a chapter by Matt Zwolinski outlining the basic advantages of a market. His
aim is not to argue that there should be unregulated markets for all goods and services. Rather, he
outlines the basic advantages of markets than any defense or criticism of a given market should
recognize. To do that, Zwolinski first explains what a market is and what a market society is. From
there, he divides the discussion into the “economic” and “moral” advantages of markets. The
economic advantages are the incentives that a market creates and the information it provides. Yet,
the moral benefits are more extensive. Markets lead to general enrichment, reduce poverty, en-
courage virtue, respect rights, and enable self-ownership.
No one doubts that people are exploited. There are clear cases when people are coerced into
labor conditions that are obviously exploitative. However, whether labor markets result in ex-
ploitation is a more contentious issue. After all, a true labor market does not force people to work, it
offers them pay in exchange for their labor and they are able to accept or reject the offer. Yet, the
DOI: 10.4324/9780367808983-19
Markets
costs of refusing such an offer can be dire. To determine whether labor markets exploit people, we
need a more general account of exploitation. In her chapter, Vida Panitch starts by distinguishing
two basic approaches for analyzing exploitation. First, transactional approaches see exploitation as a
feature of one-off interactions. For most, the terms of a particular agreement are insufficiently free or
fair. Second, structural approaches recognize exploitation as a feature of how larger social structures
situate people. Some have no choice but to enter into exploitative conditions. After comparing the
advantages and disadvantages of each, Panitch defends a particular version of the transactional ap-
proach. Her approach requires three comparisons; of how well-off each party is prior to a trans-
action, of how much each party is improved by the transaction, and of how much each party gains
relative to those that agree to similar labor. This third, “inter-transactional,” comparison posits that
exploitation exists when a person is offered terms that do not make them as better off as other offers
make those that are better situated. To substantiate and defend this account, Panitch applies it to two
potentially exploitative transactions; pharmaceutical drug testing and surrogacy.
The third chapter focuses on the fact that our understanding of a good is changed when that
good is exchanged in a market setting. The act of donating blood expresses something different
than the act of selling blood. My willingness to drive you to the airport means something different
when I charge you for it. When someone objects to the way that a market can change the meaning
of a good or an action, they raise an issue that is distinct from a simple analysis of the con-
sequences. First, “purely expressive objections” to marketization raise concerns with the way that
the meaning of a good or action changes. Second, “impure semiotic objections” to a market
emphasize the effects that are caused by changing the meaning. In “The Meaning of Markets,”
Brookes Brown examines both purely expressive and impure semiotic objections to market-
ization. In particular, she develops two new points. First, she emphasizes that the meaning of
marketization cannot always be distinguished from the benefits of marketization. Meaning is
sometimes explained by the features of markets rather than merely by social conventions. Second,
Brown emphasizes that marketizing a good can make it more difficult to express other meanings.
For example, the fact that a childcare provider is paid for care work makes it difficult for the
provider express care for a child since gestures of care might be interpreted as simple salesmanship.
All together, Brown emphasizes that the relationship between expression and marketization is not
as contingent and mutable as the debate often treats it.
As a way of coordinating productive activity, labor markets have obvious advantages. They offer
a decentralized way to quickly connect those willing to supply labor with those that want to buy it.
They also incentivize the development of skills that are needed to satisfy people’s needs and whims.
Yet, labor markets can also reproduce and reinforce existing forms of oppression. Gina Shouten
starts her chapter, “Gender and the Division of Labor,” by emphasizing the deep and persistent
inequalities in labor markets across the developed and developing worlds. She thinks through a
solution to this problem by answering four questions. First, what (if anything) is wrong with this
situation? She argues that a gendered pattern of labor activities is not necessarily wrong. What makes
it wrong is when it is the result of remediable injustice. Second, what (if anything) can we do about
it? She surveys a number of possible policy responses to this injustice and focuses on individually
allocated and non-transferable parental leave policies. Third, what may we do about it? Here,
Schouten addresses the concern that such policies would wrongfully impose a set of values. Fourth,
at what cost should we do it? Schouten ends by recognizing that policies that lessen the gendered
division of labor might not accomplish other policy goals.
Kristina Meshelski starts the final chapter of Part IV by pointing out that the good of housing is
unique. It is simultaneously a need, spatially fixed, durable over time, and heterogeneous. This
makes it particularly difficult to study, and the problems with increasing housing supply are parti-
cularly challenging. In “Housing Markets,” Meshelski identifies major injustices with the provision
Markets
and distribution of housing in the United States, then she examines proposals meant to remedy these
injustices. For each, she shows the potential problems and the difficulties of assessing success. For
example, it is commonly argued that eliminating zoning restrictions would enable housing providers
to better satisfy demand. However, developers will build for the most profitable opportunities, and
this does not ensure that those in need will gain access to housing. While rent-controls and public
housing have objectionable features, Meshelski suggests that they might be more appealing than
many think.
16
THE ADVANTAGES OF MARKETS
Matt Zwolinski
1 Introduction
For much of the 20th century, a lively debate raged between the advocates of capitalism and the
advocates of socialism. The capitalists favored an economic system based on private property and
market exchange. The socialists thought that the government ought to not only own and manage
individual firms but also to direct the entire economy in accordance with a central plan.
Today, that debate is over, and the capitalists have won. No governments currently attempt to
centrally plan their economies, and almost nobody – including most self-described socialists – thinks
that it would be a good idea for them to do so.1 Regardless of their place on the political spectrum,
virtually all economists and political theorists today grant that markets ought to play a significant, if
not dominant, role in the organization of society’s economic activities.
Of course, none of this amounts to a universal endorsement of laissez-faire. Market economies
need not be (and none in fact are) completely unregulated economies. Still, it is worth noting how
much things have changed over the last 100 years. Today, the debate is about the moral and
economic limits of markets, not about whether we should have markets at all.2 Markets are now the
default system of economic organization, deviations from which require special justification.
The purpose of this chapter is to survey the economic and moral advantages that have given rise
to the ubiquity of markets. Markets do certain things very well, and it is, therefore, no surprise that
they are so popular. Some of these advantages have to do with the goods and services that markets
bring to us – markets are a highly productive form of economic organization, and thus are valuable
from an instrumental perspective. Other advantages have to do not with what markets produce but
how they produce it – the way markets encourage individuals to treat one another, and the kinds of
character they help people to develop.
In discussing the advantages of markets, this chapter will naturally devote some attention to the
limits of those advantages. Markets might be economically efficient in certain respects, for example,
but not in others. What this paper will not discuss are the external limits of markets – the ways in
which the advantages of markets might be outweighed or counterbalanced by distinct moral or
economic considerations. Consideration of such limits is, of course, crucial for any overall assessment
of the desirability of markets, but it falls outside the scope of this chapter and will therefore be left to
other chapters in this section.
DOI: 10.4324/9780367808983-20 205

Matt Zwolinski
2 Markets and Market Economies

Before talking about the advantages of markets, however, we need to be clear about what markets
are. Historically, markets were physical places in which goods were bought and sold. Today, we use
the term market in a more general sense to refer to any institution (physical, virtual, or otherwise) in
which individuals or collective agents exchange goods or services (Herzog, 2017). Your town might
have a Farmer’s Market that resembles the markets of old. But we can also talk about your whole
town, Amazon, or the European Union as markets.
Central to the concept of a market is the concept of exchange. Markets are based on the reciprocal
exchange of one thing for another, where each party to the exchange is typically motivated by his or
her own self-interest. As Adam Smith famously described the phenomenon in his Inquiry into the
Nature and Causes of the Wealth of Nations,
Give me that which I want, and you shall have this which you want, is the meaning of
every such offer, and it is in this manner that we obtain from one another the far greater
part of those good offices which we stand in need of. It is not from the benevolence of the
butcher, the brewer, or the baker, that we expect our dinner, but from their regard to their
own interest. We address ourselves, not to their humanity but to their self-love, and never
talk to them of our own necessities but of their advantages. (Bk 1, Ch. 2)
Exchange within a market can take the form of barter – the direct exchange of goods and services for
other goods and services. But more commonly, as we will discuss in more detail below, goods and
services within markets are exchanged for money.
If a market is an institution based on exchange, a market economy is an economic system in which
markets play a dominant role in productive activity.3 To say that markets play a dominant role means
that the bulk of goods and services produced within the economy are produced for exchange, rather
than direct consumption or gift. Of course, a significant amount of non-market activity can take place
within a market economy. Families, communes, self-sufficient farmers, and charitable organizations, for
instance, produce and distribute resources, but not typically based on principles of reciprocal exchange.
Markets can be more or less competitive, depending on factors such as the number of buyers and
sellers, the level of cost involved in making transactions, the quality of information possessed by
market participants, and so on. Most of the virtues ascribed to markets in this essay apply specifically
to competitive markets. Markets need not be perfectly competitive in order to realize these virtues –
the economist’s ideal of perfect competition is better understood as an abstract modeling device than
a realistic expectation for actually existing markets.4 But they must at least be competitive enough in
the sense of avoiding monopolies and oligopolies, unreasonably high transaction costs, etc.
There are several important concepts that are closely connected to the idea of a market economy.
The first is private property. As we have seen, markets are based on self-interested exchange. But self-
interested exchange can only take place when people own something to give up, and when they
expect to have ownership rights over what they receive in return. Some private property is thus
arguably a prerequisite of markets. It should go without saying, however, that not everything in a
market economy needs to be private property. A market economy can have substantial levels of
collective or public property, though presumably these will not be the dominant form of property
arrangements in a society where market exchange makes up the bulk of economic activity. Nor
must property rights be absolute, in the sense of giving owners full control over all aspects of their
resources. Property in a market economy can, and often is, burdened by easements, duties to
maintain, to avoid nuisance, and so on.
The next important concept is that of freedom of contract. In order for markets to exist, people must
be permitted to enter into enforceable agreements with one another. This enforcement could take
206
The Advantages of Markets
the form of legal rules backed up by the coercive power of the state. But as both a conceptual and a
historical matter, state-based legal systems are not a necessary condition for the successful functioning
of markets (Benson, 1990). What is crucial is that there be some method of enforcing agreements,
whether centralized or decentralized, and that individuals be permitted to enter into a wide range of
voluntary arrangements with each other. Again, it should go without saying that even within a
market society, not all contracts need be legally permissible or enforceable.5
The final important concept is that of the price system. As we have seen, markets can be based on
either barter or money exchange. In a market economy, however, where market exchange is the
dominant mode of economic activity, barter will be relatively rare. Most commodities in a market
economy will be bought and sold for money, leading to the formation of market prices. As we will
see below, these prices play an important and often under-appreciated role in the efficient func-
tioning of markets.
3 Economic Advantages
Many of the most familiar advantages of markets are economic in nature. Indeed, the most obvious
advantage of markets is their tremendous productivity. For instance, over the last 30 years, as global
markets have expanded and communist regimes have fallen or liberalized, global poverty has di-
minished at a staggering rate. In 1990, 35% of the world’s population lived on less than $1.90 per
day, adjusted for purchasing power (World Bank Group, 2016). By 2013, only 10.7% of the world’s
population was that poor, despite there being almost 2 billion more mouths to feed due to global
population growth. And many of those mouths are not only being fed, but fed well. For the first
time in human history, over half the world’s population is now middle class or wealthier (Kharas and
Hamel, 2018).
But why are markets so productive? In this section, we will discuss some of the factors that
contribute to the incredible productive power of markets. We will begin with the familiar but
important idea that markets generate incentives. We will then turn to the less familiar but possibly
even more significant informational role played by market prices. As we will see, it was this latter
consideration that ultimately proved decisive in the capitalist case against socialism.
3.1 Incentives
Why do people work? Work is hard, and most of us have other things we’d rather be doing with our
time. So why does anyone spend their week working as a lawyer, or a janitor, or a teacher?
The answer is obvious: we do it because it pays. Sure, some people derive some intrinsic sa-
tisfaction from their jobs. But for most of us, most of the time, the overriding incentive to get out of
bed and go to work is a paycheck. In a fully centrally planned economy in which money has been
abolished, like the Soviet Union during the “War Communism” period of 1918–1921, these
monetary incentives would have to be replaced with some other sort of incentive such as the threat
of imprisonment or the withdrawal of access to certain collectively owned goods and services.6
But thinking about market incentives in terms of incentives to work leaves us with an incomplete
and partially misleading understanding. After all, not all kinds of work are incentivized in a market
system. The market provides no incentive to work in ways that produce no useful product, such as
digging a giant hole and filling it in again. Furthermore, markets incentivize behavior that can’t
really be accurately characterized as work, like selling your house when your kids have moved off to
college and you don’t need the extra rooms anymore.
The right way to understand market incentives, then, is not in terms of work but in terms of
mutually beneficial exchange. Markets incentivize people to act in ways that others will pay them for.
Sometimes that means doing physical labor. But sometimes it means selling your car or renting out a
207
Matt Zwolinski
spare room in your apartment. The important point is that markets encourage people to do things
that are valuable enough to others that they are willing to give up something valuable in exchange.
You get something you want; they get something they want.
Up until just over a hundred years ago, the mutually beneficial nature of market interactions was
poorly understood, even by such economic geniuses as Adam Smith. The reason is that prior to the
“Marginal Revolution” of the late 19th century, most economists subscribed to a faulty theory of
value which led them to misunderstand what happened when two people engaged in trade.
Economists used to think that value was something objective – that it was a property of the goods that
were being traded. Most commonly, they thought that value was a function of the amount of labor
that went into producing the good. This led them to think that exchange only made sense when two
items had equal value, or when one party had the power to compel the other to enter into an unequal
exchange.7
Today, we understand that economic value is subjective. Books, found objects, and works of art are
valuable not because of how much labor went into them (it’s possible to work very hard and produce
something with no value whatsoever), but rather because they are valued by some economic agent. And,
of course, it’s possible for different economic agents to assign different values to the same object.
Indeed, it is precisely this differential valuation that makes market exchange possible. Two parties enter
into exchange only if each party values what they receive more than what they give up. A hungry man
values a sandwich more than the $5 he pays for it. The sandwich shop values the $5 more highly than
the sandwich they sell. Both parties walk away with more value than they started with – an outcome
that would be impossible if value was an objective property of the objects being traded.
Markets thus incentivize positive-sum interactions – interactions that generate a surplus of value
and thus make it possible for both parties to be made better off. This is a crucially important point,
with implications far more significant than the mere incentive to work. For it is also possible –
indeed, all too common – for human beings to interact with each other in zero- or even negative-
sum ways. People can enrich themselves by stealing from or enslaving other human beings. But this
mode of interaction merely transfers or destroys value. It enriches one person at another’s expense,
and therefore keeps society trapped in a cycle of conflict and poverty. By making it possible to
interact with other human beings in a positive-sum way, markets make it possible for humans to live
together in peaceful cooperation and general prosperity.
3.2 Information
The wages that workers are paid in a market economy provide them with an incentive to engage in
a mutually beneficial exchange of labor for money. But wages do more than provide incentives.
They also convey information. More precisely, they convey information about the relative supply and
demand of different kinds of labor. If wages for doctors are high (because the demand for doctors
relative to the supply of doctors is high), this tells people (including college students deciding on a
career) that more doctors are needed. If construction workers in New Orleans in the wake of
Hurricane Katrina are earning more money than construction workers in California, this sends the
message to the latter group that their services are more urgently needed elsewhere.
A “wage,” of course, is just another way of describing the price of labor. What’s true of wages is
thus true of prices more generally. Prices in a market economy convey information, and this in-
formation helps to ensure that resources tend to be allocated toward their highest-valued use. The
information that prices convey, moreover, is information that is not available to any individual or
discrete group of individuals within society. It is, a Friedrich Hayek famously put it, “knowledge of the
particular circumstances of time and place” that is radically dispersed across all members of society
(Hayek, 1945). This is why Hayek, along with his mentor Ludwig von Mises, argued that economic
calculation is actually impossible under socialism (von Mises, 1981). In the absence of a price system,
208
there is simply no way to rationally compare the value of all the different commodities within an
economy, or all the different uses of the same commodity.
The informational role of prices is distinct from their incentivizing role.8 Even if we assumed that
people were altruistically motivated to work purely for the common good, prices would still be
necessary in order to convey to them how the common good should best be pursued.9 An example
may help to clarify this point. Suppose that Aria has an unusual skill that enables him to create a
high-value product, H. Alternatively, Aria, along with many other people, could also engage in the
low-skilled labor necessary to create a lower-valued product, L. Because H creates more value for its
consumers than L, those consumers will, all else equal, be willing to pay more for H than for L.
Because the producer of H can sell it at a higher price, and because there are relatively few laborers
such as Aria with the skill to produce H, the producer of H will, again all else equal, be willing to
pay a relatively higher wage to Aria. This higher wage, of course, will serve as an incentive for Aria
to take the job. But put aside the issue of incentive for the moment and focus on the problem of
information Aria faces. Even if Aria was perfectly altruistic and was willing to work for free in order
to maximize social value, he would have no way of knowing where his marginal contribution would
be the greatest without the information conveyed in prices. The wage serves to summarize in-
formation about consumer demand, opportunity costs, and relative supply and demand that no
individual or organization (including the government) could hope to obtain in any other way.
The informational role of prices thus helps markets to produce economically efficient outcomes,
where efficiency is understood in terms of maximizing outputs relative to a given set of inputs.10 For
economists, these outputs are generally understood in terms of utility or preference-satisfaction. So,
for example, if A values a resource more than B, A will (all else equal) be willing to pay more than B
for that resource. Thus, a system that allocates that resource according to market prices will (again,
all else equal) tend to distribute the resource to the person who values it most – who gets the most
utility out of it. To this intuitive idea, economists have added more formal proofs that, under certain
conditions, markets will produce efficient outcomes.11
Of course, prices do not perfectly align individual interests with the common good. It’s true that
people generally pay for what they find useful. But sometimes what is useful to one person creates
harmful spillover effects on others. Consider a factory that creates both useful products and air pollution
that makes some people in a nearby town sick. The factory responds to consumers’ preferences for more
goods by making more products. But the preferences of the townspeople for less pollution is ignored.
Similarly, the price system gives greater weight to the preferences of those with more wealth than it does
to those who have less. Someone who is poor might get much more enjoyment out of a box seat at a
baseball game than a millionaire, but if he cannot afford the ticket, his preference carries no weight.
The information conveyed by the price system is therefore not perfectly complete. Negative (and
positive) externalities and severe budget constraints thus constitute forms of market failure – areas where
markets left to their own will generate economically inefficient outcomes. When markets fail, it is
possible that government regulation could be justifiable on purely economic grounds. Government
action could, in principle, generate a more efficient outcome than market activity alone.12
Still, it is worth keeping this point in context. The fact that markets fail to be perfectly efficient is
no argument at all against the market economy. After all, no non-market system is perfectly efficient
either. So the market economy might very well be the best that we can do. And even if certain
market failures do warrant government regulation, it is worth bearing in mind that a regulated
market economy is still a type of market economy.
4 Moral Advantages
As the failure of central planning became clear in the late 20th century, most observers conceded
that socialism could not compete with a market economy on strictly economic grounds. But morality
209
Matt Zwolinski
was potentially a different matter. Even if markets are economically efficient, it is still possible that
they are deeply immoral: that they feed on or exacerbate human selfishness, that they wreck the
environment, that they generate massive inequalities, and so on. In the face of such objections, is
there a moral case that can be made for markets and the market economy?
4.1 Consequences
Moral arguments on behalf of markets, as Amartya Sen has noted, tend to come in two forms (Sen,
1985, 2). One form points to the moral significance of the beneficial consequences produced by
markets. Utilitarian arguments are one argument of this sort, but as we will see in this section, not all
arguments that focus on consequences are properly classified as utilitarian. The other form of ar-
gument draws on claims about antecedent rights and liberties. According to these arguments,
markets are morally justified not because of the consequences they produce but rather because they
are consistent with or required by individual rights. We will examine these arguments in section 4.2.
4.1.1 Enrichment
The economic arguments on behalf of markets discussed in the previous section of this chapter
already give us insight into one of the main beneficial consequences of markets: markets are
spectacularly productive. It is largely because of the expansion of markets that, starting around the
beginning of the 19th century, the world began to experience what the economist Deirdre
McCloskey has called the “Great Enrichment.”13 Despite a rapidly expanding population, GDP per
capita in much of the world has increased dramatically, up from a baseline of around $2 or $3 per
person per day, to upwards of $100 per day. On the whole, the world now produces 70 times more
goods and services than it did in 1800. That means more food and more clothes. But also, of course,
new things that no one could have dreamed of in the 19th century, from air conditioning and
universal indoor plumbing to jet planes and smartphones. Rising productivity has also meant longer
life expectancies, more leisure, more education, less disease and infant mortality and (perhaps sur-
prisingly) less pollution.
As we have seen, economists naturally tend to evaluate markets according to standards of effi-
ciency – whether allocative, Paretian, or dynamic. It is worth noticing, however, that the efficiency
of markets and their capacity to enrich us are two quite different things. Nobody thinks that
European markets during the Industrial Revolutions of the 18th and 19th centuries were anywhere
close to perfectly efficient. The natural and legal barriers to competition were extensive and have not
disappeared entirely even to this day. But just like an imperfectly efficient automobile engine can still
take you on a cross-country trip, so too can (and have) imperfectly efficient markets generate
fantastic increases in wealth.
As remarkable as the Great Enrichment has been, however, not all parts of the world have
benefitted from it equally. In 2020, GDP per capita in the United was over seven times that of
Mexico (World Bank Data, 2021). Even within wealthy countries like the United States, there is
significant inequality of wealth and income, with some struggling to make ends meet on minimum
wage jobs, while others bask in luxury. Statistics regarding aggregate economic growth do not tell us
how that wealth is distributed, either within a country or between countries. Is there anything that
defenders of markets can say about the important question of distributive justice?
4.1.2 Poverty and Inequality

We can begin by noting that, as a historical matter, some of the most important defenders of markets
have in fact framed their justifications not in terms of the aggregate wealth they create but
210
specifically in terms of their tendency to benefit the average citizen, especially the poor. Adam
Smith, for instance, held that the measure of a nation is not its overall wealth but rather the wages of
“the labouring poor,” and held it to be a matter of equity that “they who feed, cloath and lodge the
whole body of the people, should have such a share of the produce of their own labour as to be
themselves tolerably well fed, cloathed and lodged” (Smith, 1904, 80; Fleischacker, 2004). Taken
out of context, this statement might appear to be a plea for government intervention in the market
in order to redistribute wealth from the rich to the poor. In the context in which it appears,
however, it is clear that Smith intended his statement as a defense of markets as they actually function.
In a healthy, progressive, growing economy, the poor get richer. And Smith wanted to convince his
readers that this was a good thing.
Markets generate inequality, but inequality and poverty are two very different things. Inequality
is a relative measure of how one person or group is faring with respect to some other person or
group. Poverty is trickier to define but is usually understood as a more absolute measure of whether
someone has sufficient resources to meet their basic human needs. It’s possible to have a society in
which both inequality and poverty exist at high levels. But it’s also possible to have high inequality
without any poverty at all. Imagine an island populated exclusively by millionaires, billionaires, and
trillionaires.
On the domestic level, both theory and empirical evidence seem to suggest that markets generate
substantial inequalities. Theoretically, we know that markets reward some people much more
lavishly than others, and indeed that these inequalities in income are a necessary part of the in-
centivizing and informational role of the market price system. Empirically, we can see that inequality
is significantly higher in market-friendly countries like the United States (Gini index of 41.4) than it
is in more “socially democratic” countries like Denmark (28.2), Finland (27.3) or France (32.4).14
But markets also generate tremendous wealth, and this wealth has done much to dramatically
reduce poverty over time. Recall the point made earlier that in 2020, GDP per capita in the United
States was over seven times higher than GDP per capita in Mexico. This is a striking contrast. But a
contrast that reflects a far more significant reality for the citizens of Mexico is that GDP per capita in
Mexico in 2020 was seventeen times higher than it was in 1965. Since 1985, the percentage of Mexico’s
population living on less than $1.90 per day has declined from 8.3% to less than 2% (World Bank Data,
2021). And what is true of Mexico is true of other developing countries as well, especially to the extent
that those countries have become more enmeshed in networks of global trade (Bartley Johns et al., 2015).
Moreover, the relationship between markets and increased inequality on the international level is
less clear than first appearances suggest. It is true that the United States has higher levels of inequality
than Denmark, Finland, and many other countries. But Denmark and Finland are, after all, market
economies too. Indeed, the libertarian Fraser Institute ranks Denmark as the 16th and 22nd freest
economies in the world (the United States is sixth), and Denmark actually scores as freer than the
United States in several respects, including international free trade and secure property rights. The
Nordic “social democracies” are not actually socialist in any standard sense of the term. They are simply
market economies with a relatively generous welfare state.15 More pointedly – they are generous
welfare states that are made possible by the wealth generated by the market economies on which they
supervene.
4.1.3 Virtue
The material benefits of markets are striking. But markets do more than create products. They also
shape the characters of the people who participate in them.
Many critics of markets have seized upon this point, arguing that markets make people more
selfish. The idea behind this criticism is natural enough. After all, even many proponents of markets
claim that self-interest is or ought to be the primary motivating force of market participants.16
211
Matt Zwolinski
If markets reward selfishness, isn’t it natural that participation in markets would tend to make people
more selfish?
In fact, the answer seems to be a resounding “no.” Consider the results of experiments conducted
around the globe involving the Ultimatum Game. In this game, two individuals who are anonymously
paired and must figure out how to divide a sum of money between them. One of the subjects is
allowed to make a proposal to the other. They can make any proposal they like: 50/50, 100/0, or 99/1.
The other subject can either accept the proposal or reject it. If they accept it, each subject walks away
with the money as agreed upon. If they reject it, however, both parties walk away with nothing.
The selfish choice in this experiment – the choice predicted by the economic model of homo
economicus, rationally self-interested behavior – is clear. If you’re the proposer, offer the other party
the lowest amount possible greater than zero – a 99/1 split. And if you’re the receiver, the smart
move is to accept anything that’s offered to you. After all, something is better than nothing, and if you
reject a 99/1 split, nothing is precisely what you get.
And this, in fact, is just about how individuals in non-market societies tend to play the game.
Hunter-gatherers from Tanzania and Indonesia, herders from Siberia and Kenya, and subsistence
farmers from South America and Africa tend to offer little, and to accept whatever little they’re
offered (Henrich, 2020, Ch. 9). In contrast, greater exposure to market economies leads subjects to
make more even proposals, and to reject lopsided ones.
Why is this? Market economies encourage the development of market norms, and these norms
include a propensity to cooperate and play fairly with anonymous strangers. Market integration
encourages people to disregard personal preferences that might stand in the way of positive-sum,
profit-maximizing exchange. In doing so, markets change who we are, both behaviorally and
psychologically. As Montesquieu wrote in The Spirit of the Laws, “Commerce is the cure for the most
destructive prejudices; for it is almost a general rule that wherever manners are gentle there is
commerce; and wherever there is commerce, manners are gentle.” Deirdre McCloskey has argued
that it is precisely the emergence of such “bourgeois virtues” that paved the way for the Great
Enrichment of the 18th and 19th centuries (McCloskey, 2006, 2017).
4.2 Rights
In the last section, we surveyed some of the beneficial consequences that markets produce for
societies and the individuals who participate in them. But is there anything more than this to be said
on behalf of markets? In particular, can it plausibly be argued that individuals have a moral right to
participate in markets? If the government sought to prohibit “capitalist acts between consenting
adults,” would it be doing its subjects an injustice (Nozick, 1974)?
4.2.1 Property and Contract

As we have seen, markets presuppose some form of property rights. But property rights, it might also
be argued, entail some form of market. If you fully own something, part of what that includes is the
right to transfer it to someone else. If you own some lemons, you have the right to make lemonade,
and the right to sell that lemonade to others. If your neighbor, or the government, were to forcibly
stop you from doing that, they would be wrongfully interfering with your property rights.
For libertarians like Robert Nozick, markets are thus a matter of justice.17 And the justice of
markets is, ultimately, rooted in individual self-ownership. Because individuals own their bodies and
their labor, they own what they produce with that labor. Ownership of the products of labor entails a
right to give away or exchange those products, and any third-party interference in those exchanges
constitutes a violation of individual rights. Property rights entail freedom of contract, and freedom of
contract is the cornerstone of market exchange.
212
This argument can be challenged at a number of points. Left-libertarians will agree that we own
our labor but deny that this ownership can ground ownership in external resources like land.18
They, therefore, agree that it would be unjust for the government to restrict what individuals do
with their bodies, but they have no objection to the regulation or taxation of other material re-
sources. Others, like Liam Murphy and Thomas Nagel, challenge the naturalness of property rights
altogether (2002). For this group, property rights are a creation of the law, and thus when the law
“restricts” property rights (including the right to exchange), it is not infringing on individuals’ rights
at all. The “restrictions” are simply part of the specification of the rights that the government itself
creates.19 Finally, it is possible to deny that individuals have maximally extensive rights of property and
contract without denying the existence of those rights altogether. If the house you own is an
officially designated historic site, there will probably be restrictions on how you can use, modify, or
transfer it. Nevertheless, the house is still yours.20
4.2.2 Self-Authorship
Despite these objections, there seems to be something intuitively right about the idea that some forms of
economic regulation are unjust, and not merely wrong because they produce bad consequences. When
the state of Texas prevented South Asian eyebrow threaders from practicing their trade unless they
completed a lengthy and largely irrelevant training designed for cosmetologists, the problem wasn’t
merely that the government was undermining economic efficiency (Sibilla, 2015). It was doing an
injustice to the individuals it restricted by preventing them from using their skills to make a living.
One way of diagnosing this injustice is to say that the regulation violated individual self-
ownership. But that objection seems incapable of discriminating between more and less significant
restrictions on individual liberty. A small tax imposed on the products of an individual’s labor
violates self-ownership, at least according to standard libertarian theories. But such a tax seems much less
worrisome to most non-libertarians than laws that seriouisly hinder someone’s ability to earn a living.
Part of the reason is that the way one earns a living is bound up in a deeply important way with
one’s identity as a person. By denying someone the freedom to work in their chosen profession, we do
more than merely deprive them of the opportunity to earn an income. We also deprive them of the
ability to develop and express themselves in the world according to their own idea of the good life.
One important and often under-appreciated advantage of markets, then, is that they provide
a wide range of ways for individuals to be in the world. And the more extensive and open the
market, the more ways there can be. As Adam Smith noted, the division of labor is limited
by the extent of the market. And in an extensive market, it is possible for individuals to earn a
living as engineers, antique bookstore workers, part-time gig drivers, dog walkers, entrepreneurs,
and countless other options. Economic activity, as Loren Lomasky has argued, is one important
venue in which individuals define and pursue important life projects (Lomasky, 1987). And this fact,
as John Tomasi has argued, gives us good reason to consider economic liberties as morally on a par
with other basic liberties of free speech and free association (Tomasi, 2011).
4.2.3 Desert
Before we conclude this discussion of the moral advantages of markets, one final argument merits
some attention. It is commonly believed that markets reward hard work and talent, and that market
outcomes, therefore, track some underlying facts about what people morally deserve. The individual
who earns money through her labor, or the entrepreneur who creates a profitable business by seizing
some unmet opportunity, is usually thought to deserve their income. To take this income from
them and give it to someone else who has not earned it would be a kind of injustice on this view. If
213
Matt Zwolinski
markets naturally give people what they deserve, then any interference with the market order will
upset that balance, leaving some with less than they deserve and some with more.
Popular though this argument may be among the general public, it has received surprisingly little
support from scholars. The problem comes down to luck. John Rawls famously objected to the line
of argument developed above by arguing that “[i]t seems to be one of the fixed points of our
considered judgments that no one deserves his place in the distribution of native endowments, any
more than one deserves one’s initial starting place in society.” Both natural talents and one’s initial
social environment are factors over which one had no control, and for which one can accordingly
take no credit. Indeed, Rawls goes further and claims that even our choices are largely a matter of
luck, for those choices spring from our character, which “depends in large part on fortunate family
and social circumstances for which [we] can claim no credit” (Rawls, 1971, p. 104).
Rawls used this argument to support a theory of justice that required severe constraints on the
market order. Surprisingly, though, many of the most influential supporters of the market actually
agreed with Rawls about the relationship between markets and desert. In 1923, the Chicago
economist Frank Knight similarly criticized the “just deserts” rationale for markets. Milton Friedman
likewise conceded that “most differences in status or position can be regarded as the product of
chance at a far enough remove.”21 Friedrich Hayek thought that it was a pernicious mistake to
regard markets as rewarding moral merit (Hayek, 1960, Ch. 6). And Robert Nozick likewise re-
jected the idea that the distribution of wealth or income in the market follows any sort of “pattern”
(Nozick, 1974, pp. 155–160). For these individuals, markets are highly desirable, but emphatically
not because they give people what they deserve.
We can’t claim any credit for our talents, nor for the conditions of supply and demand that render
those talents remunerative (or otherwise). But for Hayek, these facts are no mark against the market
order. The function of prices (and wages), according to Hayek, is not to give people what they
deserve, but to induce socially useful behavior. Prices are forward-looking, not backward-looking.
Similarly, Nozick accepted the fact that markets don’t give us what we deserve because he thought
that they give us what we are entitled to. Individuals have a right to gifts that they are given by their
parents, or wages that they are paid by their employer, not because they deserve them in ant deep
sense but simply because the money was peacefully transferred to them by someone who had a right
to dispose of it in any way they saw fit.
5 Conclusion
This essay has surveyed some of the moral and economic advantages of markets. It has not surveyed
the many real and purported disadvantages or limits of markets, let alone tried to sort out the relative
weights of advantages and disadvantages so as to arrive at an all-things-considered judgment about
the desirability of markets. The analysis here is thus incomplete, in addition to being one-sided.
It is worth noting in closing, however, that an all-things-considered judgment about the de-
sirability of markets as such is often not what is needed in thinking about public policy. Rarely, if
ever, do voters or legislators decide whether or not to have a “market economy” or something else.
But we are often called to make judgments about particular markets: whether marijuana should be
legal or criminalized, whether homeowners in a residential neighborhood should be able to rent
their homes through AirBnB or not. In thinking about such issues, it is worth bearing in mind that
the advantages surveyed in this article (and the disadvantages ignored) can be stronger or weaker
depending on what specific kind of market we’re talking about. There may be good reasons for
allowing, extending, or liberalizing one sort of market while there are equally good reasons for
banning, restricting, or regulating another. Shifting the focus from markets in general to particular
forms and instances of markets can simplify the normative and empirical analysis in some respects,
even if it complicates it in others.
214
Notes
1 For an overview, see Lavoie, 1981.
2 See, for example, Satz, 2010, Sandel, 2012; Brennan and Jaworski, 2016.
3 The concept of a market economy is related to, but distinct from, the concept of “capitalism.” On the latter,
see Gaus, 2010.
4 The ability of imperfectly competitive markets to yield efficient outcomes has been demonstrated experi-
mentally in a laboratory setting. See, for example, Smith, 1962; Gode and Sunder, 1993; and Axtell, 2005.
5 The doctrine of unconscionability, for instance, prohibits certain agreements on both substantive and
procedural grounds. For a helpful discussion, see Wertheimer, 1996, Ch. 2.
6 Theoretically, rewards could be used instead of punishment. In practice, however, punishments were much
more common. For a discussion of War Communism in the Soviet Union, see Pipes, 2011, Ch. 8.
7 This idea was popular not only with economists but with social reformers of various political stripes. In the
United States, for example, it was developed by Josiah Warren, through whom the idea went on to have a
tremendous influence on later anarchist thinkers such as Benjamin Tucker. See Warren, 1852.
8 This paragraph borrows from Zwolinski, 2020.
9 This is the approach taken by the contemporary socialist Joseph Carens (1981).
10 The definition above captures the standard notion of static or allocative efficiency. Economists sometimes
employ other notions of efficiency such as Pareto-efficiency or dynamic efficiency. For an overview, see
Buchanan, 1988.
11 See the first theorem of welfare economics and its proof in Mas-Colell et al., 1995, Ch. 16.
12 Though perhaps not. As Henry Sidgwick noted, “It does not follow that whenever laissez faire falls short
government interference is expedient; since the inevitable drawbacks of the latter may, in any particular
case, be worse than the shortcomings of private enterprise” (1887, 414). Sidgwick’s point was developed in
the 20th century by the public choice school of economics, which studies the existence of “government
failure” as the flip side of “market failure.” See, for example, Buchanan and Tullock, 1962.
13 See McCloskey, 2017, Ch. 1. The figures in the remainder of this paragraph are drawn from her discussion.
14 The Gini Index is a commonly used measure of inequality, with higher numbers (up to 100) reflecting
higher levels of inequality.
15 For more on Social Democracies, see Jeppe von Platz’s chapter on “Social Democracy” in this volume.
16 In Adam Smith, the claim is descriptive. In Ayn Rand, the claim is normative (1964).
17 For an overview of libertarian thought on the relationship between justice and economic activity, see
Zwolinski and Ferguson, 2022. Also see Eric Mack’s chapter on “Property Rights and Justice in Holdings”
in this volume.
18 For an overview, see Vallentyne, 2000.
19 For more on Left-Libertarianism and Nagel and Murphy’s argument, see section 5 in Bas van der Vossen’s
chapter on “Property” in this volume.
20 For an influential critique of libertarian theories of property along these lines, see Fried, 2004.
21 See Knight, 1923; Friedman, 1962. For an overview and discussion of Rawls’ indebtedness to Knight, see
Lister, 2017.
References
Axtell, R. (2005) “The complexity of exchange,” The Economic Journal 115, 504, F193–F210.
Bartley Johns, Marcus, Paul Brenton, Massimiliano Cali, Mombert Hoppe, and Roberta Piermartini. (2015)
The Role of Trade in Ending Poverty. Washington, DC; World Bank Group.
Benson, Bruce. (1990) The Enterprise of Law: Justice Without the State. San Francisco, CA; Pacific Research Institute.
Brennan, Jason, and Peter Jaworski. (2016) Markets Without Limits: Moral Virtues and Commercial Interests. New
York; Taylor and Francis
Buchanan, Allen. (1988) Ethics, Efficiency, and the Market. New York, NY; Rowman and Littlefield.
Buchanan, James, and Gordon Tullock. (1962) The Calculus of Consent. Ann Arbor, MI; University of Michigan.
Carens, Joseph H. (1981) Equality, Moral Incentives, and the Market: An Essay in Utopian Politico-Economic Theory.
Chicago; University of Chicago Press.
Fleischacker, Samuel. (2004) A Short History of Distributive Justice. Cambridge, MA; Harvard University Press.
Fried, Barbara. (2004) “Left-Libertarianism: A Review Essay,” Philosophy and Public Affairs 32, 1, 66–93.
Friedman, Milton. (1962) Capitalism and Freedom. Chicago; University of Chicago Press.
215
Matt Zwolinski
Gaus, Gerald. (2010) “The Idea and Ideal of Capitalism,” in George Brenkert and Thomas Beauchamp, eds.,
The Oxford Handbook of Business Ethics. Oxford; Oxford University Press.
Gode, D. K., and 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.
Hayek, Friedrich. (1945) “The Use of Knowledge in Society,” American Economic Review 35, 4, 519–530.
Hayek, Friedrich. (1960) The Constitution of Liberty. Chicago; University of Chicago Press.
Henrich, Joseph. (2020) The WEIRDest People in the World: How the West Became Psychologically Peculiar and
Particularly Prosperous. New York; Farrar, Straus and Giroux.
Herzog, Lisa. (2017) “Markets,” The Stanford Encyclopedia of Philosophy. Edward N. Zalta, ed., available at
https://plato.stanford.edu/archives/fall2017/entries/markets/
Kharas, Homi, and Kristofer Hamel. (2018) “A Global Tipping Point: Half the World is Now Middle Class or
Wealthier” Brookings, September 27, 2018.
Knight, Frank. (1923) “The Ethics of Competition,” The Quarterly Journal of Economics 37, 579–624.
Lavoie, Donald. (1981) Rivalry and Central Planning: A Re-Examination of the Debate over Central Planning under
Socialism. New York, NY; New York University Press.
Lister, Andrew. (2017) “Markets, Desert, and Reciprocity,” Politics, Philosophy & Economics 16, 1, 47–69.
Lomasky, Loren E. (1987) Persons, Rights, and the Moral Community. Oxford; Oxford University Press.
Mas-Colell, A., M. Whinston, and J. Green. (1995) Microeconomic Theory. Oxford; Oxford University Press.
McCloskey, Deirdre. (2006) The Bourgeois Virtues: Ethics for an Age of Commerce. Chicago; University of Chicago
Press.
McCloskey, Deirdre. (2017) Bourgeois Equality: How Ideas, Not Capital or Institutions, Enriched the World.
Chicago, IL; University of Chicago Press.
Mises, Ludwig von. (1981) “Socialism: An Economic and Sociological Analysis,” trans. J. Kahane. Auburn, AL;
The Ludwig von Mises Institute.
Nagel, Thomas and Liam Murphy. (2002) The Myth of Ownership. Oxford, UK; Oxford University Press.
Nozick, Robert. (1974) Anarchy, State, and Utopia. New York, NY; Basic Books.
Pipes, Richard. (2011) Russia Under the Bolshevik Regime. New York, NY; Knopf Doubleday.
Rand, Ayn. (1964) “The Nature of Government,” in The Virtue of Selfishness. New York, NY; Signet.
Rawls, John. (1971) A Theory of Justice. Cambridge, MA; Harvard.
Sandel, Michael. (2012) What Money Can’t Buy: The Moral Limits of Markets. New York; MacMillan.
Satz, Debra. (2010) Why Some Things Should Not Be for Sale: The Moral Limits of Markets. Oxford University
Press.
Sen, Amartya. (1985) “The Moral Standing of the Market,” Social Philosophy and Policy 2, 2.
Sibilla, Nick. (2015) “Texas Supreme Court Strikes Down Useless Eyebrow Threading License” Institute for
Justice. Available at: IJ case: https://ij.org/press-release/texas-supreme-court-strikes-down-useless-eyebrow-
threading-license/ (accessed 20 July 2021).
Sidgwick, Henry. (1887) The Principles of Political Economy, (2nd. Edition). London, UK; MacMillan and
Company.
Smith, Adam. (1904) An Inquiry into the Nature and Causes of the Wealth of Nations. Edwnin Cannan, ed.,
London; Methuen.
Smith, Vernon. (1962) “An Experimental Study of Competitive Market Behavior,” Journal of Political Economy
70, 2, 111–137.
Tomasi, John. (2011) Free Market Fairness. Princeton, NJ; Princeton University Press.
Vallentyne, Peter. (2000) “Left-Libertarianism: A Primer,” in Peter Vallentyne and Hillel Steiner, eds., Left-
Libertarianism and Its Critics: The Contemporary Debate. New York; Palgrave.
Warren, Josiah. (1852) Equitable Commerce: A New Development of Principles as Substitutes for Laws and
Governments, for the Harmonious Adjustment and Regulation of the Pecuniary, Intellectual, and Moral Intercourse of
Mankind, Proposed as Elements of a New Society, American Classics of History and Social Science. New York,
NY; Burt Franklin.
Wertheimer, Alan. (1996) Exploitation. Princeton, NJ; Princeton University Press.
World Bank Data. Available at: data.worldbank.org (accessed 9 July 2021).
World Bank Group. (2016) Taking on Inequality. Washington, DC; World Bank.
Zwolinski, Matt. (2020) “Equal Pay: A Floor Not a Ceiling,” in Debating Equal Pay: Economy, Practicability and
Ethics, by Anders Örtenblad. Switzerland; Palgrave.
Zwolinski, Matt, and Benjamin Ferguson. (2022) The Routledge Companion to Libertarianism. New York;
Routledge.
216
17
EXPLOITATION
Vida Panitch
1 Introduction
Exploitation is a distinctive kind of unfair advantage-taking unique to market exchanges. Disagreement
among exploitation theorists pertains not to the defensibility of this generalized account, but to the
nature of the advantage-taking in question and its normative implications. Theoretical accounts of
exploitation tend to align with either the liberal or the Marxist tradition. Exploitation theorists who take
a broadly liberal approach seek to determine the presence or absence of exploitation by asking whether
particular market transactions between discrete parties have been marred by impaired consent or unfair
distributive outcomes. This transactional approach has its roots in a broadly Lockean account of fair
market price and a distinctly Kantian account of autonomous agency. Exploitation theorists of the
structural tradition see exploitation as resulting from, and occurring within, socio-economic systems of
hierarchy, domination, and oppression. Structuralists take the Marxist account of class exploitation as
their point of departure yet reject the labor theory of value on which it originally depended.
Both models face challenges. Transactionalists struggle to account for the institutional inequalities
that consistently make some groups more exploitable than others. Some also appeal to ideal market
conditions as a metric against which to determine the presence of wrongful advantage-taking. While
these abstract considerations may help resolve philosophical puzzles about the culpability of greedy
umbrella salesmen in a storm, for example, or water hoarders in a desert, they have proven less
helpful with respect to explaining and addressing real-world exploitative practices; yet the non-ideal
solutions that transactionalists have proposed are more problematic still. Structuralists, meanwhile,
lose sight of the agential aspects of exploitation, indicting systems as a whole, or groups within them,
rather than individual perpetrators. They also tend to explain exploitation by appeal to another
concept of social injustice that is no less complex and in need of theorization: domination, op-
pression, degradation, and the like. To collapse exploitation into one of these other equally complex
wrongs is, as Kymlicka (2002: 184) puts it, “not to gain an insight but to lose a word.”
My aim in this brief chapter is twofold. The first is to lay out and motivate objections to both
structural and transactional accounts of exploitation. The second is to defend a certain kind of
transactional account, albeit one that is properly sensitive to the most important of the structuralist’s
insights. I will argue that transactionalism must contain a commitment to assessing the fairness of
distributive outcomes not only within transactions but across them; that is, to making not only intra-
transactional but also inter-transactional assessments of fairness. Such a commitment would enable
the transactionalist to address structural concerns, as well as the problems plaguing ideal transactional
DOI: 10.4324/9780367808983-21 217

Vida Panitch
accounts and the leading non-ideal solution that has thus far failed to resolve them. I will show this
commitment to be particularly useful in identifying and addressing global exploitative practices by
exploring its implications for global clinical research and global gestational surrogacy.
2 Structuralism and Transactionalism

Theorists in the structural tradition see exploitation as resulting from and endemic to socio-
economic systems of domination and oppression. On the Marxist (1847, 1867) account, in which this
tradition has its roots, workers are exploited by a system of property rights in which they effectively
have none and are thereby forced to sell their labor power to the capitalist for less than the true value
of the commodities they produce, as determined by the amount of labor required to produce them.
Theorists attracted to the structural nature of Marx’s account of exploitation have had to propose
strategies by which to salvage it from the widescale rejection of the labor theory of value on which it
depends. For structural exploitation theorists, the solution lies in broadening the focus away from
determining the distributive fairness of micro-level wage-based interactions between workers and
owners at the point of production, and towards the macro-level structural features of unjust social
relations in which a vulnerable underclass is rendered inherently exploitable due to their mem-
bership therein.
John Roemer (1982) advocates for this kind of conceptual shift by arguing that we ought to
conceive of exploitation as a form of social parasitism inevitable under a capitalist mode of property
relations whereby one group (owners) is made better off by the existence of a second group
(workers), while the second group is made worse off by the existence of the first. For Nicholas
Vrousalis (2013), exploitation is a particular form of social domination, which occurs when one
party instrumentalizes a vulnerable other with whom he is embedded in a systemic power relation,
explicitly so as to extract a net benefit for himself. For Iris Marion Young (1990), who offers
arguably the most seminal contemporary version of structuralism, exploitation is a social by-product
of multiple interlocking systems of oppression. Oppressive social regimes deny members of some
social groups equal standing by entrapping them in relations of subordination based not only in class
but also race and gender. For Young, exploitation is the inevitable steady transfer of the social
contributions of an oppressed group to the benefit of their oppressor.
On all of these accounts, the focus is on systems of power and the relations of vulnerability they
generate between members of privileged and underprivileged social groups. As McKeown (2017: 35)
puts it, structuralism posits exploitation as “the forced transfer of productive powers from groups
positioned as socially inferior to the advantage of groups positioned as socially superior.” Structural
accounts have an easy time grounding general worries about such things as sweatshop labor, migrant
labor, and gendered domestic and care labor. But even as structuralism accounts for many of our
considered judgments about the evidently unfair advantage-taking that occurs, for example, when
global corporations open production facilities in the developing world to skirt labor laws and minimum
wage guarantees at home, or when first world pharmaceuticals enlist developing world research
subjects into clinical trials so as to avoid having to provide their control subjects the standard treatments
they would be obliged to offer at home, the structural model faces serious challenges.
The first of these challenges is that because structuralism sees exploitation as a macro level wrong,
occurring between social groups rather than between individuals engaged in discrete exchanges, the
members of these social groups can be said to be merely acting out the roles they occupy. This robs
the concept of exploitation of its power to indict unfair exchanges that occur between members of
the same social group, and worse, the power to indict individual exploiters who belong to a pri-
vileged social group and who, as such, couldn’t have acted otherwise. Although structuralism offers
us a chance to deride unjust social practices, it denies us the chance to deride the individuals who
participate in and benefit from these practices; the system is at fault instead of its beneficiaries. If any
218
Exploitation
interaction between members of a dominant and subordinate group from which the dominant
member benefits is exploitative, then we can only indict wholesale systems of oppression and the
unfair practices to which they give rise.
This challenge leads to another, which is that we already have a host of normative concepts
available for indicting wholesale systems, as structuralists are themselves all too ready to demonstrate:
injustice, domination, oppression, degradation, etc. These concepts alone constitute harms the
avoidance of which provides ample grounds for structural reform. If an exploitative practice is
wrongful because it is made possible by an oppressive institutional regime, then it is the regime of
oppression that should be reformed. It is the relevant structural injustice that does the heavy lifting of
explaining not only what exploitation is, but also why it is wrong and what should be done about it.
Structuralists who appeal to another social wrong to both explain the content of an exploitation
claim and its normative force thereby make exploitation a mere middle-man, philosophically
speaking. Accounting for exploitation by appeal to one of these other concepts empties it of its
unique rhetorical and normative force.
By contrast, transactional accounts of exploitation aim to identify both the unique normative
content and force of an exploitation claim with respect to market exchanges between two discrete
parties. Transactionalists hold that the relevant wrongful advantage-taking occurs when there is an
impairment of consent in the process by which two parties agree to interact or an unfairness in the
distributional outcome of their interaction. According to Alan Wertheimer’s seminal account (1996),
what matters with respect to the transactional process is whether one party has lied to or coerced the
other into participating. Lying is fairly straightforward, coercion less so. For coercion to occur, A must
threaten to violate B’s rights, or to render B worse off if B refuses to transact with A on A’s terms
(1996: 25–26). It is not coercive for A to make B an offer that, given her circumstances, it is not in
B’s interests to decline; nor is it coercive for A to make B the offer due to some fact about B’s
disadvantaged circumstances as long as A did not create those circumstances (1996: 27–28). Party A
must threaten to make B worse off if she refuses to interact with A in order for it to be said that B’s
rational decision-making about whether to transact with A has been impaired.
With respect to the transactional outcome, what must be determined is the fairness of the post-
transactional shares of the cooperatively produced surplus. If A benefits from the transaction while B
is made worse off by it, relative to B’s pre-transactional baseline, A has exploited B (Wertheimer,
1996: 18–32). Party A walks away with the cooperative surplus produced through her interaction
with B, while B is denied a share of what she expended time and energy to produce and is thereby
all things considered made worse off as a result of an exploitative exchange. According to
Wertheimer, when exploitation is non-consensual and harmful to one party, the evident advantage-
taking is clearly wrongful. Such transactions not only warrant the making of an exploitation claim
but also elicit the normative force thereof, meaning they should not only be described as exploitative
but interfered with on such grounds (1996: 247–277).
The complicated cases, for transactionalists, are ones in which both parties consent to and benefit
from the exchange, while some unfairness nonetheless persists. These cases constitute ones of mutually
advantageous exploitation. Many of the scenarios we typically decry as exploitative are clear instances
of mutual advantage, including sweatshop labor and global clinical research, along with more discrete
exchanges that transactionalists enjoy puzzling over in which, for example, an umbrella salesman raises
prices during a downpour, or a hiker with extra water offers to share his surplus with a fellow trekker
for an exorbitant fee. When an arrangement is non-consensual and harmful to one party, the work of
the transactional exploitation theorist is easily done; but when an arrangement is both consensual and
mutually advantageous, albeit unevenly so, the question of what to do becomes much less straight-
forward. The interesting task for the transactionalist is to determine when mutual advantage is
nonetheless exploitative and, furthermore, whether intervention is normatively justified.
219
Vida Panitch
Transactionalists offer two distinct albeit-related answers; one that can be said to constitute an
argument from ideal theory and another from non-ideal theory. If we want to say that exploitation is
wrongfully unfair when party A walks away with a disproportionately high share of the co-
operatively produced surplus relative to party B, then some metric is required to determine what
would constitute their fair shares. Wertheimer argues that in order to determine the fairness of a
transactional outcome we must appeal to idealized markets, considering what a fair price and a fair
distributive outcome would look like as determined by two rationally and equally situated parties
bargaining under ideal market conditions (1996: 230–236). The thirsty trekker and the water
hoarder would arrive at quite a different price for a bottle of water under idealized conditions in
which neither has a surplus nor faces imminent deprivation. Transactions from which both parties
benefit, but where one party’s share deviates too considerably from the ideal price, would constitute
wrongful albeit mutually advantageous exploitative exchanges.
The first problem with this approach is that ideal metrics are just that, and thereby leave us
guessing after counterfactuals. The second is that we are also left to wonder just how great a de-
viation from the ideal must be for it to be normatively wrongful. In light of the previous two, the
third problem is that while this ideal has helped shed light on imagined scenarios involving water
hoarders and greedy umbrella salesman, it has been less useful in determining exploitation in
complex non-fictionalized socio-economic practices (Deveaux and Panitch, 2017; Jaggar, 2013).
The fourth problem is that, if we use this ideal as a basis upon which to not only make an ex-
ploitation claim but to thereby justify interference in unfair arrangements, we risk preventing the
worse off from bettering their circumstances, however meagrely, and may thus, as Wood (1995) puts
it, be assigning them to “an even worse fate than being exploited” (156).
This latter worry has led Wertheimer and other transactionalists (Ferguson, 2016; Powell and
Zwolinski, 2012; Zwolinski, 2009) to turn to a non-ideal metric for assessing distributional fairness.
If an ideal metric can tell us when an unfair distribution is normatively wrongful, this doesn’t mean it
should tell us when to interfere under non-ideal conditions, lest we thereby remove an option that is
all things considered superior relative to the worse off party’s status quo. Transactionalists concerned
with this problem offer what are called non-worseness claims (NWC), according to which, so long as
party A has no obligation to interact with party B on any particular terms, and so long as B both
consents to and is made no worse off in some way relative to where she began, the interaction can
still involve unfairness but not of the sort that warrants intervention (Powell and Zwolinski, 2012;
Wertheimer, 1996; Zwolinski, 2009). The purpose of an NWC is to ensure that in our well-
intentioned drive to protect the vulnerable from unfair advantage-taking we don’t thereby deprive
them of opportunities to engage in mutually advantageous exchanges that would enable them to
improve their lot.
The problem is that because NWCs allow for any meager gain to the worse off party, however
distant that gain may be from what they are ideally owed, any mutually advantageous arrangement will
be permissible. Provided that party A had no pre-existing duty to party B, and that party B benefits
relative to her pre-transaction baseline, it matters neither how meager this improvement is, nor how
meager her share of the cooperatively produced surplus is relative to A’s. Mutually advantageous
transactions may still warrant the making of an exploitation claim if they deviate too greatly from the
distributive shares that would be agreed to by rational parties under ideal market conditions, but
provided the NWC is satisfied, they generate no normative force in the non-ideal world in which
exploitative transactions actually take place. This is a deflationary solution to the transactional task of
identifying cases of mutual advantage that nonetheless warrant intervention because, according to this
solution, none of them do.
A promising line of response to the meagerness problem left by the NWC is to argue that A does
indeed have an obligation to interact with B on specific terms, and as such must ensure that B
receives more than meager benefits. One such obligation is ostensibly that of need satisfaction,
220
Exploitation
whereby A is obliged to meet B’s basic needs. This response is made available by exploitation
theorists who argue that a duty to meet basic needs is precisely what’s violated in exploitative
exchanges (Goodin, 1985, 1988; Sample, 2003; Snyder, 2008). While Snyder (2008) takes his ac-
count of needs-exploitation to serve as a response to the meagreness problem of the NWC directly,
all needs-based accounts equally entail that an NWC would rarely if ever be satisfied by meager gains
to B because A has an obligation, at the very least, to meet B’s basic needs. So, if B not only begins
but ends her interaction with A below a threshold of basic need satisfaction, then regardless of the
benefits that do accrue to B from the interaction, B has been unfairly taken advantage of by A in a
manner that is both wrongful and that warrants intervention.
But as with all needs-based philosophizing, questions arise as to the source and the content of the
relevant obligation. Applied in the realm of exploitation theory, needs claims prompt us to ask, spe-
cifically, where the relevant obligation comes from, and whether it arises as a result of the interaction or
whether it exists prior to it. If there is a pre-existing duty to meet the needs of the vulnerable, then that
duty is borne by the better off generally, and not simply by those who engage in market transactions
with the worse off. As Goodin himself shows, the moral wrongdoing of playing for advantage against
the vulnerable is secondary to the more egregious wrongdoing of failing to have met their needs in the
first place (1985), and the normative solution to exploitation lies in institutional welfare reform to
ensure need satisfaction (1988).
For Sample (2003) and Snyder (2008) the duty to meet needs stems from a more foundational
duty of respect for persons, an imperfect duty which Snyder argues is rendered perfect when specific
interactions are undertaken with those in need. On his account, the duty to meet basic needs is
actualized by the interaction itself, which prompts us to ask why this particular duty is generated, and
whether is it appropriate to all market transactions. In some transactions, the meeting of basic needs
might be woefully inadequate, depending on the nature of the surplus produced, while in others it
might be exorbitantly high. Simply because one party is in a position to meet the other’s basic needs
does not mean they have an obligation to do so in virtue of interacting with them, particularly if this
stems from a more foundational duty of respect for persons that is borne by everyone alike and
which might be more efficiently and justly discharged through the institutions of the state (Goodin,
1988; Wilkinson, 2003). The needs-based response to the meagreness problem of the NWC either
posits the duty of need satisfaction as prior to the interaction and thus faces a similar challenge to
structuralism or sees the duty as arising from the exchange and has to explain why need satisfaction
constitutes an appropriate and non-arbitrary requirement of distributive fairness across diverse and
discrete transactions.
3 Non-Idealism and Inter-Transactional Parity

I propose in the remainder of this chapter to argue for a different kind of transactional response to
the meagreness problem generated by NWCs’. According to NWCs, so long as B consents to and
isn’t made worse off as a result of a transaction with A, and so long as A had no pre-existing duty to
interact with B on specific terms, then we should not interfere in their exchange. This claim rightly
invites a structural challenge, namely, that we shouldn’t take the wildly unjust bargaining positions
of better and worse off parties in our non-ideal world as an acceptable metric against which to
evaluate the fairness of their post-transactional shares. Given systemic vulnerabilities and the bar-
gaining asymmetries to which they give rise, we start off on the wrong normative foot if we take the
pre-transaction baseline of a vulnerable party with no bargaining power as a justifiable normative
benchmark by which to determine what she is owed.
To this, the transactionalist can respond with ideal metrics and argue that where vulnerable
parties can’t bargain for themselves, we should appeal to ideal bargainers. All mutually advantageous
transactions that deviate from this ideal could thereby be banned or interfered with to the point at
221
Vida Panitch
which the ideal is achieved. But as we’ve seen, this response overlooks that for many vulnerable
parties, mutually advantageous exploitative exchanges are nonetheless one of the few opportunities
available to them in our non-ideal world, and that the disincentivizing effects of insisting on ideal
benefit sharing would likely produce the same disadvantageous results as banning unfair exchanges
altogether.
If we accept that an unfair exchange can be exploitatively unfair yet deem it permissible on the
grounds that it is nonetheless superior to a state of affairs in which the exploitative arrangement did
not take place – either because it was banned outright or because the terms of what would have been
a fair exchange disincentivized the initiating party for whom the exchange was entirely optional – it
looks like all of our theorizing about exploitation doesn’t actually get us anywhere. Is there a
reasonable non-ideal benchmark against which to determine wrongful and impermissible unfairness,
such that we might avoid turning in circles from idealism to NWCs and back again? I will try to
show that there is.
The transactional account of exploitation holds that in order to determine the unfairness of dis-
tributional outcomes, we must perform a three-stage comparative evaluation. First, we must compare
the gains and losses accrued to each party relative to their respective pre-transaction baselines. Second,
we must assess how much A gained relative to her baseline as compared to B. And third, we must then
ask whether the gains to B render the distributive outcome unfair as compared to some ideal metric,
yet superior to a scenario in which the transaction didn’t take place at all because it was banned or
interfered with for failing to approximate the ideal. Here is where the difficulties arise. But there is an
alternative third stage comparative analysis that we could, and should, perform instead.
This alternative third stage comparative analysis is inter-transactional, whereas the prior two
comparative analyses are intra-transactional. That is, at the alternative third stage I am proposing, what
is required with respect to assessing the fairness of a mutually advantageous transaction is a type of
comparative analysis that is performed across transactions, rather than simply within them (Panitch,
2013a, 2013b). The comparative evaluations that transactionalism typically demands of us are intra-
transactional: we are asked to compare the gains to A and B, and to each relative to their respective
pre-transaction baselines. We must perform these assessments to determine the presence of mutual
advantage. If there is none, we need to proceed no further in our comparative assessment to make an
exploitation claim with adequate normative force to warrant intervention. But if there is a mutual
advantage, such that both parties gain relative to their pre-transaction baselines, even though one
party still gains a good deal more relative to her baseline than the other, then we must engage in a
third kind of comparative analysis, and specifically an inter-transactional comparative analysis.
This third stage comparative analysis must be inter-transactional because once we have estab-
lished mutual advantage within a transaction, we need to look beyond it to determine unfairness in
our non-ideal world. For reasons that we have seen, this look beyond the transaction should not be
focused on some ideal. Instead, we must look to the actual gains accrued to different party Bs across
relevantly similar transactions who take on the same set of burdens. This proposition entails that we
should assess whether party B to one transaction has been made better off not ony compared to party
A, or to the unjust pre-transactional baseline from which B began, but also compared to different
party Bs engaged in relevantly similar transactions who were better situated to bargain. This third
stage inter-transactional comparative analysis of fairness thus asks whether party A has denied party B1
the benefits that A has provided as compensation to a less vulnerable party B2 with whom A is engaged in a
similar transaction, or that A would have had to provide to B2 had the transaction taken place between A and
B2 rather than between A and B1 (Panitch, 2013a).
We make inter-transactional comparisons frequently in colloquial discussions of exploitation and
they play a large role in our pre-reflective judgments on the topic. When we evaluate exploitation
claims in non-philosophical contexts, we find it relevant to know how well the parties did not only
relative to each other and to where they began but also how well they did relative to other parties
222
Exploitation
engaged in similar transactions who took on the same set of burdens. Consider a common example,
which many of us might claim to be a quintessential case of exploitation. The Ford Motor Company
operates a plant outside Mexico City and pays its workers there a fraction of the wage it pays to its
employees in the United States and denies them the same health benefits, workweek restrictions,
and workplace safety guarantees. We are not at all surprised to hear it said that Ford exploits its
Mexican workers but not its American workers, or that it exploits its Mexican workers in a dis-
tinctively more wrongful way.
A claim of this sort doesn’t so much point to the concern that a multi-national corporation
derives all the benefits from global manufacturing, because a wage to someone otherwise un-
employed in the developing world has obvious and considerable value (Zwolinski, 2007). Instead, it
is a claim about mutually advantageous exploitation being more problematic in some cases than
others. What makes it problematic in this particular case is that a multi-national employer takes
unfair advantage of its developing world employees by denying them the wages, union rights,
medical benefits, and safety standards that it provides to its domestic employees, despite the fact that
both sets of employees take on a similar share of the transactional burdens by doing exactly the same
work for the company.
Consider another example. Respiratory Distress Syndrome (RDS) can be fatal in premature infants
born with an insufficient amount of surfactant to inflate their lungs. Treatment initially consisted of
ineffective ventilator therapy, but since the 1980s synthetic surfactants have been commonly used. In 2000,
the drug manufacturer Discovery Labs (DL) wanted to conduct a randomized trial for a new synthetic
surfactant, Surfaxin. DL did not expect Surfaxin to be superior to existing surfactants available on the US
market and wanted to use ventilator therapy as a control. The FDA denied their proposal, stating that they
could not conduct such a trial in the US where ventilator therapy was inferior to the treatment that
participants would otherwise receive and that its use would result in the avoidable deaths of premature
infants. The company thus enlisted premature infants born to poor families in South America where
ventilator therapy remained the standard treatment (Hawkins and Emanuel, 2008: 55–62).
The South American trial involved the provision of Surfaxin to infants with RDS in the ex-
perimental wing and ventilator therapy to infants with RDS in the control wing. No promise was
made to supply the new surfactant to control participants, or to their local health units, at the
conclusion of successful testing; instead, the long-term benefits of the research would be primarily
enjoyed by first-world health consumers who would now have an additional surfactant to choose
from in the market. While the control subjects were not in any strict sense denied a benefit they
would have otherwise received, there remains a stark inter-transactional unfairness between what
they were given as compared to what their developed world counterparts would have enjoyed, both
in terms of what they were provided during the trial and in terms of what they could expect to have
access to post-trial.
Take yet one more example. Prospective parents are turning in growing numbers to commercial
surrogacy clinics in developing parts of the world. Global surrogacy is mutually beneficial, as the
parents gain a child and the surrogate a wage. But the real concern becomes apparent when we
compare the wages and working conditions of global surrogates to American surrogates. The latter
are typically college graduates who earn around $50,000 USD, and who claim to prefer the work to
other well-paid options, enjoy legal protections, an opt-out clause, and the possibility of a long-term
relationship with the intended family (Busby and Vun, 2010). Meanwhile, surrogates in the de-
veloping world typically live below the poverty line, are uneducated and illiterate, earn $3,000
USD, are often compounded during their pregnancy so that their meals and activities can be
monitored, and are subject to Caesarian delivery and fetal reduction at the behest of the intended
parents and clinic staff (Pande, 2009; Twine, 2011). Comparing gains between a global surrogate and
the intended parents of the baby she carries doesn’t identify what’s exploitative about their
223
Vida Panitch
transaction, which has more to do with the lack of parity in the benefits enjoyed by surrogates in
different parts of the world who take on exactly the same set of burdens (Panitch, 2013b).
Our intuitive judgments in these cases are instructive. To determine if a particular transaction is
exploitative, we must begin by asking whether the vulnerable party has all things considered been
harmed or benefitted and whether she has gained more or less than her co-transactor relative to
where each began. But we should not stop there. We must also ask whether the vulnerable party has
been deprived of the share of benefits that a less vulnerable party to a similar transaction elsewhere
has, or would have enjoyed, for taking on the same burdens. That we should perform this third stage
inter-transactional analysis of fairness is grounded in the moral requirement that all persons be treated
with equal concern and respect, which in turn demands that no one be denied parity of benefit relative
to their share of transactional burdens on arbitrary grounds such as race, gender, or geographical location.
Clearly auto workers in the developing world are better off with a job than without one, as are most
developing world surrogates for whom the compensation can exceed what they might otherwise earn
over a 3-year period. And the parents of an infant in a clinical trial who receives a treatment it would
not otherwise have had access to are clearly not made worse off as a result of their participation in the
trial. But just because no one in these transactions is made worse off does not entail that intervention is
unwarranted. If we arrive at this conclusion, it is only because we have failed to perform a crucial third
stage inter-transactional comparative analysis of fairness, whereby we take as our non-ideal benchmark
the benefits enjoyed by parties to relevantly similar transactions who take on the same set of burdens
yet who agree to do so from a less vulnerable bargaining position.
The relevant difference between a vulnerable and a non-vulnerable party to a relevantly similar
transaction lies not in their geography, as my account so far may have implied, but in their bargaining
power. This, I propose, is the most plausible articulation of what some transactionalists refer to as the
vulnerability clause of an exploitation claim. On Valdman’s (2009) articulation of the clause, vulner-
ability lies in one party’s inability to refuse an offer, where this inability hangs on whether it would be
unreasonable of them to refuse. The concept of reasonable refusal is overbroad as a metric of vul-
nerability, as there are myriad grounds for reasonable refusal even for the worst off (Liberto, 2014).
Among these is the unfairness of the offer, which constitutes reasonable grounds for refusal in and of
itself. Genuine vulnerability depends not on the inability to refuse, but on the inability to negotiate
(Panitch, 2013b). There may be plenty of offers that a vulnerable party can reasonably refuse, but the real
worry lies with offers she wants to accept but with respect to which she is powerless to change the terms.
The unfairness of an outcome reflects one party’s inability to have bargained for better terms and
the vulnerability of a transactional party to unfair advantage-taking thus depending on her bar-
gaining power. The developing world surrogate, unlike the American surrogate, is not in a position
to bargain because if she does, she risks seeing the offer rescinded and made to someone else eager to
take her place. She is free to refuse, but not to bargain. Her vulnerability is a function of the fact that
she cannot demand higher wages, better living conditions, or perhaps most importantly, less in-
trusive medical procedures. The same can be said of sweatshop workers and auto-parts manu-
facturers in the developing world. And although trial participants cannot bargain their way out of
being randomized to the control wing of a trial they have joined anywhere in the world, prospective
participants in the developed world can insist they not be denied something they would otherwise
have had access to through their robust health care or health insurance system, unlike developing
world participants who often enlist in research trials to gain access to any treatments they can.
The question that remains is what exactly is required by inter-transactional fairness. This is a
question that warrants a lengthy answer, which I can only begin to offer here. I suggested above that
these assessments generate a parity condition, according to which consensual and mutually ad-
vantageous exploitative transactions should be considered wrongful to the extent that they exhibit a
failure of parity in their distributive outcomes. We should understand the parity requirement of
inter-transactional fairness in its economic as opposed to its moral sense, and thus as equivalence not
224
Exploitation
identity. The parity condition would in this sense require that subjects across relevantly similar
transactions in different parts of the globe receive not identical bundles of resources, but benefits that
enable equivalent levels of purchasing power in their respective currencies (Panitch, 2013a).
We might also choose to measure a vulnerable party’s benefit share not against the actual share
enjoyed by her non-vulnerable counterpart, but against the pre-transaction baseline from which her
non-vulnerable counterpart began. In other words, we could use B2’s pre-transaction baseline as a
benchmark for determining the fairness of B1’s benefit share. The developing world subject would not
have to receive the same share of gains as her developed world counterpart to satisfy the parity
condition, but a share that would nonetheless qualify as a gain to her developed world counterpart,
indeed one that the developed world counterpart would herself regard as a gain relative to the burdens
she takes on (Panitch, 2013b). What would count as a meager gain for the developed world party could
be a significant gain for the developing world party, relative to the former’s pre-transaction baseline.
What specific implications does parity have for global research and global surrogacy? When it
comes to research, the parity condition demands that research participants with little to no bar-
gaining power because of poor health status and limited healthcare access are not taken advantage of
to the benefit of either a clinical team and their sponsors or developed-world consumers who are
most likely to enjoy the fruits of the research post-trial. Parity eschews a double standard in human
subject research. If medical subjects should be guaranteed the most effective or best standard
treatment in randomized control trials, this must be understood globally, not locally. Moreover,
since the benefits of trial participation are not always enjoyed as direct access to a new intervention
during the trial but rather as long-term access to the fruits of successful medical research, clinical
teams and their sponsors must guarantee access to the fruits thereof (or to some equivalently fair
benefit should the research prove unsuccessful) to participants who would not be able to afford them
or access them through their own impoverished health care infrastructure (El Setouhy et al., 2004).
As for global surrogacy, we may assume wage increases would be in order. But if this is true, it
would only be to the point at which equivalence is achieved in terms of purchasing power parity,
and this may mean that wage increases are not in fact required in certain global contexts. We should,
however, expect to see other kinds of contractual protections guaranteed to global surrogates that are
afforded to the American surrogates who can bargain for them. These are arguably the more im-
portant benefits to focus on in the case of surrogacy, given that for many global surrogates the wage
is already equivalent to what they might otherwise be able to earn over a 3-year period. These other
benefits must include the right to refuse a Caesarian delivery or fetal reduction, as well as the
possibility of an on-going relationship with the child and its family, proper post-natal health care,
and freedom of movement during pregnancy (Panitch, 2013b).
A transactionalist who offers an NWC might remind us that her motivation is to ensure that
overly demanding principles of fairness do not disincentivize the initiation of mutually advantageous
transactions, leaving the vulnerable to a fate worse than exploitation. However, understanding the
requirements of fairness in terms of inter-transactional parity means that the costs of global research
and global surrogacy would still remain lower than their domestic counterparts. Doubling the
payment to a global surrogate would remain cheaper than paying a $50,000 wage to an American
surrogate, although it may be equivalent in terms of purchasing power and thus perfectly adequate
from the point of view of inter-transactional parity. And enlisting research participants in the de-
veloping world is easier and thus less costly precisely because of the lack of access to adequate
medical care faced by so many ailing global citizens. The incentive to initiate transactions with
developing world parties should not be affected by the demands of parity, or not to an extent that
warrants resisting interference in all unfair yet mutually advantageous global exchanges.
The demands of inter-transactional parity, while greater than those of non-worseness, are
nonetheless considerably less than those of idealism. Inter-transactional parity can thereby address the
incentive worries that motivate non-worseness claims without asking us to turn a blind eye to
225
Vida Panitch
mutually advantageous transactions in which developing world parties are denied a fair share of
benefits due to their limited ability to demand them. The argument I have offered here for a third
stage inter-transactional assessment of fairness allows the transactionalist to address structural con-
cerns about vulnerability and bargaining power without falling prey to the normative problems of
structuralism itself, while simultaneously enabling her to avoid unhelpful appeals to idealism and
deflationary appeals to non-worseness claims.
References
Busby, Karen, and Delaney Vun. (2010) “Revisiting ’The Handmaid’s Tale’: Feminist Theory Meets Empirical
Research on Surrogate Mothers,” Canadian Journal of Family Law 26, 1, 13–93.
Deveaux, M. and Panitch V. (2017) “Introduction,” in Monique Deveaux and Vida Panitch eds. Exploitation:
From Practice to Theory. London; Rowman & Littlefield: 1–13.
El Setouhy, M., et al. (2004) “Moral Standards for Research in Developing Countries from ‘Reasonable
Availability’ to ‘Fair Benefits,’” The Hastings Center Report 34, 3, 17–27.
Ferguson, B. (2016) “The Paradox of Exploitation,” Erkenntnis 81, 5, 951–972.
Goodin, R. E. (1985) Protecting the Vulnerable. Chicago; University of Chicago Press.
Goodin, R. E. (1988) Reasons for Welfare: The Political Theory of the Welfare State. Princeton; Princeton
University Press.
Hawkins J. and Emanuel E. (2008) Exploitation and Developing Countries. Princeton; Princeton University Press.
Jaggar, A. M. (2013) “We Fight for Roses Too: Time-use and Global Gender Justice,” Journal of Global Ethics 9, 2,
115–129.
Kant, I. (1959) Foundations of the Metaphysics of Morals. Trans. Lewis White Beck. Indianapolis; Bobbs-Merrill.
Kymlicka, W. (2002) Contemporary Political Philosophy: An Introduction. New York; Oxford University Press.
Liberto H. (2014) “Exploitation and the Vulnerability Clause,” Ethical Theory and Moral Practice 17, 4, 619–629.
Locke, J. (1661) “Venditio,” in D. Wooton ed. Locke: Political Writings. Indianapolis; Hackett.
Marxist, K. (1847) Wage Labour and Capital. New York; New York Labour News Company, 1902.
Marxist, K. (1867) Capital: A Critique of Political Economy, vols. 1–3. Chicago; C.H. Kerr and Company: 1906–09.
McKeown, M. (2017) “Sweatshop Labour as Global Structural Exploitation,” in Monique Deveaux and Vida
Panitch eds. Exploitation: From Practice to Theory. London; Roman & Littlefield: 35–57.
Pande, A. (2009) “Not an ’Angel,’ Not a ’Whore’: Surrogates as ’Dirty’ Workers in India,” Indian Journal of
Gender Studies 16, 2, 141–173.
Panitch, V. (2013a) “Exploitation, Justice, and Parity in International Clinical Research,” Journal of Applied
Philosophy 30, 4, 304–318.
Panitch V. (2013b) “Global Surrogacy: Exploitation to Empowerment,” Journal of Global Ethics 9, 3, 329–343.
Powell, B. and Zwolinski, M. (2012) “The Ethical and Economic Case Against Sweatshop Labor: A Critical
Assessment,” Journal of Business Ethics 107, 4, 449–472.
Roemer, J. (1982) “Property Relations vs. Surplus Value in Marxian Exploitation,” Philosophy and Public Affairs
11, 4, 281–313.
Sample, R. (2003) Exploitation: What It Is and Why It’s Wrong. New York; Rowman & Littlefield.
Snyder, J. (2008) “Needs Exploitation,” Ethical Theory and Moral Practice 11, 4, 389–405.
Twine, F. W. (2011) Outsourcing the Womb: Race, Class, and Gestational Surrogacy in a Global Market. New York;
Routledge.
Valdman, M. (2009) “A Theory of Wrongful Exploitation,” Philosophers’ Imprint 9, 6, 1–14.
Vrousalis, N. (2013) “Exploitation, Vulnerability, and Social Domination,” Philosophy and Public Affairs 41, 2,
131–157.
Wertheimer, A. (1996) Exploitation. Princeton; Princeton University Press.
Wertheimer, A. (1997) Coercion. Princeton; Princeton University Press.
Wilkinson, S. (2003) “The Exploitation Argument Against Commercial Surrogacy,” Bioethics 17, 2, 169–187.
Wood, Allen. (1995) “Exploitation,” Social Philosophy and Policy 12, 2, 136–158.
Young, I. M. (1990) Justice and the Politics of Difference. Princeton; Princeton University Press.
Zwolinski, M. (2007) “Sweatshops, Choices, and Exploitation,” Business Ethics Quarterly 17, 4, 689–727.
Zwolinski, M. (2009) “Price Gouging, Non-Worseness, and Distributive Justice,” Business Ethics Quarterly 19,
2, 295–306.
226
18
THE MEANING OF MARKETS
Brookes Brown
To know values is to know the meaning of the market

– Charles Dow
The classic complaint about markets is that they encourage harmful or unvirtuous behavior or
characteristics, violate rights, distribute goods in the wrong way, or are exploitative. This is what
people typically have in mind when they complain that child labor should be banned, or that sex
shouldn’t be available for purchase, or that the sale of blood diamonds fuels bloody wars.
A quite different concern is that trade in particular goods or services carries the wrong meaning. In
the case of such semiotic concerns, the purported problem is expressive. Market exchanges, the worry
goes, communicate an improper attitude towards marketed goods, or to associated individuals,
commodities, or practices. As Marx writes of a market system, “all that is holy is profaned …”
(Marx, 1978, 410). This wrongful expression is argued by many to generate a weighty reason against
participating in or having markets in particular goods. Consider Michael Sandel’s claim that placing
bets on the death of other persons expresses a “dehumanizing attitude,” or that the sale of children
would “express … the wrong way of valuing them,” even if the children were not mistreated
(Sandel, 2013, 349). Or take Jacob Sparks’s concern that, “when we allow the buying and selling of
certain goods, we are expressing inappropriate attitudes … towards the closely related goods that
can’t be bought or sold” (Sparks, 2017, 349). Similar claims have been advanced by Elizabeth
Anderson (1990), Debra Satz (2010), Peter Singer (1973), Margaret Radin (1987), and Walzer
(1983). In each case, the issue is with what market trades impart.
1 An Overview of Semiotic Complaints

Semiotic concerns come in roughly two forms. Purely expressive objections contend that market ex-
changes are wrong simply in light of what the transactions convey.1 You ought, the theory goes, to act
in accordance with norms that express the right attitude towards persons or other sources of value. You
behave wrongfully when your comportment communicates the wrong standpoint. Imagine, for ex-
ample, that you carelessly toss cigarette butts onto your neighbor’s lawn or permit your dog to poop in
their yard without cleaning up. In doing so, you express casual disregard for your neighbor’s interests.
This harm is independent of any further injuries that they might suffer. Your neighbor would ex-
perience the same unpleasant inconvenience if detritus blew into their yard during a windstorm or wild
animals eliminated in their grass, but they would suffer no expressive damage.2
DOI: 10.4324/9780367808983-22 227

Brookes Brown
According to purely expressive accounts, market norms are problematic because they express the
wrong attitude towards people or goods. By selling or buying a given item or permitting a market in
such goods you say something that you ought not to communicate – that the person or good should
be valued in the wrong way or is not worthy of a kind of respect that they are due. In some cases, the
troubling expression is a byproduct of a further wrongful and potentially revisable feature of the
market in question or the society in which it is embedded. If the production process of t-shirts is
recognized to be exploitative, then all-else being equal your knowing participation in the sale or
purchase of t-shirts expresses your lack of concern for the relevant abuse. Or if the market is in goods
that violate others’ rights (slavery, unjustly stolen items) your willing engagement will communicate
your disinterest in the injustice. To take part in such a market thus signals the wrong kind of attitude.
In those cases, the problematic meaning can potentially be resolved by correcting the relevant
wrongful behavior.
In other instances, the problematic meaning of market exchanges is said to stem not from some
further wrong made possible by market exchange, committed by market participants, or taken
advantage of by market actors, but instead from a (purportedly) fundamental feature of market-
ization. On these accounts, something in the structure of markets or market transactions expresses
something inappropriate even if the market is not otherwise wrongly ordered. Thus, for example,
Anthony Booth argues that the existence of a market inevitably signals that a good can be compared
to others and that a mechanism for comparison exists (Booth, 2018). Since – or so he contends –
there are cases where signaling such value-choice is wrong because it contradicts important choices
we have made about who we are as an individual or society – markets can be communicatively
problematic even when they do not generate or reflect further wrongs or harms. In a similar vein,
Jacob Sparks argues that participation in markets necessarily expresses a preference for the market
version of a good (e.g., sex) over a related good that cannot be bought and sold (such as intimacy)
(Sparks, 2017). When such an attitude proves wrongful, our engagement in markets is inappropriate,
even if the transaction does not otherwise cause injury (if, e.g., markets in sex do not harm parti-
cipants). As Elizabeth Anderson writes, “When women’s labor is treated as a commodity, the
women who perform it are degraded” (Anderson, 1990, 75).
By contrast, impure semiotic objections contend that a given market expression is wrong because it
results in downstream harm. In these cases, the wrongness of the expression is not – or at least need
not be – parasitic on some further message-independent injury – for example, that the marketed
good involves coercion or exploitation. Instead, the issue is that the expression triggers harm not
itself inherent to the communication. Imagine, for example, that my parents disown me because I
swear, or speak French.
Many semiotic complaints can be read in this way. Elizabeth Anderson, for example, contends that
the existence of markets in particular goods makes it impossible for people to value those goods in
accordance with non-marketized norms like respect, love, or reverence. Commodification, she argues,
should be restricted when doing so proves necessary to preserve people’s freedom to value different
goods in different ways. This is true, she thinks, of goods like sex, since the existence of commodified
sexuality is (in her view) incompatible with protecting separate spheres where sexual relations can be
valued as shared personal goods. In a similar vein, Deborah Satz suggests that certain non-marketized
norms such as having to wait your turn in line have come to represent a public statement of our
equality, such that violations undermine the value of equal citizenship (Satz, 2012, 107). Similarly, she
argues that given extant cultural views about gender identity, markets in women’s reproductive labor
will predictably contribute to women as a class possessing unequal social standing (ibid., 146–47).
Notice that both claims point not to a wrong in the expression itself (Anderson does not suggest that a
use-model of valuing is inherently inappropriate for sex) but to further harms that can be caused by the
signal in question.
228
The Meaning of Markets
2 The Case Against Market Meaning

Semiotic complaints raise two important ethical questions. First, are there goods in which we should
not have markets as a matter of policy? Many have argued on these grounds that we ought not
permit the sale of organs, reproduction, or sex. Second, are there markets in which we ourselves
ought not to participate whether not the markets should be permitted to exist? Does it, for example,
follow from these semiotic claims that I should not purchase a kidney?
Jason Brennan and Peter Jaworski have recently sought to answer both questions by advancing an
argument intended to undermine all semiotic objections. They contend that semiotic concerns
never justify restrictions on markets and provide no reason for individuals to avoid market ex-
changes. Instead, they hold that each of us should conscientiously reject or ignore the expressive
code according to which the exchange is communicatively problematic.
Brennan and Jaworski’s case against semiotic objections proceeds in three moves. First, they
argue that the meaning of market exchanges is contingent. Communicative meaning is dependent on
semiotic norms. In the United States, connecting the thumb and index finger into a circle while
raising your other fingers denotes approval or signifies that the speaker or object in question is doing
fine. By contrast, in Plains Indian Sign language, the gesture refers to the sun. In the Balkans, the
move communicates support for the notion of a United Macedonia. In much of the Middle East and
South America, the gesture is considered vulgar. In each case, the meaning of the movement reflects
the operative semiotic norms in a particular community. As the norm changes across context, so too
does what the expression conveys. Markets, they contend, are no different. “The meaning of market
exchanges,” they write, “… is probably just a contingent, relative social construct” (Brennan and
Jaworski, 2017, 1061).
In the second step, the productivity thesis contends that markets produce valuable consequences.
What they seem to have in mind is that markets systematically generate distributive benefits.
Roughly (or so they suggest) markets incentivize the production and allocation of resources in a
more efficient matter, often dramatically to the good. People die every year waiting for a kidney
transplant who might receive one if they were able to purchase an organ (Becker and Elías, 2007).
As Marx points out, the opening of markets, “has created more massive and more colossal pro-
ductive forces than have all preceding generations” (Marx, 1978, 477).
The third step consists of a cost-benefit analysis (Brennan and Jaworski, 2016, 69). Semiotic norms,
Brennan and Jaworski hold, have no ethical significance. If they come at the cost of other goods,
they should be altered. “The duty to comply,” they argue, “disappears or is silenced, not merely
defeated or outweighed” (ibid., 73). On these grounds, they conclude that semiotic objections
should be altogether rejected as bases for limiting markets or avoiding market exchanges. The norms
that underlie distaste for marketization have no value to weigh against the productivity thesis and
thus should be excised. Rather than banning or limiting trade, we should alter the meaning that we
attribute to such exchange. To do otherwise, they suggest, would be actively immoral (ibid., 83).
It is questionable whether the semiotic norms surrounding marketization are as mutable as
Brennan and Jaworski suggest. At the very least, it is unclear that here and now market exchanges can
mean something different than they do, even if the meaning of markets could, in theory, have been
different than they are. It may be that our norms are no longer open to revision, for cultural or
psychological reasons. As Elizabeth Anderson writes, “the proposed reformation of meaning is akin
to asking us to speak FORTRAN instead of natural language” (Anderson, 1993, 77). Features of our
cultural understanding may prove sticky given our particular history and social practices.3 Markets
have had their meaning for a very long time. Moreover, insofar as the negative meaning of market is
attributable to problematic features of extant markets – that they are, for example, exploitative, or
traffic in goods that ought not be exchanged – the semiotic norm will likely persist unless the market
can be cleansed of these troubling features. As Jeffrey Moriarty notes of Brennan and Jaworski, “they
229
Brookes Brown
need to give us reason to suppose that the ideal markets are ones that can be brought into existence,
given economic and political realities” (Moriarty, 2017, 644). This may not prove possible.
Brennan and Jaworski reject this concern as a reason to avoid participation in markets. When
confronted with a sticky market norm, they argue that we should nonetheless plunge ahead. “Must
one adhere to that code,” they ask, “refraining from actions that express disrespect or bad motives?
We don’t see why. Instead, it seems more plausible that one may conscientiously reject or ignore the
code” (Brennan and Jaworski, 2016, 72). They hold that, in such cases, the duty to comply is not
simply outweighed – it disappears altogether.
Especially given their consequentialist framework, it is puzzling how they can reach this con-
clusion. The view that the wrongness of problematic expressions drops out altogether seems to rest
on the dual notions that the only cost involved is expressive and that this cost is either weightless or
should be treated as such since it is merely the result of a stubborn resistance by at least some market
participants to adapt a more beneficial interpretive scheme. You ought not, the logic seems to be,
accede to a kidnapper’s demands. However, this way of thinking ignores potentially significant costs
that are not silenced, may reflect no wrongdoing, and seem to deserve – at the very least – inclusion
in a consequentialist calculation.
To begin with, this argument discounts impure semiotic concerns. Problematic expression can
cause further downstream harms.4 For example, those who are subject to such statements might have
their feelings hurt or come to think of themselves as not deserving of respect. These outcomes may ensue even
if the victim recognizes that the cultural meaning of the expression is contingent. On many ac-
counts, the badness of these harms is independent of the wrongness of the expression. One might
worry about them even in instances where there was no underlying wrongdoing, as in a case where
somebody experiences a bout of depression that lowers their self-respect or where somebody
mishears an utterance and mistakenly believes themselves to have been gravely insulted. The fact that
an expressive act causes some harm would seem to deserve inclusion in the calculus of our behavior,
just as much as the fact that a market exchange causes a benefit (that people get sexual experiences
they desire or kidneys they need). This is especially true given that many of the benefits Brennan and
Jaworski attribute to markets themselves result from people otherwise failing to act as they ought.
Perhaps each of us should donate our kidneys without the incentives that markets typically provide.
In some cases, the harm caused might be extensive. If my selling a piece of art will be viewed as such
an expression of cultural disrespect that my sister will be vulnerable to an honor killing, it seems flippant
to think I should simply ignore the code. Brennan and Jaworski thus cannot reasonably say that
concerns stemming from semiotic objections just disappear from the calculation of how would-be
market participants ought to behave. Many semiotic objections take just this form. Elizabeth Anderson,
for example, holds that markets in certain goods are problematic because they interfere with individuals’
ability to value goods in ways that are at odds with market norms. This can have troubling con-
sequences. “Artists,” she worries, “may pander to public taste rather than challenging it … Doctors may
perform profitable but medically unwarranted services on ignorant or demanding patients” (Anderson,
1993, 148). Perhaps in particular cases, these harms are outweighed by the value that markets generate.
But surely such concerns should be included in any calculation. If the semiotic meaning of market
exchange proves closed to revision in-practice, and if the damage caused by the existence of market
exchanges in light of these norms is sufficiently great, then society has good reason to avoid market-
ization and individuals should abjure market participation, even as we may also have reason to decry
those who make it the case that such downstream harms follow from the relevant expression.
Some critics take the problem to be more fundamental. The meaning of markets, they argue, is
insufficiently contingent to satisfy Brennan and Jaworski’s demands. Thus, there are costs that we
cannot escape simply by reimagining the structure of market exchanges. Semiotic meanings are
social in nature. What meanings are available to a given act or action depends in part on the
context in which the act is situated. This is true on a wide variety of theories of meaning.5
230
I cannot intend to communicate P to you by saying X in some context where I know X will be
reasonably taken by observers to mean something besides P. Widely accepted features of con-
ventions around X or X-related behaviors that shape whether a large number of people will
believe the behavior in question to have the relevant significance may therefore fix or limit the
meanings that can be associated with the act. To change the connotation attached to market
exchange, it may thus be necessary to alter the way that markets themselves are organized or the
manner in which people characteristically behave in such contexts so that attendant observers will
have different evidence with regards to the meaning of market-related signals.
Many proponents of semiotic objections contend that this is not possible – or at least, that it is not
possible to do so while recognizably retaining something that constitutes a market exchange. It is
conventional to take such interactions as consisting in voluntary exchanges for consideration – an
approach that Brennan and Jaworski endorse.6 As part of the exchange, the seller receives property
rights in the good or service that constitutes consideration. This need not tell us anything significant
about the interpersonal attitudes of exchange participants.7 Andrea can engage in such an exchange
with Frank while remaining utterly indifferent to him or while caring deeply about his well-being.
But it does tell us something about the participants’ attitudes towards the goods exchanged.
Specifically, it tells us that Andrea is willing to exchange the relevant good for at least some price,
that a mechanism for comparison has been created, that Andrea is willing to exchange the good in
her possession for the good to be acquired and thus (at least if she does not additionally act outside
the market context) that she values that good more than alternative versions available without
consideration, and that she believes Frank has compatible preferences.8 So long as this conventional
structure of norms remains in place, critics argue that markets may necessarily express at least this,
and (they contend) it is this expression that proves problematic.
This worry about markets has been subject to two critiques. The first questions whether what is
at issue in debates about the meaning of markets is genuinely about expression. Instead, the complaint
goes, the concern such objections raise is with what markets indicate: that participants have objec-
tionable attitudes, or desires (Jonker, 2019). It is such mis-valuing, not the expression itself that is the
cause for criticism. Indication, on this approach, is a causal relationship. The smell of smoke indicates
the presence of fire because it is causally linked to the existence of fire. By contrast, expression
requires engagement with a convention-based communicative practice, as when the billowing of
white smoke from the Vatican declares the selection of a new pope. Complaints about what markets
express are, Julian Jonker claims, really complaints about what markets indicate. The logic is that the
concern would depend on the meaning of a market exchange if the grounds of the complaint were
really semiotic. As the meaning of a communicative act is mutable, this would be alterable simply by
changing the relevant expressive norm. However, Jonker notes, in arguing against marketization,
critics suggest that the meaning of market exchanges is non-revisable. In doing so, he argues, they
reveal that their concern is really about what markets indicate. Thus, even if semiotic objections
point to a real concern, that worry is, at the very least, misunderstood. It is not about expression.
Brennan and Jaworski are even more dismissive. This debate, they suggest, is merely semantic. If
market exchanges are taken to require negative features like, for example, purely selfish motivations,
then they argue that they are defenders of schmarkets, similarly structured practices in which people
buy and sell goods and services but lack all the relevant bad attitudes. Thus, they contend, con-
ceptual claims about the nature of markets add no further support to semiotic objections. They are
merely trivial complaints about the meaning of terms.
3 The Case for (Contingent) Market Meaning

These responses, however, move too quickly. As I will argue in this section, while the structure of
markets could in theory – and perhaps even in practice – be revised so as to avoid all problematic
231
Brookes Brown
semiotic norms, we have very good reason not to do so. Indeed, the strength of Brennan and
Jaworski’s case for markets relies on our retaining certain extant features of market exchange that
generate semiotic concerns. While they may be contingent, they are highly desirable. The pro-
ductive value of markets, in addition to other goods, rests on their presence. The cost of excising
these features in favor of schmarkets that lack them would be awful.9 Claims that the meaning of
markets is non-revisable are thus best understood as arguments that they ought not be revised in ways
that will remove the relevant concerning (but contingent) meanings. If this is right, then complaints
about market meaning are both genuinely concerns about expression and not something that should
be dismissed or avoided simply by altering our terms, as Brennan and Jaworski wish to do.
Consider first the fact that markets indicate a participant’s willingness to exchange a particular
good type or token at a given rate. I will sell you a banana for a pear, or an hour of my physical labor
for 25 dollars. In the absence of other defeating conditions, a market trade at a given price signals a
general openness to swapping these goods at this rate of exchange. If I will sell you an hour of my
time for that consideration, then I will do the same for another all else being equal. This norm is
entirely contingent. We can envision a schmarket in which participants’ behavior utterly fails to track
their general willingness to exchange but instead follows some other pattern – a random toss of a
coin perhaps, or a favor to close friends.
However, such a schmarket would lack an extremely valuable property. Exchanges that reflect
such generalized willingness are a key component of a price system in which goods are efficiently
distributed in accordance with preference and availability and individuals are able to adjust their
behavior in response to the anticipated actions of others. In such a system prices reflect particular
individuals’ willingness to trade, as signaled by their actual exchanges. The price of a stay at a resort
in the Bahamas signals that these locations are coveted – which they are. The price of a rotting
tomato signals that people don’t like inedible produce – which they don’t. Such a system solves a
complex coordination problem, making effective distribution possible even in the absence of clear
knowledge of the billions of different factors that influence need and production (Hayek, 1945).
Indeed, the productivity thesis on which Brennan and Jaworski base their consequentialist defense
of markets rests on just this property. The very fact that markets are able to effectively allocate organs,
or toothpaste, or generate new acne research is a result of this contingent structure. To give up this
feature might solve several semiotic concerns, but it would drain the productivity thesis of its value.
This is equally true of other contingent market norms. Take the fact that markets are understood
to be a form of closed exchange. Once both parties have completed their end of the bargain – I have
given you an apple, you have given me a tuba – they retain no special normative relation or
obligation to each other. In this way, they differ from non-market interpersonal relations. If I go to a
restaurant, eat, and pay my check complete with tip, our special reflexive duties are complete. No
normative residue remains. This is quite different than if you invite me to dinner at your house.
Even if I arrive with a bottle of wine I seem to have acquired a further responsibility of issuing a
similar invitation in the future. By contrast, closed interactions serve the valuable role of allowing
parties to walk away without further burden. As the actor Charlie Sheen said of sex-workers, “I
don’t pay them for sex. I pay them to go away” (Ayers, 2009).
In this way, the impersonality of traditional market exchanges also facilitates valuable relations.
Sometimes we want to be seen by those around us, to be known, to have close personal contacts. And
sometimes we want to buy tampons and morning coffee without gearing ourselves up for social
pleasantries. Or we want to explore new identities, personalities, hobbies, or ways of being without the
need to expose and justify ourselves to those with whom we share intimate relational connections.
There are many reasons one might want to acquire condoms, pregnancy tests, or bibles without
disclosure to close relations. Or you might simply want to save your limited reserves of memory for the
birthdays and favorite cookie-type of long-standing friends rather than feeling obliged to recall the
wedding anniversary of your waiter, or the woman who changes your oil.
232
The impersonal nature of such exchanges comes with many additional advantages. These norms
help alleviate experiences of discrimination.10 The cashier at CVS does not know my political or
sexual orientation. When I buy a used car online nobody needs to know that I’m a woman.
Impersonal norms also make possible substitutability. I need not feel guilty if I buy my latte from a
different barista or shop altogether. I can turn down an Amazon purchase in favor of one from Target if
the price is lower. I can switch carpet cleaners if the quality drops. I can walk away from a house
purchase if the inspection reveals issues I simply do not wish to address. In this sense, impersonal market
norms make the price system more efficient, and consequently (market failures aside) encourages a
more successful and effective satisfaction of preferences. They can also enhance our freedom by freeing
us from dependency on particular others. As Adam Smith notes, “Each tradesman … derives his
subsistence from the employment, not of one, but of a hundred or a thousand different customers …
therefore, he is not absolutely dependent on any of them” (1976, 420).
This efficiency is enhanced by the fact that markets are typically characterized by a norm of
incentive-compatibility. Market actors routinely respond to remunerative offers by agreeing to behave
in ways they might otherwise dis-prefer. They move across the country to take a more profitable
job, they show up to work on weekend mornings, or stay for an extra hour. Adam Smith writes, “it
is not from the benevolence of the butcher, the brewer or the baker that we get our dinner, but
from their regard to their own interest” (1976, i.ii.2). That is to say, they act for consideration. As
Brennan and Jaworski write of kidney markets, for example, “you aren’t kind enough to give away
your extra kidney to a stranger, but you might do it for $100,000” (Brennan and Jaworski, 2016, 8).
These norms are all contingent. We have others on offer. By contrast, personal relationships are
typically open-ended. Far from closing off an exchange, the return of a favor by some consideration
deepens and extends the participants’ mutual obligations. If I give you a birthday gift after you take
care of me when I am ill, our normative relationship is not complete. Far from it. If anything, I have
accepted an invitation to further responsibility. If you call feeling down because your dog died, I
ought to pick up and stay on the line.
Our personal relations are equally un-substitutable. I do not simply swap out one friend for another
because on this day she will be able to go for a walk with slightly less chit-chat, or sub in another
spouse for the week because she is better at doing the dishes. In contrast to my relationship to the
person who sold me a candy bar, our personal relations are characterized by quite general com-
mitments. It matters to me if my brother is trying to learn Italian, if his boss is a jerk, if he is thinking
of moving to Montana. I care about whether he cheats on his spouse, keeps his promises to friends,
changes religion. Though we may receive items of value from our friends and family, we do not
typically act for consideration. Our friends are not invited to dinner because they will bring wine.
Instead, these exchanges operate as signals of ongoing care and invitations to deepened interpersonal
commitment. Entirely different signals are necessary if we wish to cease ongoing relations.
Of course, there is overlap. An unscrupulous scallywag might seek to ape friendship with an
important or wealthy person in the hope of receiving gifts or promotions. Upon repeat exposure,
you might become friends with your hair-dresser. You might feel a commitment to your local
coffee shop, even if its prices rise and its quality drops. You might be rightfully angry if your friend
accepts a dozen rides to the airport without offering anything in return. Still, the nature of our
relations and the norms that govern them are characteristically profoundly different in the two cases.
But they need not be. In this, Brennan and Jaworski are correct. We could replace these market
norms with schmarket norms that resemble those that currently govern personal relationships. Recall,
however, that what a given act can express is in part a function of how reasonable observers interpret
extant semiotic norms. To change the evidence on which they base their interpretations would require
a profound rejiggering of how we relate in market exchanges. We could choose to purchase goods
only from those with whom we have close personal relationships, or with whom we wish to signal an
interest in developing such relations. We could refuse to act on self-advantageous incentives. We could
233
Brookes Brown
feel bound to take the time to learn about the personal lives of the people who sell us soap. We could
make brand loyalty into a normative commitment rather than a marketing slogan.
All of this, however, would come at a weighty loss. We have already seen some of what is at
stake. The impersonal, closed, relations that characterize market norms are not less valuable than the
personal, open, intimate relations that characterize many non-market norms. They are simply dif-
ferent. They provide different value and play different roles in our lives. We have reason to desire
that each operate in the sphere where they provide the most good. Something of great value would
be lost if our children, spouses, and friends were utterly substitutable, if our obligations to them were
limited and closed. But something is equally amiss when our peer reviewers, police officers, or
referees abide by the norms of close personal relationships – trading favors and privileging their
friends. In advancing the productivity thesis, Brennan and Jaworski show us why the extant norms
of the market are so valuable in the sphere of exchange. They have, as Marx wrote, allowed us to
“accompli[sh] wonders far surpassing Egyptian pyramids, Roman aqueducts, and Gothic cathedrals”
(Marx and Engels, 2002, 222).
What is at issue is thus not merely a semantic debate about the features of markets. The fact that
the meaning of market exchanges is contingent does not answer the question of whether we ought to
change them. In many cases, we should not. There are often good reasons why we should not do what
would be required to alter the semiotic norms market exchanges convey. We have reason to value
markets over schmarkets.
But retaining these structures and patterns of behavior carries costs. We have already seen some of
them. Given the structure and social norms surrounding markets, market exchanges carry real signals –
including that the desired relation is impersonal, closed, substitutable, incentive-compatible, up
for comparison, and so on. They are thus often inappropriate expressions towards certain goods or
persons – just as critics have complained. Some relationships or goods do not have – and should not be
treated as having – this kind of value. It is perfectly fine to communicate to your boss that you will
leave if somebody else offers an extra five thousand dollars. It is much more troubling to tell your wife
the same. There is something deeply problematic in the declaration that you will pay a friend a
hundred dollars to come take care of you when you are ill (or in accepting such an offer) or in the
announcement that you will sell your child for $100,000. Since markets carry (desirably, if con-
tingently) real meaning, these actions are expressions of disrespect. In addition to their problematic
meaning, such market signals may carry real downstream consequences – just as critics contend.
4 Signaling Overlap in Market Exchange

Among the costs that must be considered in marketization are semiotic costs. The contingency thesis
is consistent with the existence of a range of constraints on semiotic norms. There exist not only – as
we have seen – rational limitations on the interpretation of signals in light of extant semiotic and
social norms and structures, but also constraints on combinations of semiotic norms. For a semiotic
norm to obtain, enough people in the relevant context need to believe that the action has the
relevant significance, if it is indeed to have that significance. Though any given behavior X could
have signaled Y, and any given behavior could have signaled Z, it does not follow that the same act
can successfully signal both Y and Z. Imagine that the same signal meant both “yes” and “no.”
Signal distortion can occur in any number of ways. The most relevant for our purposes is when
there exists signal overlap. Such situations are characterized by the fact that a reasonable interpreter will
have their credence in one reading of an action’s meaning undermined by the existence of competing
evidence stemming from some other norm. This can happen in any number of cases. Imagine that your
mother sends you a sweater that you greatly like. You want to tell her how much you enjoy the gift.
But you are both aware that norms of politeness would require that you express profuse appreciation
for the gift even if you found it horrendous. The overlap between the signals associated with politeness
234
and those associated with enjoyment make it impossible for you to successfully express your genuine
appreciation, or for your mother to uptake your signal with any confidence.11 Or consider the (quite
probably apocryphal story) that when John F. Kennedy was elected President, he declared to those
gathered around him, “I will never make another friend.” His point was not that he would never again
enjoy another person’s company, nor they his, but rather that he would never again have sufficiently
good evidence that an apparent signal of friendship was sincere.
As organized – and as we have reason to want them organized – markets are associated with a cluster
of semiotic norms at odds with signaling other important values. Consider caring. It is an important
signal of your care for another person that you are willing to act for that person’s benefit at a cost to
yourself. This is why it matters that you immediately fly across the country when you learn their father
died, or that you go out of your way to cook them chicken soup when they are ill, or give up your free
afternoon to watch their child when they have an important meeting. Of course, there are other ways
you might signal your relationship. You could simply say that you care. But as economists tell us, talk is
cheap. The taking on of a cost to another’s benefit serves as a much more compelling signal.
The norms of the market interfere with this signal. Recall, as Brennan and Jaworski endorse, a
market exchange consists in exchange for adequate consideration. When you act to somebody’s benefit
for such consideration, you do not do so at a cost to yourself. The exchange thus cannot effectively
signal that you care. This is quite different from saying that you do not, in fact, care, or even that the
exchange constitutes evidence that you do not care. However, the existence of such exchange does
entail that an action that would previously have operated as a significant signal of care (flying across
the country, bringing chicken soup, exposing yourself to potential illness) no longer does so. It is
reasonable to believe that you would act for the money even if you did not care a jot. The caring
signal is defeated.
The same is true of other valuable expressions, such as those involved in testimony and esteem. In
each case, marketization interferes with an act’s ability to signal the relevant attitude. Take, for
example, Brennan and Jaworski’s own choice to sell acknowledgments in their book. As they make
clear, for different amounts you can be listed in silver, gold, or platinum acknowledgments. Upon
discovering this, a reader is no longer warranted in treating a given person’s inclusion in the ac-
knowledgments as a signal that they are, in fact, being granted the kind of appreciation as an in-
tellectual contributor to a project that is typically associated with the status. Indeed, Brennan and
Jaworski seem to acknowledge this by including a distinctly separate category of acknowledgments
that at least give the impression of having not been for sale. The discovery that this too had been
marketized would similarly undermine these signals. If the practice of such sales becomes widespread
common knowledge, acknowledgments sections would – all else being equal – no longer be able to
effectively operate as signals of appreciation, even by those who did not themselves commodify
inclusion. The signal would be defeated.
There are thus trade-offs between the values produced by market exchanges and those found in
other social practices. The existence of markets in particular acts will disrupt the ability of these
actions to operate as signals of care, or esteem, or testimony. This endangers or weakens a range of
valuable social practices – friendship, expert testimony, the reactive attitudes of morality.
5 Conclusion
Cases like those discussed in section three point to real and significant losses that would accrue if we
altered the structure of markets. We have reason to prefer markets over schmarkets. But so long as
markets retain their desirable (though contingent) form, the exchange of goods in a market context
comes with predictable meanings that, in turn, block or undermine the use of such exchanges as
ways of communicating other things. Market trades in particular goods thus come at an expressive
cost and can carry problematic meanings.
235
Brookes Brown
These concerns can justify limitations on marketization, just as semiotic critics have argued.
Importantly, the worry avoids the response on which proponents of “markets in everything” like
Brennan and Jaworski have relied. Since nothing in the argument relies on the notion that markets
have any essential meaning, the problem cannot be avoided simply by noting that the meaning of
market exchanges are contingent. While that is true, we have seen that there are good reasons to
keep the structure of markets such that they will predictably have such meanings. Nor can the worry
be dismissed as simply reflecting “sour grapes” on the part of those uncomfortable with market-
ization. Indeed, the claim champions the value of markets and applauds their vast achievements. The
costs that give rise to the semiotic concerns we have explored are precisely the cost of that triumph.
Yet semiotic critics should not rest so easily on their laurels. Though the arguments we have
considered provide a reason to refrain from the market exchange and to limit the reach of mar-
ketization, precisely what restrictions might be justified remains unknown. Those who wish to
defend constraints on the marketization of particular goods like sex or organs must show that the
relevant expressive costs outweigh the very real benefits. Given the complex array of values in play,
the outcome of such arguments is likely to be highly individualized to particular goods – the costs
and benefits of markets in acknowledgments and kidneys are, after all, quite different.
In addition, the issues we have discussed reveal gaps in the state of our understanding of market
expression. To date, discussions of the communicative content embedded in markets have interacted
surprisingly little with either more general theorizing about the meaning or with empirical evidence
about our communicative norms and practices of marketization. To what extent, for example, does
the communicative content of a market exchange depend on facts about the intentions of the
participants, localized social conventions, and so on?12 How do real-world market participants
actually interpret offers of exchange? What do they aim to signal? How do different ways of
structuring markets shape the reasonable interpretation of an exchange’s communicative meaning?
What are the consequences of different background structures (e.g., a universal basic income, or
equal bargaining power)?
Supporters and opponents of marketization must each do more to answer these questions. The
strength of semiotic complaints against the marketization on the results. However, far from being a
mark against semiotic objections, these questions suggest a rich vein for future research.
Notes
1 This distinction is advanced by David Dick. David G. Dick (2018) “Impure Semiotic Objections to
Markets.”
2 Elizabeth Anderson and Richard Pildes develop a similar example. Anderson and Pildes (2000), “Expressive
Theories of Law: A General Restatement,” 1527.
3 Bigoni et al. (2014) “Amoral Familism, Social Capital, or Trust? The Behavioral Foundations of the Italian
North-South Divide”; Dan Kahan (2000) “Gentle Nudges vs. Hard Shoves: Solving the Sticky Norms
Problem.”
4 David Dick (2018) refers to such harm as impure semiotic reasons. David G. Dick, “Impure Semiotic
Objections to Markets.”
5 Binmore (2008) “Do Conventions Need to Be Common Knowledge?”; Grice (1957) “Meaning”; Searle
(2010) Making the Social World; Searle; Lewis (1983) Philosophical Papers Volume I.
6 See, for example, Mills v. Wyman, 20 Mass. 207 (Mass. 1825); Brennan and Jaworski (2015) note that in a
normal market, “both parties volunteer an exchange because both expect to profit.” Brennan and Jaworski,
“If You Can Reply for Money, You Can Reply for Free.”
7 It does presumably tell us that the participants possess a minimal level of respect such that they view each
other as sufficiently reliable exchange-partners, at least as constrained by existing practices.
8 These are consistent with the expressive value that Booth and Sparks attribute to market exchange.
9 This idea is developed in Barry Maguire and Brookes Brown (2019) “Markets, Interpersonal Practices, and
Signal Distortion.”
236
10 See, for example, Ekpo et al. (2018) “Narratives of Technology Consumption in the Face of Marketplace
Discrimination.”
11 This is likely not a coincidence. Such norms of politeness are parasitic on norms of appreciation.
12 Other debates about expressive limitations on action, such as those guiding expressive theories of law, have
to date done more to consider these concerns. See, for example, Simon Blackburn, “Group Minds and
Expressive Harms”; Matthew Adler, “Expressive Theories of Law: A Skeptical Overview”; Alan Strudler,
“The Power of Expressive Theories of Law.”
References
Anderson, Elizabeth (1990) “Is Women’s Labor a Commodity?” Philosophy and Public Affairs 19, 1, 71–92.
Anderson, Elizabeth (1993) Value in Ethics and Economics. Cambridge, MA; Harvard University Press.
Anderson, Elizabeth, and Richard Pildes (2000) “Expressive Theories of Law: A General Restatement.”
University of Pennsylvania Law Review 148, 1531–1564.
Ayers, Ian (2009) “I Pay them to Leave.” Freakonomics Blog (February 6, 2009) available at https://
freakonomics.com/2009/02/06/i-pay-them-to-leave/ (accessed June 20, 2021).
Becker, Gary S., and Julio Jorge Elías (2007) “Introducing Incentives in the Market for Live and Cadaveric
Organ Donations.” Journal of Economic Perspectives 21, 3, 3–24. 10.1257/jep.21.3.3.
Bigoni, Maria, Stefani Bortolotti, Diego Gambetta, Francesca Pancotto, and Marco Casari (2014) “Amoral
Familism, Social Capital, or Trust? The Behavioral Foundations of the Italian North-South Divide.” The
Economic Journal 126, 594 (n.d.), 1318–1341.
Binmore, Ken (2008) “Do Conventions Need to Be Common Knowledge?” Topoi 27, 1–2, 17–27. 10.1007/
s11245-008-9033-4.
Booth, Anthony Robert (2018) “The Real Symbolic Limit of Markets.” Analysis 78, 2, 198–207. 10.1093/
analys/anx144.
Brennan, Jason, and Peter Jaworski (2015) “Markets Without Symbolic Limits.” Ethics 125, 4 (July),
1053–1077. 10.1086/680907.
Brennan, Jason, and Peter Jaworski (2016) Markets Without Limits: Moral Virtues and Commercial Interests. New
York and London; Routledge, Taylor & Francis Group.
Brennan, Jason, and Peter Jaworski (2017) “If You Can Reply for Money, You Can Reply for Free.” The
Journal of Value Inquiry 51, 4, 655–661. 10.1007/s10790-017-9616-7.
Dick, David G. (2018) “Impure Semiotic Objections to Markets.” Public Affairs Quarterly 32, 3, 227–246.
Ekpo, Akon E., Benét DeBerry-Spence, Geraldine Rosa Henderson, and Joseph Cherian (2018) “Narratives of
Technology Consumption in the Face of Marketplace Discrimination.” Marketing Letters 29, 4, 451–463.
10.1007/s11002-018-9475-4.
Grice, H. P. (1957) “Meaning.” The Philosophical Review 66, 3, 377. 10.2307/2182440.
Hayek, Friedrich (1945) “The Use of Knowledge in Society.” American Economic Rview XXXV, 4, 519–530.
Jonker, Julian D. (2019) “The Meaning of a Market and the Meaning of ‘Meaning.’” Journal of Ethics and Social
Philosophy 15, 2. 10.26556/jesp.v15i2.646.
Kahan, Dan (2000) “Gentle Nudges vs. Hard Shoves: Solving the Sticky Norms Problem.” The University of
Chicago Law REview 67, 3, 607–645.
Lewis, David (1983) Philosophical Papers Volume I. Oxford; Oxford University Press. 10.1093/0195032047.001.
0001.
Maguire, Barry, and Brookes Brown (2019) “Markets, Interpersonal Practices, and Signal Distortion.”
Philosophers’ Imprint 19, 14, 1–16.
Marx, Karl (1978) “The Communist Manifesto.” in Robert Tucker edited by. The Marx-Engels Reader. New
York; Norton Press.
Marx, Karl, and Friedrich Engels (2002) The Communist Manifesto, edited by Gareth Jones. New York; Penguin
Books.
Moriarty, Jeffrey (2017) “Wanted: Positive Arguments for Markets.” The Journal of Value Inquiry 51, 4,
641–645. 10.1007/s10790-017-9613-x.
Radin, Margaret (1987) “Market-Inalienability.” Harvard Law Review 100 (1987), 1849–1937.
Sandel, Michael J. (2013) What Money Can’t Buy: The Moral Limits of Markets. 1st paperback ed. New York,
NY; Farrar, Straus and Giroux.
Satz, Debra (2010) Why Some Things Should Not be for Sale: The Moral Limits of Markets. New York, NY; Oxford
University Press.
237
Brookes Brown
Satz, Debra (2012) Why Some Things Should Not Be for Sale: The Moral Limits of Markets. First issued as an Oxford
UNiversity Press paperback. Oxford Political Philosophy. Oxford and New York, NY; Oxford University
Press.
Searle, John R. (2010) Making the Social World: The Structure of Human Civilization. Oxford and New York, NY;
Singer, Peter (1973) “Altruism and Commerce: A Defense of Titmuss Against Arrow.” Philosophy and Public
Affairs 2, 3, 312–320.
Smith, Adam (1976) An Inquiry into the Nature and Causes of the Wealth of Nations. Indianapolis; Liberty Classics
Sparks, Jacob (2017) “Can’t Buy Me Love: A Reply to Brennan and Jaworski.” Journal of Philosophical Research
42, 341–352. 10.5840/jpr2017425101.A.
Walzer, Michael (1983) Spheres of Justice: A Defense of Pluralism and Equality. United States; Basic Books.
238
19
GENDER AND THE DIVISION
OF LABOR
Gina Schouten
Whatever their actual or potential benefits, labor markets as currently enacted create and reproduce
gender inequality. Women continue to perform the vast majority of the unpaid labor necessary for
maintaining the home and caring for the family (Lachance-Grzela and Bouchard, 2010; England,
Levine, and Mishel, 2020). This inequality is ubiquitous: it persists in both urban and rural areas, in
both developed and developing societies, and regardless of structural features of the family (Davis,
2010). Meanwhile, within labor markets, women disproportionally occupy less well-remunerated job
tracks (England, 2010; England, Levine, and Mishel, 2020). This has deep and wide-ranging con-
sequences for their relative social and material standing in society (Blau and Kahn, 2017).
Researchers increasingly use words like “stalled” to describe progress toward equality (England,
Levine, and Mishel, 2020); and that was before the pandemic, which a United Nations policy brief
warns may set gender equality back by decades (UN, 2020).
Consider some representative data points from the United States.: Even pre-pandemic, in US
households with children under six, women spent about an hour a day providing physical care for
children, while men spent about 25 minutes (American Time Use Survey, 2016, 2018). Women also
do more of the housework that children create. One study found that women did six extra hours of
housework per week, excluding childcare, after the birth of a first child; the study found no increase
among men (Baxter, Hewitt, and Haynes, 2008).
If you find these data dismaying, trend lines won’t console: Compared to the 2010 data, the 2015
and 2017 American Time Use Surveys shows virtually no increase in men’s average hours per day
spent on housework or household management or time spent caring for children. Neither has
women’s time spent on household activities or childcare decreased (American Time Use Survey, 2011,
2016, 2018).
Meanwhile, these same time use surveys show that men’s time spent on paid work increased
during those years, while women’s decreased. Growth in married mothers’ labor force participation
has declined, as has growth in entry into previously male-dominated fields (Cotter, Hermsen and
Vanneman, 2004; Goldin, 2006; Bianchi and Milkie, 2010). Women now earn a majority of post-
baccalaureate degrees but remain underrepresented in high-skilled professions (Antecol, Bedard, and
Stearns, 2016). At the less esteemed end of the occupational hierarchy, occupational segregation by
gender keeps women’s pay depressed. Jobs that involve caregiving, cleaning, and feeding are vastly
disproportionally filled by women. A New York Times analysis of US census data found that over half
of what we now routinely call “essential worker” jobs are filled by women, and that one in three
jobs held by a woman is an essential worker job (Robertson and Gebeloff, 2020). Nonwhite women
DOI: 10.4324/9780367808983-23 239

Gina Schouten
are likelier than anyone else to be doing these essential worker jobs. That same analysis found that
“of the 5.8 million people working health care jobs that pay less than $30,000 a year, half are
nonwhite and 83% are women.”
The material upshots are striking: Organization for Economic Cooperation and Development
(OECD) data reveal that a significant gendered wage gap persists in nearly every economy on earth
(OECD Data, 2021). Women’s greater share of caregiving – both paid and unpaid – is crucially
connected to their persistent disadvantage in terms of earnings and other social and material status
indicators. Studies find that “child penalties,” or “the percentage by which women fall behind men
due to children,” are “gradually taking over as the key driver of gender inequality” (Kleven, Landais,
and Søgaard, 2018). The pandemic will make matters worse, as working parents – mostly, mothers –
shoulder the added childcare burdens resulting from school and daycare closures. A report from the
Pew Charitable Trust found that mothers are three times as likely as fathers to have lost their jobs
during the pandemic, in part because of occupational segregation – women-dominated sectors
like hospitality have been particularly hard hit – and in part because women have been leaving their
jobs as caregiving demands grow too burdensome to manage alongside paid work (Henderson,
2020; see also Taub, 2020; Connolly et al., 2020).
Women’s greater share of caregiving – both for pay and otherwise – and the social and material
consequences raise important questions. The four we’ll discuss in this chapter are:
1. What (if anything) is wrong with this picture?

2. What (if anything) can we do about it?
3. What (if anything) may we do about it?
4. At what cost should we do it?
Each discussion will be necessarily incomplete. I’ll discuss only a range of possible answers, em-
phasizing the interconnectedness of the questions as revealed by the most plausible answers.
1 What’s Wrong with the Gendered Division of Labor?

The most obvious answer to this question is that it’s bad for women. But that answer is under-
informative, both because the badness can be understood along any number of dimensions and because
gendered work norms can also hurt men. Focusing first on the nature of the badness for women, we
can begin with a simple observation: though caregiving work is often difficult and unpleasant, it is not
intrinsically unpleasant, and its difficultness is part of what can make it pleasant and rewarding (Tooley,
2003). But of course, the fact that caregiving work can be rewarding and meaningful under some
circumstances doesn’t establish that the gendered division of labor is unproblematic in ours, or that
those who do a larger share aren’t unfairly badly off.
A better approach to finding the badness of the gendered division of labor for women looks to
the harms and vulnerabilities that contingently accompany the kind of work women do such a large
share of or the harms and vulnerabilities that contingently accrue to them in virtue of that dis-
proportionally large share. Susan Moller Okin’s pioneering work on the gendered division of labor
illustrates this approach. Okin argues that women’s larger share of caregiving is unjust (Okin, 1989).
To make her case, she traces the ways in which social mechanisms like early gender socialization and
unequal bargaining power within domestic partnerships reinforce gendered choices that system-
atically leave women vulnerable.
First, she argues, girls are made vulnerable by their preparation for adulthood. They are socialized
to expect that if they have children, they will have primary responsibility for childcare. This ex-
pectation subsequently leads them to prepare for work and life trajectories that are less esteemed and
less remunerative than those anticipated by boys. The expectation begins with gendered socialization
240
Gender and the Division of Labor
in upbringing and is reinforced by media influences and other cultural and educational messages. For
example, teachers and guidance counselors might consciously or unconsciously tout flexible jobs to
girls and young women.
Unequal preparation then results in many more women than men choosing jobs that accom-
modate the needs of a primary caregiver: namely, needs for flexibility at work and flexibility to take
time away from work for caregiving. Our labor markets are set up such that jobs that accommodate
these needs – jobs in fields like nursing, teaching, sales, hospitality, and administrative services – tend
to be low-paying, low-prestige positions with little room for growth and advancement. In contrast,
more men make early preparation choices without the worry that their work will need to be
conducive to primary caregiving. For example, we see more men than women in prestigious
professional positions that reward long hours with high pay and advancement opportunities.
Consciously or not, girls and boys are prepared differently for adulthood, and this affects the as-
pirations they develop and the jobs they subsequently enter. Because traditionally feminine jobs are
less esteemed and remunerative, this difference amounts to enlisting women into a system of
asymmetric vulnerability relative to men.
This vulnerability deepens with cohabitation and marriage. Due in large part to gendered pre-
paration for adulthood, women who enter heterosexual partnerships don’t do so on equal footing
with male partners. For example,
answers to questions such as whose work life and work needs take priority, and how the
unpaid work of the family will be allocated … are likely to be strongly influenced by the
differences in earning power between husbands and wives. (Okin, 1989: 146)
Women de-prioritize paid work to accommodate unpaid caregiving needs and forego opportunities
to develop their earning potential or to improve their individual material standing. And even partial
breadwinner/caregiver specialization grows more entrenched over time, as inequalities within the
home and workplace powerfully reinforce and perpetuate one another.
Unsurprisingly, asymmetric vulnerability reaches its height in the many cases of marriages that
end in divorce. We might add that Okin’s observation is true as well of non-marital partnerships:
because women are asymmetrically materially dependent in partnerships and because their con-
tributions often are not adequately recognized, they are asymmetrically materially vulnerable – and
all too often they are absolutely vulnerable – when partnerships dissolve. The fact that they often
retain custody of children can exacerbate their material vulnerability. This is to say nothing of the
asymmetric vulnerability between women and men generally once we consider mothers who never
had cooperating domestic partners, but who must cooperate and negotiate with men on unequal
footing in pursuits outside of the home.
Though Okin’s book was published in 1989, gendered socialization patterns persist, and women’s
greater share of caregiving continues to render them asymmetrically vulnerable. But so far, this
doesn’t fully answer the question with which this section began: what’s wrong with the gendered
division of labor? Asymmetric vulnerability abounds. Children are asymmetrically vulnerable relative
to their parents, for example, and that’s not unjust. Okin’s answer invokes features of gendered
asymmetric vulnerability that distinguish it from children’s vulnerability: gender perpetuates socially
constructed and remediable asymmetric vulnerability.
The social and material inequalities that comprise women’s vulnerability may be conceptualized
in different ways, and subsequent feminist theorists have illuminated this by developing different
accounts of what precisely is unjust in the gender disparities we observe across societies. While
women are generally on the losing end of gender disparities, many such accounts can appreciate that
men suffer gendered harms – and particularly harms from gendered labor norms – as well. Much of
this work aims to diagnose the injustice of the gendered division of labor in a way that shows the
241
Gina Schouten
injustice to be politically actionable, or legitimately susceptible to political remediation. In other

words, these accounts of what’s wrong with the gendered division of labor are meant as foundations
on which to situate an answer to the question I’ll discuss in Section 3: what may we do about it?
Here’s a brief sample of views in this space:
S.A. Lloyd defends a principle of justice that calls for rough equality in citizens’ shares of certain
publicly recognized values, including liberty, stability, security, opportunity, and health. She argues
that her equality principle favors equalizing the burdens of reproduction (Lloyd, 1998). Martha
Nussbaum regards the gendered division of labor as unjust in virtue of causing an unjust distribution
of central human capabilities (Nussbaum, 2000); similarly, Ingrid Robeyns argues for a conception
of gender justice that counts socially caused inequalities in men’s and women’s capability sets as
unjust (Robeyns, 2007). Anca Gheaus argues for a conception of gender justice that counts society as
unjust so long as central components of good lives are unequally costly to women and men (Gheaus,
2012). Clare Chambers argues that injustice arises when individuals are socially constructed to make
choices based on preferences that are harmful to them (Chambers, 2008). Still others argue that
injustice arises when the gendered division of labor undermines civic equality (Watson and Hartley,
2018) or when it offends against free and equal citizenship in some other way (Schouten, 2019).
2 What Can We Do About It?

Possible social policy solutions include work-time regulation, incentives for workplace flexibility,
robust enforcement of discrimination prohibitions, and caregiver support provisions. Caregiver
support provisions have received a large share of scholarly attention among those writing about
policy solutions to the gendered division of labor. That’s to be expected: caregiver support seems
absolutely essential to achieving gender equality because of the importance of women’s caregiving
burden in sustaining gender inequality (Bianchi et al., 2012); but caregiver support policy also
presents practical and philosophical complications. The rest of this chapter discusses those compli-
cations. The practical complication, in a nutshell, is this: given prevailing social norms and the
embedding of those norms in social institutions, uptake of caregiver support policy plausibly will be
gendered, with more women than men availing themselves of support. If so, then caregiver support
won’t erode the gendered division of labor; indeed, it risks increasing the gender coding of caregiving
(Zippel, 2009; Blau and Kahn, 2013; Bertrand, 2018).
Consider paid caregiving leave. Suppose the state provides 12 months of job-protected paid leave to
families after the birth or adoption of a new child, and families can use the leave however they want. In
two-parent families, parents may share leave evenly, or one parent may take it all, or one may take nine
months and the other take three, etc. On this model, women will spend more leave-supported time
out of work than men will (Kleven, Landais, and Søgaard, 2018). This is true for myriad reasons. First,
although gender role attitudes have changed greatly over recent decades, prevailing norms continue to
favor women specializing in caregiving once caregiving needs arise, even if not to the exclusion of paid
work. Social surveys reveal a robust and widespread conviction “that women should work full-time
before having children and after the children have left home, while they should work only part-time or
not at all when they have children living at home” (Kleven, Landais, and Søgaard, 2018: 6; see also
Bertrand, Kamenica, and Pan, 2015; Parker, 2015; Gerson, 2010). Second, and relatedly, gender so-
cialization still prepares women and men for gendered specialization, even if not for such strict spe-
cialization as it once did (England, 2010). Finally, straightforward household economic considerations
make women’s leave-taking, on average, less costly than men’s: because more women choose less
well-remunerated careers that are more flexible to allow for caregiving, and because the effects on
earnings potential of choices to prioritize caregiving are cumulative, men on average face higher
opportunity costs of leave-taking, both in terms of career progression and (assuming wage-replacement
242
rates fall short of full replacement for high earners) in terms of forgone income (Gornick and Meyers,
2003; Kleven, Landais, and Søgaard, 2018).
In short, because gender affects policy use, gender-neutral policy might not further – and indeed
might frustrate – efforts to erode the gendered division of labor. This possibility lurks not only for
caregiver leave but also for direct cash support for parents, which subsidize parents’ withdrawal from
work to provide care, thus effectively subsidizing mothers’ withdrawal. Indeed, the worry plausibly
arises in some form for all caregiver-support amenities that are negatively priced in the labor market:
because women are likelier than men to use leaves or other family-friendly workplace amenities,
offering those amenities on gender-neutral terms won’t meaningfully erode the gendered division of
labor (Kleven, Landais and Søgaard, 2018).
A recent study using Danish administrative data from 1980 to 2013 illustrates. During the thirty-
year period studied, parents enjoyed generous job-protected paid parental leave during the early
months after welcoming a new child; and, once the child turned six months or one year old (de-
pending on the policy year in question), parents had access to high-quality and heavily subsidized
options for purchasing care outside the home. Even so, significant gender inequality persists in
Denmark (Kleven, Landais, and Søgaard, 2018). The study found that the inequality in earnings and
other economic outcomes that persists in Denmark today between women and men “is all about
children” (Kleven, Landais, and Søgaard, 2018: 32). That’s because women and men make different
choices about whether and how to use the leave on offer: women, more than men, “favor family
amenities over pecuniary rewards,” and suffer an earnings penalty as a result (Kleven, Landais, and
Søgaard, 2018: 2). Having children impacts women’s labor market pursuits, including their labor
force participation, hours worked, occupational rank, and sector.
Nancy Fraser has observed that much feminist work on gender inequality construes equity either
as treating women just the same as men or as treating women differently than men insofar as they
currently are different than men (Fraser, 1994). She finds both approaches problematic. The sa-
meness approach is problematic because treating people the same within a social arrangement already
built on the presumption that they are different will not result in justice. The difference approach is
problematic because aggregate differences between women and men are due in large part to gen-
dered patterns of choice that result from existing injustice, so treating women differently in response
to these patterns can exacerbate the injustice.
The case of caregiver support policy illustrates Fraser’s point. We’ve seen that treating women
and men the same by letting families decide for themselves who takes how much leave plausibly
won’t erode the gender coding of work. But certain ways of treating women and men differently
won’t do the trick either. Consider a policy that supports women’s greater role in caregiving, for
example by offering significant portions of parental leave specifically for mothers. This instance of
the difference approach treats gendered patterns as inevitable or salutary, thus further entrenching
those patterns and burdening those who deviate from them.
But now consider a caregiver support policy that explicitly aims to promote caregiving among
men and labor market attachment among women: paid leaves allocated to each parent individually,
on a non-transferable basis. When leaves are allocated to individuals rather than to household units,
they generate an incentive effect: each parent’s leave is forfeit if that parent doesn’t take it (Zippel,
2009; Bertrand, 2018).1 Such a policy plausibly will induce more men to take leave, and in so doing,
help to erode norms against male leave-taking, promote parental intimacy and bonding among both
leave-takers, increase labor market attachment among women, and promote gender-egalitarian
attitudes and behavior in the future generations (Cunningham, 2001; Coltrane, 2009; Zippel, 2009;
McGinn, Castro,and Lingo, 2018). In a sense, non-transferrable leaves treat women and men
identically: Both are allocated leave as individuals on a use-it-or-lose-it basis. But the policy aims
to intervene on patterns of behavior by inducing women and men to respond to leave offerings
243
Gina Schouten
in different ways: it aims to get more women to take less leave than they would under a maximally
flexible leave regime, and more men to take more leave.
Individual-allocated, non-transferrable leaves may thus be better understood as an instance of
treating women and men differently. This instance of the difference approach might seem to
presume a “universal breadwinner” model of gender equity, which aims for both women and men
to behave as men behave now: for both to sustain substantial paid work responsibilities and make
extensive use of commodified caregiving. As a vision of gender equity, the universal breadwinner
model is flawed. As Fraser argues, it denigrates caregiving work as less valuable and it fails to support
the most badly off women, who either still will not be heavily involved in work outside the home or
who will be involved in less remunerative work outside the home.
But in fact, individual-allocated and non-transferrable leaves do not presume the universal
breadwinner model of gender equity. Such leave policies do aim to support labor market attachment
among women, but not for its own sake. For one thing, these leaves also aim to promote caregiving
and time out of paid work among men. More importantly, by way of incentives for partners to use
caregiver support in gender-norm-transgressive ways, these policies aim to change social norms by
changing patterns of behavior, disrupting the mechanisms that reproduce the gendered division of
labor over time. The policies treat women’s life patterns as valuable life patterns that are not to be
disparaged or denigrated – so much so that they undertake to make men’s life patterns look more like
women’s do now, featuring a combination of caregiving and labor market participation. Rather than
a universal breadwinner vision, then, individual-allocated, non-transferrable caregiver leaves are
part of a social support package that can bring about Fraser’s “third-way” vision, which effectively
seeks gender equity through gender deconstruction: On the third way, social policy aims to “induce
men to become more like most women are now – that is, people who [also] do primary care work”
(Fraser, 1994: 611).
Fraser’s third way incorporates a plurality of distinct normative considerations, including anti-
poverty, equal respect for women and men, substantive (distributive) equality, parity of participation in
socially valued activities, and decentering of androcentric measures of social value (Fraser, 1994: 595). I
think this vision of gender equity is indeed one we ought to strive for and that individual-allocated,
non-transferrable caregiver leaves are one piece of a social support system that can push us in that
direction. This vision presents a compelling answer to the practical question of how to support car-
egiving without entrenching its gender coding. But in so doing, it raises complicated philosophical
questions, including: are individual-allocated, non-transferrable caregiver leaves a legitimate use of political
power?
3 What May We Do About It?

So far, we’ve discussed candidate diagnoses of what’s wrong with the gendered division of labor,
focusing specifically on the gendered division of labor that results from patterns of choice concerning
how caregiving and paid labor are shared among cooperating domestic partners. Then, turning to
the question of what we can do about the gendered division of labor, I presented a certain form of
caregiver support – individual-allocated caregiver leave – as one part of an institutional response that
could help to erode the gendered division of labor so understood.
Such leaves are not the only part of the institutional response we need. They would do little, on
their own, to improve the situation of those women without labor market attachment or with labor
market attachment that isn’t particularly valuable or remunerative. But these leaves are interesting to
consider because they raise worries that certain ways of using social support to erode the gendered
division of labor constitute illegitimate uses of political power. Behind these worries there often stands a
commitment to the ideal of justificatory reciprocity. Such a commitment holds that a liberal de-
mocratic state should not act on the basis of a reasonably contested ideal of the good life. Instead, it
244
should only act in ways that are justifiable on the basis of reasons that, in some sense, all reasonable
citizens can recognize as such.
This ideal does not forbid all social policy that disproportionally burdens certain values. All social
policy – and all omission of social policy – will impose disparate costs and benefits on different citizens
depending on what they value and how they want to live. Consider an example: Public funding
for compulsory education can plausibly be said to impose higher monetary costs on citizens without
children, since they pay into an education system that benefits everyone but benefits those with
children most. And the compulsory part of public education plausibly imposes higher social costs
on citizens who participate in certain tradition-based ways of life, since compulsory education in-
duces the kind of exposure to questioning and diversity that is a driver of defection, thereby making
such ways of life more difficult to sustain over time. These disparate costs may be perfectly compliant
with reciprocity, depending on the reasons we can invoke to justify them. Properly understood, a
commitment to reciprocity proscribes all reasonably contested value commitments about what
constitutes a good life as inadmissible reasons to invoke in justifying public policy. Whatever disparate
impacts they will invariably have, policies must be justifiable without invoking such contested values.
Of course, public policy choices should be informed by some values. The values that reciprocity
permits us to invoke to justify social policy include values that are part of the public political culture:
values that enjoy broad support from a diverse array of citizens. So, for example, we may not
institute a strongly progressive inheritance tax on the grounds that the interests of the rich are of
comparatively little importance or that an acquisitive worldview is fundamentally flawed; but ar-
guably we may institute such a policy on the grounds that it is necessary to ensure broad equality of
opportunity in competitions for better-rewarded social positions. This is because antagonism toward
the wealthy is not a value that enjoys broad support from a diverse array of citizens, but equal
opportunity arguably is.
Flexible family support policy can be justified on the back of values that enjoy broad support from
diverse citizens who have different ideas about what makes for a good life or a good society. We
might make the case, for example, that well-raised child generate public benefits and so the public
should support those who do the raising. But to argue for family support policy that is designed
explicitly to erode gendered caregiving norms – which we apparently need to do to justify the less
flexible non-transferrable, individual-allocated leaves – we need to argue for the erosion of gendered
caregiving norms as a policy end in its own right.
As we’ve seen, reciprocity permits broadly shared values to do the work of justifying social
policy. More strongly still, it designates certain political values as values that we can take for granted as
shared because those values encode fundamental interests we all share as free and equal citizens. These
shared values include the presumption that we ought to be treated as free and equal citizens, and that
we have fundamental shared interests both in protecting our status as free and equal citizens and in
preserving mutual respect among citizens construed as such. We can take these fundamental interests
to be common ground, despite our many disagreements.
Why do these values have such privileged status? One way to see the answer is to notice that the
values in question are what justify a commitment to liberal toleration – and reciprocity – in the first
place. We aim to preserve reciprocity precisely in order to preserve mutual respect among free and
equal citizens. In a liberal democracy, we construe ourselves and all other citizens as political equals with
a presumption of freedom to pursue the good life as each sees it. Of course, to enjoy the benefits of
living in a cooperative society with other citizens, we must all accept some constraints on that pre-
rogative. Reciprocity tells us that the constraints we face are legitimate if and because they can be
justified to us on terms that we can accept in our capacity as free and equal citizens. By acting de-
mocratically only in mutually justifiable ways, we express civic respect even to those who are con-
strained or burdened by our so acting. This means that we can’t invoke reasons that are exclusive to
some disputed view of the good and that we can’t fail to heed the reasons that are implied by the liberal
245
Gina Schouten
democratic values that justify the whole enterprise: social policy supported by the bedrock commitment
to mutual respect among free and equal citizens is social policy that we cannot legitimately fail to enact.
We can see more clearly, now, that a commitment to reciprocity does not entail a commitment
to keeping values out of politics. But it permits only those values that are implicit in our public-
political culture or implied by the fundamental liberal commitment to mutual civic respect that
underpins the very ideal of reciprocity. Because reciprocity positively demands that this latter set of
values be heeded, liberal legitimacy cuts both ways: if policies to de-gender caregiving can be
justified only by invoking contested values, then those policies are illegitimate. But if such policies
are favored by the fundamental shared interest in preserving mutual respect among free and equal
citizens, then it is illegitimate not to enact them.
Is there a case to be made on grounds of mutual respect in favor of individual-allocated, non-
transferrable caregiving leave? We can begin to explore this question by revisiting our findings from
Section I about what’s wrong with the gendered division of labor. I wrote in that section that feminist
diagnoses of the badness of the gendered division of labor have often been developed with an eye
toward identifying a politically actionable injustice: toward surfacing in the gendered division of labor a
problem that is legitimately susceptible to political remediation.
Many of the strategies for diagnosing the injustice that I canvassed there invoke distributive
injustices that gendered caregiving perpetuates. Women’s greater share of caregiving leaves them
materially worse off than men and disproportionally materially poor in absolute terms. For example,
I wrote that S.A. Lloyd defends a principle of justice that calls for rough equality in citizens’ shares of
publicly recognized values like liberty, security, opportunity, and public health (Lloyd, 1998). But
conceptually, to identify a distributive injustice sustained by the gendered division of labor is not yet
to establish that any particular thing may be done about it.
Specifically addressing the question of political remediation, Amy Baehr interprets Lloyd’s cri-
terion of sexual equality as “a robust principle of antidiscrimination” and defends it as a legitimate
component of a political conception of justice (Baehr, 2008). In a similar vein, Nussbaum observes
that the gendered division of labor gives rise to an unjust and politically actionable inequality in
capabilities: because a conception of the person as exercising various human capabilities is one that
“corresponds to human experience,” Nussbaum argues that “there is good reason to think that it can
command a political consensus in a pluralistic society” and therefore “form the core of a political
conception that is a form of political liberalism” (Nussbaum, 2000: 56): it can justify political re-
mediation consistent with reciprocity.
I agree with much of these arguments for gender equality that rest on the clear fact that patterns
of choice surrounding caregiving and paid work give rise to material inequalities, capability in-
equalities, and absolute deprivation. But I think that, in important ways, these arguments misplace
the actionable injustice of the gendered division of labor. Elsewhere I argue that the distributive,
asymmetrical vulnerability approach to diagnosing the gendered division of labor does surface unjust
and politically actionable inequalities, but these inequalities are not unjust in virtue of being gendered
(Schouten, 2016, 2019). Of course, the fact that the inequalities are gendered is no accident; but
from the perspective of distributive justice, their being gendered is not essential to their being unjust.
A woman who suffers a certain capability deficit is not more entitled or more urgently entitled than
a man who suffers the same capability deficit in virtue of her being a woman, or in virtue of her
deficit being due to caregiving. Each is entitled to the share of material goods (or capabilities, etc.)
that justice calls for on their behalf by reference to the margin by which they fall short of equality,
and none is more entitled in virtue of belonging to a social group that figures disproportionally
among those most entitled.
Conversely, I argue, when we think about inequalities that are otherwise unobjectionable, those
inequalities do not become unjust when we add an element of gender. Suppose for the sake of
argument that justice permits marginal tradeoffs between caregiving for one’s own children and
246
remuneration. Now, in a just society, caregiving parents will receive social support for time spent
doing that valuable work; I’m only asking us to suppose that at some point, that support may justly
decline relative to what the caregivers would have made had they stayed in the labor market. In that
case, the inequality itself does not become unjust if women figure disproportionally among those who
make this tradeoff. To be sure, there may be something unjust in what causes the disproportionality.
But the inequality does not become unjust because it is gendered. The gendered division of labor is
certainly implicated in a lot of distributive injustice, but it’s not essentially a problem of distributive
injustice. And gendered inequalities may be politically actionable, but they’re not distributively
unjust in virtue of being gendered.
If distributive inequality can’t explain the injustice of the gendered division of labor in a way that
licenses political remediation, social equality looks more promising. Social equality refers to re-
lationships of equal respect, equal civic standing, and non-domination among free and equal citizens.
It is because we are owed social equality in this sense that mutual respect demands for each of us a
justification for political action that we can accept: one based on reasons that we can recognize as
such. Those who do relatively more caregiving or who participate relatively less in labor markets do
not currently stand as social equals: in their work on political liberalism and gender, Lori Watson and
Christie Hartley argue that the socially egalitarian ideal of reciprocity at the heart of political lib-
eralism “requires (1) the eradication of social conditions of domination and subordination relevant to
democratic deliberation among equal citizens and (2) the provision of social conditions of re-
cognition respect” (Watson and Hartley, 2018: 160). On these grounds, Watson and Hartley argue,
the ideal of reciprocity identifies the gendered division of labor as a politically actionable social
inequality.
I think Watson and Hartley’s argument gets a lot right, but it gets something crucial wrong. The
social inequality to which they draw our attention is of limited help in justifying policy to de-gender
caregiving, because the problem it points to is not that women and men don’t share equally in
caregiving, but rather that caregiving is wrongly devalued by our social practices. The problem, that
is, is that caregiving currently is not regarded as a source of social status in its own right. Caregivers
should stand as equals, whether they are disproportionally women or disproportionately men or
equally women and men. We should restore social equality by elevating the status of caregiving, and we
can do this without persuading women to do less of it or persuading men to do more. As someone
who favors de-gendering caregiving, I think we should also work to equalize shares of it. But this
can’t be for the sake of equal standing, because equal standing surely gives us more reason to elevate
the status of roles that we currently unfairly and mistakenly devalue than to see to it that those
devalued roles are shared equally between women and men: full-time, in-home caregivers should
stand as equal citizens, and the fact that they presently don’t gives us reason to elevate the status of
caregiving, not draw women out of it and men into it. In short: the under-valuing of caregiving is a
politically actionable problem; but it’s not one that supports political action to erode the gendered
division of labor as such.
If distributive justice and social equality don’t cut close enough to what’s politically actionable in the
gendered division of labor itself, what does? The missing ingredient is the free in “free and equal
citizens” (Schouten, 2019). Imagine that we somehow achieved distributive and social equality: sup-
pose caregiving and labor market participation were fairly rewarded and equally socially esteemed. Still,
women might be socially steered into more caregiving-intensive roles and men into more labor-
market-intensive roles. The source of this steering is the socially embedded assumption that sex dictates
work specialization. And therein lies the politically actionable complaint of injustice against the gen-
dered division of labor. That source of gendered social steering is objectionable from the perspective of
free and equal citizenship.
How so? I argue that the socially embedded assumption that sex rightly dictates work speciali-
zation is objectionable from the perspective of free and equal citizenship because that assumption is
247
Gina Schouten
inimical to autonomy. An autonomous person can reflect upon, revise, and reject the social roles and
affiliations that fundamentally shape her life. Her choices in domains central to her identity are not
effectively determined by normalized and institutionally embedded assumptions that members of a
social group to which she belongs are best suited to populating particular roles. We can be autonomous
in a gendered society because social norms do not determine our choices. And while gender norms do
attach social costs to some of the options we choose among, that in itself is no problem from the
perspective of autonomy, because our options always carry contingent social and material costs. But
notice that a social arrangement can be inimical to some value without making the value impossible to
realize. This can occur when the arrangement is predicated on the assumption that citizens will not
realize or aspire to realize the value in question. Our social arrangement is inimical to autonomy not
because it makes autonomy impossible but because it presumes that citizens will behave non-autonomously:
that they will specialize by sex into caregiving and labor market roles. Even without making auton-
omous choice impossible, our social arrangement is predicated on the institutionalized assumption that
sex will dictate the work that one does (Schouten, 2019).
How does that gendered assumption infuse our social arrangement? Labor markets are still designed
for workers with “someone else at home” to specialize in unpaid caregiving so that the “working”
partner can devote himself fully to wage earning (Williams, 2000). Employers are permitted to impose
demands on workers that are incompatible with workers having serious caregiving commitments, and
success at work requires living up to such demands. The institutionalized assumption that sex will
determine work specialization explains why employers have been empowered to impose those de-
mands. (It also explains the dearth of support for substitute caregiving in the United States. Because we
assume parents – mothers – will internalize the costs of caring for their children, we have neglected to
develop social mechanisms for sharing those costs.) And, as we’ve seen, the gendered assumption that
explains this institutional arrangement continues to influence our strategies for navigating it.
The entrenched assumption that sex matters for caregiving makes it socially and materially costly
to avoid gendered caregiving. A social arrangement that presumes that sex has this importance – and
thus makes it very costly to arrange one’s domestic life in defiance of that presumption – is inimical
to autonomy. And so these institutional arrangements and social norms are objectionable on grounds
of autonomy even though individuals can autonomously choose to comply with or flout those
norms. I argue on this basis that the gendered division of labor is objectionable at the bar of mutual
respect among free and equal citizens.2 Because that bedrock liberal value of mutual respect supports
a case for political remediation, we have a compelling political case for individual-allocated, non-
transferrable caregiver support policy.
4 At What Cost May We Do It?

I’ve argued that individual-allocated, non-transferrable caregiver leaves should be construed as
aiming to realize Fraser’s third-way vision of gender equity and that they can constitute a legitimate
exercise of political power. But such caregiver leave policy is only one part of a social arrangement to
realize Fraser’s third-way vision. To see this vividly, recall the first normative consideration en-
compassed by the third way: antipoverty. Paid leave is of little immediate use to women without
paid work and of relatively less immediate use to women whose paid work involvement is relatively
less remunerative and enjoyable. In the long run, valuing caregiving enough to ask men to do it too
will benefit women and men across the economic spectrum. But in the short term, this piece of
policy will certainly not alleviate poverty.
What this means, I think, is that if we need our caregiver support policy to also serve as economic
justice policy, we should support caregivers in some other way – perhaps through the kind of
flexible caregiver resource accounts that risk exacerbating gender coding of caregiving. Why?
Because Fraser is right that antipoverty is an overridingly important cause of justice. If our policy
248
context presents us with a tradeoff between reducing poverty and eroding gender norms, justice
favors anti-poverty. In a policy context that forces caregiver support policy to double as economic
justice policy, we face just that tradeoff. If we can’t ameliorate poverty and support caregiving in a
way that erodes gender norms, we should prioritize the former (Schouten, 2020).
But Fraser’s principles are principles to guide the design of a new welfare state regime, of which
designated caregiver support is presumably only a part. Individual-allocated, non-transferrable leaves
should be part of such a vision. If we face a tradeoff between anti-poverty and gender equity un-
derstood in the spirit of Fraser’s third way, then we should object to the circumstances that force that
tradeoff. In those circumstances, antipoverty is the first order of business. But even if their inability
to ease poverty is a reason to postpone implementing individual-allocated, non-transferrable care-
giver leaves, it is not a reason to think that those leaves are unjustified. They are a legitimate policy
response to the gender inegalitarian consequences of labor markets, even if they are no silver bullet
for all gendered social ills.
Notes
1 More strongly, policies may condition one parent’s access to paid leave on her co-parent taking leave too.
See Bergmann, 2009; Brighouse and Wright, 2009; Coltrane, 2009; Zippel, 2009.
2 This characterization omits an important premise of the argument for brevity. See Schouten, 2019, Ch. 6.
References
American Time Use Survey (2011) USDL-11-0919. Bureau of Labor Statistics: U.S. Department of Labor.
Antecol, H., Bedard, K. and Stearns, J. (2016) Equal but Inequitable: Who Benefits from Gender-Neutral Tenure
Clock Stopping Policies? IZA Discussion Papers 9904. Institute for the Study of Labor (IZA). Available at:
http://EconPapers.repec.org/RePEc:iza:izadps:dp9904.
Baehr, A. (2008) “Perfectionism, Feminism and Public Reason,” Law and Philosophy, 27, 2, 193–222.
Baxter, J., Hewitt, B. and Haynes, M. (2008) “Life Course Transitions and Housework: Marriage, Parenthood,
and Time on Housework,” Journal of Marriage and Family, 70, 2, 259–272. doi: 10.1111/j.1741-3737.2008.
00479.x.
Bergmann, B. (2009) “Long Leaves, Child Well-Being, and Gender Equality,” in Wright, E. O. ed. Institutions
for Gender Egalitarianism: Creating the Conditions for Egalitarian Dual Earner/Dual Caregiver. Madison;
University of Wisconsin Press (Real Utopias Project).
Bertrand, M. (2018) Coase Lecture – the Glass Ceiling. SSRN Scholarly Paper ID 3133423. Rochester, NY;
Social Science Research Network. Available at: https://papers.ssrn.com/abstract=3133423 (Accessed: 22
March 2018).
Bertrand, M., Kamenica, E. and Pan, J. (2015) “Gender Identity and Relative Income within Households,” The
Quarterly Journal of Economics, 130, 2, 571–614. doi: 10.1093/qje/qjv001.
Bianchi, S. et al. (2012) “Housework: Who Did, Does or Will Do It, and How Much Does It Matter?,” Social
Forces, 91, 1, 55–63. doi: 10.1093/sf/sos120.
Bianchi, S. and Milkie, M. A. (2010) “Work and Family Research in the First Decade of the 21st Century,”
Journal of Marriage and Family, 72, 3, 705–725. doi: 10.1111/j.1741-3737.2010.00726.x.
Blau, F. D. and Kahn, L. M. (2013) Female Labor Supply: Why is the US Falling Behind? Working Paper 18702.
National Bureau of Economic Research. doi: 10.3386/w18702.
Blau, F. D. and Kahn, L. M. (2017) “The Gender Wage Gap: Extent, Trends, and Explanations,” Journal of
Economic Literature, 55, 3, 789–865. doi: 10.1257/jel.20160995.
Brighouse, H. and Wright, E. O. (2009) “Strong Gender Egalitarianism,” in Institutions for Gender Egalitarianism:
Creating the Conditions for Egalitarian Dual Earner/Dual Caregiver. Madison; University of Wisconsin Press
(Real Utopias Project).
Chambers, C. (2008) Sex, Culture, and Justice: The Limits of Choice. University Park, Pennsylvania; Pennsylvania
State University Press. Available at: http://www.jstor.org/stable/10.5325/j.ctt7v2d1 (Accessed: 6 December
2017).
249
Gina Schouten
Coltrane, S. (2009) “Fatherhood, Gender and Work-Family Policies,” in Wright, E. O. ed. Institutions for
Gender Egalitarianism: Creating the Conditions for Egalitarian Dual Earner/Dual Caregiver. Madison; University
of Wisconsin Press (Real Utopias Project).
Connolly, K. et al. (2020) ““We are losers in this crisis”: research finds lockdowns reinforcing gender in-
equality,” the Guardian, 29 May. Available at: http://www.theguardian.com/global-development/2020/
may/29/we-are-losers-in-this-crisis-research-finds-lockdowns-reinforcing-gender-inequality (Accessed: 24
May 2021).
Cotter, D., Hermsen, J. and Vanneman, R. (2004) Gender inequality at work. New York; Russell Sage.
Cunningham, M. (2001) “Parental influences on the gendered division of housework,” American Sociological
Review, 66, 2, 184–203.
Davis, S. N. (2010) “The answer doesn’t seem to change, so maybe we should change the question: A
commentary on Lachance-Grzela and Bouchard (2010),” Sex roles, 63, 11–12, 786–790.
England, P. (2010) “The Gender Revolution: Uneven and Stalled,” Gender and Society, 24, 2, 149–166.
England, P., Levine, A. and Mishel, E. (2020) “Progress toward gender equality in the United States has slowed
or stalled,” Proceedings of the National Academy of Sciences, 117, 201918891. doi: 10.1073/pnas.1918891117.
Fraser, N. (1994) “After the Family Wage: Gender Equity and the Welfare State,” Political Theory, 22, 4, 591–618.
Gerson, K. (2010) The Unfinished Revolution: Coming of Age in a New Era of Gender, Work, and Family. Oxford
University Press.
Gheaus, A. (2012) “Gender Justice,” Journal of Ethics & Social Philosophy, 6, 1.
Goldin, C. (2006) “The Quiet Revolution That Transformed Women’s Employment, Education, and Family,”
The American Economic Review, 96, 2, 1–21.
Gornick, J. and Meyers, M. (2003) Families That Work: Policies for Reconciling Parenthood and Employment. Russell
Sage Foundation.
Henderson, T. (2020) “Mothers Are 3 Times More Likely Than Fathers to Have Lost Jobs in Pandemic,” 28
September. Available at: https://pew.org/368K5Jm (Accessed: 24 May 2021).
Kleven, H., Landais, C. and Søgaard, J. E. (2018) Children and Gender Inequality: Evidence from Denmark.
Working Paper 24219. National Bureau of Economic Research. doi: 10.3386/w24219.
Lachance-Grzela, M. and Bouchard, G. (2010) “Why Do Women Do the Lion’s Share of Housework? A
Decade of Research,” Sex Roles, 63, 11, 767–780. doi: 10.1007/s11199-010-9797-z.
Lloyd, S. A. (1998) “Toward a liberal theory of sexual equality,” J. Contemp. Legal Issues, 9, 203.
McGinn, K. L., Castro, M. R. and Lingo, E. L. (2018) “Learning From Mum: Cross-National Evidence
Linking Maternal Employment and Adult Children’s Outcomes,”. Available at: https://www.hbs.edu/
faculty/Pages/item.aspx?num=53768 (Accessed: 28 May 2018).
Nussbaum, M. (2000) “The future of feminist liberalism,” in. Proceedings and Addresses of the American
Philosophical Association, JSTOR, pp. 47–79.
OECD Data (2021) ““Gender Wage Gap” OECD Website,”. Available at: https://data.oecd.org/earnwage/
gender-wage-gap.htm (Accessed: 5 July 2021)
Okin, S. M. (1989) Justice, gender, and the family. Basic Books.
Parker, K. (2015) “Women more than men adjust their careers for family life,” Pew Research Center, 1 October.
Available at: http://www.pewresearch.org/fact-tank/2015/10/01/women-more-than-men-adjust-their-
careers-for-family-life/ (Accessed: 2 March 2018).
Robertson, C. and Gebeloff, R. (2020) “How Millions of Women Became the Most Essential Workers in
America,” The New York Times, 18 April. Available at: https://www.nytimes.com/2020/04/18/us/
coronavirus-women-essential-workers.html (Accessed: 24 May 2021).
Robeyns, I. (2007) “When will society be gender just?,” in Browne, J. ed. The Future of Gender. Cambridge:
Cambridge University Press, pp. 54–74.
Schouten, G. (2016) “Is the Gendered Division of Labor a Problem of Distribution?,” in Sobel, D., Vallentyne,
P., and Wall, S. eds. Oxford Studies in Political Philosophy, Volume 2. Oxford University Press, pp. 185–206.
doi: 10.1093/acprof:oso/9780198759621.003.0008.
Schouten, G. (2019) Liberalism, Neutrality, and the Gendered Division of Labor. New York, NY; Oxford
University Press.
Schouten, G. (2020) “Justice and Legitimacy in Caregiver Support: Managing Tradeoffs Between Gender
Egalitarian and Economic Egalitarian Social Aims,” in Bhandary, A. and Baehr, A. eds. Caring for Liberalism:
Dependency and Liberal Political Theory. Routledge Press.
Taub, A. (2020) “Pandemic Will “Take Our Women 10 Years Back” in the Workplace,” The New York Times, 26
September. Available at: https://www.nytimes.com/2020/09/26/world/covid-women-childcare-equality.
html (Accessed: 24 May 2021).
Tooley, J. (2003) The Miseducation of Women. 1st edition. Chicago: Ivan R. Dee.
250
UN Secretary-General. (2020) “The Impact of Covid-19 on Women,”. Available at: https://www.unwomen.

org/-/media/headquarters/attachments/sections/library/publications/2020/policy-brief-the-impact-of-
covid-19-on-women-en.pdf?la=en&vs=1406. (Accessed: 24 May 2021).
Watson, L. and Hartley, C. (2018) Equal Citizenship and Public Reason: A Feminist Political Liberalism. Oxford
University Press (STUDIES IN FEMINIST PHILOSOPHY).
Williams, J. (2000) Unbending Gender: Why Family and Work Conflict and what to Do about it. Oxford University
Press.
Zippel, K. (2009) “Institutions that Support Gender Equality in Parenthood and Employment,” in Wright, E. O.
ed. Institutions for Gender Egalitarianism: Creating the Conditions for Egalitarian Dual Earner/Dual Caregiver.
Madison; University of Wisconsin Press (Real Utopias Project).
251
20
HOUSING MARKETS
Kristina Meshelski
Though they are often overlooked, struggles over housing have been at the center of many mo-
ments of social upheaval. For the people who experienced these events at the time, a lack of
adequate housing was cited as the major cause of the Zoot Suit riots, the Watts riots, and the Detroit
riots of 1967. Rent strikes were a tactic during the Great Depression, and protests of segregated
housing were as integral to Civil Rights era strategies as the more famous protests of segregated buses
and lunch counters. Frederich Engels wrote that “all oppressed classes in all periods suffered more or
less uniformly” from lack of adequate housing (Engels, 1872). Whether it is a cause or an effect of
inequality, housing is one of the many human necessities that is distributed unequally, and yet it is
profoundly different from most other goods.
Even the concept of a housing market is problematic. Economists have pointed to the con-
currence of three features that make the housing market unique: durability, heterogeneity, and
spatial fixity (Maclennan, 1977). Housing can be useful much longer than other consumer goods,
and may not depreciate with age, thus it is durable. Housing units vary greatly in all their char-
acteristics, even similarly priced units, thus it is heterogenous. And of course, housing is located in a
particular space, and cannot travel. We could add that housing is necessary for human beings and has
high transaction costs for moving and transferring (Arnott, 1987). Other goods in the economy may
have these features, but housing is differentiated by having all of these features together.
Analysis that seeks to be relevant to policy will typically focus on either home buying, renting,
homelessness, or racial disparities. In what follows, I will attempt a broad overview of all four
perspectives, focusing on the United States. Though most people in the US own their homes, there
is an increasingly large share of the population that does not. It is also striking that in the most
populous US cities (New York, Los Angeles, Chicago, and Houston are the top four) the majority
of residents rent rather than own. The most up-to-date data can be found through the United States
Census website (US Census Bureau) but useful summaries are often created by real estate companies,
for example (Szekely, 2018).1 Section 1 will summarize the problems that exist and the rest of the
chapter will survey proposed solutions. Sections 2, 3, and 4 will discuss the case for increasing the
supply of housing through what is known as “upzoning.” Sections 5 will discuss controlling the
price of housing. Section 6 will discuss using zoning to combat racial segregation. Section 7 will
discuss democratic control of housing, and Section 8 will discuss the case for making housing more
available by preventing it from being bought, sold, or rented for profit on the private market.
Ultimately, we would benefit from treating housing as a collective resource, which would require
reducing the ability of private actors to profit from it.
252 DOI: 10.4324/9780367808983-24

Housing Markets
1 Injustices in Housing
In the late 1970s in the US, homelessness, as we think of it today, became much more common.
Prior to this, unsheltered homelessness had been considered a problem confined to the past. The
federal definition of homelessness in the US included people living in boarding houses or single-
room occupancy hotels (SROs), and the homeless were thought to be only single men with mental
illness and/or severe alcoholism. But in the late 1970s and through the 1980s, it became obvious that
more and more people were sleeping in the streets, and a larger proportion of them were women
and families. Measurements of homelessness have been contentious since this time, though perhaps
less so since the 1990s. Since 2003, The US Department of Housing and Urban Development
(HUD) has tracked homelessness, and these numbers are now widely accepted in the media as the
official record of homelessness in the US, though there are also now records of homeless students
provided by the Department of Education. According to HUD, there are at least half a million
people who are homeless, but according to the DoE there are at least one million homeless students
(US Department of Education, 2020). The DoE records students living with friends as homeless,
while HUD excludes anyone in this situation as long as they are permitted to stay longer than
14 days. HUD measures unsheltered homelessness using an annual “point in time” count, or PIT,
conducted largely by volunteers and administered by local authorities, that is required to take place
in the last 2 weeks of January. (The January 2021 count was canceled due to COVID.) HUD PIT
counts are clearly flawed in their methodology and execution, but they are unlikely to report an
overcount, so they can at least provide a baseline number. According to HUD, the number of
people experiencing homelessness fell slightly between 2007 and 2016 but has subsequently in-
creased slightly since 2016 (Henry et al., 2020, 6). Most of this increase has been driven by increases
in California.
Eviction and other forms of involuntary displacement are even more understudied than
homelessness. This is because of the differences between the way records are kept and made available
to the public. Some of the best data we have nationally come from The Eviction Lab, which has
since 2017 maintained a national database of evictions and has data going back to 2000 (The Eviction
Lab). The Eviction Lab reports close to a million evictions per year nationally, but these numbers are
far from complete. For example, in California, many eviction records are not public. Eviction Lab
has a record of 47,079 eviction filings in CA in 2016, but in the same year the CA Judicial Council
reported 156,515 eviction filings (Court Statistics Report).2 Even beyond the need for accurate
numbers of how many evictions are processed in courts, we have reason to study informal evictions
that never make it to court. One study of forced moves in Milwaukee found that half of all forced
moves are informal evictions, and only 24% are formal evictions that would have appeared in
Eviction Lab data (Desmond and Shollenberger, 2015).
Interrelated with problems of unequal distribution of housing are problems of racist exclusion and
exploitation. In a larger sense, this is connected to the original colonization of the US territory. But
much more study has focused on the after-effects of slavery, or more accurately how housing policy has
often shifted to maintain the status quo when civil rights have advanced in other areas. Early attempts
to regulate land use in the US through zoning were often ways to confine immigrants into particular
areas. The first attempt to separate an area zoned for industrial use (as opposed to residential use) was a
restriction on where laundries could operate, these laws were passed in various cities in California as
part of a host of restrictions on Chinese businesses in the late 1880s. Zoning began to be used in
the north to separate different types of land use, but zoning was used in the south to designate where
whites and non-whites were allowed to own property and to keep these areas separate. When this type
of racial zoning was found unconstitutional in Buchanan v. Warley (1917), it was on the grounds of
the property owners’ right to sell to anyone, regardless of their race. One response to this ruling was the
253
Kristina Meshelski
use of racial covenants, which were contracts between the seller and buyer of homes that forbid the
new buyer from ever selling to a non-white person. Because it was a private contract, this kind of
restriction was considered constitutional. Another response to Buchanan v. Warley was forms of zoning
that segregated housing based on socioeconomic status and not race. Economic zoning was also an
effective tool for segregation. Some cities merely took the same maps which had designated the zoning
according to race but took out the reference to race and instead designated all of the white neigh-
borhoods exclusively zoned for single-family homes, rather than multifamily apartments (Rothstein,
2017). At this time, few black families could afford to live in single-family homes.
As single-family homes became more numerous and more affordable for more people in the
post-war economic boom, the Supreme Court found racial covenants to be unconstitutional in
Shelley v. Kraemer (1948). This was one of the most important victories won by the National
Assocation for the Advancement of Colored People (NAACP) in their anti-segregation legal
strategy. Unfortunately, this did not sway the Federal Housing Authority (FHA) from their belief
that racial integration brought down property values, and thus that mortgages should not be given
for black buyers in white neighborhoods. They used maps that outlined areas where mortgages were
especially risky as red, thus the practice is called “redlining.” Redlining became technically illegal
with the passing of the Fair Housing Act in 1968.
As explicit racism in the housing market became outlawed, single-family home zoning became
more and more popular. Beginning in the 1920s, single-family home zoning was promoted as a
solution to the “encroachment” of black people into white neighborhoods. But since it does not
explicitly forbid anyone from living in an area based on race, it has always been upheld by courts. As
homeownership became more widespread, restrictions that limited the number of multifamily
dwellings, specified large minimum lot sizes, or simply forbid any construction that wasn’t a single-
family home, became more popular. Many developers at the time considered this kind of zoning an
effective way to increase the value of their developments, and homeowners considered it important
to protecting their largest asset.
The value of suburban single-family homes rose in the 1970s, just as the supply of urban multifamily
housing was becoming run down and cities could no longer afford to invest in infrastructure because
their tax base had moved to the suburbs. Many cities undertook “slum clearance” initiatives that never
replaced the cleared slum housing with improved housing stock. Some temporary increased funding
for low-income home mortgages in the HUD Act of 1968 simply incentivized mortgage lenders to
buy cheap inner-city property and sell to black buyers who had few options. Because low-income
home buyers didn’t have the extra income for the necessary home repairs, these arrangements often
ended in foreclosure, and sometimes the selling of the house to another desperate buyer (Taylor, 2019).
There was a brief increase in funding for public housing as part of the HUD Act, but it was not enough
to make up for the public housing lost over the same period. The Section 8 program, which provides
rental subsidies to low-income Americans, was developed as an alternative to public housing, but it has
never had enough funding to provide for even half of all the families that qualify for the subsidy. The
difference in the quality and value of white and black housing was already too great by the end of legal
redlining for the civil rights programs of the latter half of the 20th century to overcome. And many
within and outside the government resisted these programs so they were never implemented as fully as
would have been necessary. Today, residential segregation remains high and is particularly acute for
black people. The level of segregation that blacks experience does not significantly reduce as their
income rises, whereas the same is not true of other racial groups (Massey, 1993).
2 Upzoning: Removing Regulations

Now that I have outlined these problems I will survey some of the proposed solutions. One solution
calls for less governmental regulation of the housing market and specifically eliminating most zoning
254
Housing Markets
that would restrict the size or density of residential development. Eliminating this kind of zoning is
often referred to as “upzoning” because it would allow taller buildings and more units within each
building. Some who advocate for upzoning do not oppose all forms of governmental regulation, but
I will begin with the case that sees the problem as too much regulation. This view is primarily
influenced by the work of Edward Glaeser and Joseph Gyourko (as the best example, see Glaeser and
Gyourko, 2003). They believe that housing costs should be evaluated separately from poverty. For
them, we can identify a distinct crisis in the affordability of housing only when the costs of housing
greatly exceed its costs of production. If the cost of housing exceeds what people can afford to pay,
this may be a problem of the labor market, not a problem of the housing market necessarily. They
point out that the cost of housing is either below or similar to the cost of construction in most of the
US. So for them, there is only a “housing affordability crisis” in those areas where the cost of
housing greatly exceeds the cost of construction, which are mostly in California, Manhattan, and a
few other cities. To show that zoning is driving up housing costs, they assume that if there would be
a substantial profit in subdividing land into smaller units, then someone would do this. So, the extent
to which land is not optimally subdivided shows that there is some sort of regulation preventing this
or making it less profitable. They also point out that highly regulated areas tend to also have high
housing costs. In these places, large lot sizes coexist with high prices, but generally high prices should
lead to smaller lots, so they conclude that regulation is likely preventing this response.
But, while this is suggestive, their work does not justify the policy response they recommend. First,
they cannot rule out the possibility that people who live in more expensive areas demand more land
use regulation, so high prices cause more restrictive zoning rather than the other way around (Quigley
and Rosenthal, 2005). They have designed their analysis to rule out the theory that land is scarce in
urban areas, so the lack of available land is what drives homes prices up. They assume if they rule out
land scarcity as the explanation, the next best explanation for high prices is the extensive regulation on
home building. That would mean that they do not consider that people are willing to pay more to live
in a particular location. They model all land as interchangeable, no matter where it is. Using their
method, you would find that zoning contributed to price, even in areas where we know it has not
because zoning was absent (Murray, 2021). So, they have not shown that regulation drives housing
prices up; however, the idea that zoning raises the price of housing remains influential. And its in-
fluence is not confined to economists who are skeptical of regulation in general. Democratic politicians
are increasingly promoting the idea that decreasing land-use regulations is a good idea. It may be the
one belief that unites the Obama administration, Elizabeth Warren, the Trump administration, and
many members of the California state legislature. As of this writing, a bipartisan bill has been in-
troduced in Congress, known as the YIMBY Act, which would require recipients of Community
Development Block Grants to report on their progress in removing density restrictions and other land
use regulations that have been known as “exclusionary zoning.”
3 Upzoning: Defeat NIMBYS

The influence of the YIMBY movement, or Yes in My Backyard, has been climbing. YIMBYs
believe that their movement provides an antidote to NIMBYs, who are thought to dominate local
politics and who are able to delay or stop anything from being built with their “not in my backyard”
ethos. NIMBY was first used to describe opponents to landfills and nuclear waste dumps, the im-
plication being that these people were not opposed to hazards generally they just wanted them next
to someone else. The modern usage in housing debates is the same, the person accused of being a
NIMBY is not necessarily opposed to more housing, they just don’t want it located near them, thus
they are selfish. The YIMBY movement will often portray themselves as the underdogs, but this
overestimates the power of NIMBYs. It is true that homeowners are more likely to vote, and local
politics can seem dominated by voices of those who don’t want anything to change. However, this
255
Kristina Meshelski
obscures the fact that individual homeowners don’t have the authority to stop particular develop-
ments that aren’t on land that they own. In a system where most housing is privatized, private
money will build what it perceives to be profitable, whether the neighbors like it or not.
It is possible that homeowners exert influence over zoning, and they use this influence to increase
exclusionary zoning. But the history of zoning does not support the view that NIMBYs have been
powerful actors. When single-family zoning was first popularized, developers themselves often
advocated for it. As the suburbs were first being built, developers wanted to guarantee to their
buyers that the empty land around the new development was not in danger of becoming tenement
apartment buildings in the future. In Houston, which stands out as a rare example of an American
city that never adopted almost any zoning, early zoning advocates tried to promote zoning as a
method for racial segregation, but real estate interests consistently defeated their efforts because they
felt it was an infringement on their property rights (Kaplan, 1980).
Many who believe in the undue power of NIMBYs are influenced on this point by William
Fischel’s “homevoter hypothesis” which holds that zoning comes from the demands of homeowners
looking to protect their assets (Fischel, 2005). However, Fischel’s own work does not purport to
rule out alternative hypotheses. An attempt to empirically test his hypothesis using parcels in New
York City between 2003 and 2009 did find that higher rates of homeowners voting lead to more
“downzones,” and downzoning was strongly correlated with white neighborhoods (Been, Madar,
and McDonnell, 2014). Though the homevoter hypothesis would also expect to find that high-
value parcels are more likely to be downzoned, in fact they found that high-value parcels are more
likely to change zoning in general. They are more likely to be upzoned and more likely to be
downzoned, though the downzoning was even more likely than the upzoning. The authors did not
expect an effect in favor of the homevoter hypothesis in a large metropolis with low home-
ownership rates like New York, so one interpretation is that finding any effect is good evidence for
the homevoter hypothesis. But given that homeownership is out of reach for all but the wealthiest
New Yorkers, this is also consistent with the view that a NIMBY needs to be particularly wealthy to
exert significant power over local land use. Most examples of NIMBY power in California also can
be described as the power of the wealthy. When real estate developers and homeowners were united
in preferring single-family zoning they made a powerful coalition, but when their interests diverge,
the homeowner, qua homeowner, does not have much of a chance.
4 Upzoning: Housing Costs

But what of the YIMBY claim that building more housing will bring down housing costs? There is
some evidence that particular regulations, like parking requirements, may increase the cost of housing
(Shoup, 2011). But the overall case that zoning raises housing prices remains tenuous. (For a good
sympathetic overview of the case against regulations, see Ikeda, 2018.) The case against regulations
relies heavily on analytic work, but accurate modeling of housing prices is known to be difficult
because of the previously stated ways that housing is a unique commodity. The empirical literature on
upzoning does not paint a clear picture. Kuhlmann recently found, studying a recent ban on single-
family zoning in Milwaukee, that upzoning is associated with a rise in prices, particularly for smaller
properties in inexpensive neighborhoods (Kuhlmann, 2021). This would make sense because allowing
more units to be built on the same parcel makes the parcel more valuable to property developers. Some
argue that upzoning is giving away value that should be “captured,” and so they advocate for measures
that would tax development in some way or for countering exclusionary zoning with inclusionary
zoning that requires a certain amount of developments remain affordable.
But what of the impact of upzoning on rents? So far I have been discussing home prices, but
renting is now more common than owning in America’s largest cities, and rents have been rising
faster than wages with the number of Americans paying more than half their income in rent
256
Housing Markets
growing. When new housing is built, it is typically nicer than existing housing stock, both simply
because it is new and because developers often build what is most profitable. So this expands supply
only in the higher tier submarket, and lowers rents within that submarket. Ideally then, some
housing that was previously available only to the highest earners, becomes available for tenants in the
lower submarket, in a process known as filtering. However, filtering can have inter-submarket
repercussions that are not ideal (Galster, 1996). It is also generally inefficient at helping lower-
income tenants, given how long it takes to filter down even in the best-case scenario (Zuk and
Chapple, 2016; Smith-Heimer, 1990). One recent empirical study found that construction of
market-rate apartments was associated with higher rents for nearby cheaper units along with lower
rents for more expensive units for at least two years after new construction (Damiano and Frenier).
In the cities with extremely tight housing markets, filtering tends to move up rather than down, as
higher-income tenants take over older housing stock.
Moreover, getting rid of regulations may not actually increase supply. In some cases it may, and
Dong finds that it does using a quasi-experimental method (Dong, 2021). However, there are simply
many parcels that aren’t developed to their full capacity. It is true that expensive cities tend to have
less of this capacity, but we can’t be sure that developers will always want to build to the full
capacity, as their decisions would naturally be based more on their expected profits than simply what
they are legally allowed to build. One possibility is that if it is true that increasing supply will lower
the prices of homes, or if developers believe that this will happen, then they have an obvious
incentive not to build too much lest they threaten the value of other developments. A good strategy
would be for large developers to withhold some land from the market and wait for potentially higher
returns later. There is some evidence that this is exactly what they do (Murray, 2020). Housing is
also an asset, and the more housing prices rise the more incentive there is to hold on to land with
capacity for more housing until it becomes even more profitable.
5 Price Controls
In any case, whether zoning is likely to work or not, it is not an efficient tool if the problem we aim
to solve is that housing is too expensive. One way to look at the issue would see it mainly as a labor
market problem, with the lack of adequate wages being the main culprit (Rodríguez-Pose and
Storper, 2020). But even if we do not take that view, a more direct way to approach the problem
would be rent control. Rent control enjoyed a brief period of popularity in the 1970s, Nixon
instituted a short-lived federal rent freeze as part of his price controls in 1971 and a few cities with a
tenant population that was large enough to demand it began to pass local rent control ordinances.
But in the 1990s, Massachusetts passed a statewide law eliminating their local rent control, and
California passed statewide limits to rent control that constrained what the local ordinances could
do. Mostly the rent control laws that remain in the US are in California, New York, and New
Jersey. New York City has the strongest rent controls in the US, and it is also the only city in the US
where the majority of rental housing is subject to the controls. However, even in New York, rent is
allowed to rise every year in the majority of units.
It is hard to study the effects of rent control because we have so little of it. In general, price controls
are often frowned upon by economists, because they are thought to limit supply. However, when
there is a monopoly over some crucial thing, a price regulation does make sense according to classical
theory. Many people don’t see housing as controlled by a monopoly, since there are many different
owners. And yet, no two owners can own the same exact house, and two houses in two different
locations are probably not as interchangeable as two different brands of some other good. (It should also
not be underestimated the extent to which housing could become controlled by a literal monopoly –
Blackstone is currently the world’s largest real estate company, and at one point its subsidiary,
Invitation Homes, owned the majority of rental housing in parts of Atlanta and Sacramento.)
257
Kristina Meshelski
One potential problem with rent control is that it can incentivize owners to convert their rental
properties to condos, and this can limit the overall supply of rental housing (Diamond, McQuade,
and Qian, 2019). But just as rising prices can be dealt with directly by controlling the price of rents,
so too can supply issues be dealt with by directly building public housing, rather than waiting for the
private market to provide the necessary supply. The move towards vouchers that allow low-income
tenants to pay rent to private landlords and away from public housing has been unable to meet
the housing needs of the most vulnerable. For one, Congress has never appropriated even close to
enough funds to provide vouchers for everyone that would qualify for them (this is in stark contrast
to other welfare program like food assistance that do attempt to provide for everyone that qualifies).
But also, in urban areas with rising rent, it can be difficult to find units cheap enough. And then
finally, when there are units available, landlords are not required to take the voucher. It was only last
year that California made it illegal to discriminate against tenants who use Section 8 vouchers. Still,
discrimination of all kinds is common in housing, as it is very difficult to enforce. In expensive cities,
where rent far exceeds the cost of maintaining the property, it would be cheaper for the state to
provide low-income housing itself, rather than pay the private landlord.
In general, landlords are inefficient. Building and maintaining residential dwellings is important
and necessary work, but landlords typically do not do this work. That means that rent (the money
one pays to lease housing) is also economic rent (which is what economists call the money that one
pays in excess of what is necessary for the owner to make some resource productive). Rent-seeking
would normally be considered a bad thing for a free market because it wastes resources on un-
productive activity. In so far as one believes that there is a good case for nationalizing things like
healthcare and utilities, there is a similar case to be made for nationalizing housing.
6 Exclusionary Zoning and Segregation

Returning to the question of zoning, what about the argument that single-family zoning contributes
to segregation? It is increasingly common for YIMBYs to argue that exclusionary zoning is motivated
by racism, and that should be eliminated on that basis (Hendrix, 2019). However, there is no evidence
that low-density zoning causes segregation apart from the hypothesis that low-density zoning increases
prices. Rothwell and Massey are the most prominent proponents of the argument that single-family
zoning contributes to segregation and they base their conclusions on the correlation between single-
family zoning and high prices, which is unable to show the necessary causation with segregation
(Rothwell and Massey, 2009). A better case is made by Trounstine, who details the way that wealthy
white neighborhoods use restrictive land-use policies to prevent their neighborhoods from diversifying
(Trounstine, 2020). Richard Rothstein has also been influential on this debate, and he details the way
that segregation was enforced as explicit policy in not only zoning but also public housing and federal
mortgage lending (Rothstein, 2017). In the book, one of his policy recommendations is to eliminate
exclusionary zoning. However, Rothstein’s argument about zoning in particular sits uncomfortably
with the thesis of the rest of his book, which is that because segregation was government policy it
requires positive government action to undo. Though Rothstein now seems to acknowledge that
more action is needed, he was publicly supportive of a controversial proposal for upzoning in the
California state legislature in 2018. When the bill failed and was re-written the following year, he
changed his position and noted that the bill did not increase housing opportunities for all simply by
allowing more market-rate apartments (Dillon, 2019).
It should be acknowledged that exclusionary zoning in the US is motivated by racism – certainly
historically but this is arguably the case today (Trounstine, 2018). However, it would not follow that
upzoning is anti-racist. Much of the denser urban centers in the US have seen substantial dis-
placement of black and latinx residents as these neighborhoods have gentrified. So, black and brown
people were once confined to dense housing that they could not move away from, and they are now
258
Housing Markets
being excluded from that same housing. Keeanga-Yamahtta Taylor has argued that the history of
black exclusion from the suburbs set the stage for the extraction of wealth from inner-city black
communities in the 1970s that she calls “predatory inclusion” (Taylor, 2019). Following her logic,
after this period of predatory mortgage lending, which ended in foreclosure for black residents of
inner cities, the stage was set for another period of racist exclusion as this cheap land was now
available to be re-sold to white residents. Black residents are then priced out as the neighborhood
becomes more desirable. This process is entirely independent of changes in density zoning, except in
so far as upzoning may be associated with an increase in luxury development, and this may increase
wealthy residents, and thus also increase white residents (Davis, 2021).
Some would say this is reason to think of upzoning as one part of a larger plan, so we could
support both upzoning and increased protections for long-term tenants like rent control and other
rules limiting evictions. But there are two reasons against this view – one is that the order of how
things are done matters. If upzoning has the potential to contribute to gentrification, then pro-
tections for existing residents need to be instituted and enforced before upzoning goes into effect.
So, these are not policies we can pursue separately. Secondly, under the current status quo, building
higher or denser than the zoning code allows can usually be done in what is known as “spot
zoning.” But spot zoning requires some kind of special approval from elected officials and/or
planning departments, and the need for developers to ask for approval for these projects means that
citizens, in theory, have some leverage over these projects. This is one of the only opportunities for
most people to have any say over what is built in their community. Eliminating zoning restrictions
would eliminate this small amount of control.
7 Democratic Control
Ideally, we would have much greater democratic control over land use than these zoning codes
allow. But a truly democratic system would also need to be more inclusive, and thus would need to
secure a level of housing that allows for people to relate to each other as equals. There are strong
environmental reasons for greater density in urban areas, one of them being that it reduces the need
for driving. But this goal cannot be pursued separately from the goal of inclusion, lest we create or
perpetuate wealthy city centers in which the rich walk to restaurants that their staff drive miles to
reach. (For a look at how “transit-oriented development” affects driving, and is affected by gen-
trification, see Chatman et al., 2019.)
In a larger sense, housing policy influences who is considered a member of what political ter-
ritory. In the apartment that I live in, the man who lived here before me was evicted. I’ve been here
for 10 years. Every election, I receive his sample ballot. The optimistic version of this story is that he
was taken in by friends or relatives that live in another state. If he had registered to vote in another
state, California would likely have no record of that, so they send him his ballot dutifully. The much
more likely version of this story is that a man who had no one to help him avoid a complete eviction
became homeless and never re-registered to vote. In losing his apartment, he also effectively lost part
of his citizenship. Even if he did take up citizenship in another place, it seems deeply unfair that
because he couldn’t afford to live here, he is no longer eligible to vote on our local policies. In this
case, housing policy determined who was included in future decisions about housing policy.
Unfortunately, there is not enough work that connects theoretical debates about democracy,
inclusion, and territorial rights with housing. There is the idea of the “right to the city” which is
perhaps the most influential way to conceptualize these debates among activists. But as it was
originally proposed by Henri Lefebvre, it is in the context of thinking about the city dweller as a
new political subject, someone who has a right to the city because they are who creates it (Lefebvre,
1996). But it isn’t clear whether this theoretical orientation gets us very far in articulating what the
right to the city means to those who use the phrase in the real world. And even if it does, there is a
259
Kristina Meshelski
lot left to be done. Some promising recent work that connects housing with democracy or lack of it
includes Ananya Roy’s work on “propertied citizenship” (Roy, 2003). Huber and Wolkenstein
have connected Anna Stiltz’s work on a right to occupancy with gentrification (Huber and
Wolkenstein, 2018). Cara Nine has explored the importance of the home as a form of “the extended
mind” as a way to describe the wrong of displacement (Nine, 2018). And Schafran, Smith, and Hall
have written about the normative significance of the built environment as a “spatial contract” which
provides a theoretical grounding for bringing housing policy and theories of collective agreement
together (Schafran et al., 2020).
8 Taking Housing Off the Market

Marxist approaches to theorizing about housing typically begin with the idea that surpluses of labor
and capital need to be absorbed by opening up new markets or new resources in new places – this is
what David Harvey calls the spatial fix (2004). As applied to housing markets, this is a way of looking
at why real estate became a bigger part of the economy over the course of the 20th century and why
housing is increasingly financialized. For example, we can describe the post-2008 creation of single-
family homes as an asset class for larger investors this way (Fields, 2017). In 2016 it was estimated that
global real estate made up 60% of the world’s assets (Tostevin, 2016). This perspective is a necessary
corrective to the increasing attention to zoning in policy debates. Often those debates start from the
perspective that expensive US cities have a great deal of government regulation in housing markets
because there are so many specific zoning restrictions that developers must follow. But the bigger
picture is that real estate speculation is growing and there are not many protections in place to keep
some places for people to live safe from the market. Sam Stein has emphasized the way that any
rezoning can produce enormous value for landowners, either by increasing the amount of units they
can build, a form of “privatizing the air above,” or by raising the value of existing buildings when
downzoning (Stein, 2019). From this perspective, zoning is a battlefield for different types of real
estate interests and property owners. The solution from this perspective is to de-commodify housing
(Marcuse and Madden, 2016).
Nationalizing housing is an option that is often neglected in policy debates within the US. Even
many market advocates might agree that we could stand to have more public housing than we do,
which is substantially out of line with many other developed countries. Maintenance of publicly
owned housing would be easier to pay for if we allowed mixed-income housing, in which the
higher rents paid by the richer tenants subsidize the lower rents paid by other tenants. Many mu-
nicipalities already own a good deal of land, the challenge would be funding the initial development.
Some non-profits and community groups have experimented with community land trusts, co-opts,
and limited equity cooperatives, which all provide various ways that non-state actors can own and
maintain housing as a group, and keep it off the speculative market. However, given the cost of land
in large US cities, these solutions are not able to provide the necessary scale without a very large
amount of funding to buy more land. Expanding rent control is another smaller-scale solution, as is
taxing second homes and vacant homes.
9 Conclusion
Apart from these debates about the amount of regulation to introduce into housing markets, there
are smaller-scale reforms that are badly needed and that both advocates and critics of markets could
possibly find common ground on. These are rental registries or other sunshine laws that would allow
citizens to know who owns and operates their housing. The proliferation of Limited Liability
Corporations in housing markets increasingly means that many tenants do not know who their
landlord is (Glantz, 2019). A right to counsel for tenants facing eviction would make sure there is a
260
Housing Markets
baseline level of protection from exploitative landlords. We could also reform the fines and penalties
for eviction. Having an eviction on one’s record makes it so much harder to find a rental property in
the future (especially in those tight markets where landlords can pick and choose their tenants) and
this incentivizes tenants to leave at the slightest threat from the landlord rather than try to fight their
case in court.
It is important to remember that there are more vacant homes than there are homeless people. The
reason that this is important is not that vacancy is too high necessarily, a certain amount of vacancy
allows people to move if they want to. But it is important to highlight that the current situation has
prioritized profit over housing, so much so that we tolerate some people having nowhere to go.
The privatization of housing has steep costs, and there are many benefits to thinking of housing, and
the land that it sits on, as a collective resource that we could choose to use in a different way.3
Notes
1 However, 2020 data from the American Community Survey, which is the portion of the census that collects
housing information, will not be released for the first time since it was created in 2005. This is because the
response rate was too low, and responses were too heavily biased towards higher incomes and home-
ownership, to make the data reliable. Census problems like this could have serious long-term impacts on
research in this area.
2 The California Judicial Council is required to report annually on the condition of the courts, mostly with
regard to how quickly cases move through, they do not publish more details on these eviction cases.
3 Completed with support from the CSUN College of Humanities Faculty Fellowship.
References
Arnott, Richard (1987) “Economic Theory and Housing”, Handbook of Regional and Urban Economics, 2, 959–988.
Been, V., Madar, J. and McDonnell, S. (2014) “Urban Land-Use Regulation: Are Homevoters Overtaking the
Growth Machine?,” Journal of Empirical Legal Studies, 11, 2, 227–265.
Chatman, D. G. et al. (2019) “Does Transit-Oriented Gentrification Increase Driving?,” Journal of Planning
Education and Research, 39, 4, 482–495.
Court Statistics Report (no date). Available at: https://www.courts.ca.gov/13421.htm (Accessed: 27 April 2021).
Damiano, A. and Frenier, C. (no date) “Build Baby Build?: Housing Submarkets and the Effects of New
Construction on Existing Rents,” working paper. Available at: https://www.tonydamiano.com/project/
new-con/bbb-wp.pdf.
Davis, J. (2021) “How do upzonings impact neighborhood demographic change? Examining the link between
land use policy and gentrification in New York City,” Land Use Policy, 103, 105347.
Desmond, M. and Shollenberger, T. (2015) “Forced Displacement From Rental Housing: Prevalence and
Neighborhood Consequences,” Demography, 52, 5, 1751–1772.
Diamond, R., McQuade, T. and Qian, F. (2019) “The Effects of Rent Control Expansion on Tenants,
Landlords, and Inequality: Evidence from San Francisco,” American Economic Review, 109, 9, 3365–3394.
Dillon, L. (2019) “California housing bill targeting wealthy cities could rezone nearly all of Palo Alto - Los
Angeles Times,” 22 April. Available at: https://www.latimes.com/politics/la-pol-ca-california-apartment-
construction-density-palo-alto-20190422-story.html (Accessed: 17 May 2021).
Dong, H. (2021) “Exploring the Impacts of Zoning and Upzoning on Housing Development: A Quasi-
experimental Analysis at the Parcel Level,” Journal of Planning Education and Research, forthcoming.
Engels, F. (1872) The Housing Question. Available at: https://www.marxists.org/archive/marx/works/1872/
housing-question/ch01.htm (Accessed: 18 May 2021).
Fields, D. (2017) “Constructing a New Asset Class: Property-led Financial Accumulation after the crisis,”
Economic Geography, 94, 2, 118–140.
Fischel, W. A. (2005) The Homevoter Hypothesis: How Home Values Influence Local Government Taxation, School
Finance, and Land-Use Policies. Cambridge, MA; Harvard University Press.
Galster, G. (1996) “William Grigsby and the Analysis of Housing Sub-markets and Filtering,” Urban Studies, 33,
10, 1797–1805.
261
Kristina Meshelski
Glaeser, E. L. and Gyourko, J. (2003) “The Impact of Building Restrictions on Housing Affordability,”
FRBNY Economic Policy Review, 9, 2.
Glantz, A. (2019) Unmasking the secret landlords buying up America | Reveal, Reveal. Available at: https://revealnews.
org/article/unmasking-the-secret-landlords-buying-up-america/?fbclid=IwAR2satWLseeJ9WUCeX0fzhaDmz
MZGO_9_ZSPiqVO8NBe43QBWqh_7cDhhY4 (Accessed: 19 May 2021).
Harvey, D. (2004) “The “New” Imperialism: Accumulation by Dispossession,” Socialist Register, 40. Available
at: https://socialistregister.com/index.php/srv/article/view/5811 (Accessed: 16 May 2021).
Hendrix, M. (2019) Freer housing is ‘fairer housing’—HUD should tie funding to looser zoning | TheHill, The Hill.
Available at: https://thehill.com/opinion/civil-rights/468060-freer-housing-is-fairer-housing-hud-should-
tie-funding-to-looser-zoning (Accessed: 12 March 2021).
Henry, Meghan, Tanya de Sousa, Caroline Roddey, Swati Gayen, and Thomas Joe Bednar. “The 2020 Annual
Homeless Assessment Report (AHAR) to Congress,” January 2020, U.S. Department of Housing and Urban
Development (HUD) https://www.huduser.gov/portal/sites/default/files/pdf/2020-AHAR-Part-1.pdf
(Accessed: 10 August2021).
Huber, J. and Wolkenstein, F. (2018) “Gentrification and occupancy rights,” Politics, Philosophy & Economics, 17,
4, 378–397.
Ikeda, S. (2018) “How Land-Use Regulation Undermines Affordable Housing,” Mercatus Research, Available
at SSRN: https://ssrn.com/abstract=3211656. 10.2139/ssrn.3211656.
Kaplan, B. J. (1980) “Urban Development, Economic Growth, and Personal Liberty: The Rhetoric of the
Houston Anti-Zoning Movements, 1947–1962,” The Southwestern Historical Quarterly, 84, 2, 133–168.
Kuhlmann, D. (2021) “Upzoning and Single-Family Housing Prices,” Journal of the American Planning
Association, 87, 3, 383–395.
Lefebvre, H. (1996) Writings on Cities. 1st edition. Edited byE. Kofman and E. Lebas. Cambridge, Mass, USA;
Wiley-Blackwell.
Maclennan, Duncan (1977). “Some Thoughts on the Nature and Purpose of House Price Studies,” Urban
Studies, 14, 1, 59–71.
Marcuse, P. and Madden, D. (2016) In Defense of Housing: The Politics of Crisis. London; New York; Verso.
Massey, D. S. (1993) American apartheid: segregation and the making of the underclass. Cambridge, Mass.; Harvard
University Press (ACLS Humanities E-Book).
Murray, C. K. (2020) “Time is money: How landbanking constrains housing supply,” Journal of Housing
Economics, 49, 1051–1377.
Murray, C. K. (2021) “Marginal and average prices of land lots should not be equal: A critique of Glaeser and
Gyourko’s method for identifying residential price effects of town planning regulations,” Environment and
Planning A: Economy and Space, 53, 1, 191–209.
Nine, C. (2018) “The Wrong of Displacement: The Home as Extended Mind,” Journal of Political Philosophy,
26, 2, 240–257.
Quigley, J. M. and Rosenthal, L. A. (2005) “The Effects of Land Use Regulation on the Price of Housing:
What Do We Know? What Can We Learn?,” Cityscape, 8, 1, 69–137.
Rodríguez-Pose, A. and Storper, M. (2020) “Housing, urban growth and inequalities: The limits to dereg-
ulation and upzoning in reducing economic and spatial inequality,” Urban Studies, 57, 2, 223–248.
Rothstein, R. (2017) The Color of Law: A Forgotten History of How Our Government Segregated America. Illustrated
edition. New York; London; Liveright.
Rothwell, J. and Massey, D. S. (2009) “The Effect of Density Zoning on Racial Segregation in U.S. Urban
Areas,” Urban Affairs Review, 44, 6, 779–806.
Roy, A. (2003) “Paradigms Of Propertied Citizenship: Transnational Techniques of Analysis,” Urban Affairs
Review, 38, 4, 463–491.
Schafran, A. et al. (2020) The Spatial Contract. 1st edition. Manchester, UK: Manchester University Press.
Shoup, D. (2011) The High Cost of Free Parking, Updated Edition. 1st edition. Chicago: Routledge.
Smith-Heimer, M. (1990) “The Potential for Filtering as Public Policy,” Berkeley Planning Journal, 5, 1.
Stein, S. (2019) Capital City: Gentrification and the Real Estate State. London; Brooklyn, NY; Verso.
Szekely, Balazs (2018) “Renters vs Owners: Renters Became the Majority in These Large Cities,” RENT Café
Blog, January 25th 2018. Available at: https://www.rentcafe.com/blog/rental-market/market-snapshots/
change-renter-vs-owner-population-2006–2016/ (Accessed: 1 August 2021).
Taylor, K.-Y. (2019) Race for Profit: How Banks and the Real Estate Industry Undermined Black Homeownership.
Illustrated edition. Chapel Hill; University of North Carolina Press.
The Eviction Lab (no date). Available at: https://evictionlab.org/ (Accessed: 25 July 2021).
Tostevin, P. (2016) Savills UK | What Price The World? Available at: https://www.savills.co.uk/research_
articles/229130/198669-0 (Accessed: 16 May 2021).
262
Housing Markets
Trounstine, J. (2018) Segregation by Design: Local Politics and Inequality in American Cities. New York: Cambridge
University Press.
Trounstine, J. (2020) “The Geography of Inequality: How Land Use Regulation Produces Segregation,”
American Political Science Review, 114, 2, 443–455.
United States Census Bureau (no date). “Housing” Available at: https://www.census.gov/topics/housing.html
(Accessed: 1 August 2021).
US Department of Education “Identifying and Supporting Students Experiencing Homelessness from Pre-School to Post-
Secondary Ages” April 20th 2020. US Department of Education Available at: https://www2.ed.gov/about/
inits/ed/supporting-homeless-students/index.html (Accessed: 25 July 2021).
Zuk, M. and Chapple, K. (2016) Housing Production, Filtering and Displacement: Untangling the Relationships. UC
Berkeley Institute of Governmental Studies. Available at: https://escholarship.org/uc/item/7bx938fx
(Accessed: 25 July 2021).
263
PART V
Economic Systems
Introduction to Part V
PPE is essential for determining which economic system a society should have. Any adequate view
must understand how economic systems work, how economic systems support and are supported by
political structures, and the values by which economics systems are assessed. Many PPE scholars are
motivated to pursue their own research program because of the contribution it would make to the
choice of an economic system. So, no single volume could sample all the research topics that relate
to this choice or even sample the historical perspectives that give context to debates. Instead, Part V
introduces the topic by distinguishing five kinds of economic system.
Part V starts off with the two most commonly discussed options – Capitalism and Socialism. Each
author provides their own view of how to understand these historically contested terms. After that,
two chapters describe economic structures that are often framed as alternatives to both Capitalism
and Socialism. The idea of a Property-Owning Democracy is most associated with James Mead and
John Rawls, but contemporary work better isolates the core commitments behind this economic
structure. Instead of being grounded in theory, the idea of Social Democracy comes from the
practice of existing states. However, recent work matches this political tradition with a historically-
grounded theoretical framework. After discussing these four ideal types of economic system, the last
chapter discusses what might seem like only a degraded economic condition, corruption. The final
chapter identifies how corruption can emerge as a rational solution to governmental failures and
then linger in ways that are difficult to eradicate.
Peter Boettke’s chapter explains the nature of capitalism and defends it against common 20th-
century criticisms. In particular, he is concerned with the criticism that Capitalism does not work
because the government is needed to address common market failures. He points out that the
argument for Capitalism is not that it achieves perfection. Rather, the idea is that free markets lead to
healthier, wealthier, and more free lives that the alternatives do. While socialists also express concern
for these ideals, they fail to recognize how Capitalism accomplishes these aims better than socialism
ever has. For him, Capitalism is too often confused with Crony Capitalism. When a government
redirects free-market processes to protect the privileges of the powerful, we do not have true ca-
pitalism. Boettke argues that most of the problems associated with Capitalism can actually be traced
back to this level of government involvement. Capitalism is best understood as a liberal project that
enhances individual autonomy and wellbeing by freeing an economy from those that use govern-
ment to advantage themselves.
DOI: 10.4324/9780367808983-25
Economic Systems
Often, people associate Socialism with an economy controlled by a central state. In such a system,
the central government exercises ownership over the means of production. In “Socialisms,” Samuel
Arnold argues that this is a mistake. That kind of economic structure is best referred to as “economic
authoritarianism” and not socialism. Instead, he argues that Socialism is, simply, economic democracy.
Rather than having the direction of an economy decided by the wealthy few, Socialism utilizes
mechanisms the spread control of the economy across society such as workplace democracy, partici-
patory budgeting, and investment through locally controlled banks. While Socialism takes many dif-
ferent forms – just as political democracy does – such an economic system can still utilize markets to
overcome the epistemic problems that plague central planning. For Arnold, the central reason to want
economic democracy is that it avoids alienation. Rather than having our economic lives controlled by
forces beyond our power, Socialism would enable collective control over the economy.
To most people, the choice of a contemporary economic system is presented as a choice between
Capitalism and Socialism, even while there are varieties of Capitalism and Socialism to choose
between. The next two chapters offer economic systems that do not fit squarely into either category;
Property-Owning Democracy and Social Democracy. First, Alan Thomas explains the lineage and
key features of Property Owning Democracy (POD). This economic system organizes institutions
around the idea of granting each member of society a share of capital. This dissipates the power of an
isolated capitalist class and enhances democratic control of both government and economy. Instead
of achieving economic equality through redistribution, POD emphasizes the importance of each
citizen’s initial endowment and the rules of the game that produce a distribution of advantages.
Thomas puts contemporary discussion in a more historical context and discusses the basic policies,
like demogrants, that POD uses to achieve its aims. He ends with a discussion of the challenges –
from political left and right – to whether POD is truly distinct from both Capitalism and Socialism.
In his chapter, Jeppe von Platz identifies both the institutional features and justification for Social
Democracy as a unique form of political economy. He argues that this institutional structure, which is
most associated with the Scandinavian countries, is not merely one way of trying to realize utilitarian or
pluralist ideals. Instead, there is a unique conception of justice that motivates and explains the features of
social democracy. He argues that this conception of justice developed through changes in the Marxist
ideology. Now, it is organized around the idea of society as a system of social cooperation among equals.
The core tenets of this view require that society enable people to cooperate on equal terms and fairly
share the benefits and burdens of their cooperation. This requires the protection of civil rights, de-
mocratic rights, economic rights, and equality of opportunity. It also requires the advantages of society to
be distributed according to a fair bargaining process. After identifying this ideal for Social Democracy,
von Platz shows how the view differs from both Democratic Socialism and from left-liberalism.
It is tempting to be moralistic when analyzing corruption. As a word, it seems to indicate
something that is necessarily wrongful, so it is easy to condemn those that engage in it. Yet, we
would misunderstand the phenomenon if we saw it simply as individuals abandoning the public
good for their own private gain. At times, corruption can help society, though the advantages it
gives might only be in the short run. In his chapter, Michael Munger defines “corruption” and
identifies the circumstances when corruption is “functional.” When the legal channels for securing
government services are slow, unpredictable, or expensive, bribery can emerge as a way to secure
these services or approval. For those willing to pay the bribe, the opportunity is worth it. If paying a
bribe enables someone to provide products or services to others, then consumers all gain. Refusal to
engage in bribery would be a net loss. In such a system, bribery would become normalized and it
would lose much of its stigma. Yet, even if such a corrupt system improves the provision of
government services in the short-run, it presents a long-term problem for that society. Once a
corrupt system is in place, it is hard to eliminate. Those that are part of the system can be sig-
nificantly harmed by reform. In this way, corruption might provide a way to improve certain
dysfunctional institutions, yet it also prevents the implementation of even better institutions.
21
CAPITALISM
Peter Boettke
1 Introduction
According to the Oxford English Dictionary, the term capitalism refers to:
The possession of capital or wealth; an economic system in which private capital or wealth
is used in the production or distribution of goods and prices are determined mainly in a
free market; the dominance of private owners of capital and of production for profit.
The term has descriptive as well as normative uses. Nineteenth century uses were almost universally
condemnations, discussing the barbarism and tyranny produced by the capitalist quest for profits
over concern for the health and well-being of people. The descriptive and normative uses are seen in
the work of Karl Marx (1867; Marx and Engels, 1848) as well. The capitalist system is recognized by
Marx to be a revolutionary force in history, yet he also saw the capitalist system as exploitive,
unstable, and inefficient. It was also inescapably alienating to human beings.
From the beginning, the term was used to discuss the privileges of the property-owning class who work
in concert with those in political power to simultaneously quench their thirst for power and maintain their
wealth and private advantages. Yet, when analyzing the operation of the capitalist system – a system based on
private property and freedom of contract – the analytics focus first and foremost on voluntary exchange,
entrepreneurship, the pursuit of hitherto unrecognized opportunities, and ceaseless technological innovation.
In his recent book, How Innovation Works, Matt Ridley (2021) captures the essence of the capitalist process of
entrepreneurial innovation as follows: “Innovation is the child of freedom, and the parent of prosperity”
(262). Prosperity it surely delivers, and not just for the few, but for the many. Just look at the decline in
extreme poverty throughout the world from 1820 to 2015 (Figure 21.1).
In 2015, for the first time in human history, less than 10% of the world’s population found themselves
living in a condition of extreme poverty. Undoubtedly, we continue to live in a world characterized by
inequality, and the gap between rich and poor is significant. But consider the World Bank data – 44% of the
world lived in extreme poverty in 1980, 37% in 1990, but that figure falls to under 10% between 2000 and
2015.
And that escape from extreme poverty brings with it immense improvements in the human condition –
longer lives, greater access to health and human services, better housing, better education, etc. Progress. This
cornucopia of progress was made possible by the unleashing of the forces of “capitalism.” Globalization –
with its free flow of goods and services throughout the world, with the migration of individuals to seek out
DOI: 10.4324/9780367808983-26 267

268
Peter Boettke
Figure 21.1 Share of the world population living in absolute poverty, 1820–2015. Graph by Max Rosen/Our World in Data, CC-BY 4.0
Capitalism
better opportunities, with financial capital that seeks out its highest return throughout the globe, and with
physical capital (in terms of machinery and technology) that will move to where the profits are like moth to
a flame – results in specialized production throughout the world and peaceful social cooperation among
distant and diverse peoples. The consequences of globalization are indeed that rich countries get richer, but
detailed empirical analysis demonstrates that the poor of the world improve their condition at a faster rate
than the rich (e.g., Dollar, 2005). This is how you go from 40% in extreme poverty to less than 10%. Trade,
technological improvement, and the free migration of people and ideas translate into the betterment of the
human condition. That aspect of capitalism must never be forgotten, though it so often has been since the
introduction of the term.
In The Communist Manifesto, Marx and Engels wrote that: “The bourgeoisie, during its rule of scarce one
hundred years, has created more massive and more colossal productive forces than have all preceding
generations together” (1848, 16). But the cost of this revolutionary force is continuous disruption. As they
put it: “All that is solid melts into air …” (ibid., 17). The bourgeois class loses control of the spell over
society, as the sorcerer’s apprentice might lose control of the magic. In this case, the magic is the market
economy itself, the power of profits, and the exploitation of others by the property-owning class. Power is
matched with instability and insecurity, and capitalism sows the seeds of its own demise.
But does it? Not necessarily, is what I want to suggest. Why? History is a bit messy. Capitalism un-
doubtedly delivers the goods, as exemplified by the improvements in standard of living from the “Great
Enrichment” of the Industrial Revolution (McCloskey, 2010) and from the “Great Escape” of Globalization
(Deaton, 2013). But the capitalist era has indeed been plagued by periodic business cycles, the pathology of
political privileges bestowed on favored business interests, and the unleashing of the machinery of war to
wreak havoc throughout the globe via colonialism and imperialism. So how do we disentangle the web of
capitalist relationships with the power and prestige of the political class?
This essay will proceed as follows. In Section 2, I will lay out the arguments for how the capitalist
economy works, present the conventional critique of the capitalist economy as caricatured in the neoclassical
synthesis, and suggest why that caricature is a false one. In Section 3, I will make the argument that crony
capitalism – political capitalism, state capitalism, or whichever term you choose – is not capitalism properly
understood, and when capitalism is properly understood, the criticisms of capitalism must be redirected. In
Section 4, I explain that the economic system dubbed capitalism could easily have been named “socialism”
except for fundamental confusions in the intellectual discourse initiated by the critics of commercial society.
And, Section 5 will conclude by stressing the basic point that social theorists cannot answer empirical
questions philosophically, and the question about the relationship between capitalism and betterment in the
human condition is an empirical one, and not merely an aesthetic one.
2 How Capitalism Works

A market economy is based on private property and freedom of contract. It is the nexus of exchange, and
we must always remember that an exchange, is an exchange, is an exchange – it is voluntary and mutually
beneficial. That is how wealth is created in society. Individuals are free to choose what to do with their
talents, what they produce and offer to sell, and what they buy and consume. It is this buying and abstaining
from buying that ultimately directs economic activity, and controls the flow of goods and services in a
commercial society. The market process is a continuous mutual adjustment and adaptation of plans and
activities to other market participants. It is in this seemingly constant and chaotic agitation of adjusting and
adapting in the market that individuals discover how best to pursue productive specialization and realize the
gains of peaceful social cooperation through mutually beneficial exchange.
Nothing in this process so imagined favors any participant. The economist who studies com-
mercial society does not favor capital per se over labor and is certainly not in the business of pri-
vileging the rich over the poor. One must always remember that Adam Smith’s (1776) target of
criticism was the Mercantilist system and its elaborate set of protections designed to bestow and
269
Peter Boettke
protect the power and privilege of the mercantile elite. Economic teaching questions those pro-
tections and challenges the power and privileges of the elites.
Smith’s “invisible hand” proposition (1759, 1776) did not promise the best of all possible worlds.
All that was postulated was the individuals within the right set of institutional arrangements will be guided
to exchange and bargain with one another to realize mutual gains from trade and seek out continued
refinements to the methods of production and the specialization of production. It was a theory of
exchange and production. As the proposition that the division of labor is limited by the extent of the
market states, Smith’s theory was one of economic growth and development. Remember the title of
his treatise, An Inquiry into the Nature and Causes of the Wealth of Nations (1776). Improvement was
Smith’s claim, not any postulated perfection and optimality conditions.
However, almost from the beginning critics of commercial society expressed concern with waste and
inefficiency, unscrupulous speculation and instability, and concentrations of wealth and power and injustice.
Inefficiency, instability, and injustice are charges leveled at commercial society by its critics such as Rousseau
(1755), Malthus (1798), and Marx (1867) with varying degrees of vitriol and different points of emphasis. In
this essay, I cannot feasibly survey the vast anti-capitalist literature from the ancients to the moderns. As a
result, I will skip from the classical critics to the modern critics and focus on the textbook criticisms emerging
from within economics, rather than the broader indictment found in the humanities and in the cultural
zeitgeist generally. As Shakespeare understood, sex, money, and violence always sell, and the critique of
capitalism outside of economics certainly tends to recruit those attractors of the human imagination to their
cause. I will instead focus on the underlying technical arguments that economists have developed in the
theory of market failure as the foundation for their indictment of the capitalist system.
The “End of Laissez-Faire,” as Keynes (1926) dubbed it, consisted of a wholesale rejection of Smith’s
“invisible hand.” A close reading of Keynes’s essay will reveal that this rejection is grounded in a critique of
the rationality of economic decision makers who, instead of cold calculations of their self-interest, exhibit
ignorance and weakness of will, and doubts about the functioning of the price system and spontaneous
order, which rather than leading to a reconciliation of private interest and public interests, leads instead to the
concentration of wealth and power in the hands of a few at the expense of the many. People aren’t rational
and prices do not do the job the theory says they do. As a result, modern commercial society is plagued by
pervasive inefficiencies, instability, and injustice. Government is thus conceived as the corrective to the social
problems created by unhampered capitalism.
Keynes is today mainly remembered for his pivotal role in the development of macroeconomics;
however, he also highlighted microeconomic problems and the inability of classical doctrine to offer a viable
analysis of – let alone an answer to – the social ills that plagued British society in the late 19th and early 20th
century. Those social problems would eventually be described in the Beveridge Report (Beveridge, 1942) as
the Five Giants of Poverty, Ignorance, Squalor, Disease, and Idleness, and the only power capable of slaying
the Five Giants was the central authority of the government. The market economy could not be trusted
with the task of eradicating social ills because it was, in large measure, the culprit that caused the social ills.
After World War II, Paul Samuelson integrated these critiques into what would constitute mainstream
economics from 1950 to 1980. In major ways, the Samuelsonian transformation of economics has never
been displaced at a methodological (formalism), analytical (equilibrium theorizing), and social philosophical
(government interventionism) level. This is what is taught in Econ 101 despite the rumors to the contrary
that circulate among critics of economics and so-called neo-liberalism that occupy the humanities, and a
significant segment of the public imagination. Supply and demand depictions are quickly followed up with a
discussion of a litany of so-called market failures from externalities, public goods, monopoly power, and
macroeconomic volatility. In the face of such a reality, men of goodwill turn, as Samuelson put it in the first
edition of his classic text (1948), to the authority and creativity of government action to solve the problem.
Externalities are addressed with tax and subsidies, public goods are dealt with through collective action,
monopoly power is attacked through regulation and anti-trust policy, and macroeconomic volatility is
countered with an array of fiscal and monetary policies to provide aggregate demand management and
270
Capitalism
ensure that unemployment is kept in check. In short, what contemporary Econ 101 teaches is that markets
fail, the government is the corrective, and economic science provides the tools for effective social control
through collective action.
Critical to this exercise is the development of the model of general competitive equilibrium and welfare
economics. The model supposedly demonstrates that an optimal allocation of resources can occur, but only
under a highly unrealistic and extremely restrictive set of assumptions. Only when these conditions are met,
the argument goes, can it be said that Adam Smith’s “invisible hand” proposition is said to operate. But,
since those conditions can only be met in theory and on the blackboard in classrooms (or the required
textbooks), the “invisible hand” is, practically speaking, irrelevant to the real-world of economic life in a
modern society. Instead, the energy and creativity of the vast majority of economists from the 1950s onward
was directed toward the articulation of the correct governmental response that bridges the gap between the
ideals that eradicate inefficiency, instability, and injustice, and the persistent reality of inefficiency, instability,
and injustice. As a consequence, economics was transformed from a tool of social understanding – again
remember Adam Smith’s title An Inquiry – to a tool for social control – think about the title of one of the
great classics in the transformation of economics, Abba Lerner’s The Economics of Control (1944). It is not like
the modern practitioners of the “New Economics” were bashful about their ambitions.
Why should they be? The experience of the economic malaise of the 1930s and the Great
Depression, as well as the experience during WWII of government planning, generated a sense of
urgency in the wake of a loss of faith in laissez-faire capitalism. Capitalism had already had its critics,
as we pointed out, but the experience of the late 19th and early 20th century with monied interest,
monopoly power, and industrial fluctuations had cemented the need for concerted collective action
in the minds of intellectuals, politicians, and the public. The Progressive Era legislation in the US
from 1890 through World War I was designed to address these problems, just as the New Liberalism
in the UK and Fabian ideas shaped British policy. The discerning reader might notice already a bit of
a contradiction – public policy since 1890s in the US and the UK already involved significant
deviations from the supposedly laissez-faire system, yet Keynes was declaring the failure of free-
market capitalism in 1926 and the complete intellectual bankruptcy of classical economics by 1936.
The critique of capitalism found in economics textbooks, let alone those found in the humanities,
is based in a caricature of the main arguments for the system. Its strongest advocates never couched
their defense on the welfare economics of general competitive equilibrium, and their theory of how
the system works was not focused on optimality. Instead, it was a system that generated progress and
improvement in the human condition. Capitalism gives us innovation and wealth creation. Incomes
improve and with that rising income, life possibilities improve. Rising living standards are positively
correlated with longer and healthier lives and with greater autonomy, so individuals are empowered
to be the architects of their own lives and families and communities can live peacefully and pro-
ductively amidst each other. I will return to this last claim in Section 4. But for now, let’s look at the
problems with crony capitalism, which does attract the ire of intellectuals and the general public.
3 Crony Capitalism Is Not Capitalism

Keynes, as a critique of capitalism, caught lightning in a bottle because he was able to link the modern
public’s greatest fear (mass unemployment) with its greatest resentment (the idle rich). The landed aristocracy
of the late 19th century, and the Robber Barons of the era, elicited a resentment that persists today and was
recently resurrected again in the work of Thomas Piketty (2013) and his critique of capitalism as inherently
generating gross inequalities in society. But the source of that inequality isn’t in the so-called laws of
capitalism, but in the strategic use of the state by businesses to create and protect privileges.
Crony capitalism is merely a modern name for mercantilism, and mercantilism is best understood
as a rent-seeking society. That is to say the elaborate protections and system of policies erected were
in place to secure the power and privileges of the select few favored by the Crown. In fact, the
271
Peter Boettke
counter argument to the government as corrective claim was at least in part to point out that the
government was in fact the source of the inefficiency, instability, and injustice, and thus not the
solution. It was the government that created monopoly power through granting of special privileges
to enterprises to be the sole seller of a good or service; it was government money and credit that
distorted the price signals and thus monetary calculation of alternative patterns of investment the
produces the boom-and-bust cycle; it was government policies that empowered discrimination and
blocked the social progress of individuals and groups based on race, religion, and ethnicity.
The competitive market economy understood as the dynamic adapting and adjusting process mentioned
above works to erode the monopoly power of first-movers in any market through entry and competition on
a multiplicity of margins (price, quality, convenience, etc.). McDonald’s, for example, didn’t invent the
hamburger, the French fry or the milkshake, but they figured out a way to deliver that meal more con-
veniently and at a lower cost than their competitors. In the process, McDonald’s restaurants provided low-
cost meals to millions, and first jobs for millions of teenagers through the years. No doubt, people can criticize
fast-food and its correlation with obesity and other health issues, but as that knowledge becomes more
widespread either McDonald’s will start to offer more healthy options or, more likely, will be challenged in
the marketplace by new entrants more attentive to these growing concerns. That is how markets work.
The price signals that emerge from the bargaining activity of individuals on the market will guide
exchange and production, and the profit and loss statements will lure and discipline the decisions made.
Again, the price system is not postulated to be perfect, but it is a rough guiding aid to the human mind as it
attempts to negotiate the tumultuous and uncertain sea of economic possibilities. The goal isn’t optimality,
but progress and human betterment. Today’s missed opportunity will be tomorrow’s profit for the in-
dividual and firm that can respond effectively to the opportunity. As I pointed out earlier, Matt Ridley has
argued in How Innovation Works (2021) that the ceaseless and unended process of adaptation and adjustment
spurred on by the creativity and cleverness of ordinary individuals produces a cornucopia of benefits for
mankind. “Innovation is the child of freedom,” and Ridley remarks, “the parent of prosperity” (262).
The alliance between the state and business, on the other hand, is the fountain of privilege and the
source of our social maladies and pathologies. Pick any issue of pressing concern, from gross inequality to
the threat of catastrophic consequences due to climate change, and you will find at root various gov-
ernment policies that exacerbate the problem and prevent the adaptation and adjustment by market
participants to the changing circumstances as guided by relative prices, profit-and-loss statements, and the
pattern of ownership. Water prices and housing construction in California are strictly regulated, and both
policies make California more vulnerable to the consequences of climate change. Both policies were not
enacted for some notion of the public interest, but rather to provide concentrated benefits on a well-
organized and well-informed special interest group and to disperse the costs on the unorganized and ill-
informed public. The business of concentrating benefits and dispersing costs is the governmental habit in a
nutshell. That habit is revealed not only in policies such as price controls and building restrictions, but also
from foreign trade and immigration to the deficit finance and monetary mischief. Adam Smith long ago
warned about this government habit in the form of a “juggling trick” (1776, 414) of deficit, debt, and
debasement that all governments ancient as well as modern engage in as those in governmental authority
use their position to embed power and doll out privileges.
Critics of capitalism are correct to focus their ire on inefficiency, instability, and injustice, but
they are identifying the wrong source of these social maladies. That leaves us, not only with the task
of correcting the interpretation, but also asking a different rather difficult question – whether it is
inevitable that capitalism evolve into crony capitalism. This is a question that Schumpeter asked, and
that more recently Michael Munger and Randall Holcombe have respectively asked. I have asked it
as well, and concluded in a more pessimistically optimistic way than say the cynical pessimism of
others. Yes, there are the grabbing hands of the state alongside the invisible hands of the market, but
there also might be hope in the constitutional project which can establish a set of rules that ef-
fectively constrain the predatory state. We must be willing to ask ourselves: Is our only choice
272
Capitalism
between crony capitalism and crony socialism, or is there actually a way to effectively remove the
mercantilist system and establish Adam Smith’s liberal plan for equality, liberty, and justice?
4 Capitalism as “Socialism”
Lionel Robbins argues in The Theory of Economic Policy in British Classical Political Economy (1952) that
the body of economic theory evolved alongside the evolution of liberalism as a political doctrine and
legal philosophy. Liberalism reflects a slow and persistent struggle against the pathologies of privi-
lege, whether that source is the Crown, the Church, the Mercantilist class, or the Slave trader. The
advancement of liberalism is identified with the abolition of privileges. In short, liberalism is an
emancipation project from the power and privilege exercised by a few over the many.
The argument that economists developed was one of the effectiveness of chosen means to obtain
given ends. The end of liberalism is peace and progress. The means suggested by the classical political
economist in the tradition of Adam Smith was private property and freedom of contract. They
argued that under such a system, it would be possible to grant freedom to all, rather than a select few.
Ordinary individuals, it was argued, could do extraordinary things if only given the freedom and
elbow room to follow their imaginations freely, bear the risk of their decisions, and reap the rewards
of those decisions. The alternative perspective argued instead that extraordinary people could act in
extraordinary ways to save us in dire circumstances if only they were given the power to act ac-
cordingly. One answer seeks to find a system that enables us to effectively govern with one another,
while the other seeks to find a system that empowers trained elites to govern over us.
Self-government works, according to the economists from Smith to Hayek, because the market system
compels its participants to adjust to one another and to adapt to the shifts in circumstances, including the
shifting tastes, technology, and resource availability through time. Individuals pursue productive speciali-
zation and realize peaceful social cooperation through exchange. It is this peaceful social cooperation that
emerges in the market that enables individuals to escape from the war of all against all. Nature is indeed red
in tooth and claw, but humanity can come to live better together than they ever could in isolation by
marshaling incentives and learning through social intercourse. The expansion of the division of labor taps
into the unique talents of individuals and discovers lower-cost ways of producing existing goods and services,
and it continually seeks out new goods and services that will improve the lives of their fellow citizens. Fellow
citizens that they will never know, never be able to identify, but who nevertheless come to depend on the
actions of others for their daily survival. This is how we end up with clothes on our backs and shoes on our
feet, let alone the homes we live in, the cars we drive, and the food that satisfies our basic requirements.
Capitalism serves mankind by marshaling the ordinary motivations of men in directions in which they can
only do better for themselves by improving the condition of others.
The socialist’s dream aspirations, on the other hand, intended to do well for others, but in practice
collapsed to hell on earth in the 20th-century social experiments. This should never be forgotten, though it
often is. The aspirations of socialism – advanced material production, harmony among the social classes, etc.
– are actually more effectively achieved by unbridled capitalism, whereas the effort of socialist planning
devolves into a cruel form of crony socialism, where the elites become a protected “New Class” and the
suffering of the masses ranges from frustration in the consumer sector to death in the Gulag.
That private property is the source of our sociability rather than our alienation is one of the most
important findings of economics, and it is also one of the most difficult lessons to learn for those who
think within an anti-Enlightenment longing for total revolution rather than alongside the
Enlightenment project of reason and evidence. These critics of capitalism are not interested in reason
and evidence because their position derives from an aesthetic one, and to them the picture of
capitalism is just an ugly and deformed way of humanity to live. Countering that picture is no small
task, but it is in that task that the future of capitalism will largely depend.
273
Peter Boettke
Capitalism is not Socialism as understood in our intellectual culture, and I want to make sure that
readers do not see me as making such a categorical error. What I am suggesting is that the aspirations
of many professed socialists are actually similar aspirations to those of professed liberals. There is a
severe disagreement over the means to achieve those goals. One argues for the establishment of
private property rights and the freedom of exchange, the other calls for, to varying degrees, the
abolition of private property and the establishment of production for direct use rather than for
exchange. The shared goals are the eradication of power and privilege, the emancipation of in-
dividuals from oppression, and the promoting of human flourishing. Liberalism is first and foremost
liberal, and the liberal must exhibit liberality in their attitudes and actions. We are one another’s
dignified equals, and we must interact with one another under that recognition. A politics worthy of
such creatures should strive to meet the criterion of non-discrimination and non-domination.
Almost by necessity, such a system promotes a move away from a Big State to a Big Society.
As practiced in the 20th century, Socialism turned instead to the Big State to accomplish its goals,
and, as such, I would argue betrayed the original aspirations of the left. Socialism in practice is the
militarization of the economy and society. It represents a great tragedy, as high-minded ideals for the
eradication of power and privilege do not result in a New Jerusalem as the UK Labor Party pro-
mised, do not bring a New Civilization as Sidney and Beatrice Webb promised, and certainly do not
transition us from the Kingdom of Necessity to the Kingdom of Freedom as Marx and the
Bolsheviks envisioned. It results in The Road to Serfdom as Hayek warned. The Smithian project, on
the other hand, promotes sociability among diverse and disparate peoples. Exchange turns strangers
into friends, and the free flow of goods, services, and people makes us strangers nowhere in this
world. That is the vision of liberal cosmopolitanism, and the economic system most conducive to
that world is capitalism. The dream aspirations of the socialist, in short, are best met by the operation
of radical laissez-faire capitalism – a workable utopia not concerned with the sensibilities of the
might, and focused on both respect for the universal rights of all of mankind and the unleashing of
the creative powers of a free civilization to escape the clutches of various social dilemmas that have
plagued mankind from the beginning.
5 Conclusion
If Karl Marx was the most comprehensive critic of capitalism, then Ludwig von Mises could be said to have
been capitalism’s most comprehensive theorist. In his magnum opus Human Action, Mises responds to critics of
capitalism who argue that social conditions have not improved under capitalism. “These grumblers,” he states,
do not realize that the tremendous progress of technological methods of production and the
resulting increase in wealth and welfare were feasible only through the pursuit of those liberal
policies which were the practical application of the teachings of economics. (1949, 8)
It is an ideological revolution, Mises contends, that produced the “Industrial Revolution,” and it is this
transformation that made possible the escape from miserable poverty. It was the ideas of classical
political economists that “removed the checks imposed by age-old laws, customs and prejudices upon
technological improvements and freed the genius of reformers and innovators from the straightjackets
of the guilds, government tutelage, and social pressures of various kinds” (ibid., 8). It was the ideology
of liberalism and the science of political economy that produced a switch from reverence for warriors,
conquerors, and expropriators towards the virtue in commerce and civil society.
Perhaps the leading voice for liberal capitalism in academia today is Deirdre McCloskey.
McCloskely, like Mises, places great emphasis in her work on the “Great Enrichment” that modern
economic growth produced on ideas, and in particular ideas that afforded dignity and purpose to the
bourgeois activity of commerce and the entrepreneurial act of innovation in the marketplace. The
274
Capitalism
simple act of “giving it a go” and to have one’s ideas trade-tested required a transformation of mental
attitudes and ultimately of speech. Capitalism requires a change in everyday rhetoric, and
McCloskey documents this transformation in her trilogy (2006, 2010, 2016).
Ultimately, the relationship between capitalism and human betterment is an empirical one. We must
look not just at per capita income figures but also various measures of health and human well-being. What
we cannot do is ignore the empirical record altogether. We cannot be content to answer empirical questions
philosophically. But this is precisely what our aesthetic critics of capitalism repeatedly do. Instead, a certain
mindfulness to facts must be evident in our work, even while we stress the multiplicity of relevant empirical
forms of data and methods for gathering and analyzing. Memoirs, life histories, biographies, and even
literature communicate important nuggets of information about the human condition at any point in
historical time. Aesthetics don’t, and while science is no doubt part of art, art does not need any grounding
in science. Our assessment of capitalism cannot be left to such artistic renderings divorced from the science of
reality. Once corrected for that, the ugly picture painted by Rousseau, Malthus, Marx, and Keynes may
actually turn out to be a beautiful picture of the creative energy of mankind being marshaled not through
top-down decree, but through bottom-up incentives and the discovery, use, and dissemination of contextual
knowledge to produce economic progress and improve the human condition. Capitalism has done that, and
capitalism will continue to do that if we just continue to respect, accord dignity, and grant freedom to
ordinary people so they may achieve extraordinary things and make all better off.
Acknowledgments
Thanks to Jessica Carges for her comments and editorial suggestions, and Tegan Truitt for research
assistance. The usual caveat applies.
References
Beveridge, William. (1942) “Social Insurance and Allied Services.” New York, NY; Macmillan Company.
Deaton, Angus. (2013) The Great Escape: Health, Wealth, and the Origins of Inequality. Princeton, NJ; Princeton
University Press.
Dollar, David. (2005) “Globalization, Poverty and Inequality,” World Bank Research Observer, 20, 2, 145–175.
Keynes, John Maynard. (1926) The End of Laissez-Faire. London; L. and V. Woolf.
Lerner, Abba F. (1944) The Economics of Control: Principles of Welfare Economics. New York; Macmillan.
Malthus, Thomas Robert. (1798) 1989. An Essay on the Principle of Population. Cambridge; Cambridge University Press.
Marx, Karl. (1867) 1887. Capital. Translated by Samuel Moore and Edward Aveling. Vol. 1. Moscow:
Progress Publishers. https://www.marxists.org/archive/marx/works/download/pdf/Capital-Volume-I.pdf.
Marx, Karl, and Frederick Engels. (1848) 1969. “The Communist Manifesto.” Translated by Samuel Moore
and Frederick Engels. Marx and Engels: Selected Works, 1, 98–137. Moscow: Progress Publishers. https://
www.marxists.org/archive/marx/works/download/pdf/Manifesto.pdf.
McCloskey, Deirdre N. (2006) Bourgeois Virtues: Ethics for an Age of Commerce. Chicago; University of Chicago Press.
McCloskey, Deirdre N. (2010) Bourgeois Dignity: Why Economics Can’t Explain the Modern World. Chicago;
University of Chicago Press.
McCloskey, Deirdre N. (2016) Bourgeois Equality: How Ideas, Not Capital or Institutions, Enriched the World.
Chicago; University of Chicago Press.
Mises, Ludwig Von. (1949) 2007. Human Action. Indianapolis; Liberty Fund.
Piketty, Thomas. (2013). Capital in the Twentieth Century. Cambridge, MA; The Belknap Press of Harvard
University Press.
Ridley, Matt. (2021) How Innovation Works: And Why It Flourishes in Freedom. New York; Harper Perennial.
Robbins, L. (1952) The Theory of Economic Policy in British Classical Political Economy. London, UK; Macmillan.
Rousseau, Jean Jacques. (1755) 1985. A Discourse on Inequality. Translated by Maurice Cranston. New York;
Penguin Classics.
Samuelson, Paul A. (1948) Economics. Boston; Mcgraw-Hill.
Smith, Adam. (1759) 1982. The Theory of Moral Sentiments. Indianapolis; Liberty Fund.
Smith, Adam. (1776) 1979. An Inquiry into the Nature and Causes of the Wealth of Nations. Indianapolis; Liberty Fund.
275
22
SOCIALISMS
Samuel Arnold
1 Introduction
Socialism: it’s back. Left for dead after the Cold War, revived by Bernie Sander’s 2016 and 2020
Presidential bids, and popularized by influential leftists like Alexandria Ocasio-Cortez, the concept has
gone mainstream. Socialist podcasts crack the iTunes Top 100. Teen Vogue publishes paeans to Karl
Marx. “When Did Everyone Become a Socialist?” New York Magazine asks, only half-seriously – but
still, half-seriously; their hyperbolic headline contains a germ of truth (Van Zuylen-Wood, 2019).
Socialism really has become surprisingly popular, especially among younger and more progressive
Americans. Gallup reports that 51% of Americans age 18–29 regard socialism favorably, as do 65% of
Democrats – a rather stunning development in this, the land of Reagan, Rand, and the Red Scare
(Jones and Saad, 2019).
Of course, not everyone welcomes this socialist revival. Count President Donald Trump among
the holdouts. “Here in the United States,” Trump declared in his 2019 State of the Union address,
“we are alarmed by the new calls to adopt socialism. America was founded on liberty and in-
dependence – not government coercion, domination, and control. Tonight, we renew our resolve
that America will never be a socialist country” (Pramuk, 2019).
Trump is far from alone in rejecting socialism. While the young and the progressive may be ready
to man the socialist barricades, their older and more conservative compatriots won’t be joining them
any time soon. Socialism, per Gallup, is disliked by 91% of Republicans and 72% of Americans over 65
(Jones and Saad, 2019). Add it all up, and roughly two-thirds of the country thinks that “Socialism
Sucks”, as Turning Point USA (an influential conservative campus group) puts it.
But what is socialism? And does it “suck”? This chapter addresses these questions. I’ll start with
socialism 101, explaining the basics (Sections 2, 3, and 4). As will become clear from this introductory
discussion, there are varieties of socialism: diverse values, principles, and institutions that can claim,
credibly, to fly socialist colors. (One might therefore speak more profitably of socialisms rather than
socialism: hence this chapter’s title.) Having sketched socialism’s overall shape, I’ll then fill in some of
its details. More specifically, I explore an important contrast that socialists draw between capitalism and
socialism. Capitalism allows economically powerful people – owners of capital – to call the economic
shots. Socialism, its defenders say, would be different. For – as Erik Olin Wright explains – under
socialism “the investment process and production are controlled through institutions that enable or-
dinary people to collectively decide what to do. Fundamentally, this means socialism is equivalent to eco-
nomic democracy” (Wright, 2019: 69–70).
276 DOI: 10.4324/9780367808983-27

Socialisms
“Economic democracy” – a fine phrase, perhaps, but what does it mean in practice? Don’t we
already sort of have it under capitalism? And what’s so great about it, anyway? I respond to these
questions in Sections 5 through 8. Drawing on work by Malleson (2014) and Schweickart (2011), I
explain economic democracy’s practical requirements and show that they cannot be satisfied under
capitalism. Accordingly, if we want collective, democratic control over the economy, we’ve got to
replace private with social ownership of the means of production. But should we want collective, de-
mocratic control over the economy? Yes, we should. Capitalism, I suggest, traps us in an upside-down
world where profit matters more than people; where human beings serve the economy rather than the
reverse; where we’re mastered, subjugated, by material forces that we made, but somehow cannot
control. The remedy for this condition of “alienation,” as socialists call it? Precisely the sort of eco-
nomic democracy socialism prescribes. In short, we should want economic democracy because it alone
promises to break the chains of our collective economic enslavement. Section 9 concludes.
2 Socialism as a Political Philosophy

Socialism, reports the Oxford English Dictionary, is “a political and economic theory of social or-
ganization which advocates that the means of production, distribution, and exchange should be
owned … by the community as a whole.” On this account, socialism bottoms out in a commitment
to a certain kind of economic system: namely, one in which certain key economic institutions (the
“means of production, distribution, and exchange”) are owned collectively, or socially, or some-
thing like that. Socialism thus contrasts with capitalism, which says that these key economic in-
stitutions should be (or perhaps merely may be) owned privately.
This is a perfectly good way to think about socialism, and we’ll return to it in Section 3. But
there’s another way of characterizing socialism that I want to mention first. We could think of
socialism as bottoming out in allegiance not to a particular system of property, but to a particular set
of moral or political values. On this approach, socialism isn’t so much an economic system as it is a
distinctive political morality. Its natural foil, then, isn’t capitalism (which is an economic system), but
rather the other great political moralities of our times: liberalism (in both “classical liberal” and
“liberal egalitarian” varieties) and libertarianism.
Space constraints prevent me from offering anything but the roughest sketch of socialism qua
political morality. But in broad outlines, socialist political philosophy overlaps considerably, but not
perfectly, with liberal political philosophy. All liberals embrace a theory of justice that:
i. places great emphasis on the protection of certain basic political and civil liberties: things like
the right to vote, freedom of expression, liberty of conscience, and so on; and
ii. guarantees to all citizens “adequate all-purpose means” to make effective use of these basic
freedoms (Rawls, 2005: 6).
And socialists certainly do take these to be minimal requirements of justice. So they are, in this way,
liberals. But they are also left-liberals, liberal egalitarians, because they think that a just society will not
only (i) ensure fundamental freedoms and (ii) protect everyone against economic deprivation, but
will also (iii) seek a fair distribution of economic resources, over and above the sufficientarian re-
quirements of condition ii.1
Given their commonalities, one may reasonably wonder if there’s any real daylight between so-
cialism and liberal egalitarianism. Are they just two different labels for what is, at bottom, the same
philosophical camp? I don’t think so. I’ve mentioned areas of overlap between liberalism and socialism,
but there are differences, too. Socialists typically emphasize certain concepts, like community, non-
alienation, and non-exploitation, that play at best a peripheral role in liberal thought (Cohen, 2009;
Gilabert, 2021; Arnold, 2021; Vrousalis, 2021). In light of these different emphases, we might say that
277
Samuel Arnold
you’re a liberal egalitarian if you accept the aforementioned requirements i-iii, but a socialist liberal
egalitarian if, in addition to these requirements, you work into your political morality a foundational
concern for community, non-exploitation, and/or non-alienation.2 I’ll say more about non-alienation
below, in Section 7.
3 Socialism as an Economic System 3

Jeff Bezos, founder and former CEO of Amazon, is really rich: $182 billion rich, at last count
(Zaroli, 2020). Bezos’s money doesn’t just buy him fancy stuff, such as his new $165 million Beverly
Hills mansion. It also gives him enormous power.
Who controls the American economy? Who calls the economic shots? Ordinary working people,
or wealthy capitalists like Bezos? The answer, socialists say, is the latter. Under capitalism, key
economic decisions are made not by all members of society but by the very richest – those wealthy
capitalists and investors who own what Marxists call the “means of production.”
Consider Amazon’s recent decision to build a second headquarters (Thompson, 2018). This was a
big opportunity for cities across the country; winning Amazon’s “HQ2” would mean 50,000 new
jobs and over $5 billion in local investment. So it’s no surprise that over 200 cities submitted bids,
each scrambling to assemble the most business-friendly package of tax cuts and other perks. Amazon
eventually awarded HQ2 to northern Virginia – an already prosperous area. The fact that other
regions stood in greater need of economic development didn’t matter.
From the socialist perspective, this episode nicely illustrates one of capitalism’s core flaws –
namely, its profoundly undemocratic nature. Under capitalism, we don’t make key economic decisions col-
lectively. We don’t decide together where 50,000 new jobs should be created, or $5 billion in capital
should be invested. Instead, such decisions are left to capitalists like Bezos, who can do what they
like with what they own. That’s just how capitalism works. As Erik Olin Wright explains, capitalism
is “an economic structure within which the allocation and use of resources in the economy is
accomplished through the exercise of economic power” (2019: 69). Accordingly, under capitalism
“investments in and control over production are the result of the exercise of economic power by
owners of capital” (ibid.).
Socialism would turn this picture on its head. Rather than allowing wealthy elites to control
major decisions about investment and production, socialism would empower ordinary people to
decide such things collectively. Under socialism, the “commanding heights” of the economy –
major industries and large corporations – would be brought under social ownership and control.
Would the state own everything, then? No, for two reasons. First, on most accounts, socialism targets
only the economy’s “commanding heights” – think Walmart, Amazon, and other major corporations –
not smaller-scale productive property like your local pizza place (Edmundson, 2020).4 Nor does soci-
alism have any beef with personal property: socialists don’t want to collectivize your toothbrush.
Second, socialists seek “public” or “social” ownership of the economy, which is not necessarily
the same thing as “state” ownership (Wright, 2019: 68–71). Although state ownership is certainly an
important tool in the socialist toolkit, it’s definitely not the only one. As American socialist Michael
Harrington puts it, social ownership means “the democratization of decision making in the everyday
economy”; it’s a “principle of empowering people at the base, which can animate a whole range of
measures” beyond state ownership – “some of which we do not yet even imagine” (Harrington,
1989: 197).
It’s in this “empower the base” spirit that socialists have proposed measures having nothing to do
with state ownership per se – measures like workplace democracy (which gives workers rather than
capitalists control of their workplaces), participatory budgeting (which empowers ordinary citizens
to direct their community’s spending), and social control over investment (which channels all in-
vestment funds through networks of public, democratically accountable banks). These “building
278
Socialisms
blocks of democratic socialism,” as Wright calls them, “help shift the power configurations of ca-
pitalism towards an economy animated by social empowerment” (Wright, 2019: 71). By putting
ordinary people behind the economic steering wheel, they bring us closer to socialism.5
At its core, then, socialism means economic democracy. Across the globe, 18th- and 19th-century
radicals overthrew political monarchs like King George III, replacing political dictatorship with
political democracy. Socialists say that we should complete what these revolutionaries started: we
should overthrow our economic monarchs like Jeff Bezos, replacing the economic dictatorship of
capitalism with the economic democracy of socialism.
4 What Socialism Isn’t

This definition of socialism as economic democracy is at odds with three popular, but flawed ways of
understanding socialism.
The first confuses socialism with statism; it says that socialism just means state control of the
economy – even if the state in question is profoundly undemocratic. This view regards dictatorships
like the Soviet Union, Venezuela, and North Korea as socialist, precisely because they achieved
near-total state control over the economy. But in fact, these regimes are not at all socialist. Socialism
means economic democracy, not economic authoritarianism.
A second mistaken conception of socialism conflates socialism with social democracy; it says that
countries like Denmark and Sweden, which tax citizens heavily and spend generously on social
programs like income support, education, and health care, are socialist. But socialism isn’t, at root,
about taxation or social spending – it’s about who controls the means of production. You can’t
prove that a country is socialist by referring to its tax rate or unemployment benefits. You’ve got to
look at who’s making its major investment and production decisions: wealthy elites? Then it’s
capitalism. Democratically unaccountable state bureaucrats? Then it’s statism. The people them-
selves (even if only indirectly)? Only in this third case do we have socialism.
A third mistaken conception of socialism assumes that socialism must oppose not merely private
ownership of the means of production, but also markets and profits. On this conception, socialism
necessarily seeks to replace market production for profit (e.g., Apple makes iPhones in order to make
money by selling them) with planned production for use (e.g., government planners tell the smartphone
factory to make smartphones, because people need or want them).
Admittedly, some socialists – especially those in the Marxist tradition – do conceive of socialism
in this anti-market way. But others, influenced by the economic and moral failures of centrally
planned economies like the Soviet Union’s, accept some role for markets and profit-seeking in their
socialist visions. On their view, socialism only requires a certain form of ownership, namely, social
ownership; about markets, socialists should be open-minded. Markets are just tools for commu-
nicating information and motivating economic activity. Like any tool, they should be evaluated
instrumentally. Do markets work better than planning-based alternatives? Can we harness their
informational and motivational advantages while keeping them subordinate to democratic processes?
If so, socialists should embrace them. I return to the role of markets in Section 6.
5 The Concept of Economic Democracy

Democratize the economy: this, I have suggested, is socialism’s core commandment.6 While I’ve gestured
towards various “building blocks” socialists might deploy in pursuit of their democratizing aim, the aim
itself may remain somewhat obscure, both conceptually and institutionally. What does economic
democracy really mean? And how, in a brass-tacks sort of way, could we actually implement it?
Let’s tackle the conceptual issue first. As Malleson (2014) has argued, a democratic economy
would empower ordinary people in at least two ways.
279
Samuel Arnold
First, it would grant everyone roughly equal influence over what is produced. No economy can
make everything people might want. Scarcity forces us to choose. Will we make yachts or bikes?
Baldness cures or vaccines? Tax preparation software or mediation apps? When what Malleson calls
consumer democracy reigns, each person has roughly equal say over such production decisions
(2014: 93).
Second, a democratic economy would empower us not merely as economic consumers, but also
as economic citizens: it would provide us with roughly equal influence not merely over what’s made
here and now – that’s consumer democracy – but also “over how the economy develops over time”
(Malleson, 2014: 93). Suppose our productive activities produce a surplus; we have unconsumed
resources at our disposal. How shall we direct this surplus? We could build cathedrals to our gods, as
did feudal lords. We could chase profit, reinvesting in whichever activities promise the biggest
financial return, as do our capitalists. Or we could deliberate together about these “most important
human matters,” asking in a democratic way fundamental questions like “where people want to
invest their collective energies, how they want to balance productive work vis-à-vis family life, how
much and what they want to leave to future generations,” and many, many more (Fraser and Jaeggi,
2019: 25). Only by taking this last, democratic route – only by treating the allocation of the surplus
as a public matter, to be decided by we the people – do we implement what Malleson calls “citizen
democracy” (2014: 93).
To Malleson’s two dimensions of economic democracy, let me add a third. In a democratic
economy, ordinary people would not only have roughly equal say over what gets produced and how
the economy develops over time, they would also have roughly equal say over their workplaces.
Consumer and citizen democracy must be combined with some kind of producer democracy. A de-
mocratic economy will, among other things, foster what democratic theorist Robert Dahl calls “self-
governing enterprises”: workplaces controlled by the workers themselves, on a footing of equality
(Dahl, 1986: 91–92). “Subordinated neither to a capitalist nor a bureaucrat,” such workers “decide for
themselves … the details of their productive activities,” including how to divide labor, how to dis-
tribute pay and other benefits, and how to structure the conditions of work (Corneo, 2017: 135–136).
Will these “self-governing enterprises” require endless meetings, with each worker weighing in
on every conceivable issue, regardless of expertise or inclination? No. Citizens of a political de-
mocracy typically outsource the day-to-day business of ruling to electorally accountable re-
presentatives. Citizens of a workplace democracy can do the same, delegating various issues to an
elected workers’ council.7
Socialism, then, means economic democracy – which, in turn, involves empowering ordinary
people in their roles as consumers, citizens, and producers. In an economic democracy, key decisions
about production (what gets made?), investment (how do we direct the surplus?), distribution (how
do we allocate purchasing power?), and working conditions (how do we divide labor and structure
the work environment?) are made not by capitalists, not by unelected managers, not by un-
accountable state officials, but by “the whole community of people served by, and living with or in
the economy” (Wolff, 2019: 29).8
6 Economic Democracy in Practice 9

“Sounds good in theory,” some readers might be thinking, “but how could economic democracy
work in practice?” A reasonable question, to which socialists owe an answer. Some socialists eschew
grand blueprints, pointing instead to incremental reforms that, alone or in conjunction, might nudge
the economy in a socialist direction (Wright, 2010, 2019). But others, in a bolder spirit, do propose
to remake the economy from the ground up.10 Let me detail two such transformative proposals: one
that attempts to rehabilitate planning by decentralizing and democratizing it; another that tries to
blend social control of the means of production with market production for profit.
280
Socialisms
Democratically planned socialism. Most socialists have, with good reason, abandoned central plan-
ning, which – as practically everyone now agrees – suffers from debilitating informational and
motivational problems (Hayek, 1945; Corneo, 2017: Ch. 5). It’s nearly impossible for central au-
thorities to gather good information from producers and consumers; to generate a sensible plan on
the basis of this information; or to motivate people to play their assigned economic roles diligently
and well. But perhaps these problems have more to do with excessive centralization than with
planning as such. This is the core suggestion of thinkers like Michael Albert, who advocates a
radically participatory form of planned socialism called “Parecon” (Albert, 2003).
Parecon’s basic gist is this. Imagine an economy with social ownership of productive assets and no
markets. How might such an economy solve “the allocation problem”: how might it determine
what gets made, by whom, for whom? Well, to figure out what people are willing to produce and
what they hope to consume, suppose we just ask them – aided, perhaps, by a well-designed “Parecon
app.”11 We then aggregate all these responses and compare proposed supply with proposed demand,
helped, again, by cutting-edge supercomputers. If the proposed supply and demand don’t match, we
close the gap through democratic negotiation and deliberation.12 Through such negotiation we
eventually reach a manageable number of possible plans. We put them to a vote and enact the
winner. Voila – we’ve solved the allocation problem democratically, without hierarchy or markets.
The absence of hierarchy is worth emphasizing. No one occupies any special position of au-
thority in Albert’s model. Economic decisions are not dominated by the wealthy (as under capit-
alism) or the politically powerful (as under statism). Instead, all economic decision-making is
radically democratic and open to negotiation: each person has a say over decisions that affect him or
her, proportional to the degree to which he or she is affected. Parecon may have important flaws,
but inadequate respect for democratic values would not seem to be among them. Indeed, it’s hard to
imagine a system more faithful to the core socialist commitment to bottom-up, democratic control
over the economy. This, surely, is Parecon’s chief virtue from the socialist perspective.
“But parecon purchases democratic empowerment only at serious cost to feasibility,” some
readers might complain. Others, in a less charitable spirit, might declare parecon obviously un-
workable: a total pipe-dream.13 But if socialists cannot appeal to comprehensive planning, whether
centralized or democratic, to solve the allocation problem, then they have but one arrow left in their
quiver: markets. Many important market socialist models have been developed; here, I focus on
David Schweickart’s influential “Economic Democracy” (ED for short).14
Market socialism. Schweickart’s model strategically transplants certain core elements of capitalism –
namely, markets and profit-seeking enterprises – into a broadly socialist framework featuring social control
over investment and workplace democracy.
Let’s start with ED’s capitalist elements. In stark contrast with planning-based forms of socialism,
ED solves the problem of allocation using market competition between profit-seeking enterprises.
Turning a profit is the immediate aim of production in ED: enterprises, guided by the price signal as
determined through market-mediated supply and demand, produce to make money, not (primarily)
to satisfy human needs. As Schweickart says, “profit is not a dirty word in this form of socialism”
(2011: 51). But what happens with these profits? Here an important difference with capitalism
emerges. Under capitalism, profits go to owners, not workers, who receive wages. Under ED, by
contrast, there are no wages; rather, “workers get all that remains once nonlabor costs … have been
paid” (ibid.). Precisely how workers divvy up the enterprise’s surplus is up to them.15 Empirical
evidence suggests that self-managed firms (like those in the Mondragon cooperative in Spain) opt for
a 4 or 5:1 ratio between the incomes of the highest- and lowest-paid employees: quite a dramatic
difference from the 300:1 spread typical in large capitalist corporations (Schweickart, 2011: 96).
Having sketched ED’s capitalist elements, I now present its socialist side. First, ED requires that
all enterprises above a certain size adopt a democratic structure: they must become self-governing
enterprises, controlled by their workers (ibid.: 49). Second, ED implements social control over
281
Samuel Arnold
investment. In an ED, the means of production belong to all citizens, not to the enterprises that
happen to deploy them. To reflect this social ownership, all enterprises must pay a capital assets tax,
which “may be regarded as a leasing fee paid by the workers of the enterprise for use of social
property that belongs to all” (ibid.: 52). Revenue from this tax constitutes the national investment
fund, which is the sole source of investment in ED. By changing the tax rate, society determines the
size of the national investment fund – hence, the amount of money available for investment, and
thus the overall level of economic growth and development.
Note the contrast with capitalism: under capitalism, most investment comes from private rather
than public sources. Both the volume and direction of economic development, therefore, depend
on the whims and abilities of private investors. Let me underscore this point, which may sound dry
and technical but is actually of the highest importance. Under capitalism, society’s future economic
trajectory is largely determined by the uncoordinated, self-interested decisions of wealthy investors. Capital pours
into, say, Facebook or ExxonMobil not because we’ve decided together that this is the best use of
society’s scarce financial resources, but because that’s where private investors can expect the highest
returns. Under ED, by contrast, control over this fateful issue – future economic development – is
wrested away from the capitalist class and handed to the community as a whole.16 Under ED, we
chart our economic future together by bringing finance and investment under social control – thus
enabling (in theory, anyway) “more rational, equitable, and democratic development than can be
expected under capitalism” (ibid.: 52).
But how, specifically, should social control over investment be institutionalized? There are many
options. At one extreme, society might opt for a planning-heavy system in which a democratically
accountable planning board draws up a plan for all new investment (which Schweickart estimates
would constitute about 15% of GDP) and allocates funds accordingly. For example, the planning
board might decide to prioritize renewable energy, or consumer goods, or whatever. At the other
extreme, society might prefer a laissez-faire model in which funds are channeled through public
banks to enterprises using essentially the same criterion that capitalist banks use: namely, profitability.
In this version of ED, market forces would largely determine the pattern of investment.
Schweickart himself proposes something in the middle of these two options. He envisions a na-
tionwide system of regional and community public banks funded on a per capita basis. (All investment
funds, again, come from the capital assets tax paid by all enterprises as a ‘leasing fee’ to society.) Because
these regional banks receive funding on a per-capita basis, parity is ensured across regions of similar size,
but different levels of development. If Silicon Valley has the same population as, say, the greater Detroit
region, then both areas (despite their radically different levels of development) will receive the same
amount of investment capital. The upside of this? No economic backwaters, and no neo-liberal “race
to the bottom” to compete with other areas for investment dollars.
Enterprises in need of investment (to expand production, say) apply to their area’s banks for
funds. Banks assess applications on the basis of (a) profitability, (b) job creation, and (c) any other
democratically chosen criteria, such as ecological impact. This mixed standard implies that while
profitability matters, it is not all that matters. Projects that further socially chosen goals may be
chosen over more profitable, but less socially desirable alternatives. Market socialism thus promises
to harness the dynamism and efficiency of the market without subordinating democracy, without
putting profit over people.
7 Economic Democracy Without Socialism?

Defenders of capitalism might respond to the above by granting economic democracy’s desirability
but insisting that we already have it under capitalism. While it’s true, these defenders will say, that
capitalists steer the economic ship, it’s not as if they can head in any old direction they like. Instead –
as Ludwig von Mises (1944: 20) explains – “capitalists are steersmen only, bound to obey
282
Socialisms
unconditionally the captain’s orders.” And who is the captain? “The real bosses,” writes von Mises,
“are the consumers” (ibid.). Consumers demand new products, better quality, lower prices.
Businesses that obey, survive; those that don’t, don’t. In short, market competition positions
consumers as the real power behind the capitalist throne. Capitalism, therefore, implements a kind of
“economic democracy” in which “every penny gives the right to vote” – or so von Mises concludes
(ibid.: 21).
Although there is a good deal of truth in his argument, I think that von Mises exaggerates the
degree to which capitalist markets empower people to exercise genuine democratic control over
production and investment. There are three issues here.17
First, markets respond to money, which is distributed unequally; more money means more votes.
This is not the sort of egalitarian control championed by socialists.
Second, markets do not allow people to deliberate together about investment decisions, about
what we want our economy to look like in the future. I can buy a plane ticket on Orbitz, but I can’t
use that platform to convey my deeper belief that we’d all be better off eliminating carbon-spewing
forms of transit altogether. To have that sort of conversation – to really begin to shape our society’s
investment trajectory – we’d have to leave the market behind and enter the forum: we’d have to exit
commerce and enter politics and engage in what socialists call democratic planning.
Third, capitalism’s system of private (rather than social) control over finance and investment rights
gives private investors a potent veto over democratic decision-making. Suppose we decide together to
nip our health care problem in the bud by extending Medicare to all citizens – a costly endeavor we
propose to fund via higher taxes on the wealthy. Like clockwork, gray eminences appear on business
channels, auguring economic doom: “These tax increases will destroy the economy!” Investors, fearing
a downturn, threaten to send their capital elsewhere. Markets drop, hiring declines, and the economy
begins to teeter. Left-wing politicians, realizing that they’ll pay for any economic pain at the ballot box,
moderate their health care (and tax) proposal so that it better conforms to business interests. Thus does
another progressive dream die on the vine, killed not by any intrinsic defect, but by the “capitalist
veto”: the ability of private investors to keep public policy within narrow, capital-friendly boundaries
by threatening disinvestment (Lindblom, 1982; Schweickart, 2011: 153–158).18
In sum, then, von Mises is right that markets empower consumers to some extent, especially over
decisions about which commodities to make here and now. But do markets really enable what
Malleson calls consumer and citizen democracy? Do they really give ordinary people roughly equal
control over production and investment decisions? I think not. Under capitalism, rich people have
more “dollar votes,” thus undermining consumer democracy. Nor do capitalist markets implement
citizen democracy, for two reasons. First, to share in control over how the economy develops over
time, we need to reason and deliberate together, not signal our immediate consumption preferences. We
need the forum, not the market; we need politics, not commerce. But this means we need to engage in
precisely the sort of democratic planning proposed by socialists. Second, capitalism gives the investor
class de facto veto power over policy, thus sharply limiting citizen democracy. We can hardly be said to
enjoy equal control over investment if a tiny financial elite can smother disliked proposals in the cradle.
8 Why Economic Democracy?

Only socialism, I have just argued, delivers genuine economic democracy. To which the capitalist
might reply: so what? What’s so appealing about economic democracy, anyway? One answer (of
many that we might explore) is this: economic democracy rescues us from a distinctive kind of
unfreedom experienced under capitalism called “alienation.”
Imagine you’re a businessperson in a capitalist economy. Competitive pressures force you to cater
to consumer demand. You’ve got to give customers what they want, for the lowest possible price.
So you sell A rather than B, because A flies off the shelves, while B collects dust. What if you think
283
Samuel Arnold
A is a harmful piece of junk – indeed, that the world would be better off without it? Well, tough
luck: if you want to stay in business, produce A you must. Nor does it matter much what you do.
Should you, on principle, decide not to make A, some other businessperson will fill the void.
We see here the way in which capitalism pressures market actors to operate according to economic
imperatives rather than what we might call ethical ones. Markets steer us in directions not of our own
choosing; they displace individual judgment, leaving market participants – as Roberts (2017: 56–103) has
argued – in a position analogous to that of the “akratic”: the sufferer of weakness of will, who, seeing
the better, does the worse anyway. “Fracking,” says the oil executive. “What a disaster! But if we don’t
frack, we’ll go under. So frack we must.” The executive’s individual ethical judgment about fracking
matter not. It’s an idly spinning wheel. He’s got to do what the market tells him to do.
As do we all. For “under capitalism, all questions of what we need and what we want must be
subordinated to the question of what is profitable” (Hagglund, 2019: 250). Thus subordinated, we work,
haggle, and sell, without reference to what’s good or reasonable: only what’s profitable matters.
A peculiar kind of unfreedom, this subordination to market forces. For these forces are, after all,
products of our own creation! We are like the sorcerer’s apprentice, mastered by powers we
conjured but somehow cannot control. Marxists have a term for this sort of unfreedom: alienation.
“Alienation,” writes Kolakowski (1978: 147) “means the subjugation of man by his own works,
which have assumed the guise of independent things.” Now, this “subjugation of man by his own
works” – our enslavement to market forces – sometimes turns out rather well, as per Adam Smith’s
invisible hand; other times, not so well, as during financial crises and depressions. But – as Wood
(2004) explains – the really alienating feature of capitalism is not, at bottom, that it sometimes
produces bad outcomes. No, “what is alienating is more basically that under capitalism human
beings cannot be masters, whether individually or collectively, of their own fate, even within the
sphere where that fate is a product solely of human action” (ibid.: 49).
Importantly, the problem of alienation is
not that individuals cannot do exactly what they each want to do, but that they cannot get
together and talk about what sorts of things should and should not be done, and what sorts
of reasons should and should not count as good reasons. (Roberts, 2017: 96)
Under any conceivable system, we must integrate our activities with needs and activities with others.
Barring utopian abundance, the elimination of scarcity, there will always be work to do, and this
work will have to aim at certain goods made in certain ways. Economic constraints will always be
with us. But must these constraints be taken as utterly parametric, as brute givens beyond the reach
of democratic contestation and deliberation? Must we simply accept that fracking is profitable and so
fracking there will be?
Socialism’s central message is that alienation isn’t inevitable, that we can “consciously subordinat
[e]” the “market … to a democratic society,” as Karl Polanyi (1957: 241) puts it. Regarding the
specifics of this subordination, socialists are, as we saw in Section 6, of two minds. Some think that
to subordinate market forces requires eliminating them entirely. This was, as Wood (2004: 53)
explains, Marx’s view:
The means of production must be owned collectively… [The use of these means] will, in
Marx’s communism, be performed consciously. Decisions about them will be made de-
mocratically, by society as a whole, and not by a privileged class, acting contrary to the
interests of the laboring majority and subject to the alien constraint of profit-maximization.
Contemporary advocates of this anti-market, planning-centric position include Albert (2003) and
Hagglund (2019).19
284
Socialisms
Other socialists, like Schweickart (2011), Malleson (2014), and others, think the planning-centric
position goes too far. These market socialists grant that market forces in a capitalist context displace
ethical judgment from economic life, orienting everything around the North Star of profit. But is
the best response simply to flip this relationship between the ethical and the economic, so that we
must all justify our consumption and production choices to one another, Parecon-style? Rather than
banish markets and profits entirely – rather than turn every economic interaction into an exercise in
democratic deliberation – we should enact an institutional division of moral and economic labor, or
so these market socialists suggest. Let social control of investment, as instantiated through a network
of democratically accountable public banks, handle the ethics; let market exchange between con-
sumers and worker-controlled, profit-seeking firms handle the economics.
Under market socialism, what happens to any given commodity is generally a function of Smith’s
invisible hand. Market forces generally determine, say, how much tin is mined, and where it goes.
But these market forces are not taken as brute givens. We can bring them to heel if that’s what we
democratically decide to do. Suppose tin turns out to be critical for decarbonizing the atmosphere;
our carbon-scrubbers are hungry for the stuff. Well, we can instruct our public banks to direct
investment funds towards tin mining and decarbonization efforts, and away from uses of tin that
compete with these ends. Notice that we can do this even if these funds, or this tin, could find more
profitable uses elsewhere in the economy. Perhaps Tin Sculptures Inc. sells its wares like hotcakes; it can
turn tin into gold; it submits a proposal for additional investment capital so that it can expand
production. Under the capitalist model of finance, private investors would ask only about expected
returns on investment. Profit being their only concern, capital (and therefore tin) would fly towards
Tin Sculptures rather than saving the planet. Not so under market socialism, whose public banks are
designed to serve objectives that are broader than profit. “Sorry, Tin Sculptures,” our public bankers
could say, “but society has democratically charged us with furthering ecological goals through our
investments. This tin is needed for higher uses than mere money-making.”
Market socialism thus promises to emancipate us from alienating subordination to market forces
without doing away with markets entirely. To my mind, it strikes the right balance between con-
scious, social control, and Smith’s invisible hand (Arnold, 2017). Under market socialism, we bend
market forces to our collective will, even as we harness their dynamism and power.
9 Conclusion
The problem with capitalism, it is sometimes said, is that it puts profits over people. From the
socialist perspective, that critique is spot on. Economic forces must be subordinated to democratic
social power: this, I have suggested, is socialism’s central commandment. A socialist economy would
be one that invests ordinary people, rather than economic or political elites, with roughly equal say
over their workplaces and roughly equal influence over what gets made and how the economy
develops over time. Whether such an economic vision is compatible with markets and profit-
seeking enterprise remains a contested issue among socialists, as does the feasibility of planning-
centric alternatives. But socialists of all stripes agree that so long as ownership of the means of
production lies in private hands – so long as we let capitalists steer the economic ship – we will
thereby find ourselves at the mercy of vast economic forces of our own making but beyond our
control. We will be, in a word, alienated. Only by democratizing the economy – only by putting
people over profits – can we eliminate this distinctively capitalist kind of unfreedom.
Notes
1 Different liberal egalitarians will unpack this third, “fairness” requirement in different ways. John Rawls
(1999), for instance, argues that a fair economy is one that (a) neutralizes the effect of social class on one’s life
285
Samuel Arnold
prospects and (b) arranges economic inequalities so as to maximize the economic position of the worst-off
members of society. While socialists generally find much to like about Rawls’s position, some push his
theory in a more egalitarian direction. Thus, Pablo Gilabert (2015) and Joseph Carens (2003) argue for a
Marx-inspired principle of distributive justice on which each should contribute according to his ability, and
each should receive according to his needs (broadly construed); or again, G.A. Cohen (2009) and John
Roemer (2017) support a “luck egalitarian” conception of distributive justice that permits inequalities that
arise from choice, but forbids those stemming from unchosen factors like bad luck or inauspicious social or
natural endowments.
2 To illustrate: G. A. Cohen counts (appropriately) as a socialist liberal egalitarian in virtue of embracing not just
freedom, not just an economic minimum, and not just economic fairness (in the form of luck egalitar-
ianism), but also the “antimarket principle” of “communal reciprocity,” according to which “I serve you
not because of what I can get in return… but because you need or want my service” (2009: 39).
3 This section and the next draw on (Arnold, 2019).
4 For helpful discussion of what socialists have meant – and should mean – by “the means of production,” see
Edmundson (2020).
5 Wright emphasizes that capitalism, socialism, and statism exist in pure form only theoretically. In reality, all
economies are hybrids containing capitalist, socialist, and statist elements. We should think, then, of so-
cialism and capitalism as variables, as matters of degree. An economy is more capitalist to the extent that
investment and production decisions “are governed by the exercise of [private] economic power”; more
socialist to the extent that such decisions “are governed by the exercise of social power,” that is, the power
held by ordinary people as members of civil society. See Wright (2010: 123–128) for further discussion.
6 It is not socialism’s only commandment. For discussion of other important socialist emphases such as de-
commodification, see O’Neill (2021).
7 As Giacomo Corneo explains, “There are different ways in which the workers’ collective might make
decisions. In smaller enterprises, it could that many of them are made at general meetings…In larger en-
terprises, perhaps only important strategic questions would be taken up [collectively]. As a rule, the
workforce elects a workers’ council and delegates to it far-reaching decision-making powers” (2017: 136).
8 Objection: economic democracy assigns control rights over economic decisions to multiple agents at
multiple levels. It says, for instance, that workers should decide what to make – but so too should consumers
have a say. What if what workers want to make isn’t what consumers want workers to make? What if, say,
autoworkers want to make sedans, but consumers want SUVs? How should economic democracy balance
these competing inputs? For a helpful discussion of this important topic, see Vrousalis (2018).
9 This section draws on Arnold (n.d.).
10 Why bother with speculative model-building? I agree with Malleson that “the point is not to build castles in
the sand, but to try to paint a picture of a society that is at once plausible enough to seem worth talking
about yet inspiring enough to warrant the risk, energy, and commitment that building such a society would
require” (198).
11 Why not just tap into the app: “I propose to consume everything, and to produce nothing?” Under
Parecon, a person’s consumption power depends on how hard he or she has worked to help meet the
community’s needs. So a person who produces nothing consumes nothing (unless she is disabled or otherwise
unable to work, in which case she is guaranteed a normal standard of living).
12 Albert suggests a more complex institutional structure than I can detail here. Suffice it to say that the
planning process is mediated and aided by various councils (workers’ councils, sectoral councils, neigh-
borhood councils, national councils, and so on) and an overarching “Iteration Facilitation Board” that helps
shepherd plans through the negotiating process. These various organs help structure what would otherwise
be an impossibly unwieldy process. But at its core, Parecon’s method for solving the allocation problem is
just as I describe in the main text: we ask each other what we want to consume and produce, and keep
talking until we’ve hit upon a plan acceptable to the majority.
13 Is Parecon infeasible? For discussion, see Wright (2010: 260–265). For possible responses, see Arnold (n.d.:
section 8.b), and Gilabert and O’Neill (2019: section 5.1).
14 For a recent proposal very similar in spirit to Schweickart’s, see Malleson (2014).
15 In theory ED’s workers could split the firm’s surplus equally. But given the need to outcompete other
enterprises – hence, to attract and retain skilled labor – some degree of inequality is likely to be chosen.
More productive workers, or workers with skills in higher demand, will almost certainly earn more than
their fellows.
16 As Malleson (2014: 201) puts it, when “investment is democratized and all the banks are run as democratic
institutions[,]…the power to direct the economic development of society has evolved from a private pri-
vilege [based on ownership of capital] to a basic right of all citizens.”
286
Socialisms
17 I draw in what follows on Malleson (2014: 93–97).

18 Notice that private investors really do have a veto over democratic politics in the described scenario. The demos
says: “We want X!” But X spooks financial elites. They disinvest, or threaten to. Politicians say “Ugh…if the
economy tanks, we won’t be reelected. Let’s not do X.” And so X doesn’t happen. Not because ordinary people
don’t want X – they do! But because investors don’t want X. They can block disliked policies, even over-
whelmingly popular ones, by credibly threatening to disinvest. In other words, they have a veto.
19 Hagglund (2019: 385) writes that “the means of production must be collectively owned and employed for
the sake of the democratically determined common good, rather than privately owned and employed for the
sake of profit.”
References
Albert, M. (2003) Parecon: Life After Capitalism. London: Verso.
Arnold, S. (n.d.) “Socialism,” The Internet Encyclopedia of Philosophy [online]. Available at: https://iep.utm.edu/socialis/
Arnold, S. (2017) “Capitalism, Class Conflict, and Domination,” Socialism and Democracy, 31, 1, 106–124.
Arnold, S. (2019) “What is Democratic Socialism and Why is it Growing More Popular in the U.S.?” Teen
Vogue 1 May [online]. Available at: https://www.teenvogue.com/story/what-is-democratic-socialism
(Accessed: February 22, 2021)
Arnold, S. (2021) “No Community Without Socialism: Why Liberal Egalitarianism is Not Enough,” forth-
coming in Philosophical Topics 49, 1.
Carens, J. (2003) “An Interpretation and Defense of the Socialist Principle of Distribution,” Social Philosophy and
Policy, 20, 1, 145–177.
Cohen, G. A. (2009) Why Not Socialism? Princeton; Princeton University Press.
Corneo, G. (2017) Is Capitalism Obsolete? Cambridge; Harvard University Press.
Dahl, R. (1986) A Preface to Economic Democracy. Berkeley; University of California Press.
Edmundson, W. (2020) “What are the means of production?” Journal of Political Philosophy 28, 4, 421–437.
Fraser, N. and Jaeggi, R. (2019) Capitalism: A Conversation in Critical Theory. Medford, MA; Polity.
Gilabert, P. (2015) “The Socialist Principle ‘From Each According to Their Abilities, To Each According to
Their Needs’,” Journal of Social Philosophy, 46, 2, 197–225.
Gilabert, P. (2021) “Alienation, Freedom, and Dignity,” forthcoming in Philosophical Topics 49, 1.
Gilabert, P. and O’Neill, M. “Socialism”, The Stanford Encyclopedia of Philosophy (Fall 2019 Edition), Edward N.
Zalta (ed.) [online]. Available at https://plato.stanford.edu/archives/fall2019/entries/socialism/.
Harrington, Michael (1989), Socialism: Past and Future. New York, NY; Arcade Publishing.
Hagglund, M. (2019) This Life: Secular Faith and Spiritual Freedom. New York; Pantheon Books.
Hayek, F. (1945) “The Use of Knowledge in Society,” The American Economic Review 35, 4, 519–530.
Jones, J. and Saad, L. (2019) US Support for More Government Inches Up [online]. Available at: https://news.
gallup.com/poll/268295/support-government-inches-not-socialism.aspx (Accessed: February 22, 2021).
Kolakowski, L. (1978) Main Currents of Marxism. New York; Norton.
Lindblom, C. (1982) “The Market as Prison,” The Journal of Politics 44, 2, 324–336.
Malleson, T. (2014) After occupy: Economic democracy for the 21stcentury. New York; Oxford University Press.
O’Neill, M. (2021) “Social Justice and Economic Systems: On Rawls, Democratic Socialism, and Alternatives
to Capitalism,” forthcoming in Philosophical Topics 49, 1.
Polanyi, K. (1957) The Great Transformation. 2nd edn. Boston; Beacon Press.
Pramuk, J. (2019) “Expect Trump to Make More Socialism Jabs,” CNBC.com, 6 February [online]. Available at:
https://www.cnbc.com/2019/02/06/trump-warns-of-socialism-in-state-of-the-union-as-2020-election-
starts.html (Accessed: February 22, 2021).
Rawls, J. (1999) A Theory of Justice. Revised edn. Cambridge; Harvard University Press.
Rawls, J. (2005) Political Liberalism. Expanded edn. New York; Columbia University Press.
Roberts, W. C. (2017) Marx’s Inferno: The Political Theory of Capital. Princeton; Princeton University Press.
Roemer, J. (2017) “Socialism Revised,” Philosophy & Public Affairs 45, 3, 261–315.
Schweickart, D. (2011)After capitalism. 2nd edn. Lanham, MD; Rowman & Littlefield.
Socialism. (n.d.) In Oxford English Dictionary [online]. Available at: https://www.oed.com/ (Accessed: February
22, 2021).
Thompson, D. (2018) “Amazon’s HQ2 Spectacle,” The Atlantic 12 November [online]. Available at: https://
www.theatlantic.com/ideas/archive/2018/11/amazons-hq2-spectacle-should-be-illegal/575539/
(Accessed: February 22, 2021).
287
Samuel Arnold
Van Zuylen-Wood, S. (2019) “Pinkos Have More Fun,” New York Magazine, 3 March [online]. Available at:
https://nymag.com/intelligencer/2019/03/socialism-and-young-socialists.html (Accessed: February 22, 2021).
Von Mises, L. (1944) Bureaucracy. New Haven; Yale University Press.
Vrousalis, N. (2018) “Council Democracy and the Socialisation Dilemma” in Muldoon ed. Council Democracy:
Towards a Democratic Socialist Politics, New York; Routledge.
Vrousalis, N. (2021) “Socialism Unrevised: A Reply to Roemer on Marx, Exploitation, Solidarity, Worker
Control,” Philosophy and Public Affairs 49(1): 78–109.
Wolff, R. (2019) Understanding Socialism. New York; Democracy at Work.
Wood, A. (2004) Karl Marx. 2nd edn. New York; Routledge.
Wright, E. O. (2010) Envisioning Real Utopias. New York; Verso.
Wright, E. O. (2019) How to Be an Anticapitalist in the 21st Century. New York; Verso.
Zaroli, J. (2020) “There’s rich, and then there’s Jeff Bezos rich,” NPR.org, 20 December [online]. Available at:
https://www.npr.org/2020/12/10/944620768/theres-rich-and-theres-jeff-bezos-rich-meet-the-members-
of-the-100-billion-club (Accessed: February 22, 2021).
Further Reading
Brennan, Jason (2014) Why Not Capitalism? New York; Routledge. A sharp parody of and rejoinder to G.A.
Cohen’s Why Not Socialism? that defends capitalism on moral (rather than pragmatic) grounds.
Cohen, G. A. (2009) Why Not Socialism? Princeton; Princeton University Press, Argues that – bracketing issues
of feasibility – non-market socialism is morally desirable, but concedes that socialists do not know whether
non-market socialism is feasible.
Malleson, T. (2014) After Occupy: Economic Democracy for the 21st Century. New York; Oxford University Press.
Empirically and philosophically rich development of a broadly market-socialist position with an especially
interesting defense of workplace democracy.
Schweickart, David. (2011) After Capitalism. Second edition. Lantham, MD; Rowman & Littlefield. Argues for
a heterodox form of socialism that blends profits and markets with workplace democracy and social control
over investment.
Wright, Erik Olin. (2010) Envisioning Real Utopias. London; Verso. Drawing widely from social science and
philosophy, reimagines socialism for the 21st century.
288
23
PROPERTY-OWNING
DEMOCRACY
Alan Thomas
A property-owning democracy is a utopian proposal for a market society, containing the institution
of private property, that guarantees to each citizen an allocation of capital as of right (Rawls, 2001;
Freeman, 2013; Thomas, 2017). It is a form of pre-distributive egalitarianism where the emphasis is
not on post-distributive redress to correct for unfair outcomes. Its aim, rather, is to identify that
which Philippe Van Parijs calls an upstream “control variable” which determines that subsequent
market outcomes are fair (Parijs, 1995, p.13). That control variable is an agent’s initial endowment
of capital where the words “initial” and “subsequent” are not to be given a limiting, temporal,
interpretation (Thomas, 2020a). While it is true that some theorists in this tradition emphasise giving
capital to citizens in the early stages of their lives, from baby bonds to demogrants in early adulthood,
this policy question is separate from the abstract modelling of endowments that agents bring to the
market prior to the determination of market outcomes. From Thomas Paine to Anne Alstot and
Bruce Ackerman (and Thomas Piketty) the pre-distributive tradition is associated‚ rightly, with
demogrants, but that is merely one policy from a range of policies that flow from the more basic idea
of pre-distribution (Ackerman and Alstot, 1999; Piketty, 2020; Paine, 1797/2000). This limited
interpretation of the idea has not helped in the attempt to spell out its distinctiveness as an alternative
to familiar redistributive egalitarianism (Thomas, 2020a). At the level of policy, the property-
owning democrat hands over to the discipline of macro-economics the empirical question of which
policy solutions maximise the ratio of productive to unproductive uses of capital. However, the
tradition has developed a range of core proposals, notably targeting inheritance via estate taxes, as a
form of gift-giving that concentrates capital produced productively in the past in the hands of the
non-productive. Piketty has suggested an on-going wealth tax on assets, although partly for reasons
of transparency to encourage fuller disclosure about the forms in which wealth is held (Piketty,
2014). These policies might usefully be traded off against each other; if a person has been subject to
on-going taxation of assets throughout their lifetime, then the taxation of inheritance may prove
unnecessary.
It follows that the issue of “initial” allocations – how we model economic agency on the market – is
only part of the story. Equally important is a contrast that Rawls drew between what he called an “ideal
historical process view” and an “ideal social process view” (Rawls, 2001, pp.52–5). The former is the
neo-Lockean view of Robert Nozick that Rawls rejects (Nozick, 1974). The latter identifies a generic
family of views of which justice as fairness is a part, in which we must constantly regulate the back-
ground conditions to economic transactions to ensure that justice stays “on track” (Thomas, 2020a).
We do not simply play the game and see where it leads as process and product are interconnected.
DOI: 10.4324/9780367808983-28 289

Alan Thomas
The background conditions for fairness involve the continual adjustment of “pure procedural back-
ground justice” (Rawls, 2001, p.53). That is precisely what the pre-distributive concern with capital
generation, holding, and dispersal consists in and why Rawls believed that a property-owning de-
mocracy was a specification of his ideal of justice as reciprocal fairness.
The basic idea, then, is that if these capital allocations are fixed, then other advantages of a free
market can be achieved without the bad effects we associate with unfairly determined market
outcomes. Some proponents of the ideal, such as Rawls, were particularly concerned with the
development of an alternative to the capitalist embedding of free markets (Rawls, 2001; O’Neill and
Williamson, 2012; DiQuattro, 1983). It may seem counter-intuitive to treat a view that is focused
on capital as anti-capitalist, but Rawls took his understanding of capitalism from Marx (Rawls,
2007). On that understanding, a capitalist society is one in which a minority of citizens exercise
monopoly control over those forms of productive capital that make up a society’s major means of
production. This monopoly over the means of production gives this minority faction control over
the production process. In particular, it gives them control over the terms on which they engage other
citizens to labour for them in return for a wage. Given the extent of the social power of this restricted
class, it may be able to determine whether other citizens are able to work for a wage at all. This
capitalist class has, for example, an interest in producing solely for monetary profit and thereby an
interest in reducing the cost of labour, so it may have a derived interest in maintaining a pool of
unemployed labour to keep labour costs down and to enhance workplace discipline (Kalecki, 1943).
By contrast, given that the property-owning democrat views employment as the deployment of human
capital, structurally grounded unemployment is simply capital wastage. Typically, there are over-
determined reasons in the property-owning democratic tradition for its central policies and Rawls saw
a direct argument from his relational egalitarianism, via the importance of citizens’ self-respect, to a
policy of full employment brought about by state action (Moriarty, 2009). This is a paradigmatic
instance of the convergence between Rawls’s views and the mid-century neo-Keynesianism in eco-
nomics in which he became immersed during his time as a Fulbright fellow in Oxford.
Rawls thought that all of these bad effects of capitalism could be avoided by developing a non-
capitalist alternative form of social system in which all citizens have a right to capital (Freeman,
2018b; Krouse and MacPherson, 1986; Krouse and McPherson, 1988; Thomas, 2017). This pro-
posal dissolves the capitalist class as a distinct group. According to Rawls, this proposal embeds a
market in a pre-distributive structure that patterns its outcomes as fair (Rawls, 1971/1999).
Furthermore, recent commentators have interpreted Rawls’s proposal as taking seriously the threat
to just institutions posed by the emergence of large concentrations of economic power that then
ground unfair political advantage based on the sectional interests of the wealthy (Edmundson, 2017;
Thomas, 2017). That which Thomas Piketty has called the “drift to oligarchy” is perceived, pro-
minently in the republican tradition of political theory, to be a threat to that which Rawls called the
“fair value” of the political liberties (Piketty, 2014). William Edmundson has recently suggested that
Rawls ought to have included this socio-historical datum that there is a historical tendency, in unfair
societies, for those with excessive socio-economic power to seek undue influence over the political
process in their own interest as one of the “circumstances of justice” (Edmundson, 2017, p.14,
chapter 5). Edmundson suggests the phrase “the fact of domination” as capturing this problem to
which a property-owning democracy represents one form of solution by pre-emptively dispersing
accumulations of wealth. These measures are supposed to be more radical than the insulation
strategy of permitting significant wealth inequality but insulating the political process from financial
pressure (Thomas, 2017, p.95ff).
That issue reflects the concern to make a property-owning stable over time. It is independent of
the basic predistributive justification for a property-owning democracy. That justification is that if
the basic institutional structure in which a market is embedded produces economic outcomes that
strongly influence people’s life choices, yet this is not a structure that they can meaningfully choose
290
Property-Owning Democracy
to enter or leave, then it must meet the stringent test that it does not produce outcomes against
which they can bring a reasonable complaint (Scanlon, 2018). If it is tempting to analogise this
institutional structure to the rules of a game, they are constitutive rules, and a game that we are all
forced to play (Buchanan, 1986). Rawls believed that under these constraints his pre-distributive
proposal would produce a demandingly egalitarian outcome that, in fact, was the institutional
implementation of his idea of justice as reciprocal fairness (Rawls, 2001).
If there is a distinctive strength to the idea of a property-owning democracy it is that it attracts
support from across the political spectrum (Jackson, 2005; Ron, 2008; Jackson, 2012). If it has a
distinctive weakness, it is that this very ecumenical appeal either empties the idea of any distinctive
content or makes it an essentially contested concept in W. B. Gallie’s sense (Gallie, 1955–6). The
history of the concept shows the breadth of its support in, for example, Roman Catholic social
teaching while it has also seemed to other conservatives to be the best way to combine capitalist
economic organisation with broad social democratic support for capitalism (Thomas, 2017, chapter 6).
The idea of bringing capital under social democratic control ought to place this idea firmly at the centre
of a progressive social agenda. If Rawls has been the most influential liberal thinker of the last half-
century, it is indicative that his final summary’s conclusion concerning his theory of domestic justice
was that it was only expressed by either a property-owning democracy or Mill’s version of liberal
market socialism (Rawls, 2001). The latter is also non-capitalist in the way it brings capital under social
democratic control via a system of worker co-operatives (combine with a system of complementary
consumer co-operatives). Once again, Mill’s proposal eliminates a distinct capitalist class.
However, the ideal of a property-owning democracy has also been an object of recurrent sus-
picion from the Left on the grounds that it is an attempt to capture skilled workers from the working
class and to make them “think like the bosses” by encouraging aspiration and enterprise in the guise
of a petit rentier capitalism (Dowding, 2013). Combined with Rawls’s scepticism directed at welfare
state capitalism, the political program of pursuing a property-owning democracy has seemed, to
many on the Left, to reflect a misguided possessive individualism that inherits too many of the
personal level motivations of capitalism (Edmundson, 2017). Defenders of the ideal respond that this
reflects a limitation in the progressive imagination and that Rawls’s proposal is a radical break with
our existing paradigms with the goal of securing a republic of equals (Thomas, 2017).
1 The Importance of Rawls’s Contribution

Rawls put on the agenda of recent political philosophy the very idea of what he called our “choice
of a social system” (Rawls, 2001). This is the idea that we can select, from a finite list, a generic type
of socio-economic system that represents implementations of different conceptions of justice. Rawls
also believed that it was outside the purview of political philosophy to determine anything other
than a disjunctive answer to this question. Political philosophy allows us to rule out command
socialism, laissez-faire capitalism, and welfare state capitalism from his list, but it does not make the
choice between a property-owning democracy or liberal market socialism. The choice between
these two private property, realistically-utopian, and market-involving but non-capitalist systems
was to be made by each society in the light of its own history and traditions (Thomas, 2020c).
It has been pointed out, both to the advantage and to the detriment of a property-owning
democracy, that it is unique on Rawls’s list in not having been worked up from any historical
examples (Weithman, 2013). That seems unfairly to stack the deck when it comes to the selection of
a social system because a property-owning democracy is simply stipulated to meet the demands of
justice as reciprocal fairness. It also suggests an unfair imbalance between Rawls’s utopianism about a
property-owning democracy and his scepticism towards welfare state capitalism where a historical
record of failure is held against the latter social system (Schemmel, 2015). This point, as I will note
below, is exaggerated given that there have been traditions of pre-distributive egalitarianism, notably
291
Alan Thomas
in the USA. Equally significant is the historical role played by the inegalitarian variant of property-
owning democracy, which concentrates asset holding via its ownership by an unproductive rentier
class, in the recent history of both the USA and the United Kingdom.
Rawls’s development of this ideal from the work of Nobel Prize winning Cambridge neo-
Keynesian economist James Meade put the ideal of a property-owning democracy on the agenda
(Meade, 1964, 1993). The same ideal was then endorsed by other political philosophers working in
the ostensibly rivalrous political tradition of republicanism, the primary instance being the work of
Richard Dagger (Dagger, 1997, 2006). It has seemed to some that there is the prospect of a con-
vergence here in the form of a liberal-republicanism that meets two requirements at the same time
(Thomas, 2017). First, it meets Rawls’s requirement that Marx’s critique of capitalism be evaded by
the development of a socio-economic system without a distinct capitalist class. Second, it meets the
republican requirement that a society be developed where the domination of one agent by another be
made structurally impossible (Thomas, 2017, p.6). The liberal-republican claims that the demands of
these two requirements are co-extensive. In particular, the Roman republican and the political liberal,
both committed to a political neutrality of aim, can agree on the importance of the fair value re-
quirement for the political liberties. This convergence is, once again, unsurprising as both traditions
emphasise the importance of an ideal of citizenship as an achieved status (Thomas, 2017).
The details of Rawls’s egalitarian proposal are that the contextualisation of the market in a pre-
distributive structure reduces a citizen’s dependence on income from labour and increases the degree
to which they receive a return on their productive capital. This institutional engineering achieves
several valuable egalitarian goals in indirect ways. For example, a worker with a substantial holding
of capital has a greater democratic voice in the workplace; he or she has a strengthened right of exit;
employers will be under pressure to eliminate or automate drudge labour such that all work will be
meaningful and supportive of a citizen’s self-respect (Thomas, 2017). The ambitious goal, then, is to
achieve a range of egalitarian goals by institutional means. In the background here is a relatively
submerged commitment in Rawls’s overall scheme, already noted, to full employment guaranteed
by the state (Thomas, 2020b). That commitment could be interpreted as giving a “fair value” to the
right to exit from any particular employment.
There will be considerable unlocking of human productive potential when every citizen is given
a generative productive asset and returns to labour will be considerably flattened out, if not
eliminated, in a property-owning democracy. All incentives in the labour market will function
wholly to compensate for work that is difficult or dangerous or requires a lengthy period of training
(Smith, 1998). At the macro-level, capital will be made abundant, and economic boom and bust
cycles will be stabilised in a society without the structural deficit in consumption that is the in-
evitable consequence of high levels of inequality. Meade, the original proponent of the Rawlsian
conception of a property-owning democracy, specifically cites the rationale of avoiding the capital
loss inevitable in frequent recessions as a ground for adopting his proposals (Meade, 1993). I have
noted the ambition, in a property-owning democracy, of pre-emptively dispersing capital to
guarantee the “fair value” of the political liberties; however, this is, as it were, a counterfactual
guarantee. In a well-ordered and just society that takes the form of a property-owning democracy
(or Millian liberal market socialism) the institutional setup guarantees that a class of very wealthy
citizens will not arise in the first place.
Fiscal policy in a property-owning democracy will serve aims other than redistribution, primarily
aiming at stability. Examples of such policies would include extinguishing money in the economy so
that the state can spend without generating inflation, or to stabilise an economy by damping down
positional competition, or in other ways to serve the goal of stability. Redistributive goals will have
been made redundant by the correct institutional structure. It is for that reason that Rawls stated that
the details of tax policy are not part of the theory of justice (Rawls, 1971/1999, p.279). However,
the default policy for the taxation of incomes, he also suggested, ought to be Nicholas Kaldor’s
292
expenditure tax (Kaldor, 1956). This is a proportional tax on a person’s spending and Rawls en-
dorses Kaldor’s rationale for it: if a citizen’s income has been fairly earned, then it is only when one
spends that money that a claim is made on the “common store of goods”. This tax, then, would not
interfere with the propensity to save. The limited role of progressive taxation, in Rawls’s view, was
to “forestall accumulations of property and power” (Rawls, 1971/1999, p.279). It is, then, part of a
response to the fact of domination and its rationale is both macroeconomic and political stability, not
the raising of revenue.
From all of the foregoing, it is clear that pre-distributive egalitarianism and a property-owning
democracy are related as genus to species, but it is true that the latter is, currently, the most well
worked out a proposal for a distinctively pre-distributive egalitarianism. The basic idea of pre-
distribution is to model the endowments of agents on the market so as equalise their bargaining
power with particular attention to the labour market and to the conflict between capital and labour.
We do not permit unfair market outcomes and then redress them: it is tempting to say that with the
correct model implemented, we let the outcomes fall as they may.
Unfortunately, that simple idea is too simple, as Kristina Meshelski has pointed out (Meshelski,
2019). One way to indicate the problem here is that it makes John Rawls and Friedrich Hayek
strange bedfellows producing agreement where one might predict disagreement. The tempting idea
is that if the pre-distributivist gets her institutional setup right, then she can abstain from subsequent
interference in market transactions. With the rules of the game in place, we play the game, and
whatever outcome emerges is fair. However, Meshelski points out that this under-specified idea,
while it may explain Hayek’s attraction to pre-distribution, does not fully capture that which Rawls
intended.
Hayek did indeed, in his famous volume II entitled “The Mirage of Social Justice” exempt Rawls
from his critique of a concept otherwise described by him as a “will o’ the wisp”, an “empty
formula” or “superstition” (Hayek, 1982, p.100). However, this putative agreement overlooks the
fact that while Rawls describes his conception as one of pure procedural justice this conception is
opposed to two other contrasting conceptions. Unlike imperfect procedural justice, or perfect
procedural justice, for pure procedural justice, there is no independent standard for the process to
meet. Perfect procedural justice is established as fair because it leads to the outcome we want; in
imperfect procedural justice, both process and outcome are separately (and independently) assessable
as fair. On the Hayekian misreading of Rawls, our focus is solely on setting up the rules of the game,
but that ignores the fact that, for Rawls, we must carry out an interconnected assessment of the
justice of both procedure and outcome (Meshelski, 2019, p.346). The distinction that Rawls drew –
which has already been noted – between ideal social process views and ideal historical process views
only makes sense on the basis of Meshelski’s interpretation of Rawls’s distinctions (Thomas, 2020a).
2 Convergence with the Georgist Tradition

Pre-distributivism is a broad church, and a wide variety of political outlooks share the ideal that, in a
society of freely associating producers, each economic agent deserves an initial stock of productive
assets. The simplest idea is that one’s fundamental productive asset is one’s body, a primordial item of
property that then generates more private property through one’s ability to work. One’s body is
both a productive and generative asset: its initial ownership leads to the production of other assets.
While that is one way to initiate a pre-distributive setup, another tradition looks for an initial stock
of assets that we might reasonably assume are held in common either as a result of divine providence
or as the result of a normative intuition that nature is, prima facie, an object of collective ownership.
The Georgist tradition treats the factor of production of land, broadly conceived to cover all un-
improved natural assets, as prima facie collectively owned (Kerr, 2017). An important strand of the
pre-distributivist tradition incorporates this Georgist idea. In one version of the proposal, the use of
293
Alan Thomas
all unimproved land ought to be the subject of an auction to realise a social dividend for all citizens
(Posner and Weyl, 2018). More specific versions of the proposal focus on non-renewable resources
such as the Alaska Permanent Fund that distributes a social dividend to all citizens of Alaska that is
funded by the exploitation of oil. TheNorwegian Government’s Pension Fund uses a similar model.
Such sovereign wealth funds not only supply a citizen dividend from the exploitation of the initial
non-renewable resource, but they also make long-term investments in other forms of capital to
ensure that this revenue stream continues into the future.
Mill shared Henry George’s intuitions, arguing that English absentee owners of property in
Ireland should have their estates sequestered if they did not make productive improvements to them
for the benefit of their tenant farmers: there was, therefore, a public interest in this holding of private
capital in the form of land. In Rawls’s version of a property-owning democracy, there is no direct
engagement with Georgism. However, an interesting feature of Rawls’s later discussion of what it is
to accept the difference principle in Justice as Fairness: A Re-Statement asks the deliberating citizen to
imagine a fiction, namely, that everyone’s productive talents are held on trust for the benefit of
society as a whole and the worst off in particular (Rawls, 2001). Rawls’s first principle guarantees
that if anyone owns an individual’s talents it is that individual. It is not literally true that society owns
the talents of all as a collective trust. But in his thought experiment, Rawls asks us to entertain the
fiction that we do: this is a form of fictional Georgism where the pool of talents that is exercised in
producing the productive surplus of society is viewed as if it was a common pool resource on which
we have a (fictional) collective claim on a social dividend (Thomas, 2020b).
3 Alternative Strategies: Collateral and Monetary Policy

One respect in which a property-owning democracy shows its intellectual flexibility is if one asks the
question of why it is so important to give each citizen capital in the first place. It has been noted that,
at least in Rawls’s presentation, assigning all citizens a right to capital might be taken to strengthen
their democratic citizenship via strengthening their democratic “voice” or as putting in place a
precondition for fair equality of opportunity, or as one way to implement the difference principle
(Thomas, 2017). However, others working in this tradition, such as Robert Hockett, emphasise the
importance of one function of capital as collateral (Hockett, 2021). If the idea is that an initial stock
of capital can lead to enhanced productivity, then one way to implement this is by giving people an
inherently generative asset. Another form of implementation gives them access to credit. Credit is
not cash but a promise to pay cash in the future. On the basis of increased access to credit, citizens
can widen the range of productive assets available to them. The key point is that private allocators of
credit demand collateral as part of the management of risk. One way to implement an egalitarian
allocation of capital is not, directly, to allocate capital but rather to underwrite the risks in specific
markets taking advantage of the role of the state as the insurer of last resort.
Hockett notes that, in the history of the United States, there was a tradition of populist agitation
around the issue of access to credit as the state’s extensive farming constituency depended on access
to credit to manage the fluctuation in the productivity cycle of farm output (Hockett, 2021). The
United States has a tradition of asset- based egalitarianism, whether in the direct form of land grants,
or indirectly in granting access to specific markets insured by the Federal government. In this way,
access to basic goods such as housing or education can be broadened not by allocating capital, but by
extending the market in these goods via state insurance. In the United States, the importance in the
secondary market for mortgages of the Government Sponsored Enterprises Fannie Mae (Federal
National Mortgage Association) and Freddie Mac (Federal Home Loan Mortgage Corporation) as
state-backed guarantors of liquidity and stability are one form of pre-distributive egalitarianism
following Hockett’s model. While the idea of homesteading, or land grants, may have a Georgist
justification in terms of the allocation of prima facie collectively owned capital, the idea of selective
294
state underwriting of key markets has a pre-distributive justification in terms of the fair allocation of
access to productive assets via the idea of fair access to credit. For various complex reasons specific to
the political history of the United States, this tradition is more developed there than in other jur-
isdictions. Pre-distribution, then, can take the form of activist and explicitly egalitarian monetary
policy aimed at widening access to basic assets such as education and housing. The complaint that
Rawls’s ideal of a property-owning democracy has no historical precedent is not, then, entirely fair if
we interpret the idea of asset-based egalitarianism broadly. Conversely, one could note that in the
United Kingdom there is a historically significant example of the inegalitarian version of the ideal.
The successive governments of Margaret Thatcher adopted the policy of the sale of social housing to
occupants who had previously been restricted to renting such property. The result was a massive
transfer of wealth in the form of both land and housing assets from state ownership to the private
sector that has, over time, had significantly inegalitarian consequences – particularly when it comes
to inter-generational justice. It is an important component of the asset-based egalitarian tradition
that it focuses on the importance, both from the perspective of human well-being and of macro-
economic stability, of housing assets (Ryan-Collins, Lloyd, and Macfarlane, 2017).
4 An Alternative to Redistribution?
The greatest controversy attaches to the idea that, as a form of pre-distribution, a property-owning
democracy represents a distinct alternative to the re-distributive policies of welfare state capitalism.
Critics of a property-owning democracy argue that it cannot represent a wholly distinct alternative
to the welfare state-capitalist egalitarian ideal that is so strongly represented in the political traditions
of the Modern West (O’Neill, 2012; Schemmel, 2015). This objection is both theoretical and
rooted in history and praxis. We have successful paradigms of egalitarian society in the Modern
West, reproduced in various forms across the globe and the basic idea is that ex-post redistribution
via tax and transfer can realise the aims of a just society (Piketty, 2020). Moving from defence to
attack, the defender of orthodox welfare state provision levels three charges against Rawls in par-
ticular and the property-owning democrat in general.
The first is that Rawls is in bad company and that while he did offer a critique of the welfare state
in his work that was only because he allied himself with right-wing objections to the growth of
social dependency in welfare states. Second, Rawls offered an inconsistent standard; it has already
been noted that when it comes to the choice of a social system, only one item on Rawls’s list is
stipulated and not worked up from any historical examples – a property-owning democracy. It is not
fair, then, to judge welfare capitalism by its historical record where it has been the repeated subject
of political pressure, while nevertheless offering a track record of overall success (O’Neill, 2012;
Schemmel, 2015). Rawls attacks only welfare states that have worked badly while helping himself to
an untried ideal alternative as an object of comparison, a competition that a property-owning de-
mocracy seems guaranteed to win. Finally, and most concerning, Rawls is in fact the defender of
only half a conception of justice that gives all citizens basic liberties, and a conception of fair equality
of opportunity that represents a fair starting line conception of equality (Schemmel, 2015). Given his
commitment to markets, Rawls does not then, by adopting a property-owning democracy, offer
citizens any defence from brute market bad luck. Indeed, one could view his position as a
responsibility-sensitive version of egalitarianism or even as a form of market-based meritocracy.
Turning from theory to institutions, the defender of welfare state capitalism has further com-
plaints. In practice, it is difficult to see how the institutions of a property-owning democracy would
differ in any significant way from those of welfare state capitalism (O’Neill, 2012; Schemmel, 2015).
Institutionally, and in terms of policy, a property-owning democracy is a form of incremental ex-
pansion of welfare state capitalism to include some asset-based policies as well as progressive taxation
to redistribute income. Adding baby bonds or demogrants to welfare state capitalism remains within
295
Alan Thomas
the institutional and normative framework of welfare state capitalism and can hardly represent a
radical break from it.
The property-owning democrat has to develop a response to all of these points. The first response is
that “companions in guilt” arguments are generally unreliable and the grounds for Rawls’s critique of
welfare state capitalism are not those of the New Right of the 1980s worried about the growth of a
dependent underclass. A Rawlsian concern with what is now called social exclusion does not set him
apart from other egalitarians (Freeman, 2013, 2020; Thomas, 2017). Rawls holds a narrow conception
of reciprocal justice based on productive exchange and is primarily concerned with self-sufficient
citizens who want guaranteed employment, meaningful work, and work that reinforces and does not
undermine the basis of individual self-respect (Freeman, 2018a). The property-owning democrat
concedes that welfare state capitalism and a property-owning democracy must be held to the same
normative standards. But there is reason to believe that even egalitarian welfare state-capitalist societies
that do well under the heading of income inequality are complacent when it comes to inequality of
wealth. Inequality of wealth ties into other aspects of social structure such as unequal access to elite
schooling and, more generally, the reproduction of social elites (Piketty, 2020).
The idea that Rawls defends only a “starting gate” conception of fairness because he is committed
to markets would seem to make the very mistake about property-owning democracies that Kristina
Meshelski cautioned against (above). The property-owning democrat does not push individuals out
onto the market and let social outcomes pan out as they may. There is certainly no need to tie an
agent’s capital endowment to any responsibility condition‚ we do not do that, for example, for
young people who currently inherit money from their trust fund (Ackerman and Alstot, 1999).
Property-owning democrats have, in the past, tended to tie capital endowments to specific uses such
as funding education, or business start-ups as opposed to cash grants. But that is not from any
misplaced moralism about personal responsibility, but because the rationale for these capital grants is
to improve the individual’s productive capacity for societal benefit as well as their own. As Thomas
Scanlon has pointed out, the lack of access to “start-up” funds is a major source of inequality
between the children of the affluent and the poor in our own highly unequal societies (Scanlon,
2018, pp.54–5). We know Rawls’s attitude to returns on capital based on investment in innovation:
as Paul Smith points out “as with other gambles, such profits (and losses) reflect option luck, un-
objectionable to egalitarianism (given fair equality of opportunity)” (Smith, 1998, p.229, n.35).
More generally, the property-owning democrat gives citizens access to assets that are, in Hockett’s
term, “generative” – investments that produce further wealth as opposed, for example, to giving
citizens a guaranteed income stream (as in unconditional basic income proposals) (Thomas, 2020b).
It has also been noted in this entry, in connection with egalitarian policy as applied to monetary
issues, that the kind of asset-focused egalitarianism represented by a property-owning democracy is
not as unfamiliar to the political traditions of the USA as it is elsewhere: it is not strictly true that
policies of this kind have no grounding in history. Certainly, in particular jurisdictions, a form of
capitalism has been created which seeks above all to protect the interests of the rentier-investor class
above other social interests with price stability and returns on assets as top policy priorities
(Christophers, 2020). It has already been noted that this inegalitarian form of rentier capitalism has
had a significant impact on inequality in the United Kingdom. From the transfer of social housing to
the private sector, to the use of North Sea oil revenue to fund rentier capitalism, the inegalitarian
version of a property-owning democracy is familiar to citizens of the UK. Social coalitions that can
be built up in support of generous welfare state provision can, equally well, be pressurised and
disintegrate under counter-pressures. This issue can be resolved only by case-by-case examination of
the political sociology of particular societies, but a property-owning democracy is intended to be a
radical break from these familiar forms of welfare state capitalism. The latter have co-existed, un-
happily, with stagnant wages for the majority of those in work, downward mobility, an increase in
precarity and social and political polarisation even in paradigms of social democratic welfare states.
296
Here the defender of property-owning democracy argues that superficial similarities between a
property-owning democracy and welfare state capitalism are misleading: they share no institutions or
rationales in common (Thomas, 2020a). The property-owning democrat can point to the in-
tellectual inheritance of mid-20th century Keynesianism as the “road not travelled” in the evolution
of recent egalitarianism, contesting the view that the 1970s saw the failure of something called
“Keynesianism” (Fazi and Mitchell, 2017). This was a milieu on both sides of the Atlantic that
influenced Rawls. Keynes, and Neo-Keynesianism, share Rawls’s statist focus, and the egalitarian
form of a property-owning democracy is more strongly represented in economics than it is in
political philosophy. If the immediate influence on Rawls was the work of James Meade, that work
takes its place in a lineage that begins with Meade’s supervisor, Joan Robinson, and is continued into
the present in the work of economists such as Anthony Atkinson and Thomas Piketty (Atkinson,
2015; Piketty, 2014, 2020). While the ideas surrounding monetary policy in this tradition remain
controversial, it is undoubtedly true that one strand of Neo-Keynesianism focuses on “overt
monetary financing” and the idea that the state ought to pursue valuable social goals without being
concerned about public deficits (Turner, 2015). Public deficits and private surpluses sum to zero:
given the centrality of capital to a property-owning democracy, proponents of the view will seek to
democratize the way in which monetary policy is used to pursue egalitarian goals. If it is a truism that
a monetarily sovereign state cannot go bankrupt in its own currency and is not under the fiscal
constraints of a private household, then the question becomes the extent to which monetary policy
can develop a society’s productive capacity without generating inflation. That certainly seemed to
Keynes to require a higher degree of democratic control of that productive capacity than under
actually existing capitalism and, in particular, the control of large-scale investment, a case he made in
a government report of 1928 that Keynes largely authored (Britain’s Industrial Future, Being the Report
of the Liberal Industrial Inquiry of 1928, 1928, Book II; Crotty, 2019). All of this points to the pre-
distributive conclusion that, if we get our policies right, an egalitarian society will have a greatly
reduced need for “tax and transfer” in markets pre-distributively structured so as to be fair. This
emphasis on making capital abundant reinforces the point that the pre-distributive goal is not the
construction of an institutional scheme that protects the sectional interests of the petit-rentier and
investor classes, as the non-egalitarian versions of a property-owning democracy developed by
Reagan and Thatcher openly sought to do (Christophers, 2020).
However, even if the primary rationale for this central policy goal is to enhance the ratio of
productive to unproductive uses of capital, it does not follow that the property-owning democrat
justifies this social system because of an intrinsic connection to continual economic growth. On the
contrary, Rawls explicitly envisages his two forms of non-capitalist social system, a property-owning
democracy or liberal market socialism, as compatible with the realisation of Mill’s ideal of a sta-
tionary state (Rawls, 2001).
5 Conclusion
The case for a property-owning democracy emerges from the confluence of four streams of thought:
the republican emphasis on freedom as the absence of domination, a liberal conception of justice as
reciprocal fairness, the economic theories of Keynes and his successors, and the monumental work
on the history and sociology of “regimes of inequality” in the recent work of economist Thomas
Piketty (whose ideas show a considerable affinity with those of Rawls). It is an ideal that seeks to
realise the ideal of justice in a stable way, counterfactually protected against emergent sectional
interests, to implement a non-capitalist system that nevertheless has private property and free
markets, where capital is abundant and the full range of human productive capacity is realised in a
social order that is a fitting home for the expression of our moral powers. It is, however, a radical
break with existing egalitarian paradigms and has been implicated, in its inegalitarian variants, with
297
Alan Thomas
the “turbo capitalism” implemented in the affluent West in recent decades (Luttwak, 1999). As it
struggles to break free of this history and to define itself as a genuinely novel option in progressive
thinking, we can expect it to remain an essentially contested ideal.
References
Ackerman, Bruce, and Anne Alstot. (1999) The Stakeholder Society. Yale University Press.
Atkinson, Anthony. (2015) Inequality: What Can Be Done? Harvard University Press.
Britain’s Industrial Future, Being the Report of the Liberal Industrial Inquiry of 1928. (1928) (Ernest Benn
Limited).
Buchanan, James. (1986) “Rules for a Fair Game: Contractarian Notes on Distributive Justice,” in Liberty,
Market and State: Political Economy in the 1980s. New York University Press.
Christophers, Brett. (2020) Rentier Capitalism: Who Owns the Economy, and Who Pays for It?. Verso.
Crotty, James. (2019) Keynes Against Capitalism: His Economic Case for Liberal Socialism. Routledge.
Dagger, Richard. (1997) Civic Virtues: Rights, Citizenship and Republican Liberalism. Oxford University Press.
Dagger, Richard. (2006) “Neo-republicanism and the Civic Economy,” Politics, Philosophy & Economics, 5,
151–173.
DiQuattro, Arthur. (1983) “Rawls and Left Criticism,” Political Theory, 11, 53–78.
Dowding, Keith. (2013) “Review of O’Neill and Williamson (eds) Property-Owning Democracy: Rawls and
Beyond,” Perspectives on Politics, 11.
Edmundson, William A. (2017) John Rawls: Reticent Socialist. Cambridge University Press.
Fazi, Thomas, and William Mitchell. (2017) Reclaiming the State: A Progressive Vision of Sovereignty for a Post-
Neoliberal World. Pluto Press.
Freeman, Samuel. (2013) “Property-Owning Democracy and the Difference Principle,” Analyse & Kritik, 35, 9–36.
Freeman, Samuel. (2018a) Liberalism and Distributive Justice. Oxford University Press.
Freeman, Samuel. (2018b) “Property Owning Democracy and the Difference Principle,” in Liberalism and
Distributive Justice. Oxford University Press.
Freeman, Samuel. (2020) “Replies to Critics,” Philosophy and Public Issues, 10, 173–345.
Gallie, W. B. (1955–6) “Essentially Contested Concepts,” Proceedings of the Aristotelian Society, 56, 167–198.
Hayek, Friedrich A von. (1982) Law, Legislation and Liberty: A New Statement of the Liberal Principles of Justic and
Political Economy. Routledge and Kegan Paul.
Hockett, Robert. (2021) A Republic of Owners. Yale University Press.
Jackson, Ben. (2005) “Revisionism Reconsidered: Property Owning Democracy and Egalitarian Strategy in
Post-War Britain,” Twentieth Century British History, 16, 416–440.
Jackson, Ben. (2012) “Property Owning Democracy: A Short History,” in Martin O’Neil and Thad
Williamson (eds.), Property-Owning Democracy: Rawls and Beyond. Wiley Blackwell.
Kaldor, Nicholas. (1956) An Expenditure Tax. George Allen and Unwin.
Kalecki, Michael. (1943) “Political Aspects of Full Employment,” Political Quarterly, 14, 322–331.
Kerr, Gavin. (2017) The Property-Owning Democracy. Routledge.
Krouse, Richard, and Michael MacPherson. (1986) “A ‘Mixed’–Property Regime: Equality and Liberty in a
Market Economy,” Ethics, 97, 119–138.
Krouse, Richard, and Michael McPherson. (1988) “Capitalism, ‘Property-Owning Democracy’ and the
Welfare State,” in Amy Gutmann (ed.), Democracy and the Welfare State. Princeton University Press.
Luttwak, Edward N. (1999) Turbo-Capitalism: Winners and Losers in the Global Economy. Harper Collins.
Meade, James. (1964) Efficiency, Equality and the Ownership of Property. Allen & Unwin.
Meade, James. (1993) Liberty, Equality, Efficiency. Macmillan.
Meshelski, Kristina. (2019) “Rawls’s Socialism and Pure Procedural Justice,” Ethical Perspectives, 26, 343–347.
Moriarty, Jeffrey. (2009) “Rawls, Self-Respect, and the Opportunity for Meaningful Work’,” Social Theory and
Practice, 35, 441–459.
Nozick, Robert. (1974) Anarchy, State and Utopia. Basic Books.
O’Neill, Martin. (2012) “Free and Fair Markets Without Capitalism: Political Values, Principles of Justice, and
Property-Owning Democacy,” in Martin O’Neill and Thad Williamson (eds.), Property-Owning Democracy:
Rawls and Beyond. Wiley-Blackwell.
O’Neill, Martin, and Thad Williamson (eds.). (2012) Property-Owning Democracy: Rawls and Beyond. Blackwell-
Wiley.
Paine, Thomas. (1797/2000) “Agrarian Justice,” in Bruce Kuklick (ed.), Thomas Paine: Political Writings. Cambridge
University Press.
298
Parijs, Philippe Van. (1995) Real Freedom for All: What (If Anything) Can Justify Capitalism?. Oxford University
Press.
Piketty, Thomas. (2014) Capital in the 21st Century. Harvard University Press.
Piketty, Thomas. (2020) Capital and Ideology. Harvard University Press.
Posner, Eric, and Glen Weyl. (2018) Radical Markets: Uprooting Capitalism and Democracy for a Just Society.
Rawls, John. (1971/1999) A Theory of Justice. Belknap Press.
Rawls, John. (2001) Justice as Fairness: A Restatement. Harvard University Press.
Rawls, John. (2007) Lectures on the History of Political Philosophy. Harvard University Press.
Ron, Amit. (2008) “Visions of Democracy in ‘Property-Owning Democracy’: Skelton to Rawls and Beyond,”
History of Political Thought, 29, 89–108.
Ryan-Collins, Josh, Toby Lloyd, and Laurie Macfarlane. (2017) Rethinking the Economics of Land and Housing.
Zed Books.
Scanlon, Thomas. (2018) Why Does Inequality Matter?. Oxford University Press.
Schemmel, Christian. (2015) “How Not to Criticise the Welfare State,” Journal of Applied Philosophy, 32,
393–409.
Smith, Paul. (1998) “Incentives and Justice: G. A. Cohen’s Egalitarian Critique of Rawls,” Social Theory and
Practice, 24, 205–235.
Thomas, Alan. (2017) Republic of Equals: Pre-distribution and Property-Owning Democracy. Oxford University Press.
Thomas, Alan. (2020a) “Freeman On Property-Owning Democracy,” Philosophy and Public Issues, 10, 45–68.
Thomas, Alan. (2020b) “Full Employment, Unconditional Basic Income and the Keynesian Critique of Rentier
Capitalism,” Basic Income Studies, 15: online first.
Thomas, Alan. (2020c) “Rawls On Economic Liberty and the Choice of Systems of Social Cooperation,” in John
Mandle and Sarah Roberts-Cady (eds.), John Rawls: Debating the Major Questions. Oxford University Press.
Turner, Adair. (2015) “The Case for Monetary Finance: An Essentially Political Issue,” In 16th Jacques Polak
Annual Research Conference, IMF, Washington, DC.
Weithman, Paul. (2013) “Review of O’Neill and Williamson (eds) Property-Owning Democracy: Rawls and
Beyond,” Notre Dame Philosophical Reviews. https://ndpr.nd.edu/reviews/property‐owning‐democracy‐rawls‐
and‐beyond/.
299
24
SOCIAL DEMOCRACY
Jeppe von Platz
1 Introduction
The term “Social Democracy” is most often used to describe the political and economic structure of
countries like those in Scandinavia. These societies have strong protection of worker rights and
government regulation of a capitalist economy. Accordingly, social democratic economic systems
are understood to have the following properties (e.g., Vaut et al., 2011):
• Regulated capitalism: the means of production are (mostly) privately owned, but democratic
controls on this ownership mitigate the accumulation of capital in the hands of the few and
prevent inequalities of wealth from transitioning into political inequality.
• Interventionist economic policy: the state actively seeks to stimulate consumption, increase
employment, and moderate swings in the economy (e.g., stimulus-spending in slow-growth
periods, raising interest rates in high-growth periods).
• Work as the main source of income: full-time employment provides adequate means for a
decent existence; the state works to secure full employment and equal compensation for equal
work.
• Compressed economic inequality: inequalities of wealth and income are accepted, but
compressed by the system of laws and policies (e.g., progressive taxation, empowerment of
unions, free education and training, welfare guarantees untethered from employment history).
• Regulated bargaining: the surplus-value of production is divided through an ongoing and
well-regulated bargaining process that ensures roughly equal bargaining advantage between
capital and labor (or public employer and employees).
• Market distribution: the flow of production inputs and outputs is determined by free-market
exchange. However, the state supports the provision of public goods and might produce and
distribute essential goods and services (such as infrastructure, education, and healthcare).
• Unconditional social security: all members of society are guaranteed the elements of a
decent existence (housing, food, healthcare, etc.) independently of their employment history.
To understand the attractions of this system, we need to understand the view of society and the
principles of justice that provide its normative foundations. Why, according to social democrats, is
this system superior to the alternatives?
300 DOI: 10.4324/9780367808983-29

Social Democracy
To answer this question, this essay offers an understanding of social democracy as a theory of
justice – as a member of the family that includes right-liberalism, left-liberalism, libertarianism, and
socialism.1 The members of this family are distinct at several levels of theory and practice: they offer
competing understandings of the origins and purpose of (economic) society, they issue competing
principles of economic justice, and, therefore, require (and permit and prohibit) different designs of
the basic economic structure of society. Social democracy starts from an idea of cooperation: the idea
of members of society working together to produce and distribute the various goods and services
they need to live well. This idea of working together sustains principles of justice that require
enabling all members to participate as equals in the production and distribution of goods and ser-
vices, democratic controls of the system of cooperation, and a fair distribution of both work and its
products. These principles are satisfied through the listed institutional commitments.
There are both theoretical and practical reasons why uncovering the normative architecture of
social democracy is worthwhile. By contrast with other theories of justice, social democracy has not
received much philosophical attention.2 This relative neglect is puzzling given the apparent at-
tractions of the social democratic model of society. The Scandinavian countries are peaceful and
stable, rank low on such measures as infant mortality, violent crimes, and corruption, and rank high
on measures such as happiness and freedom.3
Uncovering the normative architecture of social democracy might also have practical use. In
situations that require choices between different interests – say, full employment or economic
growth – it is useful to have clarity about what values and principles these interests express.
Moreover, in spite of the successes of social democracies, social-democratic political movements are
in disarray. Squeezed by nationalists, liberals, conservatives, and new-left progressives, social-
democratic leaders whack-a-mole their way through challenges and crises, apparently confused
about what they stand for.4
The following section offers some remarks on method. Section 3 sketches the history of social
democracy. Section 4 presents my understanding of social democracy as a theory of justice.
Sections 5 and 6 further define social democracy by contrasting it with two closely related political
ideologies, namely, democratic socialism and left-liberalism.
2 Remarks on Method
How can one argue for a particular interpretation of a theory of justice? This question is not
particular to social democracy, but it is harder. The usual approach – identifying a limited set of core
political philosophical texts and using these to articulate the general assumptions and commitments
of the theory in question – won’t work here. Where, say, liberalism has several hundred years of
political philosophy to draw from, social democracy has a limited philosophical tradition and there
are few texts that one could point to as foundational (Crosland, 1956: 52).5
One alternative would be to reverse engineer the theory of justice from the social democratic
configuration of the economic system. However, though there should be such a justificatory fit
between theory and contemporary practice, the theory must also be grounded in the history of the
social democratic parties and movements – otherwise, the construction could not claim to capture
the ideas of those who have carried the movement forward.
So, a good social democratic theory of justice justifies the institutional scheme sketched above, fits
with the reasons social democrats offer in favor of this institutional scheme, and explains how social
democracy came to embrace these principles and this scheme.
There are, then, four desiderata by which we can measure a purported claim to express the social
democratic theory of justice. First, the theory should be attractive in its own right – meaning that it
should proceed consistently from plausible assumptions to its various principles and commitments.
Second, it should defend principles by which the institutional scheme of social democratic
301
Jeppe von Platz
economies is superior to alternative configurations and, thereby, provide a principled basis for social
democratic self-understanding and clarity about how to navigate competing interests. Third, it
should reflect the actual texts and practices of social democrats. And, fourth, it should offer a part of
the explanation for the history of social democratic parties and movements. While this essay cannot
satisfy all four desiderata, I hope it illustrates how this could be done.
3 Historical Sketch
While the roots of social democracy reach deeper, social-democratic groups were at first umbrella
organizations for anti-establishment social movements and worker unions in the latter half of the
19th century.6 A wide range of movements shared these umbrellas, including syndicalists, anarchists,
Marxists, Lasalleans, trade unionists, and utopian socialists. The social-democratic coalition soon
coalesced on Marxism as a foundational ideology. From around the turn of the century and until the
end of World War I, social democratic parties became divided by questions of what ends to pursue
and the proper means for pursuing them (and what Marx and Engels said, would have said, or should
have said about these questions). This division led to a split where those closer to orthodox Marxism
and friendlier to the Russian revolution and Bolshevism were pushed out, leaving social democratic
parties that were more revisionist, reformist, and friendlier to parliamentary democracy.7 Social
democratic parties participated in the experiments in liberal democracy of the 1920s and 1930s,
with varying success. After World War II, social democracy shed most of the remaining vestiges of
Marxism, abandoning the commitment to socializing the productive system and embracing regu-
lated capitalism as an acceptable (even superior) economic order.
The history of social democracy can thus be understood as moving from orthodox Marxist
socialism into something else. Before the change, social democratic parties were organized around
the following beliefs and principles:
• Historical materialism and determinism: a society’s political, moral, and sociological

makeup is a function of its basic economic structures; social and political history is determined
by the underlying economic history.
• The primacy of economics (a corollary of historical materialism): the economic
structure of a society explains its politics and culture. Economic change is the basis of political
and cultural change.
• Class struggle: in a capitalist society, the interests of the working class and the capitalists are
opposed.
• Internationalism: solidarity is determined by class membership, not by nationality, religion, or
ethnicity. Workers are thus united across national and other boundaries.
• Periodic crises: the internal contradictions of capitalism create increasingly deeper and more
destructive economic crises.
• The inevitable immorality of capitalism: capitalism is unavoidably the site of exploitation,
alienation, misery, and unfair economic inequality.
• The desirability of economic socialism: only a socialist society will overcome the im-
moralities of capitalism. Socialism is achieved through and with the socialization of the means of
production.
• Rejection of bourgeois politics: working within the political institutions of bourgeois de-
mocracy is both foolish and counterproductive. It weakens class solidarity, normalizes capitalist
ideology, and undermines the potential for revolution.
• Revolution, not reform: the change from capitalism to socialism cannot be accomplished
gradually through reforms, but must come about through revolution.
302
Social Democracy
While these dogmas defined early social democracy, social democracy today is the result of the
process whereby all of them were rejected:
• Against historical materialism, turn to voluntarism: social democrats came to see that
historical materialism invites a passive waiting for things to change. Things did not change as
predicted. The revolution did not come. By contrast, real change was accomplished through un-
ionizing, political struggle, the creation of political parties, and the making of alliances with other
groups – including shopkeepers, small-scale capitalists, intellectuals, public employees, etc.
• Against the primacy of economics: it became clear that the economic system is as much a
function of the political system as vice versa.
• Against class struggle, turn to the common good: social democrats found both strategic
and philosophical reasons to care for the good of non-worker members of society – farmers,
small producers, academics, public servants. Eventually, social democrats aimed for the
common good, rather than the particular good of workers.
• Broadened and narrowed solidarity: solidarity pertains among the participants in social
cooperation, and thus is broadened to include members of other classes. At the same time,
solidarity narrows to those that participate in social cooperation, so international solidarity
among workers as a class is weakened.
• Crisis management: through sound fiscal policies, the tendency to crisis can be lessened.
• Embrace of capitalism: capitalism need not be unjust or immoral. In a well-regulated ca-
pitalist society, the sources of exploitation, alienation, misery, and economic unfairness are
controlled.
• Abandonment of economic socialism: eventually, social democrats abandoned the socialist
aim of overthrowing capitalism, finding that a well-regulated capitalist economy better serves
society as a whole.
• Acceptance of “bourgeois” politics: social democrats came to see that real progress can be
made through the workings of parliamentary democracy. Indeed, social democrats soon found
themselves supporting or even participating in governing coalitions.
• Reform, not revolution: change comes gradually and there is no need for a revolution. The
hope is to gradually make a democratic and capitalist society better for all its members, rather
than to do away with capitalism and democracy for the sake of emancipating the working class.
Even though social democracy can be described as a turning away from its Marxist origins, it remains
rooted in Marxist insights. Prominent among these is an understanding of economic agency and
its normative significance. On this understanding, economic agency should not be understood
merely as the pursuit of preference satisfaction through entrepreneurship, employment, investment,
and consumer choices. Economic agency, rather, is the powers exercised when we work together in
the creation and distribution of the products and services that we need.8 Normatively, all members
of society should be enabled to develop and exercise these powers, and to receive a fair share of the
benefits of production.
In sum, the history of social democracy can be viewed as the history of a metamorphosis from
Marxist socialism into a new ideology that is neither Marxist nor socialist but stays rooted in the idea
of economic agency as the powers engaged in productive work. But what did social democracy
become? What normative principles explain and justify the change?
4 Social Democracy as a Theory of Justice

While the history of social democracy goes some way towards explaining the social-democratic
institutional commitments, it leaves open the question of justification. On one understanding, social
303
Jeppe von Platz
democracy became utilitarian (or welfarist). Social democrats and right-liberals agree on the end
(maximizing overall welfare) but disagree about the means to this end. Social democrats embrace a
more Keynesian approach focused on growing the middle class and keeping economic inequality
constrained.9 On an alternative understanding, social democracy became value-pluralist; it affirms
the values of liberty, equality, and welfare, and then seeks to balance these through the social-
democratic system.10
These two understandings of social democracy, however, make social democracy vulnerable to
those criticisms that generally apply to utilitarianism and pluralism, and neither understanding seems to
fit or explain the actual history of social democracy – rooted in the experiences and challenges of the
working population and in the simultaneous rejection of liberalism and Marxist socialism. Even so,
while a single-principle utilitarian reading of social democracy seems indefensible, it is clearly true that
social democracy cares about liberty, equality, and welfare, and the history of social democracy has
been driven in part by interests in securing freedom from domination and exploitation, political
equality, equality of opportunity, an equitable distribution of wealth and income, and economic ef-
ficiency as it translates into shared material prosperity.11 Yet, instead of pluralism, we can understand
these values and struggles as part of a unified normative framework. In this section, I propose a third
understanding of social democracy that, I believe, is superior to the pluralist understanding in terms of
justificatory power and in terms of fitting and explaining the actual history of social democracy. This
understanding progresses from a social-democratic idea(l) of society through the principles of justice
that express this idea(l), to the social-democratic design of the economic system.
4.1 Working Together

Social democracy’s animating vision is of persons freely working together as equals and on fair terms
in the production and distribution of society’s goods. It is a society without privilege, unemploy-
ment, or poverty; where all political authority issues from a fully democratic process; and where all
able members contribute a fair share of work to the production and distribution of goods and
services, and each receives a fair share of the product in return.
Social democracy accepts the Hobbesian view that without society life would be “solitary, poor,
nasty, brutish, and short,” (Hobbes, 1996: 89) for we would have neither wealth, material comforts,
honest or reliable contracts, knowledge, education, culture, community, personal security, nor
medicine. But social democracy also accepts the stronger claim, defended in various ways by Hume,
Smith, and Rousseau, that in a well-designed society we can enjoy the many things that make
human lives go well by both subjective and objective standards: food and drink (and associated
pleasures), comfortable abodes, light in the dark, soft beds, the pursuit of knowledge, arts, sciences,
and technologies, freedom from violence and oppression, security of property and possessions, peace
and stability, education, and the joys of vibrant communal life.
Social democracy accepts this: we live together in political societies because by doing so we can
find access to the good stuff – material goods, knowledge, cultural enrichment, happiness, virtue,
and freedom. It is a short step from this fact to the norm that we should live together in a manner
that best provides these ends. Yet, by contrast with right-liberalism, social democracy denies that
justice is efficiency – the norms that best allow us to maximize social prosperity. Justice, rather, is the
norms that should regulate how we pursue these ends.
Varieties of right-liberalism from Hume and Smith through Hayek and Friedman take as
foundational the norm that we should live together in the manner that maximizes social prosperity.
In the right-liberal version, this norm is used to defend both conservatism and what Smith called
“the natural system of liberty” (Smith, 2008: 391). The goal of maximizing prosperity can be used to
defend conservatism if one accepts Hume’s and Hayek’s understanding of existing social institutions,
conventions, and norms as evolved solutions to the problem of how we can live together in peace
304
Social Democracy
and prosperity; we should be wary of radical or revolutionary change because it tends to destroy
more than it creates. The goal of maximizing prosperity can be used to justify the system of liberty if
the surest route to prosperity is to unleash the powers of liberty in a free-market, capitalist system
(Hayek, 1960; Friedman, 2002). For right-liberals, economic liberty and the wonders of the
spontaneous order processes that take place in a free market system – “the Creative Powers of a Free
Civilization” (Hayek, 1960: 22) – are the best and only sure route to a prosperous society. It is,
right-liberals will say, private property, free markets, and entrepreneurship that we have to thank for
the explosion of wealth and opportunities that the world has seen over the last few centuries.
Yet, social democrats reject the right-liberal moves from the norm of efficiency to conservatism
and the system of liberty. First, they deny the empirical claims that these moves rely on. Though
social democrats tend to agree that radical change is dangerous, they claim that existing systems of
institutions, conventions, and norms developed as tools of oppression as much as tools of prosperity,
so they carry limited prima facie validity.
Second, the system of natural liberty is prone to lead to disruptions, crises, poverty, exploitation,
and domination – and so the claim that it secures prosperity for all is dubious. And, more im-
portantly, where right-liberals (at least in theory) favor whatever system maximizes broadly-
construed prosperity, social democrats reject this simple consequentialism and argue that there are
constraints of justice on the process itself. These constraints of justice are not, as right-liberals tend to
maintain, tools that allow us to pursue prosperity (an indirect consequentialist theory of justice), but
are, rather, independent constraints on this pursuit. Social democrats tend to agree with right-liberals
about the creative powers of private property, free markets, and entrepreneurship, but they add that
we also must focus on the shared work that goes into the production and distribution of goods and
services. It is by the (burdensome) effort of individuals that all the goods and services of society are
produced and distributed. In other words, social democrats add the Marxist (and, indeed, Lockean)
insight that work is the source of most of the goods of society, even while they reject the Marxist
move from this insight to the labor theory of value.
While social democrats reject the labor theory of value, they nevertheless agree with the Marxist
attention to the contribution of work and workers. This contribution tends to be neglected in the
right-liberal understanding of the free-market system, with its focus on property rights, freedom of
contract, innovation, investments, risk-taking, and so on. However, contra Marxism, social de-
mocrats do not think that the working class is the privileged source of economic value or has a
special role in (democratic) politics. Instead, social democrats assert the equal position of workers in
both the economic and political spheres. The production and distribution of goods and services is
also the work of investors, bankers, bourgeois self-owners, farmers, academics, public administrators,
and service and industrial workers. There are no privileged political or economic subjects; instead,
there is (or ought to be) a division of labor in a cooperative system where all participants enjoy equal
status as cooperators and contribute what they can (or want to) in exchange for a fair share of the
product. Ideally, all freely work together as equals in the production and distribution of goods. In return,
all have an equal claim to carry only a fair share of the burden and receive in return a fair share of the product
(which is not the same as an equal burden or share).
Social democracy thus moves from the Marxist insight to a Kantian conclusion:12 from the
political and economic points of view, all members of society must be viewed as free and equal
cooperators working together in the production and distribution of the goods of society. Free, in that
they voluntarily participate in the cooperative process of production and distribution. Equal, in that
they have the same basic status as cooperators. And cooperators, since they participate in the pro-
duction and distribution of goods and services – they all, in various ways, work together and thus
have a claim to receive a fair share of the benefits of cooperation, and only carry a fair share of the
burdens (i.e., they do only their fair share of the work).
305
Jeppe von Platz
Social democracy thus rejects all consequentialist theories of justice. Instead, it tells us to max-
imize productive output by working together on terms that respect the constraints of freedom,
equality, and fairness – constraints that are not derived from the ends of society, but instead define
how we should pursue these ends. (In this sense, social democracy affirms the priority of the right
over the good.13)
Working from this social democratic vision of society, we can articulate the social democratic
principles of justice that define the system of norms governing how we should pursue the goods of
society. It is helpful to divide these principles into two. The first principle enables all to work
together as free and equal citizens. The second principle requires a fair distribution of the work and
its product.
4.2 First Principle: Enabling Social Cooperation

Social democracy affirms the equal status of all citizens as free and equal cooperators. The first
principle, accordingly, requires that all citizens enjoy the scheme of rights adequate to this status:14
• Civil rights. These include freedoms of thought and religion, security from violence and
coercion, freedom of movement, and free choice of occupation.
• Democratic rights. These include rights to form, join, and leave political parties, to vote in free
and fair elections, freedom of association, and freedom of speech and assembly.
• Economic rights. These include access to meaningful employment, to form and join unions, to
education and training, and to basic healthcare.
• Social security guarantees. These include unconditional access to adequate food and drink,
housing, emergency services, elementary education, basic health-care, and the protections and
guarantees of the rule of law.
These rights are justified as necessary to secure the status of all members of society as free and equal
cooperators.15 Citizens are free: they are able to make up their own minds about the good life and
pursue it (as long as they respect the rights of others), and they are not forced to contribute to the
production and distribution of goods but can do so for their own rational advantage (however
conceived). Citizens are cooperators: they are part of the system whereby the social goods are pro-
duced and distributed. And they are equal: first, they equally enjoy the liberty to make up their own
minds; second, they have equal access to political power through the processes of democratic
politics; and, third, they enjoy equal basic economic rights.
4.3 Second Principle: A Fair Distribution of Work and Its Product

While social democrats accept that there are economic classes in society – including capitalists and
workers – they require that at birth the prospects for membership in these classes should be roughly
equal. In other words, social democracy affirms a strong principle of equality of opportunity,
whereby it is unjust if career and economic prospects are a function of the social class, connections,
or material condition of one’s parents. Such inequalities would lead to an unfair distribution of the
work and of the income and wealth that can be earned through work (and so of both the burdens
and benefits of cooperation).
Equality of opportunity can be defined in both weak and strong forms. In its weak form, it says
merely that the effects of the social lottery should be mitigated as far as possible while respecting
basic rights (e.g., how much time a family invests in the education of children). Therefore, we
should seek an economy where the distribution of economic advantages is a function of natural
talents, efforts, choices, and luck. The strong form requires that we mitigate the effects of both the
306
Social Democracy
social and natural lotteries as far as possible while respecting basic rights.16 In this form, economic
success is a function of efforts, choices, and luck, but not of natural talents.17
The social-democratic vision of society does not by itself determine the choice between these
two principles of equal opportunity. It is clear that class membership should not reliably predict
economic opportunity – those born into working-class families should enjoy the same employment
prospects as those born into the capitalist class – but if some persons naturally possess more mar-
ketable talents, it is unclear that this presents a problem of justice for social democrats. The concern
for efficiency seems to favor the weaker principle of equality of opportunity, insofar as it motivates
families to invest in developing the talents of their children and the employment of those talents
where they are most valued.
No matter which interpretation of equality of opportunity social democrats should choose, the
result of differential efforts, choices, and luck will create inequalities of wealth and income. These
inequalities are permissible by the social democratic theory of justice as long as they are consistent
with maintaining equality of opportunity and the three equalities of status (personal, political,
economic) outlined above. However, inequalities of wealth and income that threaten equality of
opportunity or one of the equalities of status should be mitigated.
Historically, social democrats also employed a stronger principle of distributive fairness in their
critique of capitalism, namely, the principle that contributors to production should receive a share of
the product proportional to their contribution.18 This principle has ties to the Marxist analysis of
capitalist exploitation but does not rely on this theory to sustain a critique of capitalism if capitalists
are able to extract a disproportionately large share of the surplus value of production. While a
principle of distribution according to contribution is intuitively appealing, it is very hard to define
the respective contributions of capitalists and workers (and public employees and academics, and so
on), and the principle has, perhaps for this reason, played a negligible role in post-World War II
social democratic arguments.
Instead, social democracy now embraces a proceduralist view of distributive fairness.19 The
distribution of the product (i.e., of income and wealth) is just if it results from a bargaining process –
a struggle among capitalists, laborers, and so on – wherein the struggling parties engage on a footing
of equality, so that neither party can use an unfair bargaining advantage to squeeze out a larger share
of the product. Historically, capitalists had the upper hand in wage negotiations, but various social
democratic policies that empower labor counter this tendency – for example, laws that enable and
protect unionization, regularly scheduled mandatory negotiations, laws governing work-time and
working conditions, social security that is decoupled from employment, and so on.20 If the
background structure of negotiations secures a fair bargaining process, the outcome of the bargain –
the division of the product – can be seen as fair by all parties, independently of whether this outcome
conforms to some substantive principle of distributive fairness, such as the principle of distribution
according to contribution.21
5 Social Democracy vs. Democratic Socialism

Social democracy is often defined as a watered-down or reformist (rather than revolutionary) sort of
socialism. In the charitable version of this definition, social democracy is defined as democratic
socialism, understood as the pursuit of socialist ends by democratic means and thus without violence,
revolution, or the overthrow of parliamentary democracy.22 Here is a clear distinction between
social democracy and democratic socialism. Where democratic socialism asserts a necessary con-
nection between democracy and socialism, social democrats sever this connection.
There are, I think, two distinct varieties of democratic socialism, distinguishable by the normative
connection they establish between the terms: democracy and socialism. The first variety sees
democracy as a necessary condition for socialism, for only a democratic political system will allow
307
Jeppe von Platz
the transition to, and continued upkeep of, a socialist economic system. A socialist economy is then
required for the other values, “the abolition of exploitation and oppression” and “freedom and bread
for all” (Kautsky, 1964). On this account, neither democracy nor socialism are attractive in and of
themselves, but as necessary conditions to overcome the injustices of capitalism and achieve justice,
well-being, and community.23
The other variety, by contrast, sees democracy as the end and then argues that a socialist economy
is necessary for the realization of democratic society.24 Democracy is here typically understood in
republican fashion, as political equality and non-domination, and the claim is that this ideal is
realizable only in a system with a socialist economy.
One could, of course, combine these varieties of democratic socialism and claim that (parlia-
mentary) democracy is necessary for socialism which is necessary for (republican) democracy.
Whichever of these varieties of democratic socialism we have in mind, social democracy rejects
one of the established necessary relations. Against the first, social democrats can allow that de-
mocracy is necessary for socialism,25 but social democrats deny that socialism is necessary for the
achievement of a society without exploitation and oppression, with freedom and bread for all.
Against the second, social democrats deny the affirmed necessary relation between socialism and
democracy. Social democrats make this argument by first defining the ideal of democratic society
and then showing that this ideal is realizable within a well-regulated capitalist economic system. Of
course, socialists might reject both the understanding of the ideal and the claim that the ideal can be
realized in a capitalist system; or they could argue that realizability is a matter of degree and that it is
more likely to score high on the relevant scales in a socialist system. The question of the relation
between democracy and socialism is far from settled. But the question does not have to be settled; it
suffices that the question is central and clear and that social democrats and democratic socialists
provide opposing answers to it.
So, democratic socialists maintain that there is a necessary connection between democracy
and socialism and social democrats deny it. A corollary of this distinction is that the principles
and empirical assumptions of democratic socialism together require a socialist economy; the principles
and empirical assumptions of social democracy permit, but do not require a socialist economy. This
corollary underwrites the general distinction: socialist theories of justice maintain that capitalism
(systems where the economy is organized on the basis of private ownership of the means of pro-
duction) is unjust and must be abandoned in favor of economic socialism (socialized ownership of
the means of production),26 whereas social democrats maintain that capitalism can be just. Social
democracy is not devoted to the private ownership of the means of production but has found its
peace with capitalism. Why? Because capitalism is more productive than socialism and, when well-
regulated, satisfies the principles of justice outlined above. Conversely, if it turns out that the so-
cialization of the means of production is more productive and consistent with the principles of
justice outlined above, then social democracy would favor economic socialism. But social de-
mocracy carries no principled commitment to economic socialism. Since such a principled com-
mitment is a defining feature of any socialist theory of justice, it follows that social democracy is not
socialism, and, a forteriori, also not democratic socialism.
6 Social Democracy vs. Left-Liberalism

We can ask how the theory of social democracy sketched above differs from left-liberal theories of
justice such as Rawls’s justice as fairness. Rawls starts with the idea of democracy as a system of social
cooperation between free and equal moral persons (Rawls, 2001: 39). He uses this idea to define and
defend two principles of justice. The first guarantees all citizens a scheme of basic rights adequate to
secure the development and exercise of the powers of democratic citizenship (Rawls, 2005: Lecture
VIII, 2001: 44–5, 111–5). The second requires fair equality of opportunity and that inequalities of
308
Social Democracy
income and wealth are permissible only if they are to the advantage of the least well off (Rawls,
2001: 42–4). While Rawls gives us little indication of the institutional commitments these principles
may carry, he clearly favors spreading ownership of the means of production, and satisfying the
difference principle likely requires some sort of redistributive (or predistributive) policies.27 So, how
does social democracy differ from (Rawlsian) left-liberalism?
If stated in the abstract, the first principles of social democracy and justice as fairness are similar:
both require that all members enjoy an equal and adequate scheme of basic rights, and both define
adequacy by reference to the development and exercise of the moral powers of democratic citi-
zenship. Yet social democracy parts ways with left-liberalism in its understanding of these moral
powers. Rawls identifies two moral powers of democratic citizenship: the capacity for a conception
of the good and the capacity for a sense of justice. The former is exercised in personal agency, when
citizens think about and pursue their vision of the good life. The latter is exercised in political
agency, when they think about and exercise their public role as citizens in debating collective action,
voting in elections, and so on. Social democrats agree that these are indeed moral powers of de-
mocratic citizenship, but add a first-principle concern for economic agency. From the standpoint of social
democracy, the development and exercise of the powers involved in participating in the productive
system are no less essential than the powers of personal and political agency.
Because of this emphasis on economic agency, social democracy offers a different interpretation
of the first principle and expands the scheme of basic rights protected by it. Left-liberals affirm a
combination of personal and political rights (in Constant’s distinction, both the liberties of the
moderns and those of the ancients), but social democrats believe this leaves out a crucial set of basic
rights, namely those that secure the institutional preconditions for developing and exercising eco-
nomic agency. Left-liberals view economic agency as of secondary importance – second to the
importance of personal and political agency – which is why the first principle protects personal and
political agency, whereas the exercise of economic agency is governed by the second principle (and
the first principle has priority over the second). Social democrats, by contrast, give economic agency
the same importance as personal and political agency, for it is of equal importance to the realization
of the social-democratic vision of society.28 So, social democrats add to the list of basic rights a right
to meaningful work and to the training, education, healthcare, and social security necessary for
developing and exercising this right.
Next, social democracy is not committed to the difference principle (the second part of Rawls’s
second principle). Indeed, social democracy seems indeterminate with respect to the choice of
principles for regulating inequalities of wealth and income. Social democracy requires that such
inequalities be consistent with maintaining the three sorts of status equality covered by the first
principle and with ongoing equality of opportunity, and these concerns will impose limits on in-
equality. But the social-democratic understanding of society does not require that inequalities must
be to the advantage of the least well off.
Finally, Rawls argues that no capitalist system can be just, not even the sort of well-regulated
capitalism defended by social democracy (Rawls, 2001: 135–8). Without going into the details of
the argument that leads him to reject such economic systems, it is clear that this rejection marks a
distinction between left-liberalism as conceived by Rawls and social democracy as presented
above.29 Here, however, it might be helpful to emphasize that social democracy is not welfare-state
capitalism as it is usually understood, where the idea is to mitigate the exploitation, misery, and
alienation of the poor by providing a guaranteed social minimum. Such mitigation might involve
measures that resemble the institutions of social democracy (e.g., unemployment supports, guar-
anteed access to housing, healthcare, and education), and could be justified simply by the re-
cognition that absolute poverty is unacceptable in societies that can afford sufficient welfare-
guarantees or by a utilitarian concern with the destructive effects of poverty. Social democracy, by
contrast, maintains that the above outlined institutional scheme is required to realize the principles of
309
Jeppe von Platz
justice implied by the social democratic vision of society. We should not be misled by superficial
institutional similarities: the institutions of a welfare-state aim to alleviate poverty and want; the
institutions of social democracy aim to enable all to engage in meaningful work as equals, and to
receive a fair share of its product in return.
7 Conclusion
I have interpreted social democracy as a theory of justice – an ideal of society and the system of
normative principles that express this ideal. The ideal is a society where all members freely work
together as equals, sharing the work and its products fairly between them. This ideal is expressed in
principles that justify and interpret the values of freedom, equality, fairness, and efficiency. These
principles, in turn, support the economic system outlined at the beginning of this essay.
Notes
1 Some commentators – including Samuel Freeman in this volume – use “classical liberalism” to refer to what
I call “right-liberalism” and “high-liberalism” to refer to what I call “left-liberalism.”
2 There are exceptions, e.g., Berman, 2006; Meyer, 2007.
3 See World Bank (n.d.), Nationmaster (n.d.), Transparency.org (2020), UN (2020), Figure 7, Vasquez and
Porcnik (2020).
4 See, e.g., Goodwin, 2017; Stafford and Sutcliffe-Braithwaite, 2016.
5 The theoretical situation hasn’t improved much since Crosland’s remark in 1956 that “no single constant
and consistent body of socialist doctrine exists.”
6 Every country and each social democratic party have its own story and the following simplifies what is
complex. For some general studies of the relevant histories, see Berman, 1998; Przeworski, 1986; G. Eley,
2002; Hicks, 1999. The German case is especially well studied, e.g., Moran, 1965; Schorske, 1955; Berlau,
1970. For the Swedish case, see Y. Hirdman, 1988. For the Danish case, see C. Bryld, 2002, 1976; N. F.
Christiansen et al., 2014; E. Olsen, 2005.
7 In Germany this split was violent, tragic, and likely helped Nazism, see Schorske, 1955; Berlau, 1970.
8 In the 1844 manuscripts Marx appears to define humanity as a “species being” (Gattungswesen) that engages
in self-actualization through productive activities (Marx, 1964: 112–4). It is not clear how we should
understand these passages, but social democracy is not committed to any essentialist (or other) theory of
human nature – it starts, rather, with the fact that we work together to create what we need (including our
social world).
9 E.g., Miller, 2019: 597–619. Rawls’s critique of welfare-state capitalism also assumes a utilitarian justifi-
cation, see Rawls, 2001: 135–40; Freeman, 2013: 9–36.
10 See, e.g., Crosland, 1956; Gombert, Krell, and Timpe, 2009; Carlson and Lindgren, 2007; Jackson, 2013.
11 See, e.g., Stauning, Danmarks Fremtid (Fremad, 1927); Socialdemokratiet, Fremtidens Danmark:
Socialdemokratiets Politik (Socialdemokratiet, 1945); SPD, Godesberger Programm (SPD, 1959).
12 This move from Marx to Kant was famously announced by Bernstein in his 1899 Voraussetzungen des
Sozialismus.
13 On this priority, see Rawls, 1988: 251–76.
14 For argument and elaboration, see Platz, 2016: 288–308.
15 For an elaboration of the link between moral powers and basic rights see Platz, 2014: 23–44, 2016:
288–308.
16 For more on this distinction, see Samuel Freeman’s chapter on “High Liberalism,” pp. 341–352.
17 If justice is about merit, even stronger principles of equal opportunity are generated by the luck-factor, for
then we might be required to minimize the effects of luck. I don’t see social democracy embracing this sort
of luck-egalitarianism.
18 For example, Borgbjerg in “Et Fremblik” (1915) states “the first law of the social democratic constitution”
as follows: “a right for all to work […] and a right for all who work to the full product of their work”
(Printed in Bryld, 1976: 192–97, at p. 197; my translation).
19 E.g., Bryld, 1976: 87, 112–3.
20 On the importance of the distinction between empowerment of labor and the decoupling of social security
from employment, see O’Neill, 2020: 63–91.
310
Social Democracy
21 This paragraph hints at an argument I develop in “The Principle of Merit and the Capital-Labor Split,”
Economics and Philosophy, forthcoming.
22 E.g., Harrington, 2011; Meyer, 2007; Przeworski, 1986; Russell, 2019. I agree with Berman, 2006: 96–7,
200–1, who argues that social democracy is a new thing, not watered-down socialism.
23 “Socialism and democracy are therefore not distinguished by the one being the means and the other the
end. Both are means to the same end” (Kautsky, 1964: 5).
24 E.g., Harrington, 2011; Levine, 1984: 4, 5; Schweickart, 2002: 3, 5.
25 Note that this does not imply that socialism is necessary for democracy, see Kautsky, 1964: 7.
26 The requirement to socialize the means of production is standardly taken as definitive of socialism, see, e.g.
Samuel Arnold’s chapter in this volume and Gilabert and O’Neill, 2019.
27 See, e.g., O’Neill, 2012; Thomas, 2017.
28 Social democrats need not argue that Rawls’s vision of society is mistaken, but merely that he failed to see
the social-democratic implications of this vision, c.f. Platz, 2016. There are analogies between the social
democratic and market democratic critiques of left-liberalism, see Tomasi, 2012; Platz, 2020a.
29 I discuss Rawls’s argument in Platz, 2020b: 4–33.
References
Arnold, Samuel “Socialism”| Internet Encyclopedia of Philosophy’ (no date). Available at: https://iep.utm.
edu/socialis/ (Accessed: 4 February 2021).
Berlau, Abraham J. (1970) The German Social Democratic Party, 1914-1921. New York; Octagon Books.
Berman, Sheri. (2006) The Primacy of Politics: Social Democracy and the Making of Europe’s Twentieth Century.
Berman, Sheri. (1998) The Social Democratic Moment: Idea and Politics in the Making of Interwar Europe. Cambridge,
MA; Harvard University Press.
Borgbjerg, Frederik. (1915) “Et Fremblik” in Claus Bryld eds. Det Danske Socialdemokrati og revisionismen.
Copenhagen; GMT.
Bryld, Claus. (1976) Det Danske Socialdemokrati og Revisionismen. Koebenhavn; Selskabet til forskning i
arbejderbevægelsens historie.
Bryld, Claus. (1992) Den demokratiske socialismes gennembrudsår: Studier i udformningen af arbejderbevægelsens politiske
ideologi i Danmark 1884–1916 på den nationale og internationale baggrund. Koebenhavn; Selskabet til forskning i
arbejderbevægelsens historie.
Carlson and Lindgren. (2007) What is Social Democracy? A Book about ideas and Challenges. Arbetarrörelsens
Tankesmedja and the publishing house Idé och Tendens.
Christensen, Lars K., Kolstrup, Søren, and Hansen, Annette E. (2007) Arbejdernes historie i Danmark 1800–2000.
Koebenhavn; Selskabet til Forskning i Arbejderbevægelsens Historie.
Christiansen, Niels F. et al. (2014) Socialdemokratiske tænkere. Koebenhavn; Informations Forlag.
Crosland, Anthony. (1956) The Future of Socialism. London, UK; Jonathan Cape.
Eley, Geoff. (2002) Forging Democracy: The History of the Left in Europe. New York; Oxford University Press.
Freeman, Samuel. (2013) “Property-Owning Democracy and the Difference Principle,” Analyse & Kritik, 01/
2013, pp. 9–36.
Friedman, Milton. (2002) Capitalism and Freedom. Chicago; Chicago University Press.
Gilabert, Pablo and O’Neill, Martin. (2019) “Socialism”, The Stanford Encyclopedia of Philosophy (Fall 2019 Edition),
Edward N. Zalta (ed.), https://plato.stanford.edu/archives/fall2019/entries/socialism/.
Gombert, T., Krell C., and Timpe, M. (2009) Grundlagen der Sozialen Demokratie. Bonn; Akademie Für Soziale
Demokratie.
Goodwin, M. J. (2017) “Opinion | Old Labour, New Labour, No Labour,” The New York Times, 11
January. Available at: https://www.nytimes.com/2017/01/11/opinion/old-labour-new-labour-no-labour.
html (Accessed: 2 February 2021)
Harrington, Michael. (2011) Socialism: Past and Future. New York, NY; Arcade Publishing.
Hayek, Friedrich August von. (1960) The Constitution of Liberty. Chicago, Il.; The University of Chicago Press.
Hicks, A. (1999) Social Democracy and Welfare Capitalism. Ithaca, NY; Cornell University Press.
Hirdman, Yvonne. (1988) Vi bygger landet. Den svenska arbetarrörelsens historia från Per Götrek till Olof Palme.
Stokholm; Tidens Förlag.
311
Jeppe von Platz
Hobbes, Thomas (1996) Leviathan. Cambridge, UK; Cambridge University Press.

Jackson, B. (2013) “Social Democracy”. in Freeden and Stears eds. The Oxford Handbook of Political Ideologies.
Oxford; Oxford University Press.
Kautsky, Karl. (1892) Das Erfurter Programm. Stuttgart; Dietz.
Kautsky, Karl. (1964) The Dictatorship of the Proletariat. Ann Arbor; University of Michigan Press.
Levine, Andrew. (1984) Arguing for Socialism. London; Routledge & Kegan Paul
Lindgren, Anne-Marie & Carlsson, Ingvar. (2011) What is Social Democracy? A Book about ideas and Challenges.
Stokholm; Idéer och Tendens.
Marx, K. (1964) The Economic & Philosophical Manuscripts of 1844. Dirk J. Struik ed. New York, NY;
International Publishers.
Miller, Richard. (2019) “Social Democracy and Free Enterprise,” Journal of Social Philosophy, 50, 4, 597–619.
Meyer, Thomas. (2007) The Theory of Social Democracy. London; Polity.
Moran, R. (1965) The German Social Democrats and the First International 1864-1872. Cambridge, UK;
Cambridge University Press.
Nationmaster. (nd.) All countries compared for Crime > Violent crime > Murder rate per million people (no
date). Available at: http://www.nationmaster.com/country-info/stats/Crime/Violent-crime/Murder-rate-
per-million-people (Accessed: 3 February 2021)
O’Neill, Martin. (2020) “Power, Predistribution, and Social Justice,” Philosophy, 92, 63–91
O’Neill, Martin. (2012) “Free (and Fair) Markets without Capitalism: Political Values, Principles of Justice, and
Property-Owning Democracy”. in O’Neill and Williamson, Thad, eds., Property-Owning Democracy: Rawls
and Beyond. London; Wiley-Blackwell.
Olsen, Erling. (2005) Socialdemokratiets Historie. Koebenhavn; Arbejderbevaegelsens Forlag.
Platz, Jeppe von. (2016) “Social Cooperation and Basic Economic Rights,” Journal of Social Philosophy, 47, 3,
288–308.
Platz, Jeppe von. (2014) “Are Economic Liberties Basic Rights?”. Politics, Philosophy, and Economics, 13, 1,
23–44.
Platz, Jeppe von. (2021) “The Principle of Merit and the Capital-Labor Split,” Economics and Philosophy,
forthcoming
Platz, Jeppe von. (2020a) “Rawls’s Underestimation of the Importance of Economic Agency and Economic
Rights,” in Sarah Roberts-Cady & Jon Mandle eds. John Rawls: Debating the Major Questions, Oxford;
Platz, Jeppe von. (2020b) “Democratic Equality and the Justification of Welfare-State Capitalism,” Ethics: An
International Journal of Social, Political, and Legal Philosophy, 131, 1, 4–33.
Przeworski, A. (1986) Capitalism and Social Democracy. Cambridge, UK; Cambridge University Press.
Rawls, John. (2001) Justice as Fairness: A Restatement. Cambridge, MA; Bellknap Press.
Rawls, John. (2005) Political Liberalism: Expanded Edition. New York, NY; Columbia University Press.
Rawls, John. (1988) “The Priority of Right and Ideas of the Good”. Philosophy & Public Affairs, 17, 4, 251–276.
Russell Bertrand. (2019) German Social Democracy (Six lectures presented in 1896). London; Wentworth Press.
Schorske, Carl E. (1955) German Social Democracy, 1905–1917: The Development of the Great Schism. Cambridge,
MA: Harvard University Press.
Schweickart, David. (2002) After Capitalism. Rowan & Littlefield Publishers, Inc.
Smith, Adam. (2008) The Wealth of Nations. Oxford; Oxford University Press.
Socialdemokratiet. (1945) Fremtidens Danmark: Socialdemokratiets Politik. Koebenhavn; Socialdemokratisk
Forbund.
SPD. (1959) Godesberger Programm: Grundsatzprogramm der Sozialdemokratischen Partei Deutschlands. Available at:
https://www.spd.de/fileadmin/Dokumente/Beschluesse/Grundsatzprogramme/godesberger_programm.pdf.
Stafford, J. and Sutcliffe-Braithwaite, F. (2016) “Reorienting the Left,” Renewal: A Journal of Social Democracy
21, 1. Chadwell Heath: Lawrence & Wishart Ltd.
Stauning, Thorvald. (1923) Parlamentarisme eller Diktatur? Kan Parlamentslede Helbredes? Koebenhavn; Fremad.
Stauning, Thorvald. (1927) Damnarks Fremtid. Koebenhavn; Fremad.
Stauning, Thorvald. (1934) Danmark for Folket. Koebenhavn; Socialidemokratiet.
Thomas, Alan. (2017) Republic of Equals: Predistribution and Property-Owning Democracy. Oxford; Oxford
University Press.
Tomasi, John. (2012) Free Market Fairness. Princeton University Press.
Transparency.org. (2020) 2020 – CPI. Available at: https://www.transparency.org/en/cpi/2020 (Accessed:
3 February 2021)
312
Social Democracy
UN. (2020) UN World Happiness Report, 2019; https://s3.amazonaws.com/happiness-report/2019/WHR19.

pdf (Accessed 3 February 2021).
Vasquez and Porcnik. (2020) The Human Freedom Index 2020. 5. (Cato Institute) Available at: https://www.
cato.org/sites/cato.org/files/2020-12/human-freedom-index-2020.pdf
Vaut, S., et al. (2011) Wirtschaft und Soziale Demokratie. Bonn; Akademie Für Soziale Demokratie.
Worldbank. (nd.) Mortality rate, infant (per 1,000 live births). Available at: https://data.worldbank.org/indicator/
SP.DYN.IMRT.IN?most_recent_value_desc=false (Accessed: 3 February 2021)
313
25
CORRUPTION
Michael C. Munger
1 Introduction
Corruption is a process, or the result of a process, of debasement of the proper or normal state of affairs.
In modern parlance, corruption is generally the (mis)use of high office or public trust for private gain,
for oneself, for friends or family, or for a political faction.1 An older notion of “corruption,” once
central to the concept but less commonly invoked today, was the corruption of people, whether in
their roles as citizens, voters, or essential parts of society, generally by seductive, populist blandishments
of demagogues. Perhaps this older notion of “corruption” deserves a new measure of attention, but the
focus in this chapter is the misuse of public office or trust for private gain.
The difficulty lies in distinguishing between appropriate and inappropriate use of public institutions.
Though corrupt acts, debasing governance or community, are wrong in themselves, they also hasten
the arrival of a state of “corruption” for the system. Such acts can range from theft or embezzlement to
hiring practices distorted by nepotism. An important alternative is the selling of favors, by showing
favoritism in hiring or extorting bribes from prospective job-seekers. Closely related is the hold-up of
public services (permits, licenses, etc.) until an “additional fee” is offered by the permit-seeker. The
difficulty in the system is that there may not be a one-to-one correspondence between formal laws and
corruption: not all corrupt acts are illegal, and many illegal acts are not corrupt.
Miller (2018) notes that “corruption” can affect a person, eroding or perverting the character of an
individual; an institution, deforming a process or public office, or a person or persons occupying an
institution. The advantage of that approach is that corruption can be defined in terms of the dis-
figurement or debauching of an ideal type, particularly when that person or institution is essential to
the functioning of public life. Miller’s conception reflects his concern with making corruption a
philosophically well-defined category of phenomena, and so his treatment is much broader than the
standard public policy conception of corruption, which for the most part is the violation of laws or
public ethics. Because the place of Philosophy, Politics, and Economics (PPE) is between philosophy
and public policy on matters of breadth, my treatment focuses on illegal and unethical acts, usually
done covertly to escape attention, while allowing for the possibility that corruption is also the
debasement of norms, persons, or institutions in ways not cognized by anyone. In a system where
family or political connections matter more than competence, the foregone opportunities for
prosperity and growth are hard for anyone to envision, much less measure accurately.
Certain policies, such as the adoption of an expansive regulatory regime with limited state ca-
pacity to control the activities of bureaucrats and “inspectors” (Collier, 2002), are likely to foster
314 DOI: 10.4324/9780367808983-30

Corruption
systems of corruption. Other systems may be more resistant but can still suffer corruption because of
adverse circumstances or a failure to preserve institutional integrity. Much of the discussion here will
focus on the most obvious case of large-scale corruption, involving payments illegally made or
received to distort public policy or permits, but the claims here are also true about the smaller, and
more shadowy, forms of corruption as well.
The more important contribution of this chapter lies in emphasizing that corruption not only
debases people and institutions – though it does – but that, once entrenched, corruption is not
straightforwardly reversible. Dismantling debased and corrupted regulatory structures, or restoring
the integrity of breached norms, may simply be politically impossible. Public officials sought out the
coveted veto points in the corrupt system, and they come to see the “income” from bribes and
favors as something that is permanent and even deserved. As a result, what Collier (2002) called a
“big push” is required by the political system. But in systems that barely possess the resources to keep
political order as it is, a “big push” is out of the question. It is the durability, and adaptability, of
corrupt systems, which cling obdurately to power in the face of efforts at reform, that make eco-
nomic and political development so difficult.
1.1 Corruption Defined

The etymology of the word connects with “corruptus,”2 the irregular past participle of the verb
corrumpere, which literally means “to destroy, or spoil.” Figuratively, the Latin word meant to
“seduce or bribe,” in an uncouth manner. However, in a society where corruption has become
conventional (Vanderschraaf, 1998) the moral taint may be largely, or completely, removed.
Consider this example, a description of attitudes toward corruption in India:
In India, corruption is a form of patronage; a politician or bureaucrat who takes the bribe
then has to let it cascade among a series of lesser bureaucrats and elected officials, who will
make sure that the pet projects are completed. An honest politician has no goodies to toss
around … Many Indians have a sneaking suspicion – and there is anecdotal evidence to
back this up – that only corrupt politicians are effective ones. Would you really want an
honest man representing you in parliament? The neighbouring constituency may get that
new cricket stadium you were supposed to.
Honesty is more than just a matter of ineffectiveness; it seems, in the Indian context, like a
negative attitude to existence. There is no public project in India, however big or small,
whether in construction, healthcare, or education, that would stand up to a rigorous audit;
some numbers will never add up, because someone has pilfered money somewhere. You
have to tolerate a certain level of dishonesty if the wheels are to turn. An inflexible in-
sistence on honesty in Indian public life is, generally speaking, a form of moral myopia.
The new government hospital may have cost 20% more than it ought to have, but if its
opening is blocked by a public inquiry, which could drag on for years, thousands of poor
people will suffer. (Adiga, 2008)
This kind of dilemma, where corruption seems wrong but also seems to be the only way to get things
done, is what led Gordon Tullock (1996) to conclude that, at least in some cases, corrupt systems
“work out quite well.” This may be a minimal standard for “work out,” but resources tend to move
toward higher-valued uses in an efficiently corrupt system and things get done (Munger, 2017).
A conventional definition of corruption is “the sale by government officials of government property
for private gain” (e.g., Shleifer and Vishny, 1993; Lambsdorff, 2002). In countries with well-functioning
315
Michael C. Munger
institutions and substantial state capacity (Munger, 2019), such activity by government officials is not just
illegal but is frequently perceived by both citizens and the officials themselves as morally wrong.
India, and many systems with “bad” political and economic institutions, have what North,
Wallis, and Weingast (2009) called a “limited access order.” A transition to an open-access
order, where the state fosters permissionless innovation in a way that results in rapid economic
expansion, benefits the entire society in the sense that it creates a much larger total GDP and higher
average income. But the transition robs the rent-owners in the closed access order the value of their
(de facto, and sometimes even de jure) corrupt “right” to sell access. This “locks in” (Tullock, 1975;
Arthur, 1989) closed access orders, because those who benefit from corruption can’t trust those with
power in the alternative, institutionally superior regime to compensate the “losers” for their lost
rents. Worse, corrupt officials can also expect to lose power and prestige in the new rational system
based on production rather than control of veto points (Tsebelis, 2002; Mungiu-Pippidi, 2006).
But in a closed access order, with limited state capacity and lacking enforceable norms, cor-
ruption is the only way to get things done, at least in the short run. Nations that accept corruption
pay a terrible price, however, because the search for an ever-expanding source of profitable op-
portunities to withhold permission or delay services reduces the capacity of the economy to pay the
bribes in the first place, choking off the source of wealth that made corrupt practices viable. The
transition to open access orders, with rule of law and permissionless innovation (Thierer, 2016) is a
potential Pareto improvement, but only if the stakeholders in the current, violence-based closed
access order can be guaranteed a stream of rents at least equal to what they now enjoy.
But such a guarantee is usually difficult – and may be impossible – to offer. One reason is that
those outside the current ruling elite see the existing distribution of power and wealth as illegitimate,
even unjust. The incoming new ruling elite, or in the case of democratic transitions the voters who
support the new regime, are loath to offer compensation for the loss of rents that weren’t deserved in
the first place. An example in this regard is the payment of reparations, not to slaves but to slave-
owners, when slavery is abolished. (For a discussion of the U.S. case, see Ransom, 2001.)
Even if the new prospective rulers were willing to agree to such compensation, the hand-over of
power makes such an agreement unreliable: once the new regime is in place, the displaced rulers
have no reliable means of enforcing the agreement. So, while there exists a reform that could make
literally everyone better off – the open access order with rule of law – no politically feasible reform
path can lead to it. Nations with political corruption bump along at the bottom of the possible range
of growth patterns for extended periods. This explanation is important, but frustrating, because it
means the transition to “good” institutions requires either a highly contingent set of historical
circumstances, a violent civil war, or losing an external military conflict (Olson, 1982).
We are thus left with a two-part conception of corruption: in a static sense, the sale or other
private benefit to public officials of the public duties of such officials is morally wrong, because it
violates the public trust that is the foundation of the coercive power entrusted to that official. In a
dynamic sense, corruption is corrosive to the prosperity and sense of fairness of the society, because
many of the sources of wealth derive from denying to others the benefits of cooperation and vo-
luntary exchange. But the worst dynamic feature is that over time corruption is firmly locked-in,
entrenching pervasive poverty and scorn for government and officials.
2 Functional Corruption: Working Around Institutional Pathologies

We generally disapprove of corrupt acts. In fact, it appears that we have evolved brain architectures
that kindle angry reactions when we witness the violation of social norms (as has been noted by
scholars ranging from Smith, 1984, to Haidt, 2007). However, there may be several moral con-
siderations operating at once, and things that are objectively “corrupt” if no one does them may be
perfectly socially acceptable in a context where, as the saying goes, “everyone does it” (Dungan
316
Corruption
et al., 2014). That doesn’t mean that there is no such thing as morality, but it may imply that our
reaction to corruption is deeply contextual.
Further, humans have an evolutionarily adaptive facility for explaining their actions in terms that
may not only excuse bad actions but to transform morally repulsive actions into virtue (as has been
pointed out by scholars ranging from Hume, 2000 to Ariely, 2013). Merton (1938, 1949) famously
blamed corruption on the cultural corrosiveness of competition for status based on wealth.3
Individual actions can be controlled by social norms, but the constraint is fragile: if only a few people
transgress norms, shame and guilt can operate to police the violations. But if everyone begins to
perceive corruption as the way to “get ahead,” then the proscriptive norms collapse or are invoked
only in the most egregious cases.
An overarching social consideration is the effectiveness of institutions. North (1990) argued that
institutions reduce transaction costs of cooperation, and also that pathological institutions may
survive indefinitely in many societies. If there are widespread perceptions that (a) the government
and its officials raise obstacles to business and obtaining the necessities of everyday life, and (b) that
“everyone else is paying bribes,” the social opprobrium attached to everyday corruption may dis-
appear entirely. In such circumstances, Merton (1949) argues that corruption can be “functional,”
meaning that society may simply look the other way rather than sanctioning transgressions. What
makes the corruption conventional are the beliefs that bribes make the system work and that ev-
eryone else is doing it. These produce a set of expectations (Tullock, 1996; Vanderschraaf, 1998)
that are consistent with the behavior.4
One way to understand functional corruption is that the rules being transgressed suppress market
exchange; corruption allows value creation to jump the queue in the social hierarchy of sanctioned
activities. As Munger (2019) points out, this is actually the “Coase Theorem” at work. Recall that
Coase (1960) argued that economic efficiency drives the allocation of resources, unless what Coase
called “transaction costs” prevented the buying and selling of rights and services. On its face, it
would appear that a sclerotic, unresponsive regulatory and permitting bureaucracy would stifle
entrepreneurship and investment, locking the system down in inefficient patterns of resource use.
But corruption can actually solve this problem, or at least mitigate it. Suppose an entrepreneur
with a good idea for selling a useful service lives in a nation with a state-operated communications
bureaucracy. Service is intermittent, and installation, maintenance, or repair take months to arrange.
If the entrepreneur’s ideas were developed as a new industrial process, it would produce tens of
millions of sales, and quickly. If the system were inefficient and slow, but scrupulously honest, the
inefficiency would be inescapable because the state would lack the capacity to make formal infra-
structure work on a societal scale.
Coase’s insight was that participants can make side payments to buy and sell rights when the status
quo distribution of rights differs from the efficient distribution. But the status of side payments, in
this context, are straightforward: they are bribes! Our entrepreneur thinks creatively: the inefficiency
of the entire system is not a problem if an “additional fee” or “speed money” can be paid to get
access to immediate, high-quality service. Of course, immediate high-quality service is not available
to everyone, and that is corrupt. But a substantial bribe means that the telephone company, and the
fiber optic cable company, can send their best technicians to your new plant tomorrow, and your
business can open the day after.
With bribes, you can purchase priority installation, high-quality equipment, and arrange a
“contract” with service personnel – some of whom are extremely competent, because they know
how to patch together a functional connection with stolen or repurposed parts and switches. As
Mishra (2015) notes, corruption provides a vector for market efficiency incentives to be smuggled
into inert bureaucratic systems and “public” companies that normally are insulated from prices and
incentives. More simply, corruption creates a space where the Coase Theorem can function, and
resources will go to their highest valued use. The size of this “gray” or “black” market can be
317
Michael C. Munger
allowed to grow or shrink as the needs of the society change, creating wealth for many participants
and allowing citizens to obtain food, clothing, and other necessities in a way that forestalls actual
revolution while preserving the formal powers of the state apparatus.
Johnston (2005, 135–6) lists a set of conditions under which “functional corruption” will occur,
and he points out that – given that the regulatory system of inspections, permits, and fees is inert and
locked in – corruption provides a very useful way around the bottleneck. Further, Johnston notes
that the results can be satisfactory, in the sense that resources do actually move quickly to the
highest-valued uses. The problem is not one of immoral acts: “The basic pressures and tensions that
make corruption so advantageous and tempting are products not of the kinds of people to whom we
entrust public power, nor necessary of flawed institutions, but rather of governments’ basic re-
lationship to society” ( Johnston, 2005, 33).5
To summarize: corruption, when it exists, benefits some officials, and those who are able to pay
for faster or more competent services. In a system with “good” institutions corruption is nonetheless
a net harm, and violates social norms. These social norms may be sufficient to self-police corrupt
practices in many circumstances (Underkuffler, 2013).
However, in nations with “bad” institutions, corruption may be a rational response to dys-
functional or badly structured state capacity when there are substantial economic benefits to finding
a “work around.” That likely means that the systems of socially transgressive and furtive delivery of
benefits and protection from sanctions itself becomes formally institutionalized, or nearly. Both the
citizens who depend on cozy relations, and the government functionaries who depend on income,
are likely to resist efforts at reform.
Merton offers a rather breezy conclusion: “alternative structures may fulfill [the services now
delivered through functional corruption] without necessarily entailing its social dysfunctions.”
That’s true enough, but as North, Wallis, and Weingast (2009) show there is no feasible path that
connects the existing, corrupt limited access order with the kind of open access order that Merton
(apparently) has in mind. But the elites and citizens who live in the corrupt system are not ignorant
or simply confused about its problems. Having some development aid specialist come over and
lecture about “alternatives at the appropriate point” will just bring laughter. What can the nation
that depends on functional corruption actually do about it?
3 Limited Access Orders

In the “spaghetti western” movie, The Good, the Bad, and the Ugly, Clint Eastwood summarizes the
central problem of violence in state action: “There are two kinds of people in the world. The ones
with loaded guns, and the ones who dig. You: dig.”
Property and wealth are desirable to those who have loaded guns. Even if the loaded guns at first
belong only to “roving bandits” (Olson, 1993), two things become clear: (1) the economies of scale in
military technology, both because of tactics and knowledge of armor and weapons, are so sharp that
only the largest and strongest groups can protect themselves; all other groups must pay tribute or be
destroyed; and (2) the most effective way for “roving bandits” to grow larger militaries is to convert to
being “stationary bandits,” harvesting a surplus of wealth but leaving enough incentive for vassals to
remain under the protection of proto-state and produce the wealth on which the system depends.
The power of the state derives from its capacity to focus on violence (Weber, 1921).6 The dif-
ference between a government and a powerful, efficient stationary bandit is legitimacy. A group of
people was taken over and “protected” by a band of once roving but now stationary bandits are little
more than slaves. But a group of citizens who have constituted a system of property rights and
governance and have endowed it with a monopoly on the use of force are, in some important sense,
exercising their freedom. Even farmers who seem to be “enslaved” by the stationary bandit may be
much better off than if they were exposed to roving bandits. Despite the costs of supporting the
318
Corruption
military and keeping their king in silk clothes and eating rich food, the protection that the stationary
bandit/king offered allows farmers to devote all their time to raising food and livestock.
The problem the stationary bandit/king faced was largely one of transaction costs (North, 1990;
Munger, 2018): the protection that the newly formed “state” could offer allowed the creation of value,
because the population could specialize, but it was difficult to use the specialty of the state – violence –
to collect this new wealth without destroying it. The solution, discovered independently by many
different rulers in different places and at different times, was some version of the state-protected
monopoly. If there is some product (often, salt) that was essential and whose sources could be con-
trolled, the king could sell the sole right to produce and sell the product to some entrepreneur, who
was then delegated the task of, in effect, collecting taxes. But the collection of “taxes” simply took the
form of selling the product, protected from competition. The king could expect that the state-
sanctioned monopolist would pay (most) of the contracted payment because the threat of the king to
withhold the service protection from the competition was carried through simply by doing nothing.
In England, this practice took the form of “letters patent,” ultimately regularized by the “Statute
of Monopolies” in 1624 (Nachbar, 2005). Collection of taxes was expensive and required local
knowledge; the problem could be solved by giving citizens who had a lot to lose if the system failed
a franchise to collect taxes in that neighborhood.7 We are used to seeing this sort of “franchise” for
fast-food restaurants, where one can buy the right to use a brand name to attract business. The
difference in the case of tax collection is that the “franchisee” had to bid to obtain the right to collect
taxes, and then owed that lump sum to the central sovereign. Any surplus tax revenue the local
“king’s collector” could wring from the surrounding community was the collector’s to keep.
In some areas in the American colonies, this practice of local monopoly subcontracts was subject
to abuse and inflamed colonial sentiment against the Crown. In a fascinating turn of language, the
rebellion of the “Regulators” in North Carolina was an attempt to impose demands for regular-
ization and consistency in tax bills (Bassett, 1895; Troxler, 2017). The local authorities – knowing
that the British Crown cared only about receiving the promised total – had been particularly greedy
in plundering nearly anything they could find of value that could be used or sold. Almost any
“service,” from registering or transferring a deed to seeking payment of debt, was subject to arbitrary
and often capricious additional “fees.” Worse, in the case of a dispute, the outcome could be affected
by side payments, rendering the conduct of business dealings or just daily life precarious and un-
certain. The “Regulators” were private citizens who favored rule of law to govern the stationary
bandits who had purchased the right to plunder and steal with royal approval.
To break out of this kind of limited access order requires substantial state capacity. State capacity
requires the ability to raise revenue using some other mechanism, of course, and many developing
nations are permanently arrested in this particularly vicious form of Tullock’s (1975) transitional
gains trap. Controlling violence by granting political elites privileged control over substantial parts of
the economy – in effect, property rights to collect rents – “traps” the nation’s development by
making corruption the source of income for elites.
In contrast, open-access orders, which dominate the modern developed world, control the problem
of violence through open access and competition, a system often loosely referred to as “the rule of
law.” The problem is that well-intentioned attempts simply to import “good” institutions from open
access orders rarely have the intended effect. Cox, North, and Weingast put the problem this way:
Violent contests over political power have been surprisingly common throughout the
developing world, including the richest developing countries. This observation raises the
question of resistance to political reform in a more precise form: Why do developing
countries not adopt the institutional solution(s) to the problem of political violence that
developed states have adopted? (2019, 1)
319
Michael C. Munger
The answer is that a set of rules that impose and protect distributions of wealth and power within a
state that differ sharply from the de facto distribution on the ground will either be rejected in
advance or will be overturned by military action such as a coup or civil war. Cox, North, and
Weingast call this the “proportionality principle”: if an attempt to root out corruption, or otherwise
reform institutional rules and distribution of power, denies the current “owners” of valuable rents a
stream of rents with the same present value, then the current holders of de facto power will subvert
the reform. When international aid agencies give advice on development policy that threatens the
logic of distribution based on institutionalized local corruption, limited access orders resist, some-
times with violence.
An example will help clarify the problem of the limited access order and the transitional gains trap
in societies with corrupt governments. A building inspector in a large city is tasked with inspecting
building materials and practices, and issuing permits for construction and ultimately for occupancy.
In an open-access order, this activity would be animated by the application of principles of en-
gineering and training in safety regulations. In a limited access order, the state is very likely to use the
position as a source of revenue, since the threat of delay in construction or occupancy is expensive.
But the collection of “hold up” taxes is expensive, and it helps to have direct knowledge of the
reservation price of the builder. The state, lacking much capacity to control the actions of inspectors,
simply delegates the tax-collecting authority to the inspector, who pays a fee for the franchise. That
means that the salary of the inspector is negative, and may be substantially negative (inspectors in
several Indian cities have been estimated to pay their bosses millions of rupees to be able to hold
their jobs).
It is still possible to build new buildings, but the process is expensive and time-consuming. Further,
builders are likely to stint on the use of first-quality materials and techniques, since the “inspectors”
are relatively indifferent to technical problems as long as they receive the appropriate bribes. There are
occasional scandals, of course, when buildings collapse or need repair, but the system more or less
achieves an equilibrium, with a clear set of prices and payoffs.
Now, imagine that an outside reformer points out that a system based on rule of law, an open-
access order with competition, would be better for literally everyone. Having a clear, enforceable
law on building permits would both increase the quality and safety of buildings, and make the
process of constructing a building much faster and more transparent. You can imagine the reaction
of the current building inspector: “Oh, no you don’t! I paid for this job fair and square, and unless
you pay me for my ‘investment’, I’m going to sabotage the reform!”
Such fanatical support for corruption is not irrational; it’s not even especially greedy. The price of
jobs will rise until the payments from the inspectors match the opportunity cost rate of return on
investments. When reforms block the corrupt practice of accepting bribes, the current inspector
loses all of that investment, which may be his life’s savings. Many of the income-earning assets in
such systems are the result of Tullock’s (1980) “efficient rent-seeking”; the “revenue” they produce
are from selling permits and permissions of various forms, all of which make development more
expensive. The focus of this kind of “protection from state violence” is not growth, but is instead
parasitic on growth.
The problematic transition, then, lies in the ability of a state to move from a system where order
is imposed to a system where order emerges, as groups of citizens and other people outside the
nation order and reorder themselves into temporary but highly structured groups such as firms,
voluntary associations like non-profits, and other weakly connected voluntary communities. As has
now become part of the jargon, such systems of emergent order must facilitate permissionless in-
novation without extracting rents or blocking the development of the order by requiring permissions
in the first place.8 The problem, of course, is that the state’s capacity to exert violence gives it, or its
agents, the formal and informal power to exact payments in exchange for permission, permits, or
simply the withholding of threats of force.
320
Corruption
Violence jeopardizes those rents, so the individuals who receive rents have large incentives to
suppress organizations that might commit violence or foment social change. Rent “owners” have no
choice but to support the current regime.
Elites and rent-controllers will generally charge less – possibly much less – than market-clearing
prices to secure political goals and to satisfy prior promises of quid pro quo arrangements. The result
is that prices cannot carry out their function, which is the impersonal coordination of the behavior
of individuals.9 “Natural” states (as North, Wallis, and Weingast. call primitive and limited access
orders) cripple the ability of prices to coordinate impersonal activity, to enable permissionless in-
novation, and to convey information about the appropriate margins where cost, demand, and
scarcity indicate activity should be focused.
The problem, then, connects to a general argument about the efficiency of cooperation and
competition in different settings. Economists have often argued that competition is always beneficial
and that prices always “work” to allocate resources efficiently. But competition can be diverted from
the dynamic creation of products and services, resulting in effective creative destruction, and toward
rent-seeking competitions to protect inefficient economic organizations against creative destruction.10
If the form that competition takes is to seek the attention of powerful state actors in protecting artificial
rents against new entrants, and if the form that price signals take is to measure out the cost of obtaining
such protection, the actual functions of the markets are coopted and corrupted beyond recognition.11
What looks like competition is diverted into the particular means – bribes, favors, and the strategic
erection of toll booths near bottlenecks – that characterize limited access orders.
4 Conclusion
The traditional conception of corruption in the West is adapted by Thomas Aquinas from the Greek
idea of the debased nature of the physical world, and especially humans, compared to the ideal world
of forms. The corruption of “fallen” humans is part of our essence, but the striving for the ideal form
is still the animating for the good life of the individual and for the good society. Plato discusses the
corruption of even the best individuals, the philosophers, in Book VI of the Republic:
Consider the corruptions of the philosophic nature, why so many are spoiled and so few
escape spoiling – I am speaking of those who were said to be useless but not wicked – and,
when we have done with them, we will speak of the imitators of philosophy, what manner
of men are they who aspire after a profession which is above them and of which they are
unworthy, and then, by their manifold inconsistencies, bring upon philosophy, and upon
all philosophers, that universal reprobation of which we speak …
Everyone will admit that a nature having in perfection all the qualities which we required in a
philosopher, is a rare plant which is seldom seen among men …. And what numberless and
powerful causes tend to destroy these rare natures! [First] there are their own virtues, their
courage, temperance, and the rest of them, every one of which praiseworthy qualities (and this
is a most singular circumstance) destroys and distracts from philosophy the soul which is the
possessor of them. [Second] there are all the ordinary goods of life – beauty, wealth, strength,
rank, and great connections in the State – these also have a corrupting and distracting effect.
Modern nations have larger concerns than useless philosophers, but much of the work on corruption
still takes an idealized form of a government institution or public service as the basis of comparison.
In this chapter, I have tried to show how even a narrow focus on illegal payments or the unethical
use of public office for private gain can erode public confidence and trust. The deeper problem, one
that I have emphasized here because it is not always clear in the purely functional literature, is that
321
Michael C. Munger
corrupt systems are extraordinarily difficult to reform. Corruption is itself a corrupt bargain, in
which short-term improvements in efficiency, finding workarounds for inefficient systems and
limited government capacity, lock in the corrupt system and hold back innovation and prosperity
indefinitely. It is not that people don’t realize there is a better way, but simply that the “better way”
is unavailable once corruption grips the system.
Notes
1 This is the standard definition in political philosophy, usually sourced to Nye (1967, 218): “Corruption is
the abuse of power by a public official for private gain.” For a review, see Miller (2018).
2 Online Etymology Dictionary, http://www.etymonline.com/index.php?term=corrupt.
3 “Fraud, corruption, vice, crime, in short, the entire catalog of proscribed behavior, becomes increasingly
common when the emphasis on the culturally induced success-goal becomes divorced from a coordinated
institutional emphasis. This observation is of crucial theoretical importance in examining the doctrine that
antisocial behavior most frequently derives from biological drives breaking through the restraints imposed by
society. The difference is one between a strictly utilitarian interpretation which conceives man’s ends as random
and an analysis which finds these ends deriving from the basic values of the culture” (Merton, 1938, 676).
4 As Tullock notes:
While corruption usually meets with disapproval, it may have some redeeming features. It may make
possible smaller or no salary payments to officials who, if carefully supervised, will still carry out their
functions on a fee-for-service basis. The purchase of government jobs usually is thought to be corrupt,
but in some cases, it has worked out quite well. (1996, 6)
Johnston also notes the benefits of functional corruption in dysfunctional systems:
Corruption…. was a way for elites to build their political backing in society and to win cooperation
in both parliaments and bureaucracies, a way for entrepreneurs and investors to break through bu-
reaucratic bottlenecks, an informal price system in tightly regulated economies, and a cushion against
the worst social dislocations of development. (2005, 23)
5 This calls to mind Adam Smith’s famous distinction about “good” and “bad” people, as opposed to good or
bad systems. Smith was describing the British East India Company, but he could as easily be talking about
citizens with corrupt governments, or even individual corrupt officials themselves:
I mean not, however, by any thing which I have here said, to throw any odious imputation upon the
general character of the servants of the East India company, and much less upon that of any particular
persons. It is the system of government, the situation in which they are placed, that I mean to
censure; not the character of those who have acted in it. They acted as their situation naturally
directed, and they who would have clamoured the loudest against them would, probably, not have
acted better themselves. (1776; IV.vii.c.107)
6 The “consent of the governed” is an aspirational and sufficient condition for political authority, but the
control of violence is necessary and is shared by all states, legitimate or not. As Max Weber put it:
What is a “political” association from the sociological point of view? What is a “state”?
Sociologically, the state cannot be defined in terms of its ends. There is scarcely any task that some
political association has not taken in hand, and there is no task that one could say has always been
exclusive and peculiar to those associations which are designated as political ones: today the state, or
historically, those associations which have been the predecessors of the modern state. Ultimately, one
can define the modern state sociologically only in terms of the specific means peculiar to it, as to
every political association – namely, the use of physical force …
… [A] state is a human community that (successfully) claims the monopoly of the legitimate use of
physical force within a given territory. Note that “territory” is one of the characteristics of the state.
Specifically, at the present time, the right to use physical force is ascribed to other institutions or to
individuals only to the extent to which the state permits it. The state is considered the sole source of the
“right” to use violence. Hence, ’politics’ for us means striving to share power or striving to influence the
distribution of power, either among states or among groups within a state. (1921, 396–7)
322
Corruption
7 As Nachbar put it:

The involvement of guilds in the administration of a national English economy began with their role
in tax collection. An early innovation in the administration of the English monarchy was to shift the re-
sponsibility for domestic revenue collection from royal officers to select town residents. The development
throughout the thirteenth and fourteenth centuries of a middle class of tradesmen and merchants (the
burgesses) provided a group of individuals with reputations sufficient to assure the Crown that taxes would be
paid. The Crown-chartered municipal corporations were made up of those leading citizens, who
became jointly and individually liable to pay the town’s share of the royal taxes. Those charters
exempted citizens of the towns from the normal mechanism for royal taxation (such as the paying
of tolls or domestic customs) and consequently from most royal commercial regulation. The
members of the corporation fulfilled their pledge by collecting shares of the assessment from
the other citizens of the town. (2005, 1320; emphasis added).
8 For a review, see Thierer (2016), especially pp. 7–11.

9 Douglass North has always insisted that it is the impersonal coordination of human activity that makes
markets and prices valuable. See, for example, North and Thomas (1973) and North (1990). Of course, this
claim echoes the long-held view of Mises (1952; 1998), Hayek (1945), and Kirzner (1963).
10 The necessity that “property” must entail a kind of open competition that fosters “creative destruction” is
implicit in North, Wallis, and Weingast (2009). But see also Schmidtz (2009).
11 This was in fact the original argument made about rent-seeking as a kind of destructive competition that
distorts price signals and locks in inefficiency. For a review, see Tollison (2012). Murphy et al. (1993)
demonstrate the specific barriers involved with the particular form of rent-seeking that we think of as
“corruption” play out, which causes a devastating effect.
References
Adiga, Aravind. (2008) “Indians’ worst fear: the honest politician.” Guardian. (July 30). https://www.
theguardian.com/commentisfree/2008/jul/30/india.scamsandfraud.
Aidt, T. S. (2016) “Rent-Seeking and the Economics of Corruption.” Constitutional Political Economy. 27,
142–157.
Ariely, D. (2013) The honest truth about dishonesty: How we lie to everyone— Especially ourselves. New York, NY;
HarperCollins.
Arthur, B. (1989) Competing Technologies, Increasing Returns, and Lock-In by Historical Events. Economic
Journal. 99, 394, 116–131.
Bassett, John Spencer. (1895) The Regulators of North Carolina: 1765-1771. Washington, DC; Govt. Printing
Office.
Coase, Ronald (1960) The Problem of Social Cost. Chicago, IL; University of Chicago Law School.
Collier, Paul. (2002) “How to Reduce Corruption.” African Development Review. 12, 191–205.
Cox, Gary W., Douglass North and Barry Weingast. (2019) “The Violence Trap: A Political-Economic
Approach to the Problems of Development.” Journal of Public Finance/Public Choice. 34, 1, 3–19.
DeLeon, Peter. (2015) Thinking About Political Corruption. New York; Routledge.
Dungan, James, Adam Waytz, and Liane Young. (2014) “Corruption in the Context of Moral Trade-offs.”
Journal of Interdisciplinary Economics. 26, 1–2, 97–118.
Gwartney, James, Joshua Hall and Robert Lawson. (2016) Economic Freedom of the World. Vancouver, BC; Fraser
Institute.
Haidt, J. (2007) “The new synthesis in moral psychology.” Science, 316, 5827, 998–1002.
Hayek, Friedrich A. (von). (1945) “The Use of Knowledge in Society.” American Economic Review, 35, 4,
519–530.
Heidenheimer, Arnold J., Michael Johnston, and Victor T. Levine. (1989) Political Corruption: A Handbook.
New Brunswick, N.J.; Transaction.
Hume, David. (2000) A Treatise of Human Nature (Oxford Philosophical Texts), David Fate Norton and Mary J.
Norton eds., Oxford; Clarendon Press.
Johnston, Michael. (2005) Syndromes of Corruption. New York; Cambridge University Press.
Kirzner, Israel M. (1963) Market Theory and the Price System. Edited and with an Introduction by Peter J. Boettke
and Frédéric Sautet. Indianapolis; Liberty Fund, 2011. 2/24/2015.
Lambsdorff, J. G. (2002) “Corruption and rent seeking.” Public Choice, 113, 1–2, 97–125.
323
Michael C. Munger
Leibenstein, Harvey. (1966) “Allocative Efficiency vs. X-Efficiency,” American Economic Review, 56, 3, 392–415
Merton, Robert K. (1938) “Social Structure and Anomie.” American Sociological Review, 3, 5, 672–682.
Merton, Robert K. (1949) Social Theory and Social Structure. Glencoe, IL; Free Press.
Miller, Seumas. (2018) “Corruption.” The Stanford Encyclopedia of Philosophy. Edward N. Zalta ed. https://plato.
stanford.edu/archives/win2018/entries/corruption/.
Mises, Ludwig (von). (1952) “Profit and Loss, “in Planning for Freedom. South Holland, Ill.; Libertarian Press.
Mises, Ludwig (von). (1998) Human Action: A Treatise on Economics. Auburn, AL; LVM Institute.
Mishra, Satyananda. (2015) “Incompetence is the progenitor of corruption.” Economic Times. June 13, https://
economictimes.indiatimes.com/blogs/et-commentary/incompetence-is-the-progenitor-of-corruption/.
Munger, Michael. (2017) “On the Contingent Vice of Corruption.” Social Philosophy and Policy. 35, 2, 158–181.
Munger, Michael. (2018) Tomorrow 3.0. New York; Cambridge University Press.
Munger, Michael. (2019) “Tullock and the Welfare Costs of Corruption: There IS a ‘Political Coase
Theorem’.” Public Choice. 181, 1–2, 83–100.
Mungiu-Pippidi, A. (2006) Corruption: Diagnosis and Treatment. Journal of Democracy.17, 3, 86–99.
Murphy, Kevin M., Andrei Shleifer, and Robert W. Vishny (1993) “Why is Rent-Seeking So Costly to
Growth?,” The American Economic Review, 83, 2, 409–414.
Nachbar, Thomas B. (2005) “Monopoly, Mercantilism, and the Politics of Regulation,” Virginia Law Review.
91, 6, 1313–1379.
North, Douglass C., (1990) Institutions, Institutional Change, and Economic Performance, New York; Cambridge
University Press.
North, Douglass C., and Robert P. Thomas. (1973) The Rise of the Western World: A New Economic History,
North, Douglass, John Wallis, Barry Weingast. (2009) Violence and Social Orders: A Conceptual Framework for
Interpreting Recorded Human History. New York; Cambridge University Press.
Nye, J. S. (1967) “Corruption and Political Development: A Cost-Benefit Analysis,” American Political Science
Review 61, 417–427.
Olson, M. (1982) The Rise and Decline of Nations: Economic Growth, Stagflation, and Social Rigidities. New Haven,
CT; Yale University Press.
Olson, M. (1993) “Dictatorship, Democracy, and Development.” American Political Science Review. 567–576.
Ransom, R. (2001) “Economics of the Civil War”. EH.Net Encyclopedia, edited by Robert Whaples. August 24,
2001. URL http://eh.net/encyclopedia/the-economics-of-the-civil-war/
Romig, Rollo. (2017) “How to Steal a River: To feed an enormous building boom, India’s relentless sand
miners have devastated the waterways that make life there possible.” New York Times, March 1. https://
www.nytimes.com/2017/03/01/magazine/sand-mining-india-how-to-steal-a-river.html.
Rose-Ackerman, Susan. (1978) Corruption: A Study in Political Economy. New York; Academic Press.
Shleifer, Andrei, and Robert W. Vishny. (1993) “Corruption,” Quarterly Journal of Economics. 103, 599–617.
Schmidtz, David. (2009) Functional Property, Real Justice. Keynote speech given at “Property Rights in Central
and East European Countries: Developments after the Transformation Process,” 13 November, 2009. Berlin,
Germany: https://shop.freiheit.org/download/P2@230/8197/E_ELF_Schmidtz_24_4S_Internet.pdf.
Scott, James. (1972) Comparative Political Corruption. Englewood Cliffs, NJ; Prentice-Hall.
Smith, A. (1759/1984) The Theory of Moral Sentiments (Glasgow Edition). Edited by D. D. Raphael and A. L.
Macfie. Indianapolis, IN; Liberty Fund, Inc.
Smith, Adam. (1776/1991) An Inquiry into the Nature and Causes of the Wealth of Nations. Indianapolis, IN;
Liberty Fund.
Thierer, Adam. (2016) Permissionless Innovation: The Continuing Case for Comprehensive Technological Freedom.
Arlington, VA; Mercatus Center.
Tollison, Robert D. (2012) “The economic theory of rent seeking.” Public Choice. 152, 1/2, 73–82.
Troxler, Carole W. (2017) Farming Dissenters: The Regulator Movement in Piedmont North Carolina. Chapel Hill;
University of North Carolina Press.
Tsebelis, George. (2002) Veto Players: How Political Institutions Work. Princeton University Press.
Tullock, Gordon. (1975) “The Transitional Gains Trap.” Bell Journal of Economics. 6, 2, 671–678.
Tullock, Gordon. (1980) Efficient rent-seeking, in J. M. Buchanan, R. D. Tollison and G. Tullock Eds. Toward
a theory of rent-seeking society. College Station; Texas A.&M. University Press.
Tullock, Gordon. (1996) “Corruption Theory and Practice.” Contemporary Economic Policy. 14, 3, 6–13.
Underkuffler, Laura S. (2013) Captured by Evil: The Idea of Corruption in Law. New Haven; Yale University Press.
Vanderschraaf, Peter. (1998) “Knowledge, Equilibrium and Convention.” Erkenntnis 49:337–369.
Weber, Max. (1921) “Politik all Beruf,” Gesammelte Politische Schriften. Muenchen; Drei Masten Berlag. 396–450.
324
PART VI
Distributive Justice
Introduction to Part VI
Justice seems to have two levels. At the first level, it involves a kind of lawfulness. Those that break
the law are “brought to justice” and “departments of justice” faithfully execute the law. At a second
level, justice determines what the law should be. To have a just society, we sometimes need to
change the laws. So, second-level justice often requires reform of first-level justice.
Discussions of “distributive justice” focus on a particular aim for reform. It is a concern with how
the advantages and disadvantages of society are shared among its members. Some have more wealth
than others. Some have more opportunities than others. Income, political power, economic power,
cultural power, and the social bases of self-respect are all spread across society in unequal amounts.
Put simply, distributive justice identifies the appropriate distribution of these advantages.
Distributive justice is of particular interest for PPE scholars because any distribution results from
specific economic and political systems. Few think that government has the power to divide so-
ciety’s resources up in whatever way it deems appropriate, but we can structure institutions and rules
in ways that favor certain outcomes over others. As such, our distributive ideals influence which
institutions we support. In the opposite direction, the ideals we choose might depend on which
institutions – if any – could bring about those ideals. Because of these interdisciplinary links, Part VI
introduces some key issues in distributive justice that PPE scholars continue to work on.
As the first chapter points out, the idea of distributive justice itself seems to make certain as-
sumptions. Specifically, it supposes that distributions should be a primary concern. The libertarian view
of economic justice opposes this. Instead, the primary concern should be the process that explains how
people have come to have what they have. Accordingly, evaluating a distribution without attention to
the process that led to that distribution would be a mistake. In “Property Rights and Justice in
Holdings: A Libertarian Perspective,” Eric Mack better explains the core ideas that contribute to this
libertarian conception of justice. He starts with an interpretation of how John Locke believes that
anyone could come to own any external object, then he explains the historical process view of justice
in holdings by referencing the work of Robert Nozick. With the basics of a libertarian view explained,
Mack reiterates two of Nozick’s important criticisms and develops some of the nuances of the lib-
ertarian view. In particular, he recognizes that property rights have both natural and conventional
aspects and identifies five concerns that a full libertarian theory must address.
The second chapter situates a concern for distributive justice within broader trends in liberal
thought. Oftentimes, it seems like political theories put the ideals of freedom, democracy, and
DOI: 10.4324/9780367808983-31
equality in conflict with one another. It is as if we must choose to prioritize one over the others. The
political left, for example, might criticize “liberals” because the protection of individual freedoms
seems to conveniently protect the powerful. The political right might criticize the left for prior-
itizing equality over freedom. In his chapter, Samuel Freeman shows how one liberal tradition has
developed to respect all three ideals. “High Liberalism” adjusts the commitments of “Classical
Liberalism” in response to both democratic movements and socialist criticisms of capitalism. The
resulting conception of justice remains a “liberal” view because it prioritizes the protection of certain
basic liberties. However, this view protects democracy because political freedoms are as important as
other basic freedoms. High Liberalism also has three features that promote equality. It protects the
substantive equality of opportunity, uses a conception of distributive justice to regulate a market
economy, and has an expansive conception of public goods. A final feature of High Liberalism is that
it views political power as exercised in the name of the people, so particular laws and policies must
be justified on the basis of shared reasons. With each feature of High Liberalism, Freeman shows
how this conception of liberalism differs from its alternatives.
In the third chapter, the focus is on whether individual decisions should advance distributive
justice in the same way that policy should. In his influential Theory of Justice, John Rawls expressed
the view that institutions – in particular, the institutions of the basic structure of society – are the first
subject of justice. At the end of the 20th century, this view was challenged from two perspectives.
First, egalitarians like G.A. Cohen argued that Rawlsians misunderstood the nature of justice. Justice
is a basic moral ideal that applies to both policy and personal choices, so we cannot justify egalitarian
institutions without also committing ourselves to egalitarian personal choices. Second, con-
sequentialists such as Liam Murphy argue that such a view constrains our ability to make a more just
world. Individuals should work to improve the world whenever they have the opportunity to do so
and not only improve institutions. In “Institutions and Egalitarianism,” Kok-Chor Tan defends the
position that an individual’s primary duty of justice is to promote just institutions. He argues that
institutional reform is more likely to address the root causes of injustice and not only alleviate its
effects. Moreover, institutions structure our individual rights and obligations, so we cannot have a
just world without changing unjust institutions. He addresses remaining objections to the in-
stitutionalist view and even links duties of repair with obligations to change institutions.
Like the first chapter of Part VI, the fourth also criticizes the distributive justice paradigm but for
entirely different reasons. A focus on distributive justice often abstracts away from other kinds of
justice. Identifying a just distribution of resources does not, for example, make it obvious what
retributive justice and restorative justice require. In “Social Justice,” Maeve McKeown focuses on
injustices that the distributive paradigm ignores. Oppression and domination have distributive
consequences, but the wrong done by them are not fully captured by tracking who has what.
McKeown uses Iris Marion Young’s classic criticism of the distributive justice paradigm to introduce
four concerns associated with the demand for social justice; recognition, epistemic injustice, rela-
tional equality, and structural injustice. She situates each concern in the literature that raises it and
shows how each is left out from a focus on distributions.
Historically, the scope of distributive justice has been restricted to the single state. Serena Parekh
starts Part VI’s final chapter, “Justice Across Borders,” with two reasons why this is no longer
appropriate – if it ever was. First, the decisions within any particular country now have far-ranging
consequences around the globe. Second, we are more aware of the conditions of those far removed
from our daily life. For these reasons, we should recognize that distributive justice has a wider scope,
and Parekh distinguishes four ways to do this. First, Peter Singer represents a view that focuses on
individual obligations to help. Second, John Rawls gives an influential account of states having
obligations to build up the capacities of other states. Third, Thomas Pogge argues that international
practices have wronged and continue to wrong people, so our duties of justice should be to stop and
repair that past harms. Parekh herself defends a fourth approach that is modeled on the view of
structural injustice introduced by Iris Marion Young. People are connected with one another
through the globe in complex and important ways. International injustice is not merely the fault of
states and exploitative intentions. Rather, the systems of interaction reinforce and proliferate dis-
advantage. Creating a more just world requires attention and revision to this complex structure.
Parekh demonstrates this perspective by explaining two pressing issues of international justice,
women’s rights and the treatment of refugees.
26
PROPERTY RIGHTS AND JUSTICE
IN HOLDINGS: A LIBERTARIAN
PERSPECTIVE
Eric Mack
1 Introduction
Political libertarianism is the view that the fundamental norm that ought to govern the scope and
activity of coercive political or legal institutions is respect for individual liberty. Acceptable coercive
state action is essentially limited to action that protects the freedom of individuals in ways that does
not itself violate anyone’s liberty. Coercive state action that infringes on individual liberty for the
sake of making people virtuous or for the sake of preventing people from harming themselves is not
permissible. Libertarians also maintain that respect for liberty crucially includes respect for people’s
rights to acquire property and to exercise discretionary control of their justly acquired property.
Coercive interference with people’s economic lives is as objectionable as coercive interference with
people’s lifestyle or religious choices. Thus libertarian theorists reject doctrines of distributive justice
that seek to vindicate the coercive redistribution of income or wealth that individuals have acquired
through rights-respecting processes.
Of course, a full defense of the libertarian perspective on property rights and justice in holdings is
an enormous project. This essay can merely point to some of the key contentions and arguments that
defenders of this perspective have offered. More specifically, it will focus on some of the key
arguments offered by John Locke on behalf of the existence of robust private property rights and key
arguments offered by the neo-Lockean Robert Nozick on behalf of his “historical entitlement” view
of justice in holdings and against “distributivist” doctrines that call for state action to redistribute
income or other economic holdings to achieve distributive justice. It will also go beyond Locke’s
and Nozick’s arguments by outlining the place and implications of a natural right of property within
a plausible libertarian stance on property rights and justice in holdings. Section 2 will focus on
Locke’s arguments. Sections 3, 4, and 5 will focus on Nozick’s historical entitlement view and his
critique of distributivist doctrines. Section 6 will deal with the natural right of property. Section 7
will take note of five complexities that need to be adequately addressed in any acceptable libertarian
doctrine of economic justice.1
2 Locke on Just Initial Acquisition

Locke’s initial affirmation of natural rights is cast as an affirmation of a right to freedom – a right to
control one’s own person and legitimate possessions and not be subjected to the will of others.
Locke typically explicates this right in terms of what losses or impairments may not be inflicted upon
DOI: 10.4324/9780367808983-32 329

Eric Mack
one. According to Locke, one’s life, limbs, health, and goods may not be taken or damaged or
removed from one’s discretionary control – at least not without one’s consent (Locke §6). “[E]very
man has a property in his own person: this no body has a right to but himself” (Locke §27). Moreover,
Locke says that, in virtue of our self-propriety, we each have a right over our own labor. This right
to one’s own labor is the key to Locke’s doctrine of just initial acquisitions.
Locke begins his account of the just initial acquisition of extra-personal objects by arguing that
raw natural material – the earth – is originally unowned. Hence, it is permissible for individuals to
engage in the acquisition of raw material without first getting the consent of any other individuals.
Locke’s key argument here is directed against the idea that to begin with all raw natural material is
jointly owned by mankind at large. Locke argues that if there were such an original joint-ownership,
no one would ever permissibly acquire or even use any bit of natural material. For, everyone would
have to consent to that acquisition or use before it could be permissible and such universal consent
would never be forthcoming. “If such a consent as that was necessary, man had starved, notwith-
standing the plenty God had given him” (Locke, §28). If no one had the right to acquire or even use
any bit of raw material without first getting everyone’s permission, no one would really have a
natural right to do anything to promote her commodious self-preservation. For one does not have a
natural right to do something, if one needs everyone else’s permission to do it. Since, according to
Locke, everyone has a natural right to pursue her commodious self-preservation, it cannot be that
anyone needs the consent of others before acquiring or even using any bit of raw natural material.
Suppose you come across a bit of unowned raw material – say some nice flexible vine. Over the
course of a couple of hours, you fashion the vine into five snares which you set out in hopes of
catching some unowned rabbits. Having done this you leave the immediate area with plans to return
the next morning to check your snares. But while you are gone someone else comes by, sees the
snares, thinks that he would like to have those five nifty devices and would like to avoid the trouble
of making them. So, he gathers them up and heads off to another area where he plans to make use of
them to snare muskrats.
Locke claims that this second party is a thief who in taking those snares without your permission
violates your rights. The reason, according to Locke, is that you have “mixed” your labor with or
“annexed” your labor to the raw material that you fashioned into the snares. Hence, when the
second party makes off with the snares without your permission he makes off with your labor. Since
you have a right to that labor, the second party violates your rights in making off with the snares.
You have a right to the snares precisely because your labor is invested in them.
The labour of his body, and the work of his hands, we may say are properly his. Whatsoever
then he removes out of the state that nature hath provided, and left it in, he hath mixed his
labour with, and joined to it something that is his own, and thereby makes it his property.
(Locke, §27)
Critics of Locke read “mixing” and “annexing” too literally. They suggest that Locke’s view is that,
if one mixes say 10 ounces of labor with 10 pounds of vine, one ends up with 10 pounds and
10 ounces of property. However, the appropriate reading of Locke is that he is talking about one’s
right to one’s human capital and one’s investment of that capital by way of fashioning some pre-
viously unowned material for the sake of some ongoing purpose. It’s the ongoing purpose that
explains why the human capital that is deployed is invested and not merely frittered away. Locke’s
crucial claim is that the party who makes off with the snares that you have fashioned expropriates a
portion of your human capital. Because this violates your rights, it makes sense to say that because
you have fashioned those snares, the snares are your property.
In a way, Locke’s point is simply that when one makes off with the five snares that have been
fashioned through another’s investment of her time, effort, and skill one inflicts a bit of slavery on
330
Property Rights and Justice in Holdings
that productive individual. What one does is morally on a par with putting a gun to the head of that
productive individual and coercing her to provide one with those snares. To so coerce another is to
inflict a bit of slavery upon her. As inflicting slavery violates another’s rights if anything does, making
off with the five snares violates the rights of the individual who has fashioned them through the
investment of her time, effort, and skill.
For Locke, the key philosophical question with respect to property rights is how initial property
can come into existence. He clearly believes that individuals can also acquire property rights through
voluntary transfer from current right-holders. Indeed, he takes the prospect of such transfers to be
the key incentive that individuals have to find new and better ways to produce goods that others will
be eager to purchase (Locke, §48). Yet, Locke never attempts to provide a philosophical account of
just transfers. Nor is it clear how plausible a labor-mixing theory of just transfers would be. Locke also
clearly holds that individuals who have been unjustly deprived of just holdings may rightfully extract
restitution from those responsible for those deprivations.
3 Historical Entitlement
Locke’s doctrine of just initial acquisition is a process doctrine. What makes a given initial holding
just is that it has arisen through a process in which one permissibly transforms some raw material
through one’s investment of one’s human capital. What matters is one’s acquisition through the
right sort of process, and not the magnitude – in weight or volume or economic value – of what one
has acquired. Nozick’s historical entitlement view generalizes this Lockean focus on entitlement-
conferring processes rather than on the absolute or relative size of the holdings that result from
acquisitive processes. According to Nozick, any given holding is just if and only if that individual has
acquired the holding through entitlement-conferring initial acquisition, through entitlement-
conferring transfer from someone who was entitled to it (through just initial acquisition or just
transfer), or through just restitution imposed upon individuals who have acquired one’s just holdings
through actions that are not entitlement-conferring (Nozick, 150–3).
In fact, in ordinary workaday life, almost all of us adopt this process perspective. If you ask me
why I have a right to the shiny rock I use as a paperweight, I will say that I found that object while
wandering around an unowned field and polished it up a bit, or that I purchased it (with funds that I
had justly acquired) from someone who had found and polished it, or that I had retrieved it from a
burglar who had taken the paperweight from my home in the previous week. If you accept the
factual claims involved in my answer, you would, lin ordinary life, accept my right to the paper-
weight. You would not expect me to defend my right to that object by saying that everyone else has
an equally nice paperweight or that my having the paperweight is part of an overall distribution of
physical objects that maximizes aggregate wealth or utility. You would not expect me to justify my
possession of the paperweight by showing that it is part of the overall pattern of holdings which
ranks highest among all the available patterns of holdings.
In contrast, all distributivist views concerning justice in holdings take justice to be a matter of
identifying and promoting the most highly ranked of the available overall distributions. It is the
distribution that matters, not the processes by which individuals acquire their possessions. In this
discussion, I will follow most distributivist doctrines in taking real income and, hence, absolute
purchasing power as that which is to be distributed. Nozick’s case for his historical entitlement view
is almost entirely an argument from elimination. One should adopt the historical entitlement view
because it alone is free of the deep defects that plague all noteworthy distributivist views.
Nozick distinguishes between two sorts of distributivist doctrine. There are end-state doctrines
and patterned doctrines.2 According to any end-state theorist, there is an arithmetical feature of all
distributions of income and alternative distributions are to be ranked by the degree to which they
realize that arithmetical feature. Of course, different end-state theorists favor different arithmetical
331
Eric Mack
Table 26.1 Three alternative distributions for a society consisting of individuals (or representatives of income
classes) A, B, and C
Individual A Individual B Individual C
Distribution D1 7 9 8 Most equal distribution

Distribution D2 8 12 10 Highest minimum income
Distribution D3 6 5 42 Highest aggregate
features as the basis for ranking available distributions. Table 26.1 shows three alternative dis-
tributions for a society consisting of individuals (or representatives of income classes) A, B, and C.
The egalitarian end-state theorist ranks alternative distributions according to the degree to which they
embody equality and, therefore, favors D1. The Rawlsian advocate of the difference principle ranks
alternative distributions according to how large the payoff is for the person with the lowest payoff (or for
the representative person within the lowest income group) and, therefore, favors D2. The fan of the
maximization of aggregate income favors D3. What all “end-staters” have in common is the supposition
that the information presented in a table of this sort is all one needs to rank distributions D1, D2, and D3.
However, this shared supposition is highly implausible. Suppose that, if it hadn’t been for B’s
nasty hobby of destroying productive resources and equipment, a distribution of <25, 22, and 65>
would have been available for A, B, and C. Surely that information is a highly relevant reason not to
select D2 – the remaining available distribution that is best for B. Or suppose that every produced
good or service that is available to any member of this small society – for example, all the clothing,
food, shelter, and medicine – has been produced by C. Surely that information is a highly relevant
reason not to favor D1 or D2. For both these distributions provide one-third of the population with
more purchasing power than the party who is solely responsible for there being anything to pur-
chase. The point here is the economic conduct or misconduct that people have engaged in is
relevant information for making judgments about justice in holdings or income, but all such in-
formation is excluded from the tables on the basis of which end-staters rank alternative available
distributions. Thus, any end-state ranking of distributions of income must be radically defective.
End-state theories especially seem to presume that somehow the problem of production has been
solved (or has never existed). It is a given that certain distributions of purchasing power – and,
hence, certain distributions of goods and services – are available. So, all that remains is to solve the
problem of distributing what is essentially Manna from Heaven or Gifts of Nature in accordance
with the right theory of how Manna or Gifts should be distributed. Doing justice to or avoiding
injustice toward productive contributors is simply not an issue. Economic justice is at root entirely a
matter of how the bounty of God or nature is to be distributed.
In contrast to end-staters, patterned theorists recognize that the ranking of alternative distribu-
tions must appeal to information that is not provided in the sort of table offered above. Patterned
theorists hold that the distribution of income should track the pattern in which some estimable
feature obtains among the individuals who will be subject to that distribution. A patterned theorist
might hold that the crucial feature is aesthetic sensibility. We have to find out what the pattern of
aesthetic sensibility is among the members of the relevant society and the best or just distribution
will be the distribution that most allots income to individuals in accordance with their degree of
aesthetic sensibility. Other patterned theorists will favor different features – for example, moral
sensitivity or moral merit. Or perhaps income should be distributed among individuals in accord
with their degree of economic productivity.
Three problems are obvious for patterned theorists. First, which feature should be selected or, if
some combination of features should be selected, how should those features be weighed against one
332
another? Second, how are the appropriate authorities supposed to ascertain the extent to which the
selected feature or features are present in individuals? How many people – with what degree of
moral merit – will have to be put to work in the Department of Moral Merit Measurement? Third,
what sort of surveillance of individuals – by the authorities, their family, their neighbors, or their
business associates – will have to be conducted in the attempt to acquire this information?
The patterned view that income should be distributed among individuals in proportion to their
economic productivity may appear to be similar to the historical entitlement stance that an individual’s
production of a given economic good itself creates an entitlement of that person to that good. Yet this
appearance is misleading. This patterned view does not start with an identification of those modes of
acquisition that are to be recognized as entitlement-conferring and go on to equate productivity with
acquisition that is achieved through those entitlement-conferring means. Most people who are in-
tuitively attracted to the idea of distribution in accord with productivity will not be happy if a person’s
degree of productivity is determined by the market payment she receives for the goods or services she
provides. They will instead be drawn to the idea of gauging productivity on the basis of the (supposed)
true social importance of what is produced rather than on the basis of the flaky whims of consumers. If,
instead, the degree of productivity of each individual is measured by the income attained by her in the
marketplace (including the income from gifts and charitable donations), the view under consideration
will have morphed into a non-patterned historical entitlement view.
4 How Liberty Upsets Patterns

Nozick devotes two important sections of Anarchy, State, and Utopia to criticisms of end-state and
patterned views about justice in holdings. The first of these is the commonly cited but rarely suf-
ficiently appreciated section on “How Liberty Upsets Patterns” (Nozick, 160–64). Here I offer a
reconstruction of Nozick’s most basic critique. This critique focuses on the deep tension within the
distributivist program between the end of establishing distributive justice and the commitment to the
ongoing application of a favored distributivist principle.3 I then briefly consider Rawls’s response to this
Nozickian critique.
Let us take Rawls’s difference principle as our example of a distributivist rule. The advocate of that
principle tells us that basic structure1 ought to be implemented because (according to the best experts
that money can buy) the operation of that structure will yield distribution D1 which will more fully
satisfy the difference principle than any other available distribution. The Nozickian critique asks, what
next? Is the implementation of structure1 a one-time solution or must there be further applications of
the difference principle leading to continual adjustments of the basic structure?
Table 26.2 below illustrates for a four-party (or four-income class) society the likely reality of the
ongoing application of the difference principle in the ceaseless endeavor to achieve distributive
justice. Even if D1 initially arises under structure1, certain individuals (e.g., A and C) will almost
certainly find mutually advantageous ways to cooperate with one another which are entirely permissible
under basic structure1; and the upshot of this cooperation will be D2 rather than D1. However, ac-
cording to the reigning distributive principle, D2 will almost certainly be unjust. For, given the
emergence of D2, almost certainly (the experts will judge that) a new and improved basic structure3 will
convert D2 into D3 in which the minimum income party will have a higher income than the worst-
off party under D2.
Justice will demand (or seem to the experts to demand) that structure1 be replaced by structure3
which will (we are supposing) initially give rise to D3. However, it is almost certain that some parties
(say, C and D) will engage in mutually advantageous interactions that are both not anticipated by the
experts who design structure3 and yet are entirely permissible under structure3. These actions will
transform D3 into D4. However, like D2, D4 will be (judged to be) unjust. For D4 will (or seem to)
be convertible into D5 through the subsequent adoption of structure5 in which the income of the
333
Eric Mack
Table 26.2 The likely reality of the ongoing application of the difference principle in the ceaseless endeavor to
achieve distributive justice in a four-party (or four-income class) society
Anticipated Just Distributions and Actually Arising Distributions Party A Party B Party C Party D
D1 – distribution that initially arises under basic structure1 and is 9 11 18 17

expected to maximize the minimum payout and, hence, be just.
D2 – distribution that subsequently arises under structure1 via rule- 15 11 23 17
compliant but unanticipated actions by A and C. D2 is not just
because a revised structure3 will (it seems) convert D2 into D3.
D3 – distribution that initially arises under structure3 and is 13 14 19 17
D4 – distribution that subsequently arises under structure3 via rule- 13 14 23 21
compliant but unanticipated actions by C and D. D4 is not just
because a revised structure5 will (it seems) convert D4 into D5.
D5 – distribution that initially arises under structure5 and is 15 15 20 18
worst off is (expected to be) greater than the income of the worst off in D4. Thus, justice will
demand that structure3 be replaced by structure5. And so on and on.
The problem is that the ongoing application of the difference principle requires the ongoing
transgression of the holdings that prior applications of that principle have deemed to be just. The
operation of the reconstructed basic structures will violate many of the legitimate expectations that
arise under prior structures that have themselves been implemented in the name of justice. This is
because successive (non-Stalinist) attempts to achieve distributive justice through the institution of a
new (or significantly revised) basic structure will always leave room for widespread, unanticipated,
and mutually voluntary interactions that will lead to distributions (e.g., D2 and D4) which have to be
judged to be unjust because some yet further systemic structural reset will yield a distribution (e.g.,
D3 and D5) that (experts expect) will better satisfy the favored difference.
The only way for the stalwart follower of the difference principle to escape this conclusion is to
deny the justice of the holdings and legitimacy of the expectations that arise under basic structures that
turn out to need reordering. The implication of this move, however, is that just holdings and legitimate
expectations will not exist until a basic structure is implemented that thoroughly takes account of future
human beliefs, insights, circumstances, opportunities, preferences, ambitions, and the economic activity
that will arise from them. It may be a while before such a basic structure is attained.
In Political Liberalism, John Rawls claims that Nozick’s Liberty Upsets Patterns criticism does not
undermine the difference principle. According to Rawls, this is because Nozick’s criticism mis-
takenly takes the difference principle to demand continuous intervention to reorder holdings
whenever they depart – however momentarily – from the distribution that best satisfies the dif-
ference principle. Rawls says that Nozick does not appreciate that the difference principle (along
with the other principles of justice) is actually supposed to guide society in its choice of a basic
structure of institutions and rules that will foster an overall and temporally extended distribution of
income that maximizes the income of the worst off (Rawls, 283).4
Nozick’s own presentation of his Liberty Upsets Patterns criticism does lend itself to this objection
because his examples suggest that the advocates of distributivist principles are eager to bounce upon and
nullify every particular momentary deviation from their favored pattern. However, the reconstruction
of that criticism offered here anticipates Rawls’s objection by treating the difference principle as a guide
to the continual readjustment of basic structures. Might Rawls avoid this criticism by maintaining that a
basic structure that best satisfies his difference principle can now (or soon) be established once and for all
so that holdings proclaimed to be just under such a permanent basic structure will never (or rarely) turn
334
out to be unjust and subject to infringement under a new and improved basic structure? Strikingly,
Rawls rejects this contention. He insists that ongoing resets of the basic structure are unavoidable.
The need for a structural ideal to specify constraints and guide adjustments does not depend
upon injustice [i.e., people acting contrary to the rules established by the current basic
structure]. Even with strict compliance with all reasonable and practical rules, such ad-
justments are continually required …. A conception of justice must specify the requisite
principles and point to the overall direction of political action. In the absence of such an
ideal form for background institutions, there is no rational basis for continually adjusting the
social process so as to preserve background justice, nor for eliminating existing injustice.
(Rawls, 1993, 284–5, emphasis added)
Thus, Rawls’s insistence that ongoing changes of basic structure will be needed for the sake of “the
overall direction of political action,”5 supports the reconstructed contention that the difference
principle (qua distributivist principle) turns on itself by requiring a succession of basic structures that
mandate, in the name of justice, transgressing holdings that have been established as a matter of
justice under prior basic structures.
5 Redistribution and Property Rights

I turn now to the almost explicitly Lockean criticism of distributivist views that Nozick offers in his
section on “Redistribution and Property Rights” (Nozick, 167–74). Individuals do all sorts of things
that contribute to the total income of the societies they inhabit. They labor for wages. They work
for salaries. They invest portions of their savings. They organize and reorganize production. They
patronize all sorts of businesses. They invent new ways of satisfying their own needs and the needs of
others. They enhance their human capital and that of their children. They serve as helpful role
models for others seeking to find productive roles for themselves. And so on. Part of the result of all
these individual and joint decisions, actions, efforts, and endeavors is a certain aggregate economic
output – a certain “total social product.” End-state and patterned principles call for a redistribution
of a portion of this total product. Therefore, the institution of such a principle gives “each individual
an enforceable claim” on at least some portion of the decisions, actions, efforts, and endeavors of the
individuals who have contributed to this total social product. But, says Nozick, this expropriation of
people’s actions, efforts, and decisions is morally unacceptable.
Whether it is done through taxation on wages or on wages over a certain amount, or

through seizure of profits, or through there being a big social pot so that it’s not clear
what’s coming from where and what’s going where, patterned principles of distributive
justice involve appropriating the actions of other persons …. This process whereby they
take this decision from you makes them a part-owner of you; it gives them a property right
in you. Just as having such partial control and power of decision, by right, over an animal
or inanimate object would be to have a property right in it.End-state and most patterned
principles of distributive justice institute (partial) ownership by others of people and their
actions and labor. (Nozick, 172)
Why, according to Nozick, is it unacceptable to institute and enforce a principle that gives others
partial ownership of you? Nozick’s answer is a somewhat muted, arms-length invocation of self-
ownership. “These [end-state and patterned principles] involve a shift from the classical liberals’
notion of self-ownership to a notion of (partial) property rights in other people” (ASU 172). The
335
Eric Mack
unspoken next Lockean step of the argument has to be: because people are self-owners, principles
that involve a shift to (partial) property rights in others must be false.
In contrast to Locke, Nozick does not invoke self-ownership to establish an individual’s property
right to this or that particular holding. Rather the right of self-ownership operates more globally
against the institution of distributivist programs. At least to some degree, any such program treats
people’s productive actions and decisions as though they are the workings of a machine or animal
that is owned by the programmers and is stamping out widgets or laying eggs. This manner of
treatment of people is unacceptable because, unlike machines and chickens, people are self-owners.6
In contrast to the distributivist perspective, a just array of holdings among a given set of people is
simply the array that has arisen through particular individuals acquiring holdings in accord with
entitlement-generating processes. The justice of the resulting array of holdings will be the unin-
tended consequence of individuals engaging in just modes of acquisitions and eschewing non-just
modes. If all individual holdings have been acquired in accordance with the rules of just acquisition
and just transfer, any imposed redistribution of those holdings will violate at least some people’s just
property rights. If some holdings have arisen through non-just procedures, the appropriate action is
to institute the just restitutive processes. The goal of just restitution is to nullify the effects of non-
just modes of acquisition. It aims to engender, if it can be ascertained, the condition that would have
come into existence had non-just processes not been carried out. The goal is not to bring the overall
array of holdings among the relevant people into accord with some end-state or patterned program.
6 The Natural Right of Property

Both Locke and Nozick are often understood as affirming a natural right of self-ownership as the
fundamental natural right. This reading is surprising as Locke does not allude to self-proprietorship
until his discussion of property in chapter 5 of the Second Treatise and Nozick does not mention self-
ownership until his oblique invocation of it in his section on “Redistribution and Property Rights”
in chapter 7 of Anarchy, State, and Utopia. Moreover, when Locke offers his key arguments for
natural rights in chapter 2 of the Second Treatise, he seems to be affirming a less specific, broader,
fundamental claim to freedom, that is, a claim to being allowed to act on one’s own choices and not
be subject to the will of others. Similarly, in his key arguments for moral side constraints in chapter 2
of Anarchy, State, and Utopia, Nozick promotes moral side constraints that broadly protect individuals
in their peaceful enjoyment and discretionary control over their own lives.
I believe that the sensible stance for natural rights theorists is that the right of self-ownership is part of
the proper articulation of a broader and yet more basic moral claim to be allowed to live one’s own life
in one’s own chosen ways (as long as those ways are compliant with respect for the like rights of
others). Another part of that articulation is a natural right to others’ reciprocal compliance with
agreements they have entered into with one. And a third crucial part is the natural right of property.
This is not a natural right to any particular holding or to any share of holdings. Rather it is a right to
make things one’s own and to exercise discretionary control over what one has made one’s own. It is
the right to acquire possessions and to enjoy and peacefully derive benefits from them. This is the third
right in the classical liberal trinity of natural rights, viz., the rights of life, liberty, and property.7
This natural right of property is part of the proper articulation of the basic claim to be allowed to
live one’s life in one’s own chosen way because human beings do not live their lives within the
spaces defined by the outer surfaces of their skin. Rather, we live our lives through the pursuit and
attainment of goals beyond our bodies. And almost all of these goals involve both as means and ends
the purposeful transformation of extra-personal objects. To live our lives successfully it is at least
almost essential that we attain secure control over parts of our physical environment and that we can,
as we choose, modify, and re-contour the array of objects over which we have morally protected
control. General security for people’s possession and enjoyment of the fruits of their talents and
336
efforts and peaceful transactions is a crucial condition for living our lives in peaceful, cooperative,
and mutually advantageous ways. Failure to accord people that security is a violation of their rights
to live their own lives in their own chosen ways.
Here, however, is a complication. In many cases, abstract natural rights – like other abstract moral
principles – will successfully guide action only if they take the form of rather specific, concrete rules
concerning what actions persons must eschew and what actions persons may or must perform in
various different circumstances. The natural right of property tells us that it ought to be possible for
individuals to establish property rights through initial acquisition and through voluntary transfer
from others. It ought to be possible for people to know who holds which rights and just what
constraints on others’ conduct are correlative to those rights. However, that abstract right does not
tell us precisely what sort of “mixing” of labor is needed to establish an initial property right or what
sort of transfer from others confers a right over the transferred object. The abstract right does not tell
us how much drainage from one’s property or disturbing fumes from one’s factory or casting of
shadows across another’s balcony are violations of property rights. It tells us that one can abandon
one’s property rights. But it does not tell us what in particular constitutes abandonment of this or
that sort of property.
What is needed for abstract rights like the right of property to serve their purpose is much more
concrete delineations of what procedures are entitlement-engendering, of what upshots of one’s
actions constitute violations of others’ property rights, of when the non-exercise of one’s rights
amounts to abandonment, and so on. Such delineations are necessary if people are to know what
boundaries they can insist others respect and what boundaries they must themselves respect. Such
delineations are needed for there to be good fences among neighbors – fences that facilitate peace
and voluntary cooperative interaction.
The abstract natural right to property enables us to distinguish between morally eligible sets of
concrete conventional rules that operationalize that abstract right and morally ineligible sets of concrete
conventional rules that inhibit the establishment of property rights or foster the violation of rights that
are insecurely established by rules within those sets. However, there are innumerable different eligible
sets of conventional rules – eligible practices of property – and no armchair philosophical analysis will
select any one of those possible concrete conventional realizations as the one demanded by the abstract
right of property.
Suppose that one eligible set of fine-grained rules has come into existence through the evolution
of custom, judicial decision, and even legislation. Why would this conventional practice of property
have moral authority even though it is only one among many eligible practices that might have
come into existence? The answer is that the eligible conventional practice that actually arises will be
the practice that actually instantiates the natural right of property in a concrete, finely grained way
that this abstract right has to be realized in order for people effectively to make things their own and
exercise discretionary control over what they have made their own. If we are so lucky as to find
ourselves operating within such an eligible set of conventional rules, respect for people’s natural right
of property requires respect for the entitlements they acquire in accordance with the rules of that
practice. Although the particular rules may be conventional, what gives them their moral force is
that they provide the needed concretization for the abstract right of property.8 (Of course, there may
be objectionable elements within a mostly acceptable practice. If so, the natural right of property
calls for the revision of those elements.)
7 Five Complexities Concerning Property Rights

We need to take note of five ways in which a sound libertarian doctrine about justice in holdings is
more complex than has so far been indicated.
337
Eric Mack
Just Restitution. The libertarian perspective does not endorse whatever distribution of income
has arisen within mixed economic orders like those that exist in the United States and Britain. Many
existing coercive state policies such as tariffs, occupational licensing, zoning regulations, and im-
migration restrictions have benefited some individuals – often relatively rich individuals – at the
expense of other individuals – often relatively poor individuals. To the extent that the benefited and
the harmed individuals can be identified along with the degree of benefit or harm those who have
benefited should be required to make restitution payments to those who have been injured. Indeed,
those who knowingly or recklessly have promoted or enforced those policies should be held
criminally liable. Similarly, following the American Civil War, former slaveholders and their
henchmen should have been required to compensate those that they had held in slavery and they
ought, as a matter of justice, to have been punished for their moral crimes. Just as past permissible
acquisition can give rise to current entitlements, past impermissible acquisition can give rise to valid
claims to restitution.
Nevertheless, this is a long way from endorsing proposals for wide-ranging enforced reparation
payments from members of certain racial or ethnic groups to members of other racial or ethnic
groups on the basis of injustices inflicted, say, two or more generations in the past. The most obvious
problem for such proposals is the difficulty of identifying which descendants of victims of injustice
are now worse off and to what extent than they otherwise would be and which descendants of those
who originally inflicted or benefited from the injustices are now better off and to what extent than
they otherwise would be. It is crucial to note that a case has to be made both for the transmission of
claims to compensation for past injustices suffered and for the transmission of obligations to com-
pensate for those past injustices. Moreover, the focus has to be upon validating assertions of individual
claims and individual obligations because an injustice inflicted on some member of a given group
does constitute an injustice done to the group as such and, hence, to every (subsequent) member of
that collectivity. And an injustice performed by some member of a group does not constitute an
injustice performed by that group as such and, hence, by every (subsequent) member of that col-
lectivity. Libertarians strongly reject the notions of collective victimization and collective guilt.
This is not the place to detail the difficulties that attach to establishing claims of corrective justice
on the basis of somewhat or quite temporally remote injustices. Still, it is worth mentioning one
important and overlooked consideration in determining whether certain individuals are or have
been beneficiaries of injustices performed by others. Let us consider again the horrendous case of
slavery in the United States. It is sometimes claimed that a significant proportion of the white
inhabitants of non-slave states benefited from slavery because in one indirect way or another their
incomes derived in part from participating in a network of commerce that included slaveholders and
their Northern business associates. However, this sort of claim ignores the opportunity costs borne
by such supposed beneficiaries. The opportunity costs are the gains that would have accrued to those
people had slavery been abolished decades earlier and the former slaves been allowed to seek their
own economic well-being on an equal footing with everyone else. Because those former slaves
would have been markedly more economically productive than they were as slaves, the gains to
individuals participating in a network of commerce that included those former slaves would have
been greater than their indirect gains under the perpetuation of slavery. Hence, those supposed to be
net indirect beneficiaries of slavery were actually net losers.
The Non-Diminished Opportunity Proviso. Both Locke and Nozick hold that acquisitions that
worsen the situation of others in certain ways do not engender unencumbered property rights (Locke,
§27; Nozick, 178–82). I take the best formulation of such a proviso to say that acquisitions will yield
unencumbered property rights only if no one’s opportunity to make use of her productive powers to
advance her chosen ends is on net diminished from what it would be were the acquired objects left
unowned. So, an individual who does what would ordinarily yield an unencumbered property right to
a donut-shaped parcel of land does not acquire a right to forbid another person who resides in the hole
338
of the donut to cross that individual’s land. Nor may all the employers in a given area join together to
exclude some individuals from gainful employment. And so on (Mack, 1995).
The Dire Straits Proviso: individuals who, without significant fault, find themselves in dire straits
such that in order to stay alive they must do what would ordinarily be a violation of another’s property
rights are not obligated to die. If someone must choose between freezing to death in a freak wilderness
blizzard or breaking into a remote wilderness cabin, it is permissible for that person to break into the
cabin. What is permissible is the least intrusive life-preserving action. Because it is less intrusive to break
in and subsequently compensate the cabin owner than to break in and not compensate, only breaking-
and-compensating is permissible for those who can subsequently compensate.
The Voluntary Joint-Ownership: libertarian theorizing about justice in holdings often focuses
entirely on private property rights understood as the rights of particular individuals to discrete pos-
sessions. However, nothing in a sound libertarian doctrine excludes voluntary joint-ownership.
Individuals are entirely free to merge their discrete possessions into bundles to which all become joint
shareholders. And individuals are entirely free to engage in joint-action through which joint entitle-
ments are engendered. The crucial limit here is that joint-action can generate entitlements only
through the same sort of processes that yield individual entitlements when carried out individually.
The Attenuation of Rights: for libertarian theorists in general, and especially for natural rights
theorists, prototypically moral rights are claims that are protected by property rules. This means
rights are “no trespass” signs. Others are morally forbidden to cross the moral boundaries that one’s
rights specify. An attenuated form of rights are moral claims that are merely protected by liability
rules. This means that the relevant boundaries may permissibly be crossed as long as the crossing
party duly compensates the holder of the (attenuated) right. One can hold, however, that although
the default form of moral rights (including property claims) is their being protected by property
rules, under certain special conditions, a valid entitlement can take the weaker form of a claim
protected by a liability rule.
Suppose I see that a shed in your backyard is on fire and that the fire will soon spread to your
house. In the next few minutes, I can rush into your yard, tear down the shed, and save your house.
Now, you have an entitlement to that shed which normally means that I may not tear it down
without your consent. But suppose that I have no way to contact you and seek your consent. It is
plausible to think that requiring your consent makes sense on the assumption that this consent is
feasible but that, in cases in which your consent is not feasible, to require your consent is to lock you
into the position that you disfavor. So, in such cases, my crossing the boundary defined by your
entitlement may be permissible as long as you are duly compensated for that crossing – as you would
be if your house is saved from the fire. The point here is not to work out the details of this stance
but, rather, to indicate that under certain special circumstances, an entitlement may take the less
stringent form of a moral claim protected by a liability rule (Mack, 2011).
Through the addition of a natural right of property and a recognition of these five further
complexities a libertarian view about the justice of holdings becomes richer, more nuanced, and, I
believe, better equipped to meet philosophical objections directed against it.
Notes
1 I thank Mary Sirridge and Christopher Melenovsky for their helpful comments on early drafts of this paper.
2 Later in his discussion, Nozick uses “patterned” to denote doctrines of either sort (Nozick, 1974, 160).
3 See Mack, 2002, 79–91, and Mack, 2018, 84–8.
4 Advocates of alternative distributivist principles might similarly insist that their principles are not to be
directly applied on a moment-to-moment basis but, rather, to the selection of a basic structure (or a series of
readjusted basic structures).
5 See Mack, 2018, 124–9.
6 Also see Mack, 2018, 76–92.
339
Eric Mack
7 For a much more extensive discussion, see Mack, 2010.

8 For more on valid conventional rules as concretizations of abstract moral rights, see Bryan, 2017, and Mack,
2018, 130–7.
References
Bryan, Ben. (2017) “The Conventionalist Challenge to Natural Rights Theory,” Social Theory and Practice. 43, 3
(2017), 569–587.
Locke, John. (1980) Second Treatise of Government, ed. C. B. Macpherson. Hackett Publishing.
Mack, Eric. (1995) “The Self-Ownership Proviso: A New and Improved Lockean Proviso,” Social Philosophy
and Policy. 12, 1, 186–218.
Mack, Eric. (2002) “Self-ownership, Marxism, and Egalitarianism, Part I,” Politics, Philosophy, and Economics. 1,
1, 75–107.
Mack, Eric. (2010) “The Natural Right of Property,” Social Philosophy and Policy, 27, 1, 53–79.
Mack, Eric. (2011) “Nozickian Arguments for the More-Than-Minimal State,” Cambridge Companion to
Anarchy, State, and Utopia, Ralf Bader and John Meadowcroft eds. University of Cambridge Press.
Mack, Eric. (2018) Libertarianism. Polity Press.
Mack, Eric. (2022) “Natural Rights,” The Routledge Companion to Libertarianism, eds. Benjamin Ferguson and
Matt Zwolinski. Routledge.
Nozick, Robert. (1974) Anarchy State and Utopia. Basic Books.
Rawls, John. (1993) Political Liberalism. Harvard University Press.
340
27
HIGH LIBERALISM
Samuel Freeman
1 Classical Liberalism and the Evolution

of the High Liberal Tradition
The term “classical liberalism” is here used broadly to refer to an ideal type of liberalism that
originates in the 17th and 18th centuries and develops to full fruition in the 19th and early-20th
centuries – with Locke and Kant being major philosophical advocates within the social contract
tradition, and Hume, Adam Smith, Jeremy Bentham, and John Stuart Mill in his earlier economic
works being major advocates within utilitarianism. Classical liberalism involves the following the-
oretical essentials (rarely realized in practice): (1) basic rights and liberties that include freedom of
religion, association, thought and expression, personal freedoms of tastes and pursuits, and basic
economic freedoms of contract and transfer combined with robust rights of private ownership of
property, including productive resources; (2) freedom of occupation and choice of workplace with
formal (legal) equality of opportunity and “careers open to talents”; (3) competitive economic
markets with minimal regulation of contracts and exchanges; (4) government’s role in providing a
narrow range of public goods, including national defense, public security, public health measures,
highways, ports, and other means necessary for free commerce, and publicly funded basic education;
(5) a social safety net that provide means of subsistence for those unable to work or otherwise
provide for themselves; and finally (6) recognition of the public nature of political power, which is
impartially exercised for the public good.
High liberalism is the progressive democratic development of the classical liberal tradition from
the mid-19th century up to the present.1 It has three main philosophical sources: first, high liber-
alism developed as a liberal response to democratic ideas and movements starting with the French
Revolution and influenced by Rousseau’s works on social equality and democracy. Second, high
liberalism was inspired by 19th-century ideas of autonomous self-determination and individuality
developed within German Idealism, and by J. S. Mill and others. Third, high liberalism was further
prompted by socialist critiques of laissez-faire capitalism and classical liberalism, and responses to
these critiques by J. S. Mill, T. H. Green, John Dewey, and others.
The progressive development of high liberalism is foreshadowed in the early 19th century French
classical liberal Benjamin Constant’s distinction between the “liberties of the moderns” and “liberties
of the ancients.” The former included personal freedoms of belief and conduct and rights of private
property, while the latter included rights of citizens (then only white males) to political participation
DOI: 10.4324/9780367808983-33 341

Samuel Freeman
and influence. High liberalism combines and extends the basic personal “liberties of the moderns”
together with equally basic political liberties, grounding both in a democratic conception of free
citizens who are social and political equals.
High liberalism is primarily distinguished from classical liberalism, including liberal libertarianism,
in the following respects:
A. High Liberalism regards equal political rights of participation as essential to social equality and thus
as included among the basic rights and liberties. Moreover, it guarantees citizens fair oppor-
tunities to exercise equal political influence, regardless of wealth or social class.
B. Apart from basic freedoms of occupation and rights of personal property sufficient for in-
dependence, high liberals regard economic liberties, ownership, and control of productive
economic resources as non-basic rights. The terms of ownership and control of the means of
production are to be democratically determined according to an appropriate conception of
distributive justice that mitigates inequalities in order to better enable citizens to meet basic
needs, effectively exercise basic liberties, and maintain their social and political equality.
C. High liberalism guarantees not simply formal equality of opportunities to compete for open
positions but substantive or fair equal opportunities for all citizens. This enables members of society
to develop and educate their innate capacities so that they can experience the benefits of culture
and compete on fair terms for educational, employment, and other social positions.
D. High Liberalism accepts a predominately competitive market system for the allocation of pro-
ductive resources, including labor. However, it regulates markets to prevent the concentration of
wealth and market power, provide occupational health and safety measures, prevent the mis-
treatment and unfair exploitation of employees, and guarantee conditions enabling workers to
organize into labor unions and worker cooperatives.
E. High liberals endorse a non-market criterion of distributive justice that both (1) guarantees all so-
ciety’s members rights to all-purpose means sufficient to satisfy basic needs, effectively exercise
basic rights and liberties, take advantage of opportunities, and maintain personal independence,
and (2) limits economic inequalities to the degree necessary to maintain the social and political
equality of citizens.
F. High liberals have an expansive conception of public goods, corresponding with its conception of the
shared common interests of free and equal democratic citizens.
G. High liberals have a broad conception of the role of public political power in maintaining justice and
promoting the public good. Moreover, high liberals interpret public political power to require the
public political justification of laws and their application, which is to be conducted in terms of shared
public reasons reflecting the common interests of free and equal democratic citizens.
In the following sections, I elaborate on each of these distinctive differences between high liberalism
and classical liberal and libertarian positions.
2 Basic Rights and Liberties, Effective Freedoms and

the Principle of Basic Needs
Liberalism is not about “maximizing” liberty as such; were that its aim, private property would hardly
be possible since it constrains the freedom of everyone except owners. Liberals, however, do make a
presumption of liberty, which can be outweighed only for certain kinds of reasons. Basic liberties are those
liberties that can be restricted only for reasons of protecting other basic liberties and in order to maintain a
system of basic rights and liberties fully adequate to the common fundamental interests of citizens. Basic
liberties then cannot be restricted for the sake of majority will, economic efficiency, greater social
welfare, or perfectionist values. By contrast, non-basic rights and liberties can sometimes be legitimately
342
High Liberalism
restricted for these reasons (as with zoning restrictions on land use, or financial, corporate, and securities
regulations). Basic liberties are primary among the institutional conditions necessary to realize the high
liberal understanding of the freedom, equality, and independence of persons as citizens in a democratic
society.
The equal basic rights and liberties commonly recognized by high liberalism as necessary to
sustain persons as effectively free, equal, and independent members of the society include:
A. The personal liberties. These include (a) liberty of conscience in moral, philosophical, and re-
ligious convictions regarding questions of value and morality, the meaning of existence, and
how one ought to conduct one’s life. This liberty includes the freedom to practice and act on
one’s conscientious religious, moral and philosophical convictions, consistent with others’ basic
rights and liberties; (b) freedom of thought and feeling regarding all subjects and its expression
in speech, inquiry, discussion, art, etc.; (c) freedom of the person to act on their beliefs, satisfy
their tastes and preferences, and to pursue their primary aims, life-plans, and occupations; (d)
freedom of association – the freedom to unite, publicly or privately, with others in groups and
in personal associations of love, friendship, and other interpersonal relationships, with a right of
privacy in intimate, family and other associational relations; (e) the right to hold personal
property, as is necessary for individual independence and self-respect; without the security of
personal possessions and residence, one cannot effectively pursue aims and plans; (f) rights to the
physical security and psychological integrity of the person, with the protections of the rule of
law (due process, equal protection, right to fair trials and against self-incrimination, no arrest
without reasonable cause, no unreasonable searches and seizures, and so on).
Basic rights and liberties are institutional freedoms and guarantees that are legally protected and that
have priority over other political considerations. Both equality of basic rights and liberties and in-
dependence of the person requires freedom from external interference while exercising one’s legitimate
rights and liberties. But freedom from interference is not enough. Also required is access to adequate
resources enabling citizens to effectively exercise basic rights and liberties and take advantage of formal
opportunities to occupy social and political offices and positions. A principle of basic needs is for high
liberals a precondition for the effective exercise of equal basic liberties. The legal guarantee of liberal
freedoms is of little if any value to persons who do not possess adequate means and capacities for their
effective exercise. A homeless person legally may be equally free as the wealthy to sleep under bridges
and scavenge others’ refuse for their sustenance, but otherwise formal freedoms are hardly of any value
to them. If the formal liberties and opportunities recognized by a liberal society are to be effectively
exercised and taken advantage of, then individuals must have at least their basic needs met and real
prospects to adequately develop human capacities so they can take advantage of diverse opportunities
within the liberal culture and democratic society. High liberals then recognize that a precondition for
the effective exercise of both basic and non-basic rights and liberties is a constitutionally guaranteed
social minimum of economic, educational, and other social resources.
B. The Political Liberties: high liberals contend that democratic rights and liberties of political
participation are equally fundamental and inalienable as are the basic personal rights and lib-
erties. Basic political rights and liberties include equal rights to vote, hold office, politically
assemble, publicly express political opinions and grievances; form and join political parties, and
special protections given to freedom of political speech and expression. These equal political
rights are not simply formal on a high liberal view. They are to be combined with the further
substantive guarantee of measures that promote fair opportunities for equal political partici-
pation and influence. The significance of political rights of participation to high liberalism is
discussed more fully below in §III on political equality.
343
Samuel Freeman
Contemporary classical liberals agree with high liberals on most, if not all, of the foregoing formal
personal liberties, though they often interpret them and their significance differently. Classical
liberals do not endorse a right to a social minimum but see it as undeserved public charity.
Moreover, they do not regard equal political rights as basic or having the same priority as personal
liberties. Political rights are largely seen as instrumental to protecting basic liberties; democracy is the
least bad form of government so long as basic personal and economic liberties are protected from
majority rule. Classical liberals’ rejection of basic political liberties marks their first fundamental
difference with high liberal views. Instead of democratic political liberties, classical liberals regard as
basic robust economic rights of private property in productive resources and freedom of economic
contract, transfer, and exchange.
C. The Economic Liberties: for High liberals, freedoms of occupation, freedom of movement, the
right to hold personal property, and the guaranteed social minimum are basic economic rights
and liberties broadly construed. These economic rights are basic because they are basic personal
liberties too, subsumed under freedom of conscience, freedom of the person to form and pursue
their life plans, and the necessary conditions for the effective exercise of basic liberties and
individual independence. Other economic rights and liberties, including private ownership of
means of production and rights of exchange and transfer are non-basic; they can be restricted
or regulated for reasons of fairness, economic efficiency, and the public good, and are subject to
requirements of economic justice that do not apply to the basic personal and political liberties.
This marks the second fundamental difference with classical liberalism, which sees extensive
private economic rights and liberties as equally basic as personal liberties.
One reason for this discrepancy is that high liberals have a more complex conception of property and
its dual function. The rights of personal property are basic since they are necessary for individual
security, independence, self-respect, and the effective exercise of the basic rights and liberties that
enable individuals to pursue their fundamental purposes, relationships, and life plans. By contrast, non-
personal property – involving large-scale ownership and control (whether private or public) of means
of production that other persons are employed to work with – does not bear the same crucial relationship
to individuals’ personal security, independence, self-respect, and effective exercise of their liberties and
capacities for freedom. Indeed, high liberals regard the classical liberal doctrine of laissez-faire as a major
impediment to the full independence of free and equal persons as it encourages employee subservience
and substantial inequalities that undermine political equality, fair equal opportunities, and self-respect.
The default “take it or leave it” laissez-faire labor contract typical of minimum wage employment ties
workers into submissive work conditions with no discretion or self-control. The fact that workers
consent to servile work conditions requiring suspension of personal liberties is little solace when all
positions open to them come with the same restrictions.
High liberals contend that extensive economic rights and liberties are not among the basic rights
and liberties of free and equal persons in a democratic society. Instead, economic rights of ownership
and control, economic freedom of contract, and the structure of the economy are to be determined
by independent considerations of economic and distributive justice, grounded in fairness and re-
ciprocity (Rawls), or the promotion of individual welfare (Mill and restricted utilitarians), or other
egalitarian values.
Classical liberals argue that robust capitalist economic liberties are basic since necessary to pre-
serve both personal and political liberty.2 But social democratic and regulated capitalist welfare states
do quite well without unfettered capitalism. Moreover, for a right or liberty to be truly basic in the
sense required by the priority of equal liberty means it cannot be restricted for the sake of economic
efficiency or social welfare, but classical liberals themselves (unlike orthodox libertarians) often
recognize the need to restrict property and contract rights in order to maintain efficiency and the
344
High Liberalism
fluidity of competitive markets. Finally, on high liberal views, for a liberty to be basic, it must be true
that for reasonable and rational persons generally, a basic liberty is normally essential to each person’s
full and informed exercise of their moral and rational capacities for agency and social life – to their
“individuality” (Mill, 1859), or their “moral powers” (Rawls, 1999), or other liberal conceptions of
autonomy and self-determination. It is difficult to see how access to laissez-faire rights and liberties
in a capitalist society is essential to the full and informed exercise of the capacities for rational moral
agency and social cooperation of the large majority of employed people in society, especially for
workers who do not enjoy rights of ownership and economic control of their means of production.
This is especially true when that capitalist society devalues democracy and the fair value of political
liberties and imposes no restrictions on economic inequalities.
3 The Liberal Significance of Political Equality and

the Integrity of Democracy
There are several reasons for the political equality of democratic citizens. Having rights of political
participation and influence are normally necessary for citizens to protect their other rights and
interests. Individuals without political rights are the first to have their interests compromised and
rights infringed. Moreover, the exercise of equal political rights and liberties and citizens’ active
involvement in political thought and discussion furthers the full and informed exercise and devel-
opment of individuals’ social and deliberative capacities, particularly their moral sense of right and
justice (Rawls, 1996, Ch. 8). Finally, even when political rights are left unexercised (as so often
among the least advantaged) political equality is among the necessary conditions for social equality and equal
respect among citizens in a democratic society. A democratic society’s members regard themselves not
simply as morally free and equal persons but also as equal citizens who have legitimate claims to be
recognized as socially equal with other members of society. As such, they should have the same basic
rights, liberties, and formal access to opportunities. Political equality is then a condition of main-
taining the social equality of members of a democratic society; it encourages mutual respect among
society’s members for one another as equals and is, therefore, a condition of citizens’ self-respect
(Rawls, 1999, §82).
Equal rights of political participation require that governments guarantee that citizens who are
similarly motivated and able to have roughly an equal chance to influence government and occupy
positions of political authority irrespective of their social and economic class (Rawls, 1996, 358).
Political equality does not require strict equality of political influence, which is unrealistic given
many differences of talents, circumstance, interests in politics, personality, and renown (Dworkin,
2011, 388). But political equality does require a good faith effort to substantially mitigate the ar-
bitrary influence of economic inequality on elections and the political process. It also requires that
government avoid political dilution of voting power by political gerrymandering, voter suppression
measures that limit access to polls and discourage citizens from voting, and trumped-up allegations of
voter fraud. Political equality further requires the public financing of political campaigns, with
measures that restrict contributions to and expenditures by political campaigns and political parties.
Government has an obligation to maintain the integrity of the public political forum so that it is not
distorted by lies and misinformation tactics by political propaganda that undermines confidence in the
political process. Freedom of political speech should not protect the freedom to persistently lie and
otherwise intentionally deceive citizens with a purpose of undermining democracy. This violates the
political freedom and equality of citizens. Concerted misinformation and malicious lies undermine the
integrity of the political process and deprive voters of the fair value of their political rights – their equal
opportunity to influence the political process and occupy positions of authority. The onslaught of
political propaganda that now characterizes American politics – encouraged and orchestrated by a former
President – is characteristic of fascist politics and is a gross violation of citizens’ equal opportunities to
345
Samuel Freeman
influence a fair and informed political process. The liberal justification of freedom of thought and
political expression is itself undermined by an ongoing campaign of malicious political slander and
propaganda. Those responsible should be civilly penalized for their malicious lies and defamation.
Constitutional democracy cannot survive fascist propaganda onslaught.
4 Substantive Equality of Opportunities

Equality of opportunity is also a fundamental condition of social equality. But liberals have different
conceptions of the background conditions necessary for it. There are three distinct versions: (1) the
classical liberal conception of formal equality of opportunity with careers legally open to talents; (2) the
high liberal conception of substantive or fair equality of opportunities for educational development and
employment, with substantive opportunities to take advantage of the benefits of liberal society and
culture; (3) the luck egalitarian conception of perfect equality of opportunities for welfare, which seeks to
equalize the effects of differences in natural talent, fortune, and misfortune, by redressing unequal
outcomes and compensating for arbitrary disadvantages.
1. Formal equality of opportunity fits with the classical liberal conception of non-coercion and basic
economic liberties. According to this ideal, there should be no legal requirement of dis-
crimination on grounds of race, religion, gender, sexual preference, or other characteristics that
are not relevant to meeting the qualifications required for social positions. Allegedly, “the
Constitution is color-blind” when this ideal is realized, and it forbids legally enforced dis-
crimination and segregation. But since economic liberties of contract and rights of property are
basic rights for classical liberals, the government has no authority to prohibit private employers,
businesses, or accommodations from discrimination and refusal to do business with racial,
ethnic, religious, sexual, and other minorities. As Milton Friedman argued, in opposing the
predecessor of the 1964 Civil Rights Act, liberal freedom requires “cooperation without
coercion”; hence compelling employers and businesses to contract with and give equal con-
sideration and service to people they do not want to deal with is a violation of basic economic
liberties and also basic freedoms of association (Friedman, 1962, Ch. 6). Formal equality of
opportunity also forbids legally enforced preferential treatment that benefits racial, ethnic, and
gender minorities in order to achieve the social equality that the government has long denied
them ever since the origins of reputedly democratic governments and societies.
Like their purely formal interpretation of basic liberties, classical liberal equality of opportunity carries
no substantive guarantee of the worth of formal equal opportunities for open positions. There is no
political guarantee of the effective exercise of rights to take advantage of opportunities to compete for
and occupy social positions and enjoy the cultural benefits of a liberal and democratic society.
2. High liberal substantive equality of opportunity gives effective force to the formal equalities of
opportunity guaranteed by law in a liberal society. Because economic rights and liberties are not
basic for high liberals, they can be delimited for reasons in addition to protecting and main-
taining other basic rights and liberties. Economic rights and liberties can then be regulated and
restricted to guarantee conditions of social equality for all members of society. A precondition
of social equality is that citizens have real and not merely formal opportunities to apply and
compete for open positions; that they be treated fairly and without discrimination in economic
transactions and be accepted as equal members of society. There is then no violation of basic
economic liberties in requiring private employers and businesses to give equal consideration and
engage in business transactions with job applicants and customers regardless of their racial,
ethnic, national, religious, or sexual classification.
346
High Liberalism
Moreover, Friedman’s argument that the basic freedom of association is infringed by Civil Rights
Acts prohibiting private racial discrimination in education, business, and employment mistakes the
reasons for freedom of association. This basic liberty is a personal, not an economic, liberty. It is
necessary to sustain and exercise freedom of conscience and the practice of religious, philosophical,
moral, and evaluative convictions and tastes. Also, citizens in their personal, private, and intimate
associations should be free of government interference when they decide whom they befriend and
associate with. These extensive personal freedoms of association do not extend to economic con-
tracts. Individuals normally have the non-basic liberty to enter economic contracts with whom they
choose, but they may not refuse to do so for reasons that undermine the fair opportunities and social
equality of other citizens (Freeman, 2018, 41).
Substantive equality of opportunity also guarantees citizens social and economic benefits to
develop and educate their capacities so that they can lead not only productive but also meaningful
lives and take advantage of the many professional and cultural opportunities in a diverse liberal and
democratic society. Publicly funded education, rather than being justified on grounds of economic
efficiency, is a social right among high liberals, guaranteed by substantive equality of opportunities,
as are prenatal and child development programs, job training and retraining, and universal health
care for all members of society (Scanlon, 2018, Ch. 5). Moreover, many public goods and services
are justifiable as among the conditions for achieving fair equal opportunities for all citizens: including
public schools, substantive funding for higher education, public health measures, public funding for
infrastructure including highways, railroads and public transportation, communications networks,
electricity, funding scientific research and discovery, and so on. The fundamental idea behind
substantive equality of opportunity is that the members of society should all have fully adequate
resources to enable them to develop their capacities and take advantage of the diverse educational,
employment, and cultural opportunities within a modern liberal society. This is not simply a
condition of social equality, but also among the conditions for citizens’ effective exercise of their
basic liberties.
3. Perfect Equality of Opportunity: the high liberal conception of equal opportunity does not
however aim to achieve the impossible, namely, perfect equality of opportunities to compete
for educational, employment, and cultural positions. Rawls says there is no way to approximate
perfect equality so long as the family exists (TJ 448); for family life has enormous consequences
for the development of children’s capacities, skills, interests, and emotional well-being. Some
form of the family is necessary to nurture, raise and care for the young in order that any society
can perpetuate itself across generations. Individuals inevitably will have different life prospects
for happiness, well-being, and success or failure because of their family backgrounds, as well as
their natural and social differences of birth and class, and accidents of fortune and misfortune
they are not responsible for. The aim of luck egalitarian equality of opportunity is to com-
pensate those who are disadvantaged in these respects, which normally requires taxing or
disadvantaging others who have undeserved advantages they are not responsible for. Perfect
equality of opportunity for welfare is not really so much about establishing liberal equality of
opportunities for education, employment, culture, and political influence. Rather, it is about
equalizing individuals’ opportunities for happiness and well-being, so far as this is feasible. This
is not a distinctively liberal or democratic position, but a more general egalitarian welfarist
position that can be applied to any society, liberal, or illiberal.
5 A Regulated Market Economy

Predominant reliance on free competitive markets for the allocation of productive resources, including
labor, is characteristic of liberalism. Competitive markets reinforce freedom of occupation and fair
347
Samuel Freeman
equality of opportunity and decentralize the exercise of economic power (Rawls, 1999, 241).
Competitive markets also normally mitigate economic waste and enable the efficient allocation of
productive resources. Finally, liberals rely on markets and other consensual transfers as the primary
instrument for the distribution of income and wealth among members of society.
There are at least two respects in which liberals differ in the role assigned to competitive markets.
First, high liberals reject the classical liberal position that free markets and other consensual transfers
provide the fundamental criterion for determining the just distribution of income and wealth – a
consequence of classical liberals’ assumption that complete rights of property and its transfer are basic
rights and liberties. Classical liberals normally endorse a conception of pure procedural justice that holds
that the distributive outcomes of competitive markets and other consensual transfers are pre-
sumptively just, barring force and fraud. High liberals do not reject pure procedural justice, but they
do reject the laissez-faire conception of full property and transfer rights that ground the classical
liberal conception. High liberalism’s second difference with respect to the role of markets is that they
reject the classical liberal presumption of laissez-faire freedom of contract; instead, they endorse the
democratic regulation of markets and contractual relations when needed to maintain health and
safety of workers and the public and realize other public goods, and when required by a high liberal
criterion of distributive justice.
J. S. Mill, John Rawls, and the economist James Meade all distinguished between the allocative
vs. the distributive role of markets. Mill says that laws of economic production are of universal
applicability, whereas laws of distribution are conventional. The laws of production “partake of the
character of physical truths,” whereas “the Distribution of Wealth… is a matter of human institution
solely. The things once there, mankind, individually or collectively, can do with them as they like”
(Mill, 2006, 199).
Similarly, Rawls distinguishes between the role of market prices in allocating productive resources
versus their role in distributing income and wealth. “The [allocative function] is connected with their
use to achieve economic efficiency, the [distributive function] with their determining the income to be
received by individuals in return for what they contribute.” In using the market to allocate productive
factors, “prices are indicators for drawing up an efficient schedule of economic activities …. It does not
follow however that there need be private persons who as owners of these assets receive the monetary
equivalents of these evaluations” (Rawls, 1999, 241). High liberals assume that the efficient allocation
of productive resources does not require that the final distribution of income and wealth also be
determined by market distributions resulting from efficient allocations of resources.
Classical liberals including Hayek challenge the separability of market functions (Hayek, 1960,
96), contending that the allocative efficiency of markets necessitates also relying on markets as the
fundamental standard for the distribution of income and wealth, and that redistribution for purposes
other than remedying externalities undermines the efficiency of markets. This is surely an empirical
question of the effects of taxation and redistribution on allocative efficiency and the elasticity of
supply and demand. Also, Hayek appears to presuppose that a utilitarian standard of maximizing
overall economic welfare underlies the fair distribution of income and wealth, and that this can only
be realized through invisible hand market distributions. For high liberals, economic efficiency and
maximization of overall wealth is not the proper standard to assess the just distribution of income
and wealth or the regulation of market transactions.
6 Distributive Justice and the Social Minimum

High liberals reject unregulated consensual transfers of laissez-faire property rights as providing the
baseline and fundamental standard for determining the just distribution of income and wealth. Free
markets and unregulated transfers are prone to too many morally arbitrary contingencies to provide
the fundamental standard for the just distribution of income, wealth, and economic powers and
348
High Liberalism
prerogatives. Among these are arbitrary differences in natural talent, social starting positions, re-
lationships, individual accidents of misfortune, fluctuations of the supply and demand for labor and
productive resources, climate change and environmental disasters, and many other forms of good
and bad “brute luck.” As a result, the gradual tendency of the “invisible hand” favors oligopolistic
accumulation and is away from rather than towards distributive justice (Rawls, 1996, 267). Rather
than endorsing the arbitrary outcomes of markets and consensual transfers as just, high liberals in-
stead endorse, first, a right to a guaranteed social minimum that is more than sufficient to meet
citizens’ basic needs, enabling them to effectively exercise their basic liberties and take advantage of
employment and cultural opportunities. Second, high liberal egalitarians add that distributive justice
requires the further mitigation of economic inequalities in order to (a) maintain political equality and (b)
substantive equal opportunities, and furthermore (c) meet requirements of a separate egalitarian
standard of distributive justice. Among these are Rawls’s difference principle, the liberal luck
egalitarianism of Ronald Dworkin, G. A. Cohen, John Roemer, Philippe van Parijs, and others; and
restricted utilitarian or prioritarian principles that give added weight to the position of the least
advantaged (Richard Arneson et al.).
a. The Social Minimum: classical liberals and some liberal libertarians can recognize a limited social
minimum – a “social safety net” adequate to meet at least basic subsistence needs as con-
ventionally understood. But this is not a political right or entitlement required by justice.
Instead, the social minimum might be conceived as a duty of public charity towards “the
deserving poor” due to their misfortune. An alternative position, voiced by Thomas Hobbes, is
endorsed by classical liberals such as Hayek: a safety net is needed to maintain peace and
lawfulness, to prevent widespread theft and other crimes and unrest among the poor. Still other
classical liberals such as Friedman argue that public charity is a public good as all benefit from
the absence of the consequences of extreme poverty such as starvation, threats to public health,
or unsightly destitution. In any case, the basic needs of beneficiaries of public charity are thinly
defined, and the social minimum is not determined by the high liberal standard that it be
adequate to enable people to effectively exercise their basic liberties and take advantage of
substantive opportunities. Instead, the neo-liberal basic minimum is restricted to provide
adequate means of subsistence, while not creating disincentives to work at the minimum wage.
For high liberals by contrast, membership in society carries with it a guaranteed entitlement to a
social minimum – a property right to a sufficient level of economic resources. As discussed earlier, this is
essential to maintain high liberal ideals of effective equal freedoms – citizens’ capacity to effectively
exercise basic liberties, maintain their individual independence and sense of self-respect, and take
advantage of substantive opportunities. The guaranteed social minimum is also among the condi-
tions of the social and political equality of democratic citizens. The exercise of basic liberties is
effective only when citizens have educational and cultural opportunities to freely develop their
capacities and are potentially capable of choice and pursuit of a wide range of occupations and
permissible conceptions of the good.
There are multiple means and social programs to provide the guaranteed social minimum high liberals
defend: generous educational benefits for all and universal health care are essential. Some advocate a
universal basic income; or a means-tested minimum income with income supplements and family al-
lowances for children; or means-tested in-kind benefits (food vouchers, publicly funded housing or
housing vouchers, etc.). There are numerous possibilities and combinations of such entitlements.
b. Distributive Justice, the Mitigation of Inequalities, and Permissible Inequalities: the basic social minimum
is guaranteed to all members of society, regardless of whether they work or not. It is an essential
condition of citizenship and being a free, equal, and independent member of society. For persons
349
Samuel Freeman
not fully capable of exercising basic liberties or taking advantage of opportunities, members of any
society have a basic human right to have their essential needs met. The chronically ill, mentally
and physically incapacitated, and others with serious disabilities should have their basic needs met,
regardless of the society they live in, liberal or illiberal. A basic social minimum is then among the
primary moral duties of justice that any society owes its members.
But beyond these guarantees to meet basic needs and enable the effective exercise of basic freedoms,
there remains the pressing question of the fair distribution of income and wealth among society’s productive
members, who cooperate and do their fair share to contribute to the joint social product. How is a
society (whether domestic or global) to determine the distributive shares owed to those who
contribute their fair share of time, efforts, and savings to the joint production of income and
economic wealth that results from economic cooperation and activity among the members of that
society? This is the question of distributive justice in the narrow sense (Rawls, 2001, 61). This question
motivated the classical liberal utilitarian economists who argued for laissez-faire organization of the
economy with rewards distributed to economic agents according to the (marginal) value of their
contributions. This question also motivated the socialist critique of laissez-faire capitalism. It claimed
that the capitalist class contributes little more than pure ownership of means of production, which is
but a convenient legal fiction, and that only labor produces economic value when combined with
inanimate productive resources; consequently, workers should be rewarded according to their ef-
forts, including labor time. The high liberal position takes a third path that is different from both
classical liberal and traditional socialist positions. It says that while economic agents are indeed to be
rewarded for their efforts and other contributions including savings and risks undertaken, these
questions are to be determined primarily with an eye to maintaining fairness and the social and
political equality of members of society.
Rawls’s difference principle is a notable example. It says that economic agents are duly rewarded for
their efforts and contributions when they participate in the processes of an economy that is orga-
nized to satisfy a principle of fair reciprocity. The difference principle says in effect that distributive
shares of income, wealth, and economic powers and prerogatives are fully just when they are the
outcome of an economy that is organized so that its least advantaged economic agents are better off
(in terms of their share of income, wealth, and economic powers and prerogatives) than they would
be in any alternative economic system. Rawls contends that the economic system that achieves this
goal is either a property-owning democracy or liberal socialism. The former widely distributes
ownership and control of productive wealth among all working members of society, enabling them
to gain a share of income (profits, dividends, interest) from ownership of means of production.
Liberal socialism permits public ownership of major productive resources, which are managed by
free associations of workers who bid for use and control, and earn income from their labor and
dividend shares of the firm’s earnings. What is distinctive about these two economic systems is that
they require the dissolution of concentrations of capital and the elimination of a capitalist class that
exclusively controls vast amounts of productive resources and economic wealth in society.
Understood in this way, capitalism does not satisfy Rawls’s conception of economic justice because
of the vast economic inequalities it permits and the concentration of economic wealth, powers, and
prerogatives in the capitalist class. Rawls contends this is true even of welfare-state capitalism with
generous social welfare programs as it also involves large concentrations of wealth and control of
productive resources by a capitalist class.3
Ronald Dworkin, by contrast, defends welfare state capitalism from a different egalitarian position –
equality of resources – a form of liberal “luck egalitarianism.” Dworkin says that political equality is to
be guaranteed by neutralizing the effects of wealth on politics, and equality of opportunities; moreover,
starting positions for those similarly skilled and motivated are to be equalized by public education and
other measures. Inequalities in resources such as income and wealth that individuals are not responsible
350
High Liberalism
for but that result from “brute luck” – such as their natural talents, social class, and life’s accidental (mis)
fortunes – are to be equalized so far as possible. More advantaged individuals are to be taxed for
undeserved gains that result from undeserved good fortune, and those who suffer undeserved natural
and other disadvantages due to brute bad luck are to be compensated by guaranteed social insurance
measures. Beyond that, economic inequalities that result from “option luck” – including individuals’
choices among free-market options – are permissible within a competitive market system. Dworkin
provides a liberal justification for the capitalist welfare state that is an alternative to the traditional
restricted utilitarian argument (Dworkin, 2011, Ch. 16).
High liberal utilitarians have long argued in favor of a welfare state on grounds that the least
advantaged gain more utility than the more advantaged from a given sum of income or other
resources; there is then a prima facie case for moderating substantial inequalities, sufficient to provide
an adequate social minimum to meet the basic needs of the least advantaged. This focus on a
sufficient social minimum is a version of so-called “sufficientarianism.” By contrast, “prioritar-
ianism” is an alternative restricted utilitarian position that goes one step further. It advocates giving
increasingly greater weight to the welfare and needs of those less advantaged – with the welfare of
the least advantaged counting the most – in the calculation of the sum (or average) of overall utilities
in society (or the world) (Parfit, 2012). Prioritarianism justifies a greater basic minimum and social
insurance than traditional liberal utilitarianism arguments informed by the sufficiency view.
7 The High Liberal Conception of Public Goods

Public goods for political purposes are goods or services that benefit all (or at least most) members of
society, but because of their “indivisibility” – individuals can benefit from them whether or not they
pay – public goods must be publicly provided through taxation; for there are insufficient incentives
to adequately provide public goods through private transactions and market exchanges. Standard
examples include national security, protection of persons and property, the rule of law with judicial
resolution of legal disputes, lighthouses, street lighting, highways, ports and canals, public health
measures, and so on. What counts as a public good largely depends on an account of the interests of
members of society. If interests are defined purely in terms of subjective desires, then as some
libertarians argue, there are no public goods that benefit everyone, only private goods tailored to
individuals’ particular desires. Even peace, security, and the rule of law are not in the subjective
interests of many misanthropes, anarchists, mercenaries, bandits, and fascists who thrive on others’
woes, social unrest, piracy, or rampage.
The idea of public goods must therefore abstract from the subjective desires of evil, misanthropic,
or irrational people and presuppose a normalized conception of persons and their objective interests.
Economists normally assume a model of a rational self-interested “economic man” to define public
goods, who is motivated to maximize its self-focused utility. Similarly, classical liberals define public
goods by a narrow range of fundamental interests that all rational and normally self-interested
persons are presumed to have – security of self, family, and property; personal and economic
freedoms; and acquiring income and wealth.
High liberals have a different conception of persons and their fundamental interests that informs a
richer conception of public goods and the duties of government to provide for them. The funda-
mental interests of free and equal persons normally involve not just personal interests, but also certain
moral and social interests. For J. S. Mill, our fundamental interests include security, freedom to
pursue one’s good, “self-development” of one’s “higher faculties,” and “self-government” ac-
cording to the rules of justice (Mill, 1861, Ch. 5; Mill, 1859, Ch. 3). For Rawls, the fundamental
interests of “free and equal moral persons” are defined in Kantian terms as the full and informed
exercise of their capacities for practical reasoning and social cooperation: (a) the rational capacity to
form, revise and pursue a conception of the good, and (b) the social capacity to understand, apply,
351
Samuel Freeman
and act on requirements of justice. Such fundamental interests inform high liberal accounts of the
common good of democratic citizens and therewith the public goods needed to realize the common
good. If all citizens have a fundamental interest in justice and in the development of their rational
and moral capacities for social cooperation, then the public institutions necessary to realize and
maintain justice and the common good are primary among the public goods. For example, to ensure
fair opportunities for equal political influence, the government is obligated to maintain the public
good of a fair political process with a public political forum for debate designed to truthfully inform
citizens of political issues. Universal health care and extensive educational opportunities, including
retraining to maintain full employment, are also public goods necessary to a substantive equality of
opportunities.
8 The Public Nature of Political Power and Political Justification

Through Public Reasons
The idea that political power is a fiduciary power, held in trust and to be impartially exercised only
for the public good, is a fundamental tenet of liberalism (Locke, 1689, 1690). Public political power
so conceived contrasts with political power as a private power held by those who possess and
exercise it as they choose; or alternatively a private power whose exercise is negotiated and paid for
in exchange for compensating benefits. The former position is characteristic of absolutist positions,
while the latter view is typical of feudalism and orthodox libertarianism.
The idea of the public nature of political power implies that its exercise is to be publicly justifiable
to liberal citizens on terms that they can reasonably and/or rationally accept. Public justification is
conceived differently by liberals. Hobbesian classical liberal views contend that it rests on a com-
promise among conflicting personal interests and/or non-public reasons individuals have based in
their differing moral, religious, and philosophical views.4
The high liberal position conceives of public justification in terms of a democratic ideal of free and
equal citizens with shared fundamental interests in maintaining their individual freedom and their
moral status as equal persons. Such fundamental interests underpin the idea of shared public reasons
among democratic citizens. On the high liberal account, the exercise of political power is legitimate only
if it can be publicly justified to persons in their capacity as free and equal democratic citizens, in terms
of shared public reasons all accept that reflect their fundamental interests as such citizens.5
The idea of political justification through public reasons, though complex, is familiar to government
officials who avoid explicitly appealing to “special interests” and certain kinds of reasons. Religious
reasons are not public reasons, nor are controversial philosophical and moral reasons about which free
and equal liberal citizens seriously disagree. A liberal supreme court would not justify constitutional
decisions by the philosophical position that human autonomy is the sole source of all values, morality,
and justice. Likewise, a liberal court would not legitimately appeal to religious doctrines to assess laws
challenging the right to abortion, contraception, or same-sex marriage. These controversial positions
cannot serve as the base for the public political reasons relied upon to publicly justify legislation or ad-
judication of constitutional and other political conflicts. In requiring that political authority be ex-
ercised only for the public good, the public nature of political power requires that the political
justification of laws and adjudication of controversies be conducted in terms of shared political values and
public reasons that reflect common interests and the common good of free and equal democratic citizens.
Among the political values of public reason are the freedom and equality of citizens; basic rights and
liberties, equal opportunities, a social minimum; “establishing justice,” “domestic tranquility,” the
“common defense,” and the “general welfare;” the common good, equal respect for citizens, eco-
nomic reciprocity, equal rights of women and racial and ethnic minorities, and so on. The protection
and furtherance of these and other political values of public reason are essential to the public justifi-
cation of the exercise of public political authority in a liberal and democratic society.
352
High Liberalism
Notes
1 I first used the term “high liberalism” in an early 1990s essay subsequently published as Freeman (2001). See
also my “Capitalism in the Classical and High Liberal Traditions,” (Freeman, 2011). Both appear in
Freeman (2018) as Chapters 1–2.
2 For example, see Friedman (1962) and Tomasi (2012).
3 See the Preface to A Theory of Justice, revised edition (Rawls, 1999), as well as Rawls (2001, §§18–22,
41–2, 49).
4 See Gauthier (1986) and Gaus (2011) for contemporary Hobbesian accounts of classical liberalism and its
public justification.
5 See John Rawls’s “liberal principle of legitimacy” (1996, 136–37; 216–17).
References
Dworkin, Ronald. (2011) Justice for Hedgehogs, Cambridge MA; Harvard University Press.
Freeman, Samuel. (2001) “Illiberal Libertarians: Why Libertarianism Is Not a Liberal View,” Philosophy and
Public Affairs, 30, 2, 105–151.
Freeman, Samuel. (2011) “Capitalism in the Classical and High Liberal Traditions,” Social Philosophy and Policy,
28, 2, 19–55.
Freeman, Samuel. (2018) Liberalism and Distributive Justice, New York; Oxford University.
Friedman, Milton. (1962) Capitalism and Freedom. Chicago, IL; University of Chicago Press.
Gaus, Gerald. (2011) The Order of Public Reason. New York; Cambridge University Press.
Gauthier, David. (1986) Morals by Agreement. Oxford, UK; Oxford University Press Milton.
Hayek, Friedrich. (1960) The Constitution of Liberty. Chicago, IL; University of Chicago Press.
Locke, John. (1689) A Letter Concerning Toleration, London; Awnsham Churchill.
Locke, John. (1690) Second Treatise of Government, London; Awnsham Churchill.
Mill, John Stewart. (1861) Utilitarianism, Fraser’s Magazine, 63 (October‐December).
Mill, John Stewart. (1859) On Liberty, London; J. W. Parker and Son.
Mill, John Stewart. (2006) The Principles of Political Economy (7th ed.). Indianapolis; Liberty Fund.
Parfit, Derek. (2012) “Another Defense of the Priority View,” Utilitas, 24, 03, 399–440.
Rawls, John. (1999) A Theory of Justice (Rev. ed.). Cambridge MA; Harvard University Press.
Rawls, John. (2001) Justice as Fairness: A Restatement. Cambridge, MA; Harvard University Press.
Rawls, John. (1996) Political Liberalism. New York; Columbia University Press.
Scanlon, T. M. (2018) Why Does Inequality Matter, Oxford, UK; University Press.
Tomasi, John. (2012) Free Market Fairness: Princeton, NJ; Princeton University Press.
353
28
INSTITUTIONALISM, INJUSTICE,
AND PERSONAL RESPONSIBILITY
Kok-Chor Tan
UNIVERSITY OF PENNSYLVANIA
People have valuable personal pursuits, such as individual projects and associational and relational
concerns, that need not be motivated by egalitarian considerations or be equality promoting. How
should professed egalitarians balance their commitments to egalitarian justice and personal demands
and concerns? The institutional approach to egalitarian justice or institutionalism offers the form of a
response to this problem: individuals have the duty of justice to support and maintain egalitarian
institutions but are free to pursue their personal and associational ends within the rules of these
egalitarian institutions.1 This reconciliation of justice and personal pursuits by means of a division of
domain is well conveyed in John Rawls’s remarks that “within the framework of background justice
set up by the basic structure, individuals and associations may do as they wish insofar as the rules of
institutions permit” (Rawls, 2001, 213). In this respect, Institutions have a certain normative pri-
macy over personal ends: just institutional rules specify the terms within which individuals may
rightly pursue their personal and relational ends.
But how does institutionalism apply in the non-ideal situation where just institutional arrange-
ments are absent (or not sufficiently present)?2 Without the benefit of a “framework of background
justice” what is the proper space for personal pursuits? And what does just action require of in-
dividual actors? One institutionalist proposal is as follows: individuals, in this case, have the duty of
justice to bring about more just arrangements and that when they have done their share in this regard,
they are free to pursue their personal ends. This primacy of institutional action reflects Rawls’s
comment individuals have the “fundamental” duty of justice “to further just arrangements not yet
established, at least when this can be done without too much cost to ourselves” (Rawls, 2001, 115).
Critics of institutionalism, however, will point out that this prioritizing of the duty to further just
arrangements (when they are absent) is further evidence of the implausibility of the institutional
approach. One line of objection goes that privileging institutional reform in circumstances when
more good can be achieved by direct interpersonal actions “fetishizes” institutions; it confuses means
for ends (Murphy, 1999, 282, 286). Another line of objection holds that associating the duty of
justice to that of furthering just arrangements suggests an unacceptably lax view of individuals’ duties
of justice (Cohen, 2000,2008; Berkey, 2015). It gives the impression that furthering just ar-
rangements is all that individuals are required to do by way of justice even when they can do more to
reduce injustice through direct personal action.
Both institutionalists and their critics agree that in the context of unjust institutional arrange-
ments, those with more resources are not justly entitled to these resources. These individuals simply
have more than their fair share in this case, both sides to the debate will agree. Where they disagree is
what should be done in the name of justice. The institutionalists think that the primary duty of
354 DOI: 10.4324/9780367808983-34

Institutionalism and Personal Responsibility
individuals, in this case, is to do their part to bring about better arrangements (and this is consistent
with the requirement that individuals use their unjust shares toward this end), whereas the critics
think that the aim is to bring about the best outcome and that there is nothing special about
attending to institutions in this respect.
My goal in this chapter is to clarify how institutionalism applies to the non-ideal case, and that it
can do so without being fetishistic about institutions or being too lax about justice. This project of
defending the application of institutionalism is clearly not a complete justification of institutionalism.
It is only a second-stage argument, an attempt at an application that presumes the institutionalist
ideal. But since some critics take the inapplicability of institutionalism to the non-ideal case as an
independent reason to reject it, it is worth showing why the privileging of institutional reform is not
as straightforwardly implausible as critics suggest (e.g., Murphy, 1999, 281, 286).
Although I referenced Rawls above (and will do so elsewhere), my purpose is not Rawls
exegesis. My defense of institutionalism in the non-ideal case should be on its own merits and not
that it is what Rawls would say necessarily. Also, I will use the term “justice” mostly to refer
specifically to egalitarian justice. I would argue that institutionalism can serve as an approach to
social justice more broadly. But for the sake of focus here, I limit my defense of institutionalism
to egalitarian justice.3
Before proceeding, there is a possible response to the above objections that I will not pursue. This
is the response that the special domain of direct personal action that the critics want to preserve is in
fact a non-category because personal actions motivated by injustices will ultimately (even if in-
directly) have political or institutional implications. But I think it is reasonable to grant to the critics
that some personal actions do not have institutional reform as their intention, and that whatever
“trickle-up” effects they might have in terms of impacting institutional arrangements are tenuous.
For instance, private charitable donations of food to people hit by a natural disaster are not
institution-focused in the way, say, direct public pressure for more effective governmental response
is. Similarly, volunteering time as a parent in an under-served public school is not institutional in the
way that agitating the school district and government for more equitable educational funding is.
There will be gray areas naturally: an organized and coordinated and publicized parental engagement
with a struggling school can perhaps have trickle-up institutional effects whether that is intended or
not. But the critics’ objection is motivated so long as there are clear cases of direct personal action as
opposed to actions aimed at furthering just arrangements.
1 The Institutional Ideal

To begin, I will outline the institutional approach in its ideal form and note some of its main
motivating considerations. One well-known reason why justice ought to be concerned with in-
stitutions is because of the pervasive and profound impact the basic societal institutions have on
persons’ life prospects. The central political, economic, and social institutions of a society, what
Rawls calls its basic structure, determine persons’ fundamental rights, entitlements, and responsi-
bilities. Given this impact of institutions on people’s life prospects, they must necessarily be regulated
by the demands of justice. In this way, as Rawls puts it, the basic structure is “the primary subject of
justice” (Rawls, 2001, 7).
This important reason of the profound and pervasive impact of institutions on life prospects,
however, only shows why it is necessary for justice to be concerned with institutions. It does not tell
us why an institutional focus is sufficient. That is, if we are concerned with the institutions of society
because of their profound and pervasive effects on persons’ life prospects, why should we limit our
attention to institutions when there are other features of social life that could have similar kinds of
effects? (e.g., Cohen 2008, 378–9).
355
Kok-Chor Tan
The second motivating reason explains this exclusive focus on institutions. This is the pre-
sumption of value pluralism (e.g., Freeman, 2006, 137–8; Scheffler, 2010, 134–5). While egalitarian
justice is an important value or good, and a value that has a certain regulative primacy over personal
pursuits, it is not the only value that matters or the dominant value that explains and justifies other
value commitments. Valuable personal pursuits of the various kinds that individuals typically engage
in, for example, are not valuable only because they are instrumental for promoting egalitarian justice.
Nor do we take the value of these pursuits to be derivative of egalitarian principles. For example, the
value of friendship and the activities associated with friendship are not wholly reducible to or
justifiable by some egalitarian distributive principle. While the primacy of egalitarian justice means
that the kinds of personal pursuits we can have, and how we realize them, have to be limited by the
requirements of egalitarian justice, it does not follow that these pursuits are valuable only by re-
ference to some ideal of egalitarian justice.4 Justice sets the limit of the good but it does not ground
the good, so to speak.
In light of this pluralism of the good, the end of justice is not to direct all permissible personal
ends toward the cause of equality, but the different and more modest one of establishing the
background conditions in which persons can pursue their various (including nonegalitarian) personal
ends freely and equally in relation to others. Rawls’s comment that “justice draws the limit and the
good shows the point” (Rawls, 1993, 174) can be interpreted in this way: while justice defines how
personal ends are to be pursued and which ends may be pursued (i.e., justice sets the limit), personal
ends are what give meaning and purpose to people’s life (i.e., the good shows the point). On this
view, a conception of egalitarian justice that does not provide reasonable space for various personal
pursuits defeats the purpose of justice.
There is, therefore, the need to demarcate the boundary of justice and personal pursuits. The
institutional ideal, by limiting the site of justice to institutions, provides a feasible method by which
to demarcate the division between justice and personal demands (see Rawls, 2001, 556). It provides
an identifiable target for individuals to direct their responsibilities of justice while respecting the
space within which they may permissibly pursue their personal ends.
The institutional approach thus presents a way of compartmentalizing different moral domains
(i.e., the domain of justice and the domain of personal life) that is reflective of the pluralistic
structure of value. Tensions between reasons of equality and reasons of personal pursuits within an
agent’s soul are resolved by making these reasons domain-specific.
A third motivating reason for institutionalism has to do with the public (and in this respect
collective) character of justice. Justice is a public project in two respects. One is that the enforcement,
coordination, and maintenance of distributive equality is necessarily a common project. The rea-
lization of justice is not something that individuals can on their own effectively bring about without
social coordination. Uncoordinated efforts on this front are not only inefficient but can also cancel
each other out. But more importantly, the requirements of societal justice are not within the rights
of individuals to privately enforce. On the contrary, the right to use coercive powers to enforce the
requirements of justice is a shared collective power. The problem of coordination, enforcement, and
maintenance of justice is thus an important reason for establishing institutions with the capacity and
proper standing to ensure justice.
But the second respect in which justice is public and collective is perhaps more basic. What justice
specifically requires of each of us is not something we are able to unilaterally determine. Rather the
terms of (distributive) justice, that is the terms of rightful possession (who rightly owns what) are, as
Kant would put it, determined through being “united with the will of all in a public lawgiving” (Kant,
1993, 78). That is, to generalize this Kantian point, only under a public institutional order “can
something external be mine or yours” (Kant, 1993, 77). The reason for this is that outside any “civil
condition” characterized by a shared and public system of lawgiving, what we each are entitled to by
way of justice is undetermined or at most provisional (i.e., open to reasonable contestation). For
356
instance, even if we allow that first occupancy gives you the right of possession over a piece of land, this
right is inconclusive. While you have the right when you physically stand your ground since physically
removing you will amount to a violation of your moral right, what happens should you vacate the space
temporarily? Do you still retain a right of possession of an external good when you are no longer in
physical possession of it? For the institutionalist, there is no natural, pre-institutional answer to this
question that is not reasonably disputable.5 We need, therefore, a public system of rules in order to
specify the terms of people’s rights of ownership and others’ corresponding responsibilities of justice. By
reference to a common institutional order, under-determined or provisional rights are made more
determinate or conclusive. Thus, given this public aspect of justice, institutionalism shows the way we
can arrive at an understanding of our rights and specific responsibilities of justice.6
To summarize, institutionalism is motivated by the following considerations: we focus on in-
stitutions because of their profound and pervasive impact on people’s lives, and we focus exclusively
on institutions in recognition of value pluralism. Institutions are necessary, not only for the purpose of
coordinating and enforcing the demands of justice but also for specifying our responsibilities of justice.
A description of the institutional ideal does not, of course, justify it.7 But my purpose, to re-
iterate, is not to defend the ideal but to defend its application to the non-ideal case where just
institutional arrangements are absent. The circumscribed but important challenge I am addressing is
that whatever the merits of the institutional ideal, it has to be rejected because it cannot be sensibly
applied to the non-ideal case. In any case, in my examination of this challenge below, I will ela-
borate more on these features of institutionalism and offer some partial considerations in their favor.
2 Institutional Fundamentalism
Let’s turn first to the objection that prioritizing the duty to further just institutions is misplaced.
How can it be right, Liam Murphy asks, that we ought to focus on institution building, as a duty of
justice, even in cases when we “could do more to reduce inequality, alleviate suffering or whatever
by direct action [or interpersonal duty]” (Murphy, 1999, 281). For example, of two choices (as-
suming here that only one course of action is available), that between providing assistance directly to
a homeless person (interpersonal action) or agitating the city government for accessible housing for
all (an institutional response), why assume that the latter is the more basic duty?
Similarly, Amartya Sen writes that “[o]ne could hope that strangers badly in need would have
some direct claim to just considerations by others at home and abroad, not merely through ‘the
obligation to support just institutions’” (Sen, 2009, 413). To insist that our duty of justice is ne-
cessarily institutional even in these types of cases amounts to what Sen calls a kind of institutional
“fundamentalism” (ibid., 82).8
With this concern in mind, let us clarify why institutionalism takes the furthering of just ar-
rangements to be a primary duty of justice. The first point to recall is the familiar one, noted above,
that institutions have a pervasive and profound impact on people’s life options. When institutions are
unjust, individuals are not allotted their fair share of resources or opportunities. Direct personal
action to assist the disadvantaged within unjust arrangements, without also attending to the back-
ground injustice, can at best address the symptoms of injustice and not the source of injustice. An
interpersonal response will be, at most, palliative rather than curative we might say. A person in-
stitutionally denied certain rights by unjust arrangements can be assisted directly and thereby rescued
from the bad consequences of the unjust arrangement. But this hardly resolves the injustice because
the role of institutions with respect to justice is not merely instrumental but also constitutive.
Institutions aren’t merely a means for realizing justice – they are the source of injustice and responding
to the effects of injustice without addressing institutional failures confuses a source for a mere means.
Second, institutional rules specify what just actions are required of us. Recall the public character
of justice under the institutional ideal, as noted earlier. Direct interpersonal action in the absence of
357
Kok-Chor Tan
just background institutions can be blind to what justice requires. Without just institutions, in-
dividuals cannot be sure if their personal acts of assistance are in the service of justice. The point here
is not that institutions play an instrumental role in coordinating our response to injustice, to make
sure that our efforts don’t cancel each other out, and so on. Institutions do play this role, but if this is
all institutions do, then the “fundamentalism” charge stands. The more important idea, rather, is that
institutions determine what justice entails, not just how to most efficiently achieve justice.
To illustrate with a simplified example: a person who is unfairly advantaged by an unjust dis-
tributive arrangement may be motivated personally to re-distribute her holdings to her less well-off
compatriots. But what proportion of her holdings does she really owe to others? Assuming after
transferring her wealth to more needy members of society, she remains still comparatively very well
off. Does she owe them still more? What is she really entitled to hang on to? In the absence of just
institutional rules, these questions can’t be meaningfully addressed. Particularly, it prevents us from
saying with conviction that she owes more to society as a matter of justice. Moreover, to whom and
to what causes should the wealthy philanthropist in a society with unjust rules contribute to? Why to
this particular cause and not another, why assist this particular set of individuals and not another?
Without clear institutional rules, we cannot say with clarity what and how much the unjustly well-
off owes and to whom.
Thus, rather than being fetishistic, prioritizing the creation of just institutions is necessary for
justice. It addresses the source of injustice (the pervasive and profound impact of institutions on
states of affairs), and it aims to specify the requirements of just actions that are under-determined
while just institutional rules remain absent.
The primacy of institutional justice does not deny that addressing the effects of injustice through
personal action can sometimes be urgently required. The institutionalist does not insist, should there be
a real trade-off, that it is better to participate in a rally for better state welfare than to provide direct
assistance to a starving person. But what they will say is that even if helping the person in dire straits is
the better thing to do, this personal direct action has a very different objective. An interpersonal duty
cannot be substituted for an institutional duty without loss from the standpoint of justice. The inter-
personal response in special cases of this kind has to be seen as a second-best solution though warranted
under the circumstance. Thus, it is open to the institutional view that sometimes it is better, all things
considered, for a person to act on an interpersonal duty than an institutional one. But it is still the case,
from the standpoint of justice, the most important thing to do for any individual, is that she acts in-
stitutionally. Commendable personal actions can’t replace the need for structural changes and reforms.
Thus, the motivation behind institutionalism is not as counter-intuitive as the critics implied, that
it is not a simple blunder of confusing means and ends. But I claim, further, that institutionalism in
fact better reflects the systemic and social quality of justice than the non-institutional alternative.
A society that is institutionally unjust but in which everyone acts charitably and decently at the
interpersonal level is still a society lacking in justice. Personal good may bring about a desirable state
of affairs in this case, but it is hard to conclude that justice thereby obtains.
To see this, imagine a caste or hierarchical society where members of the privileged class take on
commendable interpersonal action to counteract the effects of their unjust societal order. However,
none of them is working towards the abolishment of the hierarchical institutional structure of their
society. While we can acknowledge the good that is realized from personal kindness, we will still say
that this is still an unjust society, the outcome notwithstanding. From the standpoint of justice, an
adequate response to unjust institutions (like a caste-based society) is not personal goodwill or in-
dividual acts of beneficence, but the collective and institutional duty of abolishing and replacing the
unjust order. No doubt it is better that people in an unjust society take on interpersonal duties to
mitigate the effects of the injustice than if they did nothing at all.9 But their collective duty of justice
remains unfulfilled since nothing in what they do is directed at bringing about the end of this
injustice.10 What is wrong with this is that members of the subjugated group remain subject to the
358
arbitrary power of the superior class; without institutional correction, if the members of the op-
pressed group do experience a tolerable day-to-day existence, that is because of the whim and fancy
of the powerful. The institutional view thus aligns with the republican concept of freedom (as
developed in Pettit, 1997) that freedom has to be understood as independence from the arbitrary will
of another. The institutional approach recognizes that individuals remain fundamentally unfree as
long as their institutional order is unjust and that whatever free space they do enjoy rides on the
mere goodwill of the dominant.
Rainer Forst expresses this general idea as follows: justice is “a ‘relational’ matter; it does not first
inquire into subjective or objective states of affairs but into relations between humans” (Forst, 2014, 24),
specifically political and power relations. The debate between institutionalists and their criticism
turns fundamentally, then, on how we ought to conceptualize justice, which “picture” (to use
Forst’s term) of justice we ought to adopt. My above remarks of course don’t amount to a full
defense of the institutional approach or the relational “picture” of justice. But they suggest that the
institutional view, far from being counter-intuitive, in fact, more plausibly reflects the systemic and
relational character of justice.
Ultimately, the disagreement between the institutionalists and their critics turns on which
“picture” of justice we should adopt, a debate that I have not entered into here. What I claim only is
that there is nothing straightforwardly implausible about the institutional approach and that its view
that people’s primary duty of justice is to further just arrangements is not as counter-intuitive as
critics have charged. If institutionalism is to be rejected, it cannot be because it has no sound
application in the non-ideal case.
3 Evading the Burdens of Justice?

G.A. Cohen famously criticizes the institutional approach for evading the burdens of justice.
According to Cohen, a just society ought to have more than just institutions. A just society should
also be one whose members internalize an egalitarian consciousness or ethos that regulates and
motivates their personal conduct within the rules of just institutions. In a truly just society, on Cohen’s
view, personal choices and conduct “in the thick of daily life” must also be informed by egalitarian
principles (Cohen, 2000, 4).
Cohen’s criticism of the institutional ideal has been the subject of much philosophical discussion.
But leaving aside his critique within ideal theory, one might think that Cohen’s basic point about
the slackness of institutionalism has force in the non-ideal situation. The institutionalist claims that
residual inequalities, when background institutions are just, are not unjust inequalities that individuals
must address as a matter of justice. But the same cannot be said of inequalities in the context of
unjust institutions. Here the inequalities seem objectionable since they derive from institutional
injustice. Thus, it might seem that in the absence of the parameter-setting rules of institutions, an
egalitarian ethos is all the more important as there are no clear institutional rules to frame or regulate
individual conduct. Moreover, since injustice will result in more deficiencies in people’s needs, it
seems mistaken to hold that individuals do not have claims of justice against one another at the
interpersonal level in the context of unjust institutions.
Given this concern, let me consider how the institutionalist can maintain the division of domains
between justice and personal life in the absence of just arrangements. The first point to note is that
institutionalism does not say that the space for personal pursuits is fixed whether or not just in-
stitutions are present. The institutional ideal that persons may do as they wish within the rules of just
institutions presumes that there are just institutional rules. People are in conformity with their duty
of justice when they support and comply with these rules, and so in this context, they may pursue
their personal and relational ends secure in the knowledge that justice is maintained. But where
there are no just institutions, the institutional constraints will take a different form. The requirement
359
Kok-Chor Tan
now will be that individuals have the duty of justice to create just institutions in this case, and only
within this additional constraint are they free to engage in personal pursuits.
Thus, it does not follow that the permissible scope of personal pursuits stays fixed regardless of
background conditions of justice. In fact, instead of simply complying with and supporting just
institutional rules as in the ideal case, people in the non-ideal case are now also required to take the
more active and additional step of helping to establish just institutions. We can expect this additional
demand of justice to be more exacting than the demand to support and comply with just rules.
Hence, since the duties of justice set the parameters for personal pursuits, we can expect, conse-
quently, that the space for personal pursuits will be reduced in the context of injustice. For a simple
example, time that could be rightly devoted to personal pursuits when there are just institutions may
have to be dedicated to the cause of furthering just institutions when these are absent. Thus, al-
though the duties of individuals in both contexts of injustice and justice are formally identical in the
sense that they are directed at institutions and that they draw the boundary between personal life and
the demands of justice institutionally, they will not be equivalent. The content of their institutional
duties differs substantively depending on the degree of background justice or injustice.
The greater and more serious the institutional failure, the larger each of our share to repair and
create better institutions. It is reasonable that in times of ordinary injustice that we get to retreat to
our personal space after we have done our part to repair unjust arrangements. But in times of
extraordinary injustice, in a legitimacy crisis, we can imagine, consistent with the institutional view,
that the task of eliminating and replacing unjust structures with just ones will require more if not all
of our time and attention. That is, the institutional approach provides a target for just actions, and
that target provides a means of demarcating the demands of justice and the prerogatives of personal
life. But the line of division between justice and personal pursuits is not fixed once and for all but
will depend on the context of the case.
When Rawls writes that the natural duty of justice “also constrains us to further just arrangements
not yet established, at least when this can be done without too much cost to ourselves” (Rawls, 1971, 115),
the (added) italicized clause “without too much cost to ourselves” might appear to be a laxity that is
inherent in the institutional view. But this concern would be misplaced. The “personal-cost lim-
itation” is octagonal to institutionalism. We can imagine, for example, an institutionalism that says
that just actions require that we do all we can, give our all, to create justice institutions, and only
when that is done may personal pursuits be rekindled. That is hardly an undemanding duty of justice
even though it is institution-focused.
So, if there is any questionable laxity in the Rawlsian dictum, it is with the “without too much
cost” clause, not because it is institution-focused. However, it is not clear that such a limitation
renders justice too lenient. Most reasonable conceptions of justice and just actions observe limits on
what persons can be required to do.11 There is nothing inherent to institutionalism that fixes the
personal cost limit in this place or that, more stringently or less, as compared to the critic’s alter-
native. No doubt, how we are to draw the limit of duty in this way is a difficult matter and one for
which there is likely to be no simple principle. But again, this is not a problem peculiar to the
institutional approach.
In fact, I will suggest, that institutionalism provides clearer guidance on how to divide up the
demands of justice and the prerogative of personal life even in the non-ideal case. It offers a
structure for reconciling these competing demands: engage in personal pursuits only when that is
consistent with your doing your fair share to bring about just institutions. It is true that even here,
what qualifies as one’s “fair” share needs to be worked out. But the institutional approach
identifies the site where action is needed and thereby provides at least this method of dividing up
responsibilities. The alternative view in contrast, that people should do what they can, in all
spheres of their lives, to bring about justice sounds too broad to provide meaningful guidance.
Institutionalism provides a focus – it shows where the work is that needs to be done. This
360
compartmentalization of moral domains at least provides a reference point for determining what
“without too much cost” entails.
People will naturally continue to have needs that can generate moral demands of beneficence even
after I have done my share of institutional duty. Should these additional demands not morally move me?
In response, it is important to clarify that the institutional approach in both its ideal form and its non-
ideal extension does not deny that there can be moral demands of (interpersonal) beneficence distinct
from the (institutional) demands of justice. Even in a just society, a fellow citizen can get injured on the
street; a neighborhood can encounter financial challenges; an acquaintance may suffer from sudden bad
health, and so on. These kinds of situations impose demands of beneficence on us. The institutional
approach does not say that justice exhausts the domain of morality, that there are no moral relations
outside those regulated institutionally. And if demands of beneficence are operative in just societies,
they are operative in unjust societies too. The key point is that these are not demands of justice.
This is not merely a semantic point. When we call a duty a duty of justice, we accord it a certain
primacy over other moral demands. Beneficence, in contrast, is generally believed to be one value
among others that can be subject to some kind of personal discretion or triage (in the sense that the
actor has to decide which of her duties are more important). Thus, much substantively rides on
whether we characterized our interpersonal demands of the above sorts as demands of beneficence
or demands of justice.
Indeed, there is a further consideration against making justice a private matter. This is the worry
that it wrongly lets the state off the hook. The institutional response – working to establish better
governmental arrangements – is how we collectively ensure that the state does that which it is
supposed to do and to hold it publicly accountable. For example, public education is a governmental
or state responsibility, and agitating for better educational institutions is how we hold governments
to account and compel them to discharge their duty. To take this matter into one’s own hands, in
contrast, gives states the option of free-riding on the private actions of individuals. The state’s
responsibilities are taken up by non-institutional actors, effectively thus privatizing what is supposed
to be a governmental function. Thus, in response to the charge that institutionalism amounts to an
evasion of justice, the institutionalist can turn the tables on her critics: if there is any evasion, it is
they who are guilty of allowing the state to evade its burdens of justice.12
To sum, I have tried to allay the charge that institutionalism is too facile in the non-ideal case via the
following points: (i) institutionalism does not fix the space for personal pursuits regardless of context,
but will in fact recalibrate the division between justice and personal pursuits in the absence of just
arrangements; (ii) the duty to further just arrangements is not per se undemanding (i.e., the charge that
institutionalism per se lets people off the hook too easily is misguided); (iii) there are duties of bene-
ficence beyond the duty to further just arrangements, but beneficence is not the same as justice not only
in name but also in substance and (iv) finally, to the extent that we believe certain functions are properly
the job of governments, limiting personal responsibility to institutional duties helps prevent the state
from free-riding on personal action.
But aren’t there certain classes of duties of justice that can arise because of injustice, such as duties of
reparative justice? If so, couldn’t these be additional duties of (reparative) justice that go beyond the
duty to further just institutions? And, in addition, to return to the earlier “fundamentalism” charge,
couldn’t reparative duties require direct personal action? Thus, the reparative obligations that
background injustices generate seem to support the two central objections that justice should not be
limited to institution-building. I will briefly comment on this challenge from reparative justice next.
4 Reparations and Unjust Institutions

Does the reasonable presumption that injustices ought to be corrected or redressed pose difficulty for the
institutional claim that the primary duty in response to institutional injustice is to create better institutions?
361
Kok-Chor Tan
To address this problem, we should first note that reparative duties for personal failures of justice
and reparative duties for institutional failures will take very different forms. If I violate a requirement of
justice towards you, then I can owe you reparations as a matter of justice, and that this reparative action
can be direct and personal. This is a case of personal non-compliance, not a case of institutional failure.
But a reparative duty in response to institutional failures requires institutional action. Suppose that
institutional injustice has resulted in my being unjustly advantaged while others are short-changed
under the arrangement. What am I to do by way of compensation? I know that I need to give up my
unjust entitlements. But what exactly must I give up by way of reparative justice? And even if I, for
some reason, am able to determine the exact amount of my unjust holdings that I must renounce, to
whom do I transfer this to? It seems that the reparative duty here cannot be one that is discharged
interpersonally but one that must be mediated by shared institutions.
Taking on a reparative duty personally to address what is a collective failure of injustice in-
troduces problems that we normally associate with vigilante justice. When the injustice is personal
and specific, no worries concerning impartiality and appropriateness of the recipient of the duty
arise. If I violate a just rule and injure a particular person in the process, I know to whom I owe the
duty, and in some cases some idea of what is owed. But where injustice is institutional – where there
is a class of individuals entitled to reparations as result – taking matters into my own hands raises the
following: which individuals do I compensate? How can I decide these questions on my own?
In these sorts of cases, reparative duties must take the form of repairing and furthering just
arrangements, including creating mechanisms of reparative justice. The basic idea, again, is that
injustices due to institutional failures must have an institutional character, and not be in the form of
responses that are subject to the vagaries of personal choices and the contingencies of partial interests,
and that are not subject to public accountability.13 Thus, institutional injustices can result in the need
for duties of reparative justice. But these duties to compensate for institutional injustice must be in
the form of institutionally directed actions. The worry that institutional action does not account for
reparative duties per se is misplaced.
5 Conclusion
The institutional approach permits personal pursuits on the condition that these fall within and do
not undermine the rules of just institutions. In the absence of just institutions, one might think that
all bets are off, and no personal pursuits are permitted until institutional justice is restored. But such
an account would be morally impoverishing. A better alternative is to say that personal pursuits are
permissible on the condition that people are doing their part to promote just institutions. In the
context of injustice, then, the pursuit of personal ends can remain legitimate so long as people pursue
these ends while discharging their duty of justice to create better institutions. I have tried to
argue that while there can be other moral demands on people besides this institutional demand
under unjust circumstances, the duty to create just institutions is both necessary and sufficient with
respect to the cause of justice.
A complete defense of the institutional approach, among other things, will have to justify the
picture of justice that institutionalism presupposes. But the institutional approach should not be
rejected solely on the charge that it has no sensible application in the absence of just institutions. An
institutional focus, both as a requirement of justice and as a method of identifying the scope of one’s
duty of justice, is not as implausible as some of its critics believe.
Acknowledgments
I am indebted to audiences at the University of Virginia (Fall 2012); Georgia State; Binghamton
University; Penn; Fatih University, Istanbul (in Spring 2013); Columbia University; Goethe
362
University, Frankfurt; Yale University; and the University of Durham. I benefitted also from ex-
tensive comments from several individuals, at the above events and in private exchanges, including
Rainer Forst, Darrel Moellendorf, Eszter Kollar, Jon Elster, David Johnston, Bjorn Gomes, Michael
Kates, Colin Bird, Nicole Hassoun, Robert Guay, Collin Anthony, Eric Boot, Daniele Botti,
Shmulik Nili, Dan Halliday, Pietro Maffetone, and Elizabeth Kahn (with apologies to others I may
have inadvertently omitted). I thank Chris Melenovsky additionally for his astute comments and
editorial suggestions.
Notes
1 For some accounts of institutionalism see Rawls, 1971; Tan 2004, 2012; Freeman, 2006, 137–8;
Scheffler, 2010, 134–5; Melenovsky, 2014, 2016; Hodgson, 2012.
2 It is worth noting that the non-ideal case I am dealing with is different from the non-ideal case of non-
compliance. In the latter, just institutions are present, but individuals fail to comply with their rules. For a
discussion on this non-ideal scenario, see Kates, 2014.
3 Iris Young investigates the related problem of responsibility in cases of structural injustice (Young, 2011). See
McKeown’s chapter “Social Justice” in this volume for more detail. Young, however, has a different focus. She
is concerned with the unjust cumulative effects of individual actions that are performed under purportedly just
institutional arrangements. Her basic point, against Rawls, is that justice has to be concerned with more than
institutions since just actions under just arrangements can have unjust outcomes. We thus need to go beyond
Rawls’s focus on the basic institutions of society and be attentive also to “structure,” understood more broadly
to include patterns of interactions and choices, including those performed under just institutional rules. The
problem I am attending to here concerns our duty in the case where formal institutions are unjust. So our
questions are different. As for Young’s thesis, the institutionalist will note that should interpersonal activities
permitted under existing institutional arrangements result in unjust outcomes, then there is actually a failure of
institutional justice. If supplementary and reparative institutional forms are available for addressing these unjust
outcomes, especially when experience allows us to predict these outcomes, then a commitment to institutional
justice should compel improving our institutions to respond to the injustice. For example, if implicit racial biases
in hiring decisions tend towards unfair decisions regularly even when there are clear rules against discrimination,
then the institutional response will require that we formalize some affirmative action procedures etc. Our
institutions fail to be just otherwise. So Young’s worry that just institutions can result in unjust outcomes
(because of social structures) mischaracterizes the real problem. When institutions allow for predictable and
preventable unjust results, then these institutions cannot be called just.
4 See here Rawls’s remark: “Within the limits set by the principle of right, there need be no standard of
correctness beyond that of deliberative rationality” (2001, 564).
5 Rousseau (2008) gives voice to this reasonable objection: “Even those whom industriousness alone had
enriched could scarcely base their property on better titles. No matter if they said: It is I who built this wall;
I earned this plot by my labor. Who set its boundaries for you, they could be answered; and by virtue of
what do you lay claim to being paid at our expense for labor we did not impose on you?” (2008, 172). Like
Kant after him, Rousseau says that possession becomes a property right only in the political association.
6 Do institutions only specify or do they in fact create duties of justice? In some respects, institutions create
responsibilities that would not otherwise exist – for example, the requirement to pay taxes, the responsibility
to vote, and so on. But I take these new requirements or responsibilities to be ways of specifying our
preexisting general (and natural) duty to do justice.
7 I do more to defend the institutional ideal in Tan, 2012, Part I; Tan, 2004.
8 Although Sen is not necessarily denying that justice requires the right institutions, he objects to the priority
or what he calls the “foundational role” given to institutions (84–5).
9 Even here, one might object that direct personal acts of benevolence and charity are actually unjust because
they make possible the maintenance of the unjust institutional order. They sustain rather than correct
injustice in part by letting the state off the hook. I thank Jon Elster for this point.
10 In cases where individual actions are aimed at the larger goal of abolishing the system, then these will count
as institutional responses. To the extent that engaging in activities such as rescuing and liberating enslaved
individuals (thereby changing their status) are in effect challenges to the slave institution, these would count
as institutionally directed activities. For example, the Underground Railroad was fundamentally an effort to
abolish slavery, an operation that works against rather within the system. I thank Rainer Forst for this
suggestion.
363
Kok-Chor Tan
11 Even Cohen, for example, accepts that there are personal prerogatives that permit departures from the
demands of justice (Cohen, 2008, 388–90).
12 See Becker and Lindsay, 1994. Becker and Lindsay discuss how governments get to “free ride on private
sector public good production” (277). I thank Jon Elster for this point and the reference.
13 This is why the issue of personal philanthropy in unjust societies raises challenging ethical questions.
See Riech, 2018; Berkey, 2018 for contrasting views.
References
Becker, Elizabeth and Cotton M. Lindsay (1994) “Does the Government Free Ride?” The Journal of Law &
Economics, 37/1, 277–296.
Berkey, Brian (2015) “Double Counting, Moral Rigorism, and Cohen’s Critique of Rawls: A Response to
Alan Thomas,” Mind, 124/495, 849–874.
Berkey, Brian (2018) “The Institutional Critique of Effective Altruism,” Utilitas, 30/2, 143–171.
Cohen, G. A. (2000) If You’re An Egalitarian, How Come You’re so Rich? Cambridge, MA; Harvard University Press.
Cohen, G. A. (2008) Rescuing Justice and Equality. Cambridge, MA; Harvard University Press.
Forst, Rainer (2014) Justification and Critique. Cambridge; Polity Press.
Freeman, Samuel (2006) Justice and the Social Contract. New York; Oxford University Press.
Hodgson, Louis-Philippe (2012) “Why the Basic Structure?,” The Canadian Journal of Philosophy, 42, 303–334.
Kant, Immanuel (1993) The Metaphysics of Morals, tr. Mary Gregor. Cambridge; Cambridge University Press.
Kates, Michael (2014) “Individuals and the Demands of Justice in Nonideal Circumstances,” Social Theory and
Practice, 40/3, 388–408.
Melenovsky, C. M. (2016) “Incentives, Conventionalism, and Constructivism,” Ethics 126, 3, 549–574.
Melenovsky, C. M. (2014) “The Basic Structure as a System of Social Practices,” Social Theory and Practice, 39, 4,
2014.
Murphy, Liam (1999) “Institutions and the Demands of Justice,” Philosophy and Public Affairs, 27.
Pettit, Philip (1997) Republicanism: A Theory of Freedom and Government. Oxford; Oxford University Press.
Rawls, John (1971) A Theory of Justice. Cambridge, MA; The Belknap Press of Harvard University Press.
Rawls, John (1993) Political Liberalism. New York, NY; Columbia University Press.
Rawls, John (2001) Justice As Fairness: A Restatement. Cambridge, MA; Harvard University Press.
Riech, Rob (2018) Just Giving: Why Philanthropy Is Failing Democracy and How It Can Do Better. Princeton, NJ;
Rousseau, Jean-Jacques (2008) The Discourses and Other Early Political Writings, ed. Victor Gourevitch.
Scheffler, Samuel (2010) Equality and Tradition. New York, NY; Oxford University Press.
Sen, Amartya (2009) The Idea of Justice. Cambridge, MA; Harvard University Press.
Tan, Kok-Chor (2012) Justice, Institutions and Luck. Oxford, UK; Oxford University Press.
Tan, Kok-Chor (2004) “Justice and Personal Pursuits,” The Journal of Philosophy, 101, 7, 331–362.
Young, Iris Marion (2011) Responsibility for Justice. Oxford, UK; Oxford University Press.
364
29
SOCIAL JUSTICE
Maeve McKeown
For the last 50 years, debates about justice in political philosophy have focused on distributive justice
– the distribution of resources between citizens – or, in John Rawls’s famous terminology, the
distribution of “the benefits and burdens of social cooperation” (Rawls, 1971: 4). But this way of
thinking about justice has its limitations. In particular, it fails to identify the many kinds of social
injustice that afflict contemporary advanced capitalist societies, including (amongst other things)
sexism, racism, homophobia, transphobia, and ableism. Looking at the social movements that have
emerged over the last 50 years reveals claims about injustice relating to identity, difference, status,
and unequal power relations. Take any “new left” social movement of the 1960s–1980s (the
Women’s Liberation Movement, the LGBTQI+ movement, the disability rights movement,
Indigenous Rights movements, etc.), or some of the most prominent movements today, such as
Black Lives Matter, movements for decolonization, or trans rights, and you will see that these groups
have broader demands that are not captured by the idea of distributive justice.
These social movements demand recognition of difference, they highlight the inability of members
of certain social groups to be heard and taken seriously (epistemic injustice), they demand relational
equality, and they have highlighted the problem of structural injustice. These social movements do not
necessarily tell a story about what justice looks like, rather they identify and aim to rectify injustice.
Philosophers have tried to systemize the ideas of these social movements, in order to support these
movements’ quest for justice. In this chapter, I will look at the topics of recognition, epistemic
injustice, relational egalitarianism, and structural injustice, but first, I start with the feminist critical
theorist Iris Marion Young’s (1990) critique of the “distributive paradigm” of justice, which chal-
lenges John Rawls’s theory of distributive justice in a way that captures these different strands of
alternative social justice theory.
1 Critique of the Distributive Paradigm

In her influential 1990 book Justice and the Politics of Difference, Iris Marion Young wrote an excoriating
critique of what she called “the distributive paradigm” of justice, which continues to inspire philo-
sophers of a more critical bent to this day. Young argued that, while redistribution of resources was
certainly an important concern for justice, thinking about justice solely in terms of distributions of
resources failed to account for the range of injustices in the contemporary United States.
Distributive justice theorists focus on the distribution of resources between individuals, which as-
sumes that individuals are social atoms, rather than members of groups involved in power relationships
DOI: 10.4324/9780367808983-35 365

Maeve McKeown
(Young, 1990: 18). But the “bundles of resources” (to use a phrase that is popular with distributive
theorists) that individuals have are due to their membership of social groups, as women, as Black
people, as disabled people, etc. Because it focuses exclusively on individuals’ resources, the distributive
paradigm does not question the relationships between social groups that lead to these unequal dis-
tributions. The distributive paradigm also does not question the institutional framework that decides
upon the allocation of resources – it assumes there is a centralized state bureaucracy, which parcels up
the bundles and allocates them to individuals, without the individuals having a say over resource-
allocation (Young, 1990: 20).1
What’s wrong with this picture is that various issues that are intuitively considered to be issues of
justice are excluded from the framework. Young (1990: 22–4) focuses on three. First, decision-
making power. Who gets to decide on what the resources are and how they are distributed? And
why do some agents have the power to determine the shape of others’ lives? For example, what if a
large corporation decides to close its plant in a small, rural town, rendering half the population
unemployed; is it just for a corporation to hold the livelihoods of these people in its hands? Second,
the division of labor. Distributive theorists are concerned with the distribution of the top jobs that
have high salaries and status. But why is there a division of labor in the first place that divides jobs
along these lines, into professional and non-professional jobs; is that just? Young gives the example
of feminist critiques of the division of labor. On the one hand, feminists make a distributive justice
critique – why are so many of the high-salary, high-status professional jobs occupied by men? Justice
demands that more women get these jobs. On the other hand, feminists also critique the fact that
many low-salary, low-status jobs are associated with women, such as caring and cleaning, and indeed
jobs that earn no money whatsoever – domestic labor and childcare in the home. This is a non-
distributive claim: it questions why certain roles and occupations are feminized, which leads to the
third issue not addressed by distributive justice theorists – culture. The dominant social group in
society is viewed as the cultural norm. Their worldview, attitudes, and tastes are perceived as
normal. The worldview, attitudes, and tastes of other social groups are perceived as abnormal or
deviant. This leads to two related problems: these groups are simultaneously marked by the
dominant groups as “Other,” associated with various demeaning stereotypes, and they are also
rendered invisible, their voices and different perspectives are ignored.
Young argues that instead of focusing on an “ideal theory” of distributive justice, the starting
point for justice theory should be actually existing forms of injustice. In contemporary capitalist
societies, these are domination and oppression. Domination is the “institutional constraint on self-
determination,” meaning that individuals and social groups do not get to determine the conditions
of their actions (Young, 1990: 37). Most people today are dominated because the conditions of their
actions are determined by states, which they have minimal influence over, and by large corporations,
over which they have no say. The counterpoint to domination is more democracy; increasing the
power that individuals have to determine the conditions of their lives in the public-political sphere
and the workplace. Oppression is “the institutional constraint on self-development,” meaning that
some people are prevented from developing their talents, from learning skills, and from commu-
nicating effectively with others by institutions and social structures (Young, 1990: 37). While most
people in advanced capitalist societies are dominated, not everyone is oppressed. Members of some
social groups have opportunities to become members of the professional class, to develop themselves
as individuals, and pursue their ambitions. Groups that do not have these opportunities are op-
pressed. Perhaps most famously, Young argues that oppression manifests in at least five forms
(Young, 1990: Ch. 2).
First, exploitation. Some social groups expend their energies and labor power for the benefit of
other social groups, undermining the oppressed group’s capacity for self-development. For instance,
women expend their time and energy on domestic labor, childcare, and emotional labor, freeing up
men for more creative pursuits. People of color are often trapped in “menial” labor, both in the
366
Social Justice
developing world and within advanced capitalist societies. They work in unskilled, low-paying
occupations in which they lack autonomy and facilitate the work of others or serve others. Second,
marginalization. Marginals are people who are excluded from the capitalist economy, including
racialized people and also the old, the young who can’t get first jobs, and many people with dis-
abilities. Being marginalized is not only a distributive justice issue, the marginalized are bored, feel
useless, and lack self-respect. Third, powerlessness. The powerless are the “non-professional” class
who do not have the opportunity to exercise power over others, power is always exercised over
them. They lack authority and status. Fourth, cultural imperialism – the culture of the dominant
group is the norm and everyone else is “Othered” and silenced. Fifth, violence. Members of certain
social groups are vulnerable to random acts of violence, both physical and verbal, including women,
people of color, and LGBTQI+ people. Members of these groups know they could be victims of
violence at any time. This violence is legitimized by society because the perpetrators almost always
get away with it.
Tackling oppression might not only require redistribution of resources in some instances (e.g.,
exploitation, marginalization), but it also requires “cultural revolution” (Young, 1990: 152).
Difference must be recognized, accommodated, and even celebrated. Ingrained ways of thinking
and acting must be challenged in order to tackle unjust social relationships between groups.
Everyone must be able to have a say in the public-political sphere, which requires attending to and
respecting people’s differences. Decision-making power, the division of labor, and culture are all
sites of contemporary injustice, but they are not captured by the distributive justice framework.
Redistributing resources is inadequate to tackle domination and oppression; redistribution only
addresses one part of a much bigger problem.
2 Recognition
Young’s analysis highlights the themes that I will discuss in the next three sections of this chapter:
recognition, epistemic injustice, and relational equality. All of these topics are concerned with what
can broadly be called “status inequality” – the idea that what matters for justice is the equal status, or
social standing, of all individuals and social groups.
Young’s analysis also represents a bridge between two different philosophical traditions – Anglo-
American and continental philosophy. Where Anglo-American political philosophy tends towards
questions of rights, duties, and social contract theory, continental philosophy tends towards ques-
tions of power, the self, and recognition. Recognition has been a foundational concept in con-
tinental philosophy and its roots go back to Fichte and Hegel.
Hegel famously theorized the problem of recognition through the master/slave dialectic. Briefly,
the dialectic is as follows: in the beginning, the self sees itself in a mortal fight with the other. In this
struggle for life or death, the self faces three options: eliminate the other, dominate the other, or let
the other be (Williams, 2013: 7). In the master/slave dialectic, the master dominates the other. Both
the master and slave preserve their independence, but in an unsatisfactory way: the slave becomes
thing-like and servile, an instrument to serve the master’s ends, whereas the master’s independence is
related to coercion, domination, and the threat of death at the hands of the slave. Neither is truly
free; both are dependent on the other. This dialectic can be resolved when both parties recognize
the other. This paves the way for mutual affirmation and reconciliation.
Philosophers working in this tradition understand justice through the lens of recognition. They
argue that the self is not formed in isolation; it is formed in relation to others. Other people, and
society more generally, have the capacity to misrecognize individuals and social groups, which
constitutes an injustice.
Philosopher and psychiatrist Frantz Fanon discussed this problem in relation to colonialism. He
describes a scenario where he, a Black man from Martinique, was walking along a street in Paris in
367
Maeve McKeown
the 1950s when a White child pointed at him and said to her mother “Mama, see the Negro! I’m
frightened!” (Fanon, 2008: 84). Fanon says that the child not only objectified him, but he also
objectified himself. He immediately associated himself with a range of stereotypes associated with
Africans in the White colonial imagination. He writes:
I took myself far off from my own presence, far indeed, and made myself an object. What
else could it be for me but an amputation, an excision, a hemorrhage that spattered my
whole body with black blood? But I did not want this revision, this thematization. All I
wanted was to be a man among other men. (Fanon, 2008: 85)
Fanon goes on to argue that this sense of being an object and being inferior to White people has
been internalized by Blacks. Part of achieving equal status is not just achieving equal rights but
achieving an internalized sense of equality. Black people need to see themselves as equal. And White
people need to recognize them as equal. For Fanon, the battle for recognition prefigures the pos-
sibility of ethical relations between groups (Gordon, 2007: 6). If a group is seen as not fully human,
then it will not be included in any kind of distributive justice or moral scheme. This group therefore
must fight, including by use of violent means, to be seen as on a par with the dominant group.
Politics (and violence) precedes ethics.
Charles Taylor developed a politics of recognition in the context of multicultural policymaking
in Canada in the 1990s. Taylor writes:
The thesis is that our identity is partly shaped by recognition or its absence, often by the
misrecognition of others, and so a person or group of people can suffer real damage, real
distortion, if the people or society around them mirror back to them a confining or de-
meaning or contemptible picture of themselves. Nonrecognition or misrecognition can
inflict harm, can be a form of oppression, imprisoning someone in a false, distorted, and
reduced mode of being. (Taylor, 1994: 25)
The politics of recognition ask us to recognize the unique identity of different individuals, which
depends on their membership of particular social groups. Failure to recognize the unique identities
of members of minority groups amounts to assimilating everyone to the dominant group. Because
the self is bound up with group identity, this leads to political questions about special protections for
minority groups. It is on this basis that Taylor argues that justice requires that group identities ought
to be preserved and given special protections. On Taylor’s view, recognition works out politically
by giving certain minority groups different, collective rights, for example, First Nations Canadians
should be afforded certain group rights that are not given to other Canadians (such as rights to
territory or hunting, or some political autonomy), and some minority groups can exclude others or
enact policies designed to preserve their cultural integrity (such as Quebec requiring immigrants to
send their children to French-speaking schools).
Recent Indigenous scholarship has seen pushback on Taylor’s perspective. For example, Glen
Coulthard (2014: Ch. 1) argues that Taylor’s politics of recognition fails to recognize the revolu-
tionary anti-colonial implications of Fanon’s interpretation of the master/slave dialectic. Fanon
highlighted that in the context of colonialism, the master is not dependent on recognition from the
slave, and that recognition-politics in this context amounts to the colonizer bestowing rights on the
colonized serving to bolster the underlying power structure. Furthermore, the colonizer merely
deciding to recognize the colonized, such as the Canadian state deciding to recognize First Nations
peoples, ignores the need for a (violent) struggle for recognition. It, therefore, cannot achieve the
kind of psychological liberation of the colonized that Fanon envisaged. However, psychological
368
Social Justice
liberation in the colonial context doesn’t have to depend on violence; Coulthard suggests, following
Fanon, that the slave (the colonized) should “turn away” from the master (the colonizer) and focus
instead on self-recognition.
In the 1990s–2000s, the politics of recognition mostly applied in theory and in practice to debates
about multiculturalism and Indigenous group rights. The struggle for recognition is perhaps most
visible today in the trans rights movement. As psychoanalyst Jacqueline Rose (2016) writes in her essay
on the trans movement, “Whatever stage of the trans journey or form of transition, the crucial question
is whether you will be recognized as the other sex… no human can survive without recognition. To
survive, we all have to be seen.” The trans movement’s emphasis on misrecognition of gender raises
new political questions to previous debates; instead of generating demands for group rights, some
trans scholars and activists aim for recognition within existing group categories of sex and gender,
whereas others destabilize and demand transformation of these categories (Bettcher, 2020).
However, while recognition is thought to be vital to justice within continental philosophy, there is
debate about the extent to which redistribution is also needed. Is recognition alone sufficient to
constitute justice, or must it be combined with redistribution of resources? Nancy Fraser (1997) argues
that the politics of recognition can be in tension with the politics of redistribution. For instance, the
struggle of the working class is not for recognition of difference, but for economic parity, which
eliminates difference. Even though struggles for recognition and redistribution often go hand in hand,
it is useful to keep them analytically distinct. On one end of the spectrum, the working class requires
redistribution to overcome socioeconomic injustice like exploitation, economic marginalization, and
deprivation. On the other end of the spectrum, LGBTQI+ people are struggling for recognition and
overcoming cultural or symbolic injustice, like domination, disrespect, and nonrecognition. Groups
enduring both forms of injustice face “the redistribution-recognition dilemma” because they both need
to claim and deny group specificity. These groups include women and African Americans. Both groups
face economic and cultural injustice. Achieving economic justice seems to deny difference, but
achieving cultural justice seems to require affirming it.
The solution is to look for transformative, as opposed to affirmative remedies. Affirmative po-
licies aim to correct inequalities without addressing the underlying framework that generates them
in the first place. Transformative remedies aim to restructure the underlying generative framework.
An affirmative redistributive policy, like the welfare state, can in fact create recognition-based in-
justices: the welfare state singles out the poor as requiring help which stigmatizes them and creates
hostility towards them (Fraser, 1997: 25). A transformative approach to material inequality aims to
change the underlying class structure by severing the link between employment and consumption
through universal welfare programs, progressive taxation, full employment, public ownership, and
democratic decision-making. An affirmative cultural remedy is like Taylor’s proposal – special rights
for certain minority groups. But a transformative remedy aims to deconstruct group identities,
challenging everyone’s sense of identity.
Not everyone agrees with Fraser’s position. Axel Honneth provides perhaps the most systematic
attempt to understand justice in terms of recognition, claiming contra-Fraser, that recognition is the
fundamental problem. Honneth claims that the lack of distributive justice signals disrespect and the
humiliation of those affected and that is the source of the injustice (Fraser and Honneth, 2003: 125).
Iris Marion Young also challenges Fraser’s view, claiming that social movements for feminism, anti-
racism, and LGBT liberation conceive of “cultural recognition as a means to economic and political
justice” (Young, 1997: 148). But Fraser has hit back in recent years, arguing that the problem of
global justice highlights that redistribution can be independent of claims for recognition. She argues
further that the problem of global justice has highlighted a third vector of justice: redistribution,
recognition, and representation (Fraser, 2010). The issue with the latter is the frame of justice: who is
included in social justice, citizens of a nation-state or all of humankind?
369
Maeve McKeown
3 Epistemic Injustice
Another form of injustice has received significant philosophical attention in recent years – epistemic
injustice. Epistemology is the theory of knowledge, and epistemic injustice is the injustice of failing
to recognize certain speakers or groups as credible sources of knowledge. This aspect of injustice taps
into recognition theory and also Young’s category of cultural imperialism – the silencing of minority
groups.
Despite being a relatively recent philosophical debate, claims about epistemic injustice have
history, deriving from a longstanding tradition in Black feminism and postcolonial thought. The idea
that subalterns cannot be heard was famously highlighted by Gayatri Spivak (1997) in her classic
postcolonial text “Can the subaltern speak?” from 1985. But long before that, Sojourner Truth, a
former slave who fought for emancipation, proclaimed “Ain’t I a woman?” in a 1851 speech at the
Women’s Rights Convention in Ohio, illuminating the erasure of Black women’s claims and needs
from the first-wave feminist fight for suffrage. This theme of the silencing and denial of the
credibility of Black women threads throughout Black feminist thought, including bell hooks’ book
Ain’t I a Woman? (1987), and Patricia Hill Collins’s argument that Black women in the US are
systematically denied credibility (Collins, 2000).
Charles W. Mills identified the problem of “white ignorance” in philosophy and society (Mills,
2007). Despite innovations in Marxism and feminist standpoint theory, which emphasize that
knowledge is situated and shaped by social group membership, mainstream epistemology has relent-
lessly understood knowledge as individualistic, and it assumes the “neutral” perspective, which is, in
fact, the perspective of White, propertied men (the people who were allowed to do philosophy for
most of the discipline’s history, at least in the West). But this problem is not confined to academic
philosophy. As Mills highlights, W. E. B. Du Bois identified the problem of “double-consciousness” in
his 1903 book The Souls of Black Folk, which is that Black people see themselves both from their own
perspective and from the perspective of White people; they have to know the White population’s way
of seeing them in order to survive (Du Bois, 2007: 8). White ignorance of other perspectives has been
instrumental in serving the interests of White people. For instance, it allowed Whites to perceive lands
as “empty” even though thousands of people lived there. Nowadays it enables Whites to see the world
through the lens of “color blindness,” where people of color are supposedly of equal standing, but this
assimilation erases the need for material and symbolic repair of past injustice; material repair meaning
monetary reparations and symbolic repair being recognition of the past and continuing harms com-
mitted against African-Americas, such as an apology for slavery or the dismantling (or at least significant
reform) of the police and criminal justice system. Mills adds that Black testimony has always been
“whited out” (Mills, 2007: 34). For example, during slavery Blacks were not allowed to testify in court
against Whites “because they were not seen as credible witnesses,” and enslaved people’s narratives in
writing or in speeches had to have a White authenticator who wrote the preface or introduced them as
a credible speaker (Mills, 2007: 32).
Miranda Fricker systematized these ideas in her book Epistemic Injustice (Fricker, 2007). She iden-
tifies two aspects of epistemic injustice. Testimonial injustice is the failure to hear the testimony of a
particular speaker and to deny their credibility. For example, the enslaved people that Mills referred to.
Kristie Dotson has added that “epistemic violence” occurs when listeners fail to communicate in a
reciprocal way with a speaker based on their own pernicious ignorance (Dotson, 2011).
Another stark example of testimonial injustice and epistemic violence is the case of the Yorkshire
Ripper in the UK. The Ripper, Peter Sutcliffe, was found guilty of murdering 13 women and
attempting to murder 7 others between 1975 and 1980. Sutcliffe was interviewed by the police
9 times and released before he was finally caught on a minor traffic charge. The police were
convinced that the killer was a hoaxer called “Wearside Jack” who sent them a tape where he
confessed to the murders and spoke with a North-Eastern English accent. But many women who
370
Social Justice
were attacked by Sutcliffe told the police that the voice on the tape wasn’t their attacker’s voice, that
he was a local man with a Yorkshire accent, and they provided multiple photofits that closely
resembled the killer. Joan Smith, a journalist who covered the murders at the time concludes,
“‘experts’ don’t listen to women” (Smith, 2020). In this case, failure to perceive women as credible
speakers, and therefore ignoring what they had to say, had lethal consequences.
The second branch of epistemic injustice, according to Fricker, is hermeneutical injustice. This is
where the society does not have the collective hermeneutical resources to recognize something as an
injustice. In other words, instead of ignoring the speech of a particular individual (testimonial in-
justice), the society, in general, does not recognize the claims of an oppressed social group because it
has no language or theoretical framework to comprehend it. The example Fricker cites is sexual
harassment. Before the term “sexual harassment” was coined in the 1970s, it had been going on
as long as workplaces existed, but it was never discussed nor addressed. Having apt terminology
allowed people to identify the problem and then to do something about it. Perhaps a more con-
temporary example is the movement to “decolonize” institutions. Before the term “decolonize”
existed in its current usage, many institutions were based on colonial norms and frameworks, but
without any conscious awareness of that fact (at least among the members of dominant social
groups). Decolonizing provides the necessary language to draw attention to certain facts, like that
museums in Europe are full of artifacts and artworks looted from former colonies and that university
and school curricula fail to integrate the perspective of the colonized.
In a pioneering book The Epistemology of Resistance, José Medina argues that epistemic injustice can
be addressed by creating “epistemic friction” – generating opposing narratives to dominant cultural
discourses (Medina, 2013). Some people gain “meta-lucidity” – an insight into the ways in which the
dominant society fails to recognize, listen to or understand the claims of the oppressed – and these
meta-lucid individuals can promote epistemic friction and build up a following. An example he gives is
Rosa Parks. She knew that segregation on buses in the Jim Crow South was an injustice. This
knowledge led her to act – not giving up her seat on the bus to a White person. Her action opened
other people’s eyes and inspired many others to participate in the struggle for civil rights.
Epistemic injustice is the area of justice theory that has received the most attention from phi-
losophers in recent years. It highlights that claims to injustice can refer not only to the distribution of
resources but also to the ways in which individuals and social groups are perceived and (mis)treated,
much like the recognition framework that preceded it.
4 Relational Egalitarianism
Theories of recognition and epistemic injustice have highlighted varieties of social injustice that are
not captured by the distributive justice paradigm. However, most theorists of justice agree that
equality is a key component of justice and that equality has a distributive or material component. So,
what should that look like?
After Rawls’s theory of distributive justice, another theory of distributive justice emerged,
dubbed by Elizabeth Anderson as “luck egalitarianism.” Luck egalitarians argued that Rawls failed to
consider personal responsibility: Rawls argued that resources should be redistributed to the “worst-
off” in society, but he failed to consider how those people became the worst-off in the first place;
maybe it has something to do with the choices that those individuals have made. They argue that
egalitarian justice demands that individuals are compensated for bad “brute luck,” meaning whatever
is outside of their control, for example, being born with a disability. However, individuals have to
bear the consequences of bad “option luck,” for example, if you choose to gamble your house in a
card game and lose, that’s your fault and the state should not intervene to help you, even if you end
up homeless. Therefore, the “worst-off” in society is disaggregated into those who deserve support
from the state and those who do not.
371
Maeve McKeown
However, some egalitarians responded to luck egalitarianism by pointing out that it fails to
adequately address what matters when talking about equality, and that it has potentially unjust
consequences. In a famous article, “What is the Point of Equality?” Elizabeth Anderson argued that
in compensating individuals for so-called bad brute luck, such as being born disabled, ugly, or
otherwise talentless, luck egalitarians would have to create invasive bureaucratic procedures that
judge individuals according to demeaning criteria and would intrusively investigate individuals’
personal choices. Luck egalitarianism could, in effect, create a two-tiered society with the “lucky” –
the talented, good-looking, able-bodied people – compensating the pitiful, unlucky people – the
untalented, ugly, disabled people – with cash pay-outs for their perceived flaws. Furthermore, those
who have made unfortunate choices at some point in their life, are left to bear the consequences of
this choice into perpetuity, so a luck egalitarian society leaves no room for making mistakes or taking
risks. People who make bad choices or take risks are liable to social exclusion and ostracism. Such a
society creates relations of disrespect (Anderson, 1999).
Anderson argued that what matters for equality is, therefore, social standing, or what she calls
“democratic equality.” A just society should aim to create relations of respect between citizens.
When it comes to the distribution of resources, the state should adopt the capabilities approach,
championed by Amartya Sen (1979) and Martha Nussbaum (2000). Anderson (1999: 317–8) argues
that the state has to ensure that every individual has sufficient resources to realize their capabilities
as a human being, as a participant in a system of cooperative production, and as a citizen of a
democracy. With this baseline level of distributive justice in place, the society will ensure that people
have access to sufficient levels of functioning that enables them to stand as social equals over the
course of their lifetime. Anderson’s intervention in the debate has led to a strand of literature known
as “relational egalitarianism,” which seeks to address how the just society would create equal social
standing and considers distributive justice a means to that end. Christopher Lebron has added an
amendment to Anderson’s view. He claims that Anderson’s vision of democratic equality still
perceives the disadvantaged in society as recipients and fails to address the responsibilities of the
privileged. He argues that the privileged people in the society should cultivate the virtue of trying to
see the world through the point of view of the other, in order to generate a human point of view of
equality (Lebron, 2014).
5 Structural Injustice
Anderson made a further point, which is that luck egalitarians focus on luck, but no social
movement ever has made arguments about justice or equality from the perspective of luck. Social
movements care about socially-caused injustice. Even a classic luck egalitarian case of bad luck –
disability – is much more complicated than they suppose. The disability rights movement is based on
the social model of disability, which argues that society is disabling because the material infra-
structure and social and institutional rules are designed with able-bodied people in mind. The
problem is not disabled people’s bodies, the problem is the society in which they live. From this
perspective, the society needs to change to accommodate people with various kinds of disabilities,
rather than compensating individual disabled people for their “bad luck.”
The question arises, however, how does socially caused injustice come about? It will not be
possible to remedy socially caused injustice without some account of its causation. This brings us
back to where we started, with Iris Marion Young. In her late work, Young argued that instead of
focusing on individuals, theorists of justice need to think about justice structurally. Rawls didn’t do
this sufficiently because he focused on the “basic structure” of society, or what Young (2011: 70)
calls “a small sub-set of its institutions.” Instead, Young argues that thinking structurally is a “way of
looking” at society and requires taking a macro view.
372
Social Justice
Young (2011: 52–64) argues that “structural injustice” is the result of four interlocking processes.
First, “objective constraint”: the material world in which individuals live is objectively constraining.
For example, towns and cities in the US are designed with cars in mind, so people who do not have
cars or can’t drive for whatever reason are disadvantaged. The material infrastructure constrains some
individuals and enables others. Further, the social and legal norms of a society also constrain some
individuals and enable others. Second, society is made up of social positions: sociologists think of
society as a multi-dimensional space (Blau, 1977) or a field (Bourdieu and Nice, 1977) with in-
dividuals positioned in relation to each other. These positions condition and place limits on our
interactions. This macro view also helps identify structural inequalities that persist over time. Third,
individuals reproduce structures through their actions. For example, every time I buy a T-shirt from
a budget online retailer, I reproduce the injustice of sweatshop labor (McKeown, 2018: 497).
Fourth, all of these social-structural processes have unintended consequences. For instance, I don’t
intend to exploit sweatshop workers, but the combination of my purchase of clothes, with the
practices of multi-national clothing corporations seeking to reduce costs, and low-income countries
seeking economic growth, results in groups of people – mostly young women of color – being
exploited (McKeown, 2017).
The result of structural injustice is that certain groups are rendered vulnerable to domination or
oppression (Young, 2011: 52). So structural injustice is the underlying cause of political inequality; it
is the source of domination or oppression of various social groups. Because structural injustice is the
outcome of so many complex processes, Young argues that no one is to blame for it. Instead,
everyone connected to structural injustice should take up their “political responsibility” to work
collectively to overcome it. This includes states, corporations, and individuals.
While I am broadly supportive of Young’s view, I argue that powerful agents, meaning agents
with the capacity to remedy structural injustice, such as multi-national corporations and states, bear
moral responsibility to address structural injustice (McKeown, n.d.). Continuing with the example
of sweatshop labor, I do not have the capacity to change the structural processes that cause the race
to the bottom in global garment supply chains – the lowering of wages, the excessive working hours,
the lack of fire and building safety measures, etc. But some agents do have that capacity – the largest
multi-national corporations (MNCs) in the industry. According to the McKinsey “State of Fashion
2019” report, a mere 20 corporations make 97% of profits in the global garment industry (a situation
which has not significantly changed as a result of the COVID-19 pandemic – if anything, these
corporations have consolidated their power) (Amed et al., 2018: 11, 2020: 111). These corporations
set the rules of the game through private-private partnerships (partnerships between private cor-
porations with no state involvement, such as the Ethical Trading Initiative) which determine how
the industry will be regulated, and they act in various ways to maintain their domination over
workers, such as lobbying. Moreover, large garment MNCs demonstrated their capacity to improve
working conditions for garment workers after the Rana Plaza factory collapse in 2013, in which
1,134 people lost their lives. They implemented the Bangladesh Fire and Building Safety Accord; a
legally-binding set of regulations that led to the inspection of over 2000 factories and the re-
mediation of fire and building safety issues in 90% of cases. However, the Accord only lasted for 5
years (2013–18), with a 3-year extension to 2021; it is now being phased out. The Accord only
addressed fire and building safety, not wages, working hours, or workers’ rights to unionize; and it
only applied in Bangladesh. MNCs could have done a lot more to improve the conditions for
workers in Bangladesh and they could implement similar measures across the whole sector, but they
fail to do so. For that reason, I argue they are blameworthy for the perpetuation of structural
injustice. Moreover, understanding the ways in which powerful agents act to maintain unjust
structures provides insight into how structural change can occur.
The value of the structural injustice approach is that it reveals the economic, social, and political
forces that combine to render some social groups vulnerable to oppression and domination. It has
373
Maeve McKeown
also opened up a debate about who bears the responsibility to remedy the injustice that is embedded
in these structures, acknowledging the necessary interconnection of individuals, institutions, and
socio-political-economic structures. This is in contrast with the theorists who place the burden of
responsibility for justice solely on particular institutions (Rawls, luck egalitarians), or on individuals
(utilitarians).
6 Conclusion
The distributive paradigm of justice is impoverished, or at least incomplete, and it is a deep shame
that it has dominated political philosophy so thoroughly over the last fifty years, only now releasing
its grip (to some extent). As Katrina Forrester puts it, political philosophy has been living “in the
shadow of justice,” meaning the shadow of Rawlsian distributive justice specifically. She argues that
“The Rawlsian framework came to act as a constraint on what kind of theorizing could be done and
what kind of politics could be imagined” (Forrester, 2019: 275).
Why was it so popular? Possibly, as Charles Mills argues, because White men completely dominated
the profession until recent years, and they didn’t see or experience these other kinds of injustice
outlined here and so didn’t consider them relevant. Claims to epistemic injustice, in particular, have
developed since scholars from minority groups have become more integrated in the discipline. Another
reason is the artificial distinction between Anglo-American and continental philosophy, with Anglo-
American philosophers focusing on distributive justice and continental philosophers focusing on
recognition, with the two branches rarely talking to one another. That is one of the reasons why Iris
Marion Young’s work has been so innovative in the realm of social justice, because she worked within
both frameworks. Katrina Forrester offers a historical explanation, which is that Rawls was committed
to the kind of thinking that was dominant in his youth – the post-war New Deal era – and that luck
egalitarians were influenced by the right-wing Reagan-Thatcher era of the 1980s, with its overweening
emphasis on personal responsibility. Another reason might be the emphasis on “ideal theory” – the-
orizing about justice in the ideal – which edges out forms of injustice that are relational and the result of
past injustice.
Whatever the explanation, it is now apparent that social injustice can take many forms. In this
chapter, I have highlighted four: recognition, epistemic injustice, relational inequality, and structural
injustice. Social justice, therefore, is not reducible to one single issue; it is multifaceted and a work in
progress theoretically and in reality.
Note
1 Of course, Rawls did not only talk about the distribution of wealth, but also the distribution of the social
primary goods, including things like the social bases of self-respect. However, Young acknowledges this and
claims that another problem with the distributive paradigm is that it applies the logic of distribution to things
that cannot be distributed like self-respect, rights or power. None of these things are entities that a state can
distribute, they are relationships. I footnote this point because it will be of interest to the reader who is more
engaged with contemporary political philosophy and will be concerned that Young made a strawman out of
Rawls (which is an open question), but for the lay reader, the points in the body text are sufficient to
capture Young’s view.
References
Amed I., Balchandani A., Beltrami M., et al. (2018) The State of Fashion 2019: A Year of Awakening. New York.
Available at: https://www.mckinsey.com/industries/retail/our-insights/the-state-of-fashion-2019-a-year-
of-awakening.
Amed I., Berg A., Balchandani A., et al. (2020) The State of Fashion 2021: In search of promise in perilous times.
New York. Available at: https://www.mckinsey.com/industries/retail/our-insights/state-of-fashion.
374
Social Justice
Anderson E. S. (1999) “What is the Point of Equality?,” Ethics 109, 2, 287–337.

Bettcher T. (2020) “Feminist Perspectives on Trans Issues,” The Stanford Encyclopedia of Philosophy. Fall.
Blau P. (1977) Inequality and Homogeneity: A Primitive Theory of Social Structure. London; Collier MacMillan Publishers.
Bourdieu P. and Nice R. (1977) Outline of a Theory of Practice. Cambridge; Cambridge University Press
Translated by Richard Nice.
Collins P. H. (2000) Black Feminist Thought: Knowledge, Consciousness, and the Politics of Empowerment. Second
Edition. New York; Routledge.
Coulthard G. S. (2014) Red Skin, White Masks: Rejecting the Colonial Politics of Recognition. Minnesota; University
of Minnesota Press.
Dotson K. (2011) “Tracking Epistemic Violence, Tracking Practices of Silencing,” Hypatia 26, 2, 236–257.
Du Bois W. E. B. (2007) The Souls of Black Folk. Oxford; Oxford University Press.
Fanon F. (2008) Black Skin, White Masks. London; Pluto Press.
Forrester K. (2019) In the Shadow of Justice: Postward Liberalism and the Remaking of Political Philosophy. Princeton;
Fraser N. (1997) Justice Interruptus: Critical Reflections on the “Postsocialist” Condition. London; Routledge.
Fraser N. (2010) Scales of Justice: Reimagining Political Space in a Globalizing World. Cambridge; Polity Press.
Fraser N. and Honneth A. (2003) Redistribution or Recognition? A Political-Philosophical Exchange. London; Verso.
Fricker M. (2007) Epistemic Injustice: Power and the Ethics of Knowing. Oxford; Oxford University Press.
Gordon L. R. (2007) “Iris Marion Young on Political Responsibility: A Reading Through Jaspers and Fanon,”
Symposia on Gender, Race and Philosophy 3, 1, 1–7.
hooks bell (1987) Ain’t I am Woman? London; Pluto Press.
Lebron C. (2014) “Equality from a Human Point of View,” Critical Philosophy of Race 2, 2, 125–159.
McKeown M. (2017) “Sweatshop Labor as Global Structural Exploitation,” in Deveaux M. and Panitch V. eds.
Exploitation: From Practice to Theory. London; Rowman and Littlefield Publishers.
McKeown M. (2018) “Iris Marion Young’s ‘Social Connection Model’ of Responsibility: Clarifying the
Meaning of Connection,” Journal of Social Philosophy 49, 3, 484–502.
McKeown M. (n.d.) With Power Comes Responsibility: The Politics of Structural Injustice. London; Bloomsbury
Academic.
Medina J. (2013) The Epistemology of Resistance: Gender and Racial Oppression, Epistemic Injustice, and Resistant
Imaginations. Oxford; Oxford University Press.
Mills C. W. (2007) “White Ignorance,” in Sullivan S. and Tuana N. eds. Race and Epistemologies of Ignorance.
New York; SUNY Press, pp. 11–39.
Nussbaum M. C. (2000) Women and Human Development: The Capabilities Approach. Cambridge; Cambridge
University Press.
Rawls J. (1971) A Theory of Justice. Revised Edition. Cambridge, MA; The Belknap Press of Harvard University
Press.
Rose J. (2016) “Who Do You Think You Are?,” London Review of Books 38, 9. Available at: https://www.lrb.
co.uk/the-paper/v38/n09/jacqueline-rose/who-do-you-think-you-are.
Sen A. (1979) “Equality of What?,” in The Tanner Lecture on Human Values. https://tannerlectures.utah.edu/_
resources/documents/a‐to‐z/s/sen80.pdf.
Smith J. (2020) “Covering Peter Sufcliffe’s Crimes, I Saw That Women Weren’t Listened to – And They Still
Aren’t,” The Guardian, 13 November. London. Available at: https://www.theguardian.com/
commentisfree/2020/nov/13/peter-sutcliffe-crimes-women-police-investigation-murders-misogny.
Spivak G. C. (1997) “Can the Subaltern Speak?,” in The Post-Colonial Studies Reader. London: Routledge,
pp. 24–29.
Taylor C. (1994) “The Politics of Recognition,” in Amy Gutmann ed. Multiculturalism. Princeton; Princeton
University Press, pp. 25–73.
Williams R. R. (2013) Recognition. International Encyclopedia of Ethics. Hoboken, NJ; Blackwell Publishing Ltd.
Young I. M. (1990) Justice and the Politics of Difference. Princeton; Princeton University Press.
Young I. M. (1997) “Unruly Categories: A Critique of Nancy Fraser’s Dual Systems Theory,” New Left Review,
147–160.
Young I. M. (2011) Responsibility for Justice. Oxford; Oxford University Press.
Further Reading
Anderson E. S. (1999) “What Is the Point of Equality?,” Ethics 109, 2, 287–337. (The founding article of the
relational egalitarianism literature.)
375
Maeve McKeown
Forrester K. (2019) In the Shadow of Justice: Postward Liberalism and the Remaking of Political Philosophy. Princeton;
Princeton University Press. (An historical overview of justice theory in the post-war era.)
Fraser N. and Honneth A. (2003) Redistribution or Recognition? A Political-Philosophical Exchange. London; Verso.
(Fraser and Honneth’s views on the relationship between redistribution and recognition.)
Mills C. W. (2007) “White Ignorance,” in Sullivan S. and Tuana N. eds. Race and Epistemologies of Ignorance.
New York; SUNY Press, pp. 11–39. (An important point of reference for social epistemology.)
Young I. M. (1990) Justice and the Politics of Difference. Princeton; Princeton University Press. (Classic critique of
the distributive paradigm of justice.)
376
30
JUSTICE ACROSS BORDERS
Serena Parekh
1 Introduction
How we distribute important economic and social goods in society is a perennial topic of philo-
sophical, political, and economic discourse. Philosophers refer to these as questions of distributive
justice. Discussions of distributive justice usually focus on justice within a particular country. This is
because countries are the units in which goods get distributed. Canada, for example, decides how
much inequality will exist in their society through their tax policy, how much the government will
subsidize university education, and how strongly they will enforce non-discrimination policies in
employment, etc. Questions of economic justice have traditionally been understood as domestic
in nature.
Increasingly, however, globalization has forced us to think about questions of distributive justice
globally and consider our obligations to people beyond the borders of a particular country. There is a
good reason for this. Globalization has meant that countries have an increasing economic and
political interdependence. As a result, the decisions made in one country can have deep ethical
implications for many others. Sometimes the effects are obvious: when the United States (US)
invaded Afghanistan in 2001 after the 9/11 attacks, it created, among other things, hundreds of
thousands of refugees who fled the country, creating a refugee crisis in Pakistan and other countries.
Sometimes the effects of domestic policies are more subtle and require more effort to see them. For
example, in 2008 there was a global food crisis that caused many of the poorest people in the world
to experience severe food insecurity. This was caused, in part, by the US’s policy of subsidizing
biofuel production which created economic incentives to shift agricultural production from pro-
ducing food to producing grains for biofuel. The scarcity in grains for food meant higher prices and
this made it difficult for the poor around the world to afford food (Ackerly, 2018: 33). Domestic
decisions big and small can have large ethical implications.
Further, globalization has provided the tools to allow us to know more about the implications of
our actions around the world. Hundred years ago, a citizen of the UK might have little idea about
the levels of poverty in Sudan or the impact of energy consumption on the climate nor could they
watch a famine unfold in Yemen via videos on their Twitter feed. But in the 21st century, such
information is widely available and this fact – that we know about the suffering of others across our
borders and the ways that we are connected to it – gives rise to a new way of thinking about what
we owe to each other, where nationality is not the only consideration. Distributive justice requires a
broader analysis given this increase in knowledge and awareness.
DOI: 10.4324/9780367808983-36 377

Serena Parekh
Philosophers and theorists have responded to this new reality and its implications for justice across
borders in many different ways. This chapter will give an overview of the contours of the debate
about what justice across borders entails. Though the debate is expansive and covers a wide range of
topics and positions, I identify four different approaches to the topic. I refer to the first two ap-
proaches I examine as the foundational approaches because of how influential they have been in
subsequent debates. Peter Singer is often credited with beginning this discussion of global justice
with his 1972 article, “Famine, Affluence, and Morality.” John Rawls puts forth an equally influ-
ential view about the more limited obligations wealthy liberal democracies have to poorer countries.
Though they disagree in their conclusions, there is a shared assumption for both of these thinkers
that whatever we owe to the poor, the obligation is not due to the fact that we are trying to rectify a
harm. That is, wealthy countries did not cause poor countries to be poor or deliberately harm them
in some way. Rather, justice requires considering how much economic help we need to give poor
countries just because they are poor and their people are in need.
The third approach I examine questions this assumption and starts from the view that wealthy
countries have in fact played a causal role in the poverty and human rights violations of people in
poor countries via historical actions such as colonialism and neocolonialism, and contemporary
economic policies, such as trade and tax policies. Thomas Pogge and others argue that we have
structured our global economic institutions in ways that systematically harm the poor. In ad-
dition to our economic policies harming the poor, our political decisions to interact with de-
veloping countries in certain ways, have contributed to their impoverishment. Many of the root
causes of poverty – corruption, weak economic structures, lack of democratic culture – can be
traced to political policies implemented in Western states and supported by the global com-
munity. Obligations to the poor are not a matter merely of transferring resources to “help” poor
states, but a matter of stopping the harm that we’re doing via our economic and political
institutions.
The fourth approach begins with this insight: that countries are deeply interdependent through
our global economic and political institutions but takes it one step further beyond transnational
exploitation. Philosophers like Iris Marion Young and Brooke Ackerly, among others, argue that we
ought to see our interconnection even more broadly. Young argues for the existence of a kind of
global structural injustice, where the injustice stems not just from unfair institutions like the ones
Pogge identifies, but also from norms and everyday practices that countries as well as individuals
uphold. Injustice itself, as Ackerly calls it, requires critical examination of these norms and social
structures.
In my view, this is the perspective we need in order to think about justice across borders from a
Philosophy, Politics, and Economics (PPE) perspective, especially if we are interested in under-
standing what our roles ought to be, as citizens of wealthy countries or as individuals concerned
about injustice we see in the world. I’ll illustrate why we ought to take structural injustice and global
interdependence seriously through looking at two issues – women’s rights and refugees. It will
become clear that justice across borders requires that we take seriously the perspectives of structural
injustice and global interdependence on economic, political, and ethical levels.
2 Foundational Arguments
Discussions of justice across borders often begin with the following observation: we live in a world
of profound inequality, with some living in luxury that is almost unimaginable and others, the
majority of the world’s population, living in poverty, often extreme and debilitating. The moral
question quickly becomes what, if anything, do those who possess wealth owe to those who have
very little and often not enough to live a life of basic dignity? Whose responsibility is it to help the
poor and why?
378
Justice Across Borders
One of the most influential answers to this question comes from Peter Singer. He argued that we
have a duty to donate to charities that can help save the lives of those living in extreme poverty
around the world. This obligation is based on the principle of humanity: if someone is suffering
greatly and you can relieve or eliminate that suffering at relatively low cost to yourself, you ought to
do so. If you fail to help, you have done something morally wrong (Singer, 1972).
Singer bases his view on an intuition he believes is widely held: that it is wrong not to help
someone in extreme need when we are able to without sacrificing anything significant. To illustrate
the strength of this intuition, he asks us to imagine walking past a shallow pond and seeing a child
drowning. You realize that you are in a position to save the child and the only cost to you would be
wet shoes and perhaps being late for work. The suffering that would be caused by the child
drowning (to the child, to their parents, to the community) would be immense and because you
could stop this without risking anything of comparable moral worth you ought to do so. Indeed,
you would be monstrous if you choose to have dry shoes over saving the child.
In his view, this is how people in wealthy countries stand to those in poverty: we have
the capacity to easily reduce or eliminate the suffering of the poor without sacrificing anything of
value ourselves. Our proximity to the harm does not matter. “It makes no moral difference whether
the person I can help is a neighbor’s child ten yards from me or a Bengali whose name I shall never
know, ten thousand miles away” (Singer, 1972: 231–232). Based on this, Singer concludes that those
of us living in wealthy countries that have money to spend on luxuries like going to the movies have
a duty to donate substantially to the poor who live outside our countries because they are the ones
most likely to be in extreme need. Charity is an obligation of justice across borders.
The obligation to donate to charity to save the lives of the poor is rooted in the principle of the
Good Samaritan, which is sometimes called the principle of humanity. We have an obligation to
help under these circumstances even if we played no role in causing their harm or suffering. What is
interesting about Singer’s view is that he takes a widely shared intuition – that when people are in
desperate need and we can help them, we ought to, regardless of whether they are fellow citizens –
and shows that it leads to a conclusion that in fact goes against how many people think about our
obligations to the poor. Donating money to fight poverty becomes not merely an act of charity that
we are free to do or refrain from doing, but an obligation: “we ought to give the money away, and it
is wrong not to do so” (Singer, 1972: 235). The implication of this, namely that we are not entitled
to our disposable income but have a duty to donate this to fight poverty, is morally demanding.
A second highly influential account of our obligations to the poor across our borders comes from
John Rawls. Rawls discusses the obligations of countries that are liberal democracies, rather than
individuals, to poor countries that are “burdened” by unfavorable conditions and as a result are poor
(Rawls, 1999). Rawls opposed the view that wealthy, liberal democracies have obligations to re-
distribute wealth on a global scale and should aim to create a more egalitarian world order. He did,
however, hold that “well-ordered societies” have duties to help burdened countries become
“decent,” in the sense of having strong political and economic institutions.
Well-ordered peoples have a duty to assist burdened societies. It does not follow, however,
that the only way, or the best way, to carry out this duty of assistance is by following
a principle of distributive justice to regulate economic and social inequalities among
societies. (Rawls, 1999: 106)
Our obligations to poor countries do not require us to eliminate poverty or global inequality since,
“great wealth is not needed to establish just (or decent) institutions” (Rawls, 1999: 107), which for
Rawls is the goal of his duty of assistance.
This way of thinking about our distributive justice obligations is rooted in Rawls’s understanding
of poverty.
379
Serena Parekh
A society with few natural resources and little wealth can be well-ordered if its political
traditions, law, and property and class structure with their underlying religious and moral
beliefs and culture are such as to sustain a liberal or decent society. (Rawls, 1999: 106)1
Because, in his view, poverty is due to these internal features of a country, well-ordered societies
should aim to change these conditions. The goal of international duties of aid is to change political
cultures, not to relieve poverty through direct aid.2
Though Singer’s and Rawls’s approaches lead to very different obligations to the global poor,
they do have a few things in common. Importantly, wealthy countries and individuals have duties to
help the poor across their borders, though for different reasons. For both, the reason poor countries
are poor is either irrelevant or not the fault of the wealthier countries. For Rawls, poverty is caused
by a poor country’s political culture. For Singer, our obligations to help the poor are not rooted in
anything wealthy countries may have done or not done in terms of causally contributing to global
poverty; what matters is our ability to help. For both, wealthy countries stand in the position of the
Good Samaritan who can offer help to the needy when they determine it’s appropriate but cannot
be held responsible for causing the harm of poverty.
3 From the Good Samaritan to Institutional Interconnection

Many non-philosophers, I think, share Singer’s and Rawls’s assumptions about the relationship
between wealthy and poor countries. Many in wealthy countries see themselves as being in a po-
sition to help the poor because of our resources and ability to help, not because we have caused the
poverty. Some would agree with Rawls that poor countries are poor because of internal causes –
allowing rampant corruption, improperly managing their natural resources and having cultures that
don’t value hard work. Those who don’t share this view may find themselves agreeing with Singer
that the reason poor countries are poor doesn’t matter when their need is so great and we can help
them without really inconveniencing ourselves. When wealthy countries generously provide de-
velopment aid and individuals donate to charities that help the poor, they are not doing it to make
up for a past harm or to compensate for an unjust economic policy that might have contributed to
the poverty. Wealthy countries are unconnected from poor ones in these ways.
Though this picture that I’ve described is widely held, both by philosophers and many in the
general public in wealthy countries, there are good grounds for challenging it. Much contemporary
global justice scholarship has contributed to debunking this way of viewing relations between
countries and showing that the profound interdependence of countries gives rise to a very different
way of understanding justice across borders. In the next section, I explain a few strands of this
critique and demonstrate why, in my view, scholars that want to understand justice across borders
must take seriously the ways that the economic and political policies, norms, and practices of those
of us living in wealthy countries have caused and sustain the poverty, inequality, and displacement of
those living in poor countries. Whatever distributive justice obligations we have to the poor across
our borders must flow from this understanding.
One of the strongest critiques of both ways of approaching justice across borders comes from
Thomas Pogge. In his view, one of the main errors of Rawls’s approach is his belief that poor
countries themselves are solely responsible for their poverty. He is critical of Rawls’s assertion that
the causes of poverty are “purely domestic” and unconnected to history or contemporary global
economic and political institutions. The view that global poverty is due to solely domestic factors is
empirically inaccurate and leads to a moral error. The moral error is that we are only responsible as
bystanders, not because we are causing the harm in question. It is more accurate to say that the West
has directly shaped global poverty by structuring international rules so that they favor rich countries
and disadvantage poor countries. Once we understand global poverty through this lens, we see that
380
our strongest duties are not to provide charity or development aid but are in fact much stronger: we
have duties to stop hurting the poor.
Though Pogge agrees that historical injustices such as colonialism and enslavement have con-
tributed to the status quo and resulted in the current state or profound inequality, his argument
against the purely domestic poverty thesis, as he calls it, does not rest on this. In his view, the current
global institutional order, with its various political and economic structures – largely influenced by
wealthy countries and created for their benefit – plays a large role in the causes and persistence of
severe global poverty.
Our world is characterized by considerable international economic interaction regulated by

an elaborate system of treaties and conventions about trade, investments, loans, patents,
copyrights, trademarks, double taxation, labor standards, environmental protection, use of
seabed resources, and much else … Had these rules been shaped to be more favorable to
the poor countries, their people would now be much better off. (Pogge, 2004: 4)
Rather than being a neutral set of policies that allow countries to work together, Pogge shows how
they systematically benefit wealthy countries and directly contribute to global poverty. For example,
wealthy countries are able to use their power and influence in trade negotiations to put protectionist
measures in place on goods such as agriculture and textiles and impose tariffs on imports from poor
countries, thus preventing poorer countries from competing in these lucrative markets. In this way,
the global rules for international trade allow wealthy countries to maintain their historical advantage
and prevent poorer ones from catching up.
One reason that some philosophers oppose the view that distributive justice can be applied at a
global scale is because there is not the same level of coercive political and economic institutions as
the domestic level. International institutions, though influential, don’t impact individuals in the same
ways that domestic institutions do and therefore don’t require the same kinds of moral justifications
(Blake, 2001, 2013). Pogge’s work takes pains to show that this is not the case, and that because
global political and economic institutions systematically disadvantage less powerful and wealthy
countries, these too should be subject to constraints of justice. The causes of a country’s poverty are
directly related to the structure of the global order. While domestic factors certainly play a role in the
persistence of poverty, these factors are heavily impacted by the rules of the international order.
Justice, then, requires us to reform these institutions in ways that contribute to alleviating poverty,
even if that goes against a wealthy country’s perceived national interest. This is very different from
an obligation to donate money (Singer’s view) or help countries achieve just institutions (Rawls’s
view). In fact, we should be thinking about our obligations as
not merely “assisting” the poor abroad, but mitigating the effects of unfair rules that bring
us unjust gains at their expense. We are not redistributing from the rich to the poor, but
offsetting an unjust redistribution from the poor to the rich – re-redistributing, if you like.
(Pogge, 2004: 17)
4 Structural Injustice
Pogge’s work helps us to see that countries are deeply interconnected in the 21st century and justice
across borders requires us taking this interconnection seriously. Building on this insight, Iris Marion
Young has created a framework that helps us to understand – and take responsibility for – the ways
that injustice can arise due to our global interconnection. While Pogge traces global problems like
poverty to the global institutional order, Young holds that we ought to interrogate global structures
and find ways to address what she calls global “structural injustice.”
381
Serena Parekh
The structural view stresses the ways that the advantages that are enjoyed by affluent citizens in
wealthy countries are connected in deep and sometimes surprising ways to the disadvantages suffered
by others not just because of the economic structure of the global order (as Pogge highlighted) but
also because of the legal rules, social norms, and other policies and practices that virtually everyone
contributes to and that many benefit from. Structural injustice
exists when social processes put large categories of persons under a systematic threat of
domination or deprivation of the means to develop and exercise their capacities, at the
same time as these processes enable others to dominate or have a wide range of oppor-
tunities for developing and exercising their capacities. (Young, 2006: 114)
Young encourages us to see injustice in a new light, where injustice is the result of a pattern of
action that occurs within social, economic, and political structures that benefit some and dis-
advantage others.3
The reason the structural injustice perspective is helpful when it comes to justice across borders is
because it captures some of the ways that injustice operates globally that are not captured by more
traditional conceptions of justice. For example, Young argues that injustice can result not only from
someone deliberately and knowingly harming another person (what she refers to as the liability
model) but also as a consequence of many different individuals and institutions pursing their own
interests, goals, and ends and acting according to widely accepted norms. In fact, her view forces us
to take seriously the way that injustice is often unintended and the outcome of ordinary, everyday
actions that result nonetheless in a system that puts some under conditions of vulnerability and
privileges others (McKeown, 2018).
To illustrate structural injustice at a global level, Young asks us to consider the kind of injustice
involved in “sweatshops.” Sweatshops are factories that manufacture goods in poor countries, goods
that are mostly exported to wealthier countries. They are known for employing mostly young
women who work under what are often abhorrent conditions – long hours, poor ventilation, and in
unsafe factories – and in ways that can lead to disability and even death. Because of global poverty
and global trade, many multinational corporations set up factories in poor countries where the cost
of labor is cheaper and other regulations, such as labor standards and environmental regulations, are
easy to get around. Young thinks that many will see these working conditions and the exploitation
of these women for the sake of producing cheap clothing as an injustice, yet it is not one that can be
traced to any ill intention or deliberate policy on the part of the garment manufacturers, the local
government, the multinational corporations or their consumers. Each entity was acting in ways that
are considered acceptable – increasing tax revenue, maximizing profit, buying clothes cheaply – and
no one intended the injustice. Yet despite this, there is an injustice that deserves a remedy. The
structural injustice account can help us to understand why this is an injustice and to think about how
we can begin to address it.
What does justice across borders require on the structural injustice account? For Young, justice
begins by taking seriously the various ways that people are connected to the injustice and then taking
political responsibility to make conditions more just in the future. Young develops the social
connection model to help us understand that though we may not have intended a particular harm,
our connection to the harm is sufficient to ground a responsibility to work towards justice. The
injustice of sweatshops is connected to norms of global capitalism that make it morally acceptable to
drive labor costs as low as possible; to international agreements that allow multinational companies
to work in poorer countries with very little labor oversight in order to maximize their profit, and to
the demand of consumers in wealthy countries to buy clothes as inexpensively as possible. We can
see how individuals, the government, corporations, NGOs, and various other entities contribute to
the structures that place some women in a position of vulnerability where taking a job in a
382
sweatshop is their best option. Young helps us to see global injustice as directly connected to us via
our participation in these unjust structures and as a result, we can see the injustice in a new light.
In order to address structural injustice Young urges us to avoid the backward-looking kind of
responsibility that seeks to find an individual liable for the harm. For example, after the Rana Plaza
disaster in 2013, where a sweatshop in Bangladesh collapsed killing 1,132 people and injuring 2,500
more, many tried to find someone to blame: the factory owner, building inspectors, corrupt gov-
ernment officials. Though it was important to find those who were culpable, this does little to
address the deeper structural injustice. Focusing on finding someone to blame for a particular tragedy
may lead us to avoid asking about the deeper, structural causes that led to the situation in the first
place. It would allow us to ignore the reality that there are many other sweatshops with many other
women and girls working in similar – though not yet tragic conditions – around the world (Ackerly,
2018: 84). Finding someone to blame for a particular tragedy will do little to address structural
injustice and make conditions better for poor workers in the future.
By contrast, responsibility for structural injustice is based “not simply membership in a group or
nation, but rather the fact that we participate by our actions in the operations of institutions that
together sometimes produce injustice” (Young, 2011: 123). Her account of structural injustice does
not focus on finding someone to blame, but instead on taking responsibility to transform conditions
so that they are more just in the future; it is a forward, not backwards, looking account.
Responsibility for Young is rooted in her “social connection” model of responsibility, which holds
that “all those who contribute by their actions to structural processes with some unjust outcomes
share responsibility for the injustice” (2011: 96). Once we realize our connection to sweatshops,
through, for example, purchasing clothes from multinationals that contract with them, and the role
we play in supporting the norms of global capitalism that allow so many to live in conditions of
vulnerability, we can begin to think about ways that we can transform these conditions.
As a result, what justice across borders means under conditions of structural injustice will be
different from the previous views analyzed. Justice for women working in sweatshop conditions will
not be realized with charity or development aid to make institutions more just, as Singer and Rawls’s
approaches indicate, since it is an injustice that is sustained and perpetuated by larger structural forces
related to global capitalism and global inequality. While addressing structural injustice requires that
we look at global institutions and policies, such as trade and taxation, that contribute to this, as
Pogge’s account asks us to do, this is only part of what we must do to address structural injustice.
The structural injustice account requires that we acknowledge that global injustices such as
sweatshops are connected to underlying structural problems, such as global poverty and gender
inequality, and deeply entrenched norms about global capitalism, human exploitation, and en-
vironmental degradation that create the structural conditions that allow some people to be exploited
and others to benefit. Young’s analysis requires that we find ways to make these structures more just
in the future. To do this, we cannot focus only on the immediate problem – such as by prosecuting a
negligent factory owner, donating money to a relief fund or even boycotting a particular company –
but that we rethink our norms, values, institutions, and policies so that they do not put some under
threat of domination and deprivation while benefiting others. Justice across borders is much more
challenging under this view but it helps us to understand how we may begin to address the core of
global injustice in order to genuinely transform it.
5 Global Gender Justice

I agree with others who have come to reject the standard view of global justice, the view which
holds that we are independent units that can decide when and how much to “help” others whose
poverty or suffering is unconnected to us. Both on an institutional level and on the level of
social structures, we – as members of a particular country and as individuals living in a globalized
383
Serena Parekh
world – are deeply interdependent and our obligations arise from this interdependence. As Pogge
and Young, among others, have demonstrated, the poverty and human rights violations experienced
by many around the world can be traced to institutions, practices, and norms that people in
wealthier countries support and benefit from, and this relationship is the ground of our responsibility
to those experiencing injustice outside of our countries. Justice across borders is rooted in our global
political, economic and social interdependence. Let me illustrate this point with two examples.
Take the issue of women’s human rights. Discussions were often framed as whether and to what
extent supposedly enlightened Westerners should help “Third World Women” (Mohanty,
2003;Khader, 2019) and challenge the backwards cultural practices that kept them oppressed. Susan
Moller Okin is among the most well-known proponent of this view, arguing that traditional cultures
are the primary reason some women around the world have not achieved equality and continue to
experience often brutal human rights violations. This approach shares the underlying assumption of
the traditional view of global justice, that Western countries can see themselves as bystanders to the
injustice in question, as causally unconnected, and simply in a position to offer help. This view
regarding women’s rights has been challenged by many feminists, notably Alison Jaggar.
In Jaggar’s view, it’s incorrect to treat problems like female poverty and women’s human rights
abuses as due primarily to local cultural traditions because it assumes that these practices are in-
dependent and unconnected to Western culture or global economic institutions. The injustices
suffered by poor women around the world must be understood as connected to global economic
structures. Though globalization had the promise to create the conditions that would allow women
to achieve economic equality and raise their social status, it often brought the opposite by increasing
the gap between the rich and poor, intensifying environmental exploitation through privatization of
public resources and ultimately creating “a system hostile or antagonistic to women” (Jaggar,
2001: 301).
For example, institutions like the IMF and World bank have for years imposed neoliberal
economic policies that require poor countries to implement economic changes that have sig-
nificantly hurt women. For example, structural adjustment programs often require that countries
increase agricultural exports and open their markets to heavily subsidized imports from Western
countries,4 so locally grown agricultural products are unable to compete with heavily subsidized
foods from richer countries. Yet many women in poor countries had traditionally survived through
small scale subsistence agriculture. After these policies were imposed, this was no longer possible,
and women were pushed to work in cities and shanty towns where they struggled to survive in the
informal economy, due to low wages and poor working conditions. Neoliberal economic policies
often insist that poor countries cut back on social services, such as public health and education,
which affect women more than men because they are often responsible for caring for children and
other family members.
To connect this to Young’s view of structural injustice, it’s easy to see how poor women might
choose to work in sweatshops, regardless of the exploitation, because all other ways of surviving
have been undermined due to these economic policies and norms around gendered labor. Jaggar
demonstrates clearly that the injustice and poverty experienced by woman in poor countries is
deeply interconnected to global institutions, led by Western countries, and to economic policies
they impose. Justice across borders requires reforming these policies and institutions and taking
seriously the outcomes for poor women and poor countries. No amount of charity will eliminate
poverty as long as these structural conditions are in place.
6 The Refugee Crisis

My own work on the contemporary global refugee crisis demonstrates a similar point. Like Pogge,
Young, and Jaggar, I argue that we ought to reject the bystander view assumed by Singer and Rawls
384
regarding how those of us in relatively wealthy liberal democracies should respond to the global
refugee crisis (Parekh, 2020a, 2020b). I reject the view that holds that such countries have not caused
the refugee crisis and so are free to take in refugees or not, depending on their own values, needs,
and political commitments. This, I argue, misunderstands the global refugee crisis and the causal role
that Western states have played in what I refer to as the crisis for refugees – the inability of most
refugees to access the minimum conditions of human dignity while they are refugees – which, I
suggest, must be understood as a kind of structural injustice. Understanding justice across borders for
refugees requires that we take seriously the role of Western states in creating many of the injustices
and human rights violations that refugees experience once they have left their home countries and
are seeking refuge. Justice will require not merely helping refugees in the short term, either by
donating to charities that work with refugees or through resettlement but by reforming the global
refugee protection system so that it can be more just in the future and provide, at minimum, basic
dignity for refugees.
The global refugee crisis is often framed as a crisis where Western countries are being asked to
take in large numbers of refugees and asylum seekers from poor countries that are often dissimilar
religiously and culturally. Like Singer’s analogy of the child drowning in the pond, Western states
see themselves as bystanders who are able to help refugees but who are not causally connected to the
crises that caused people to flee. When the refugee crisis is framed in this way, justice across borders
implies that the way we ought to help refugees is by allowing at least some refugees into our
countries. In the view of some, justice requires taking in many refugees; for others, values like
freedom of association and political or cultural self-determination mean that states are free to take in
few if any refugees.5 Whatever position you take, there is a tacit agreement that Western countries
are bystanders and free to decide for themselves what help to give refugees and the only way to help
refugees is by admitting them to our countries.
This approach misunderstands the injustice refugees are experiencing. The real crisis for refugees
is that once they have fled their home countries, the vast majority of refugees cannot get refuge, that
is, they cannot access the minimum conditions for human dignity while they wait for a more
permanent solution such as returning home or being resettled. This outcome must be understood as
something that Western states have created and sustained, even though it was not intentional. The
crisis for refugees is that all the options the international community allows them to access – refugee
camps, urban settlements without any aid, or dangerous journeys to seek asylum – do not provide a
minimal level of human dignity. This might not be such an injustice if refugees remained in this
situation for a short period of time before being resettled. But in the 21st century, refugees remain
refugees for on average 17 years and have less than 1% chance of being resettled in a new country
(Parekh, 2020a: 4–5). That the vast majority of refugees will never receive refuge is, in my view, a
profound injustice.
The situation just described ought to be understood as a structural injustice. The outcome, that so
few refugees are able to access refuge, was not intended by any one state or institution but resulted
from many different states acting according to their own political and economic interests (such as
border security) and according to widely accepted political principles (such as national sovereignty).
The result is sustained by xenophobia and attitudes around nationalism that hold that we can treat
non-nationals in ways that we consider illegitimate to treat fellow citizens (e.g., by taking away their
children when they seek asylum, as was done in the US in 2018). Even though we cannot hold any
one individual or country liable for the overall outcome, it remains an injustice that must be
addressed.
Justice across borders for refugees will require transforming the current norms, institutions, and
practices. It will require that we rethink and reformulate the structures that are in place to respond to
refugees from political, economic, and philosophical perspectives. We can start with the outcome –
that so few refugees have meaningful access to refuge – and reconsider ways that we structure the
385
Serena Parekh
options for refugees. For example, rather than consign refugees to camps, we can work with
countries that host refugees and try to come up with ways that would allow refugees to live with
dignity and autonomy, such as by supporting programs that allow refugees to work. In addition, we
can reimagine resettlement where more countries are involved and resettlement is seen as re-
quirement, not a matter of discretion (Carens, 2015). With this moral grounding in place, reimaging
options for refugees will require an understanding of the economic impact of refugees as well as how
such options might become politically viable.
7 Conclusion
To briefly conclude, how should we understand justice across borders from a PPE perspective?
Regardless of the conclusion, you reach about what is owed to the poor in other countries or to
people experiencing systematic human rights violations, it is clear that the work done in philosophy,
politics, and economics must inform each other, especially on topics like global justice. Philosophers
and theorists must understand the intended and unintended outcomes of economic and political
policies, both domestic and global, and include these in our normative theory. Similarly, political
scientists and economists must also think about the ethical implications for people abroad for what
may appear to be a merely domestic policy, such as refugee resettlement or domestic agricultural
subsidies. While there may be a debate about how exactly we are interconnected and what moral
obligations these interconnections give rise to, it is clear that justice across borders requires a
structural, interdependent approach. PPE as a discipline is well suited to this kind of approach and
has the potential to make an impact on how we understand and respond to issues of justice across
our borders.
Notes
1 More specifically, Rawls writes: “the causes of the wealth of a people and the forms it takes lie in their
political culture and in the religious, philosophical, and moral traditions that support the basic structure of
their political and social institutions, as well as in the industriousness and cooperative talents of its members,
all supported by their political virtues…. the political culture of a burdened society is all-important …
Crucial also is the country’s population policy” (Rawls, 1999: 108).
2 For a defense of this Rawlsian approach in this volume, see Kok-Chor Tan’s chapter, “Institutionalism,
Injustice and Personal Responsibility,” in this volume.
3 For more on Young’s view of structural injustice in this volume, see Maeve McKeown’s chapter, “Social
Justice,” and Ann Cudd’s chapter, “Feminist Theories” in this volume.
4 US and Europe spend $350 billion on farm subsidies, which is six times what they spend on aid (Jaggar,
2005).
5 See Parekh (2020a), Chapter 3, for a detailed discussion of this debate.
References
Ackerly, B. (2018) Just Responsibility: A Human Rights Theory of Global Justice. Oxford; Oxford University Press.
Blake, M. (2001) “Distributive Justice, State Coercion, and Autonomy,” Philosophy & Public Affairs, 30, 3, 257–296.
Blake, M. (2013) Justice and Foreign Policy. New York; Oxford University Press.
Carens, J. (2015) The Ethics of Immigration. New York; Oxford University Press.
Jaggar, A. (2001) “Is Globalization Good for Women?,” Comparative Literature, 53, 4, 298–314.
Jaggar, A. (2005) “‘Saving Amina’: Global Justice for Women and Intercultural Dialogue,” Ethics and
International Affairs, 19, 3, 55–75.
Khader, S. (2019) Decolonizing Universalism: A Transnational Feminist Ethic. Oxford; Oxford University Press.
Mohanty, C. T. (2003) Feminism without Borders: Decolonizing Theory, Practicing Solidarity. Durham; Duke
University Press.
386
McKeown, M. (2018) “Iris Marion Young’s ‘Social Connection Model’ of Responsibility: Clarifying the
Meaning of Connection,” Journal of Social Philosophy, 49, 3, 484–502.
Parekh, S. (2020a) No Refuge: Ethics and the Global Refugee Crisis. New York; Oxford University Press.
Parekh, S. (2020b) “Reframing the Refugee Crisis: From Rescue to Interconnection” Ethics and Global Politics
13, 1, 21–32.
Pogge, T. (2004) “‘Assisting’ the Global Poor,” in Chatterjee, D. K. ed. The Ethics of Assistance: Morality and the
Distant Needy. Cambridge; Cambridge University Press, pp. 260–288.
Rawls, J. (1999) Law of Peoples. Cambridge, MA; Harvard University Press.
Singer, P. (1972) “Famine, Affluence, and Morality,” Philosophy and Public Affairs, 1, 3, (Spring), 229–243.
Young, I. 2006. “Responsibility and Global Justice: A Social Connection Model,” Social Philosophy and Policy,
23, 102–130.
Young, I. (2011) Responsibility for Justice. Oxford; Oxford University Press.
Further Reading
Jaggar, A. (ed.) (2014) Gender and Global Justice. New York; Polity.
Tan, K. (2017) What Is This Thing Called Global Justice? New York; Routledge.
Wisor, S. (2016) The Ethics of Global Poverty: An Introduction. New York; Routledge.
387
PART VII
Democracy
Introduction to Part VII

Political Science already looks at both the theory and practice of democracy from varied perspectives,
but the conflicts between Economic and Philosophic approaches make a PPE analysis quite distinct.
First, the 20th century saw the emergence of Public Choice Theory, which applied many of the lessons
and methods of economic analysis to the study of democratic decision-making. In the hands of people
like James Buchanan, this grounded skepticism towards collective decision-making. This conflicts with
20th-century Philosophy’s celebration of democratic decision-making as a form of government
grounded in an ideal of equality. Second, Philosophers often emphasize the deliberative side of de-
mocracy; it is ideally a forum for productive exchange and reflection. Economists, by contrast, focus on
the procedures of democracy that give each an equal vote. Finally, there are differences in how the
disciplines assess democracy. In Economics, the emphasis has been on aggregated welfare, with Social
Choice Theory trying to identify methods for aggregating individual perspectives into a single measure
of social wellbeing. By contrast, there is a general skepticism of welfarism in Philosophy, and democracy
has been justified by appeal to contractualism, egalitarianism, and basic rights.
Part VII introduces themes in the PPE analysis of democracy. It starts with three chapters that
identify challenges to democracy; a challenge to its moral basis, a challenge for picking the right
procedures, and a challenge for realizing democratic ideals in practice. After that, a chapter looks at
extending democratic ideals beyond its typical purview. The last talks about social conditions linked
with successful democracies. Taken together, these chapters show differing directions that PPE
work can take the study of democracy.
The first chapter in Part VII challenges assumptions about democracy’s superiority. Even if we
think that typical democratic practices are better than other existing political systems, that does not
mean that democracy is the best system. Jason Brennan’s chapter, “In Defense of Epistocracy,” argues
for one alternative. Often Epistocracy is associated with a form of government in which greater
political power is given to those that score well on some assessment of their knowledge. Such an
assessment thereby enables “rule by the knowledgable.” In the chapter, Brennan argues for a so-
phisticated version of Epistocracy that does not exclude people from voting. “Enlightened Preference
Voting” is a system in which all voters take a survey that identifies their voting preferences,
knowledge, and demographics. These surveys are used to determine how the informed members of
different demographic groups have voted. This information is then used to figure out how each
person, given their demographic features, would have voted if they were informed. The results of the
DOI: 10.4324/9780367808983-37
Democracy
election are then determined based on how everyone would have voted if they were informed.
Brennan argues for this system by, first, showing why common assumptions that bolster democracy
are mistaken. He then shows that the failures of democracy are not solved when large numbers of
people vote. Because of this, current democratic practices subject people to injustice by violating a
simple “Competence Principle,” and Enlightened Preference Voting would be a more just alternative.
He ends the chapter by explaining how such a system would work and how it could be organized.
In the second chapter, “Voting Rules,” Itai Sher evaluates different methods of voting by their
formal properties. In so doing, he introduces key axioms of social choice theory and shows how they
apply. For example, he explains why majority rule is the unique procedure that satisfies anonymity,
neutrality, and positive responsiveness in binary elections. He then shows that no single voting
procedure satisfies the commonly defended axioms for multi-candidate elections. In so doing, Sher
demonstrates the range and depth of formal work that has been done to evaluate voting procedures.
The third chapter moves away from the formal properties of democracy and identifies the social
conditions that empower all members of a democracy. Typically, democracies grant each adult a
right to vote, the vote of each person has equal weight, all citizens are eligible to run for office, and
people are allowed to freely express their views. For real democracies, votes must be accurately
counted in periodic elections and the government should act in ways that respond to the election’s
results. In Thomas Christiano’s chapter, a democracy that meets these conditions is only “minimally-
egalitarian.” A truly egalitarian democracy would not only ensure that each vote is equally effective;
it would ensure that each person has equal access to informational power. Having good information
enables people to ensure that government works for them, so unequal information leads to a
government that treats people unequally. Since the cost of seeking information about policy is high
and the benefit to any one person is low, most people acquire information passively. As such, we
must ensure that public institutions make it easier for people to become informed and engaged
voters. In his chapter, Christiano explains how information is spread and received across a de-
mocracy and how institutions affect the distribution of informational power.
The fourth chapter extends enthusiasm about political democracy into the workplace. Today,
there is a general consensus in favor of democratic forms of government. It seems appropriate that
our public life should be responsive to the needs and values of the public, and democratic structures
seem to do this best. Yet, most people willingly submit to non-democratic authority during the
workweek. The decisions made by one’s employer can have much more direct effects on one’s
wellbeing, but most people never doubt the legitimacy of authoritarian workplace structures. In her
chapter, Lisa Herzog examines the arguments used against democratizing the workplace.
Specifically, she looks at claims that property rights, contractual rights, the advancement of worker
wellbeing, or greater productivity favor non-democratic workplaces. She argues that each argument
is lacking. While this provides a defense of workplace democracy, she ends by explaining why we
should not expect occupational self-determination to be a social cure-all.
Instead of focusing on democratic social structures, the final chapter emphasizes the social conditions
associated with successful democracy. In “Social Trust,” Cook and Reidhead examine the different
effects that the level of social trust has in a society. “Social Trust” usually refers to the level of trust that
members of society would put in strangers, but it also refers to trust that the people have in public
institutions. The chapter covers two different research areas. First, it identifies what effects social trust has
on the larger society. They examine the effects of trust on cooperation, economic performance, and
government performance. Second, the chapter identifies what contributes to higher and lower social
trust in society. Following the work of Robert Putnam, there is a widespread belief that ethnic diversity
lowers social trust. Cook and Reidhead break down the various studies on that issue and conclude that
ethnic diversity itself as a lower effect than neighborhood socio-economic status. They also show that
inequality and government corruption seem to have a greater effect on social trust than ethnic diversity
does. They conclude with suggestions for what future research on the topic should accomplish.
31
IN DEFENSE OF EPISTOCRACY:
ENLIGHTENED PREFERENCE
VOTING
Jason Brennan
In June 2016, a slight majority of British voters voted to leave the EU. Economists widely believed –
and still believe – this will harm the very citizens who voted to leave. Should they have gotten
their way?
Maybe not. The polling firm Ipsos Mori discovered that the British public was systematically
misinformed about the basic facts relevant to the decision. Leave voters believed that EU immigrants
constituted 20% of the UK’s population; Remain voters estimated 10%. The correct figure is about
5%. On average, both Leave and Remain voters overestimated by a factor of 40–100 what the UK
pays in Child Benefits to members of other EU countries. Both vastly underestimated the amount of
foreign investment from the EU and vastly overestimated the amount from China (Ipsos, 2016).
Both Leave and Remain voters got the basic facts wrong, but the more wrong a person was, the
more likely they were to vote Leave. It is plausible that if the populace were better informed,
Remain would have won.
In general, in most democratic elections and referenda, citizens are ignorant and misinformed
about the basic facts. Their mistakes are systematic and their worldviews unsophisticated. They
process information in deeply irrational ways. Only a minority have stable political beliefs or opi-
nions. We ask them to choose leaders and, in some cases, choose laws, but they rarely have any clue
what they are doing.
Democracies generally outperform other forms of government we have tried. Just why that is so
is disputed (Acemoglu and Robinson, 2013; Jones, 2020). However, most other historical forms of
government – from monarchies to oligarchies to one-party states – primarily existed to enable
government elites to exploit the masses. In the same way, a bungling and inept mother who means
well is better than an abusive mother. The bar is low.
This chapter outlines an alternative political system called epistocracy. Epistocracies retain most of the
features of modern social democracies, including liberal constitutions limiting government power,
the separation and devolution of powers, frequent contested elections, contestatory forums, and the
like. But epistocracies apportion political power on the basis of knowledge and political competence,
in order to reduce the harm caused by the ignorant, misinformed, and irrational electorate.
1 Voter Behavior 101

In primary school, many of us learn a particular theory of how democracy functions. Let’s call it the
“popular sovereignty model.” The theory goes as follows: first, it posits that citizens have various
DOI: 10.4324/9780367808983-38 391

Jason Brennan
concerns and goals, some selfish and some not; second, they learn how the world, politics, and
economy works. They form political beliefs and begin to advocate various political policies because
they believe those policies will realize their goals; third, they examine the candidates and parties on
offer, and tend vote for the best match with a real chance of winning; fourth, since everybody does
that, the winning candidates or parties will tend to match what the majority wants. Thus, the
policies and laws that are implemented after the election tend to reflect the ideological preferences of
the winning coalition of voters. Finally, and fifth, if leaders do a bad job, voters punish them by
voting them out in the next election. (The technical term for this behavior is “retrospective voting,”
which refers to when voters decide whether to keep or remove an incumbent based upon their past
performance.)
Unfortunately, the popular sovereignty model is wrong. Or, more precisely, it describes a tiny
minority of citizens, perhaps as few as one in ten (Achen and Bartels, 2016; Kinder and Kalmoe,
2017). Political scientists, psychologists, and economists have studied voter behavior for over sixty
years. They’ve conducted thousands of studies and amassed a huge amount of data. Their findings
are largely uniform and depressing. In general, voters are ignorant, misinformed, and biased.
However, there is tremendous variance. Some people know a lot, most people know nothing, and
many people know less than nothing because they are systematically mistaken (Campbell et al., 1960;
Delli-Carpini and Keeter, 1996; Converse, 1964; Friedman, 2006; Caplan, 2007; Somin, 2013).
For instance, during election years, most citizens cannot identify any congressional candidates in
their district (Hardin, 2009, 60). Citizens generally don’t know which party controls Congress
(Somin, 2013, 17–21). During the 2000 US Presidential election, while slightly more than half of all
Americans knew Gore was more liberal than Bush, significantly less than half knew that Gore was
more supportive of abortion rights, more supportive of welfare-state programs, favored a higher
degree of aid to blacks, or was more supportive of environmental regulation (Somin, 2013, 31).
Most citizens are aware that there is an ideological difference between liberals and conservatives but
cannot identify what this difference is (Kinder and Kalmoe, 2017).
Voters are not merely ignorant, but many are misinformed. They make systematic mistakes about
basic economic theory (Caplan, 2007) and about how political power functions (Caplan et al.,
2013). The American National Election Studies, conducted every other year, often test basic po-
litical knowledge. The bottom 25% of voters often perform worse than chance (Althaus, 2003).
Citizens are also epistemically irrational. They suffer from cognitive biases which prevent them
from processing information in a reasonable or truth-tracking way. Strong emotions cause them to
reason poorly. They tend to look for and accept evidence that confirms their pre-existing beliefs but
dismiss or ignore evidence that contradicts what they believe. They quickly rationalize and dismiss
bad behavior on their side but interpret even good behavior from the other side in a negative way.
They tend to assume those they disagree with are stupid and evil. They try to twist evidence to claim
it supports whatever they want it to (Tajfel and Turner, 1979; Tajfel, 1981, 1982; Kahneman and
Tversky, 1973, Kahneman, Slovic, and Tversky, 1982; Rasinki, 1989; Bartels, 2003; Arceneaux and
Stein, 2006; Lodge and Taber, 2013; Westen et al., 2006; Westen, 2008; Kelly, 2012; Haidt, 2012;
Chong, 2013; Lodge and Taber, 2013; Taber and Young, 2013; Erison, Lodge, and Taber, 2014).
Information matters. The policies people prefer depends in part on how informed they are.
When controlling for the influence of sex, race, and income, highly informed citizens have sys-
tematically different policy preferences from ignorant or misinformed voters (Althaus, 2003).
However, most citizens lack stable political beliefs or ideologies. While many citizens label themselves
conservative or liberal, or attach themselves to political parties, only a small minority, fewer than one in
five, have stable beliefs over time, or have real political opinions. A large segment of the population is
politically agnostic (Converse, 1964; Barnes, 1971; Inglehart and Klingemann, 1976; Arian and Schamir,
1983; Converse and Pierce, 1986; Zaller, 1992; McCann, 1997; Goron, 2005; Zechmeister, 2006;
Lewis-Beck et al., 2008; Achen and Bartels, 2016; Kinder and Kalmoe, 2017; Mason, 2018).
392
In Defense of Epistocracy
Political affiliation is largely not about belief or policy. As Anthony Appiah (2018) says, “People
don’t vote for what they want. They vote for who they are.” Citizens vote largely on the basis of
partisan loyalties grounded in their identities, which do not track ideology, sincere policy pre-
ferences, or their interests. Rather, partisan attachments usually result from accidental, historical
connections between certain identity groups and certain political movements and parties. In the
same way that people from Boston root for the Patriots to demonstrate their fidelity to their group,
Boston Irish people vote Democrat, Southern Evangelicals vote Republican, and so on. Citizens
often change their expressed “beliefs” to fit their party; they rarely choose a party on the basis of
shared beliefs (Cohen, 2003; Mutz, 2006; Iyengar et al., 2012; Kahan et al., 2013; Somin, 2013;
Iyengar and Westwood, 2015; Achen and Bartels, 2016).
Citizens are bad at retrospective voting (Healy and Malholtra, 2010). Retrospective voting de-
mands many voters. They must know who was in power, what they did, what they could have
done, how to evaluate what they did versus what they could have done, and finally whether the
challengers are likely to be any better. In fact, it appears that voters at best tend to punish or reward
incumbents for the last sixth months or so of economic performance. However, as Achen and
Bartels say, if the incumbents were not responsible for those outcomes, this is little better than
kicking the dog because one had a bad day at work (Achen and Bartels, 2016).
In political science and economics, the dominant explanation for why citizens behave so poorly is
that democracy incentivizes them to do so. Because individual votes count for so little, citizens
generally have no incentive to be informed, no incentive to correct their errors, and every incentive
to indulge their worst biases. They are ignorant, misinformed, and biased because the expected costs
of acquiring information and overcoming their biases exceed the expected benefits. They are
rationally ignorant and rationally irrational.
2 We Don’t Make It Up in Bulk

Most voters are ignorant, misinformed, irrational, tribalistic followers. Still, in some cases, large
groups of people can be wise as a collective even though the individuals within those crowds are not
wise. For instance, when we ask people to guess the number of jellybeans in a jar, most individuals’
answers are mistaken. But their mean guess is quite accurate. Indeed, the more people we add, the
more accurate the mean becomes.
Perhaps democratic voting has the same features, in which the group is wise even though the
individuals within the group are not. Three popular mathematical models are often invoked to
defend democracy in just this way:
1. The Miracle of Aggregation: ignorant voters will vote randomly. Accordingly, they will cancel
each other out, leaving the well-informed minority to decide the election. A large electorate
composed mostly of ignorant voters performs like an informed electorate.
2. Condorcet’s Jury Theorem: if, on average, individual members of a group decision are more likely to
get the right answer than the wrong answer, then as the group becomes bigger, the probability
that the group will select the right answer approaches one. (However, if instead individual
members are on average more likely to get the wrong answer than the right answer, then as the
group gets larger, the probability the group will select the wrong answer approaches one.)
3. Hong-Page Theorem: Under certain conditions, when groups are making a collective decision,
increasing the cognitive diversity of members of the group better enhances the reliability of the
group as a whole than increasing the reliability of individual members of that group does.
Each of these theorems relies upon a mathematical model. The important question is whether the
models correspond to what happens in real-life democratic decision-making.
393
Jason Brennan
It’s worth noting that the mathematics of the third theorem are highly controversial.
Mathematician Abigail Thompson claims that the proof of the Hong-Page Theorem rests on several
identifiable mathematical errors. She further claims that the mathematical stand-in for “diversity” in
the theorem does not correspond to anything that we might call “cognitive diversity” in the real
world. She also argues the proof is not generalizable (Thompson, 2014). Similarly, Paul Quirk,
among others, claims that the “proof” depends upon a series of computer experiments “strongly
biased toward that result [that diversity trumps ability] and argues that it tells us nothing about
decision-making in real-world political settings” (Quirk, 2014, 134). Further, as many have noted,
one reason why “diversity trumps ability” in the Hong-Page theorem, as groups become larger, is
that the theorem in effect models large groups as including the most elite performers and deferring to
them when they are right. Philosopher David Wallace notes that the theorem simply assumes that
whenever smart agents get stuck, there is always another person who can and will improve the
group’s decision. The theorem is supposed to prove this, he says, but in fact Hong and Page bake
these assumptions in as premises. Thus, their result is trivial (Leiter, 2019).
Let’s put these worries about the Hong-Page theorem aside. Instead, consider what Hélène
Landemore says when discussing all three theorems:
The main problem with the optimistic conclusions about group intelligence … is that in
some way or another they rely on the assumption that there is a symmetrical distribution
(random or otherwise) of errors around the right answer (Miracle of Aggregation) or that
errors are negatively correlated (Hong and Page). (2012, 195)
In short, all three theorems can be used to defend group intelligence only when citizens make
randomized rather than systematic errors. If citizens’ errors are not randomly or symmetrically
distributed, or if citizens tend to make systematic mistakes, tend to follow one another’s opinions, or
tend to be systematically misinformed and unreliable, then these mathematical theorems cannot be
used in support of democracy. (Landemore herself nevertheless supports highly optimistic conclu-
sions about how well democracy performs.) Further, for these theorems to “work” as defenses of
democracy, they require voters to be actually trying to solve a largely-agreed-upon problem, rather
than doing something else.
As discussed in the previous section, the empirical literature supports the following claims:
1. Citizens do not form their ideas or decide how to vote independently and separately. They
follow one another, and in particular, tend to parrot whatever their party happens to say.
2. Citizens vote for largely non-cognitive and non-ideological reasons. They are cheering for their
team, not trying to discover the right answer. But the theorems require that voters are trying to
solve a problem, not simply using their votes in this expressive manner.
3. Most citizens have very unsophisticated mental models of politics and very low levels of in-
formation. However, the Hong-Page theorem requires that individual problems solvers have
somewhat complex mental models.
4. Citizens make systematic errors in reasoning and are systematically mistaken about a wide range
of basic political facts and more advanced social scientific knowledge.
These points are fatal to “wisdom of the crowd” defenses of democracy. Voters’ errors compound
rather than cancel. They lack the kind of cognitive diversity and basic sophistication of the Hong-
Page theorem. Because people are, on average, more likely to give the wrong answer than the right
answer, if Condorcet’s Jury Theorem applies to democratic decision-making at all, it would predict
that large democracies tend to make bad decisions.
394
3 Incompetent Rule Is Unjust

Imagine a capital murder case. Suppose that the trial proceeds as normal, with both the prosecution
and defense presenting their arguments, evidence, and so on. However, suppose the jury has any or
all of the following features:
1. They are ignorant. They pay no attention to the facts of the case. They refuse to read the
transcript. They flip a coin and find the defendant guilty.
2. They are misinformed. The jury deeply misunderstands the facts of the case. For instance, they
have clearly false beliefs about where the defendant was during the murder, what the de-
fendant’s relationship to the victim was, and so on. Their false beliefs explain why they found
him guilty.
3. They are irrational. They pay attention to the facts of the case, which indicate rather clearly that
the defendant is innocent. However, the jurors process information in a deeply irrational way,
and so conclude he is guilty.
4. They are malicious, selfish, or acting in bad faith. They find the defendant guilty because he is a
member of a disliked religious group, or because he owns a rival restaurant, or because they
took a bribe.
5. They are tribalistic. They find the defendant guilty because they are just the kind of people who
vote guilty every time, regardless of the facts.
In these cases, if we knew the jury decided in any of these ways, it would be wrong and unjust to
enforce their decision. Their decision would lack authority. Indeed, in some US states, if a defendant
who demonstrated the jury made a decision on any such grounds would be entitled to a retrial.
What seems to explain these intuitions are the following: the jury is charged with administrating
justice. They act as representatives of society as a whole. They will impose their will upon a possibly
innocent person. Their decision is high stakes and can deprive a person of property, freedom, and
even life. In situations like this, a minimal condition for the decision to be legitimate and authoritative
is that the jury decides competently and in good faith. If they are incompetent as a body in general or
if they make this particular decision incompetently and in bad faith, then it would be wrong to
enforce their decision.
This point generalizes to other political decisions. Many political decisions are high stakes, can
greatly affect other people’s welfare, alter their life prospects, and deprive them of life, liberty,
property, and happiness. The people making these decisions are usually charged with acting on
behalf of the common good and are supposed to aim for just outcomes. Thus, I think juries, judges,
police officers, presidents, legislators, bureaucrats, governors, and even the voting public is constrained
by what I call the Competence Principle:
The Competence Principle
It is presumed to be unjust and a violation of a citizen’s rights to forcibly deprive them of

life, liberty, or property, or significantly harm their life prospects, as a result of decisions
made by an incompetent deliberative body, or as a result of decisions made in an in-
competent way or in bad faith. Political decisions are presumed legitimate and authoritative
only when produced by competent political bodies in a competent way and in good faith.
In short, the idea is that a minimal condition of a political decision being authoritative and legitimate
is that it must be made by a reliable body or decision-making process, in a competent way, and in
good faith.
395
Jason Brennan
To make my argument work, I do not need to defend some precise theory of political com-
petence. Any plausible theory of competence and good faith would agree that the jurors are in-
competent or acting in bad faith in one to five above. If one is ignorant or misinformed despite
salient information being available, if one sticks to the same beliefs come what may, or if one forms
beliefs almost entirely on the basis of non-evidentiary factors, then one acts incompetently.
Notice that the electorate’s decisions have the same morally salient features as jury decisions:
1. Electorates are charged with making morally momentous decisions, as they must decide how to
apply principles of justice, and how to shape many of the basic institutions of society. They are
one of the main vehicles through which justice is to be established.
2. Electoral decisions tend to be of major significance. They can significantly alter the life pro-
spects of citizens, and deprive them of life, liberty, and property.
3. The electorate claims sole jurisdiction for making certain kinds of decisions over certain people
within a geographic area. The electorate expects people to accept and abide by their decisions.
4. The outcomes of decisions are often imposed involuntarily through violence and threats of
violence.
This is strong presumptive reason to hold that the Competence Principle applies not merely to
juries, judges, presidents, and the like, but even to the voting electorate as a whole.
How the group votes matters. Voters sometimes directly choose policy. Other times, they choose
representatives who in turn create policy. If voters choose badly, they can cause serious injustices.
They can choose leaders who will implement destructive tariffs, run up the public debt, leave the
poor behind, start unnecessary and unjust wars, ignore or exacerbate existential threats, or mis-
manage criminal justice, among other things.
When democracies make bad choices, this is not the moral equivalent of a single person making
poor choices for herself. When the democratic majority or winning plurality makes bad choices,
they impose their will upon the losing minorities, residents unable to vote, future generations, and
foreigners who must live with the consequences.
4 Enlightened Preference Voting

There are many possible ways to mitigate the harmfulness of bad voting. Here, I’ll discuss one
possibility.
Ample work in political science and economics shows that low-information and high-
information voters have different policy preferences. Martin Gilens notes that even within a single
party, the high-information and low-information members disagree. For instance, high-information
Democrats are more pro-free-trade and less militaristic than low-information Democrats (Gilens,
2012, 106–11).
However, we also know that information is correlated with various demographic factors. A
persistent finding over the past sixty years in the US is that, when it comes to basic political
knowledge, the rich know more than the poor, men know more than women, whites know more
than blacks, and so on. In short, the more privileged tend also to be better informed (Delli-Carpini
and Keeter, 1996, 135–77). Thus, one might worry that if informed people disagree with the
uninformed, these demographic factors rather than information might confound or even drive
the results. As high-information voters also tend to be rich, perhaps it’s their income rather than
the information that makes them pro-free-trade.
Fortunately, researchers are already aware of this and have developed research methods that allow
us to test the effect of information, while controlling for demographics, and vice versa. The basic
method goes as follows:
396
1. Give everyone a test of some aspect of political knowledge. Find out what they know.
2. Collect information about their demographics. Find out who they are.
3. Survey them on their opinions, beliefs, etc. Find out what they want.
Once we have all three sets of data, we can assess the independent effect of knowledge or the
independent effect of demographics, all while controlling for confounds. Further, we can statistically
estimate what the public would have wanted if things changed. This method allows to estimate, for
instance, what an otherwise identical but all-female or all-male public would want. Most im-
portantly for our purposes here, it allows us to estimate what a demographically identical public
would want if it had gotten a perfect score on the knowledge test. Call this the publics’ enlightened
preferences (Althaus, 2003; Caplan, 2007).
Althaus summarizes the method as follows. First, one collects the kind of data discussed above, by
asking people a wide range of policy and ideology questions while also collecting their demographic
data and assessing their levels of basic political knowledge, such as whether they can identify their
representatives or the unemployment rate. Then:
Estimates of fully in-formed opinions are generated by assigning the preferences of the
most highly informed members of a given demographic group to all members of that
group, simultaneously taking into account the influence of a wide range of demographic
variables. For instance, if policy preferences of well-informed respondents from union
families differ from those of ill-informed respondents from union families, then this ap-
proach assigns the mix of fully informed preferences to all respondents from union families.
But instead of considering only the bivariate relationship between union membership and
policy preferences, this method looks at union respondents who are women, from a certain
income level, who live in eastern states, are married, own homes, of a certain age, and so
on. If the most informed people sharing all these characteristics have different preferences
from the least informed people, then their mix of fully informed preferences is assigned to
everyone who shares their demographic characteristics. (Althaus, 2003, 548)
In short, what the method does is survey a wide range of people with the same demographic identity
or characteristics, and then determines how people with those characteristics vote when they are
fully informed, as measured by the accompanying quiz of basic political information. Most people
are a mix of demographic identities, of course, but the method simultaneously allows researchers to
estimate how strongly different identities affect preferences, and thus allows researchers to estimate
what a demographically identical public would vote or would believe if it were fully informed.
I suggest we use this method to produce better political outcomes. The procedure goes as fol-
lows. On election day, everyone is allowed to vote, including children. When they vote, though,
they must do three things:
1. Take a forty-question, closed-book quiz on basic political knowledge.

2. Tell us their demographic factors. (Perhaps this can be set ahead of time on a voter ID card.)
3. Tell us their opinion on whatever the election is about, for instance, which candidate or party
they support, or which position they take on a referendum.
Afterward, all the voting data is anonymized and made public. The government then calculates – using
methods that can be checked by any major newspaper and many statistically savvy researchers – what a
demographically identical public would have wanted if it had gotten a perfect score on the quiz. In
short, we calculate the electorates’ enlightened preferences and implement those instead of their actual,
unenlightened preferences. Call this enlightened preference voting.
397
Jason Brennan
Keep in mind that the quality of the candidates on the ballot, the quality of the policies they
espouse, and the ways parties are organized are not exogenous factors. They depend significantly
on the kind of voting system used and on the quality of the voters themselves. Parties want to
win, and so the positions they push and candidates they support depend on what they believe will
help them win. Enlightened preference voting will not merely tend to ensure that we select the
better choices on the ballot; it will tend to ensure that the choices that have made it on to the
ballot are already better.
Now, one might argue this counts as a form of democracy rather than epistocracy. After all, no
one is excluded. Citizens are not required to pass the quiz to earn the right to vote. It’s not
exactly true that the more knowledgeable receive more votes than the less knowledgeable.
Rather, everyone has equal input, and we use these inputs to estimate what an informed but
otherwise the identical public would have wanted. We can mathematically estimate afterwards
that different citizens had different average or marginal effects, but even this is somewhat artificial.
For this reason, a committed democrat who likes this idea might insist that enlightened pre-
ference voting is a form of democracy. Rather than excluding some people or elevating some
above others, it is simply a better method for extracting the hidden wisdom of the crowd. On the
other hand, in enlightened preference voting, the “people” do not get their way; rather, the
ruling group is a hypothetical electorate statistically derived from the actual electorate. This
counts against calling the system democracy. Still, rather than resolve this definitional debate
here, I will simply note the issue.
One virtue of this system is that it allows us to test to what degree various political outcomes result
from demographic bias. We can simulate whether, for instance, an all-black or all-female polity would
have chosen differently. With such information, we could in principle correct for problems that arise
when small minorities have their (enlightened) preferences thwarted time and time again.
There are good questions about how to design this system. Who decides which questions go on
the quiz? Who decides what the demographic categories will be?
This might matter less than one would suspect. After all, as of now, various political scientists and
economists have employed the enlightened preference on different data sets, using different groups
of people, different demographic categories, and different tests of competence and knowledge. So
far, they tend to generate similar results despite these differences: the enlightened public is more free
trade, more in favor of interactions with foreigners, more civil libertarian, and more in favor of tax
increases to offset the deficit (Althaus, 2003; Caplan, 2007; Gilens, 2012).
Regardless, I recommend that in order to reduce the amount of political gaming and rent seeking
that might corrupt the system, we let the people design it. Allow elections to proceed as normal.
However, three weeks or so before the election, we randomly select five hundred citizens. They are
paid to spend a weekend together deliberating to choose the questions which go on the quiz. They
also can revise the demographic indicators.
This may seem paradoxical. I have argued citizens are largely incompetent to choose policy or
leaders. They lack basic political information. Why then would they be competent to design the
quiz? They know an informed citizen needs to know who is in power, what they did, how to
assess what they did, whether people are getting richer or poorer, healthier or less healthy, or
whether crime is up or down. Here, the problem is not that citizens lack a good grasp of what the
right questions are. Rather, the problem is that they lack the answers. The question “what counts
as an informed citizen” is an easy one where the crowd can produce a good answer, even if though
most in the crowd are not informed. Further, the citizens who design the poll can piggy-back off
of things like the American National Election Studies or the US citizenship exam. Further, one
advantage of this system is that, with only 500 citizens choosing what appears on the test, they will
have stronger incentives to do their jobs well.
398
In the real world, I expect this system to be flawed, just as in the real world, any democratic
voting system is flawed. The question is not whether it will be perfect. The question is whether it
will be better. Democracy has many virtues compared to the systems we have tried. It also has a
systematic flaw: it spreads power out widely, and in virtue of doing so, incentivizes those who hold
that power to use it unwisely. We have a moral obligation to fix this problem as best we can.
References
Acemoglu, Daron, and James Robinson. (2013) Why Nations Fail. New York; Currency.
Achen, Christopher and Larry Bartels. (2002) “Blind Retrospection: Electoral Responses to Draught, Flue, and
Shark Attacks.” Prepared for presentation at the annual meeting of the American Political Science
Association, Boston.
Achen, Christopher and Larry Bartels. (2016) Democratic Theory for Realists. New York; Princeton University Press.
Althaus, Scott. (2003) Collective Preferences in Democratic Politics. New York; Cambridge University Press.
Appiah, Anthony. (2018) “People Don’t Vote for What They Want. They Vote for Who They Are,”
Washington Post, August 30. https://www.washingtonpost.com/outlook/people-dont-vote-for-want-they-
want-they-vote-for-who-they-are/2018/08/30/fb5b7e44-abd7-11e8-8a0c-70b618c98d3c_story.html.
Arceneaux, Kevin, and Robert M. Stein. (2006) “Who Is Held Responsible When Disaster Strikes? The
Attribution of Responsibility for a Natural Disaster in an Urban Election,” Journal of Urban Affairs, 28,
43–53.
Arian, Asher, and Michal Schamir. (1983) “The Primarily Political Functions of the Left-Right Continuum,”
Comparative Politics, 15, 139–158.
Barnes, Samuel H. (1971) “Left, Right, and the Italian Voter,” Comparative Political Studies, 4, 157–175.
Bartels, Larry. (2003) “Democracy with Attitudes,” in George Rabinowitz and Michael B. MacKeun eds.
Electoral Democracy. New York; Oxford University Press.
Campbell, Angus, Philip E. Converse, Warren E. Miller, and Donald E. Stokes. (1960) The American Voter.
New York; John Wiley.
Caplan, Bryan. (2007) The Myth of the Rational Voter. Princeton; Princeton University Press.
Caplan, Bryan, Eric Crampton, Wayne A. Grove, and Ilya Somin. (2013) “Systematically Biased Beliefs about
Political Influence: Evidence from the Perceptions of Political Influence on Policy Outcomes Survey,” PS:
Political Science and Politics, 46, 760–767.
Chong, Dennis. (2013) “Degrees of Rationality in Politics,” in David O. Sears and Jack S. Levy eds. The Oxford
Handbook of Political Psychology, pp. 96–129. New York; Oxford University Press.
Cohen, Geoffrey. (2003) “Party over Policy: The Dominating Impact of Group Influence on Political Beliefs,”
Journal of Personality and Social Psychology, 85, 808–822.
Converse, Philip. (1964) “The Nature of Belief Systems in Mass Publics.” In D. E. Apter ed. Ideology and
Discontent. London; Free Press of Glencoe.
Converse, Philip, and Richard Pierce. (1986) Political Representation in France. Cambridge MA; Harvard
University Press.
Delli-Carpini, Michael X., and Scott Keeter. (1996) What Americans Know About Politics and Why It Matters.
New Haven; Yale University Press.
Erison, Cengiz, Milton Lodge, and Charles S. Taber. (2014) “Affective Contagion in Effortful Political
Thinking,” Political Psychology, 35, 187–206.
Friedman, Jeffrey. (2006) “Democratic Competence in Normative and Positive Theory: Neglected
Implications of ‘The Nature of Belief Systems in Mass Publics’,” Critical Review, 18, i–xliii.
Gilens, Martin. (2012) Affluence and Influence. Princeton; Princeton University Press.
Goron, Paul. (2005) “Party Identification and Core Political Values,” American Journal of Political Science, 49, 882–897.
Guerrero, Alexander. (2010) “The Paradox of Voting and the Ethics of Political Representation,” Philosophy
and Public Affairs, 38, 272–306.
Haidt, Jonathan. (2012) The Righteous Mind. New York; Pantheon.
Hardin, Russell. (2009) The Economics of Ordinary Knowledge. Princeton; Princeton University Press.
Healy, Andrew and Neil Malholtra. (2010) ““Random Events, Economic Losses, and Retrospective Voting:
Implications for Democratic Competence,” Quarterly Journal of Political Science, 5, 193–208.
Inglehart, Ronald, and Hans Klingemann. (1976) “Party Identification, Ideological Preference, and the Left-
Right Dimension among Western Mass Publics,” in Ian Budge, Ivor Crewe, and Dennis Fairlie ed. Party
Identification and Beyond, London; Wiley.
399
Jason Brennan
Ipsos Mori. (2016) “The Perils of Perception and the EU” June 8, 2016. https://www.ipsos.com/ipsos-mori/
en-uk/perils-perception-and-eu.
Iyengar, Shanto, Guarav Sood, and Yphtach Lelkes. (2012) “Affect, Not Ideology: A Social Identity Perspective
on Polarization,” Public Opinion Quarterly, 76, 405–431.
Iyengar, Shanto, and Sean J. Westwood. (2015) “Fear and Loathing Across Party Lines: New Evidence on
Group Polarization,” American Journal of Political Science, 59, 690–707.
Jones, Garrett. (2020) 10% Less Democracy. Stanford; Stanford University Press.
Kahan, Dan, Ellen Peters, Erica Cantrell Dawson, and Paul Slovic. (2013) “Motivated Numeracy and
Enlightened Self-Government,” Behavioral Public Policy, 1, 54–86.
Kahneman, Daniel and Andrew Tversky. (1973) “Availability: A Heuristic for Judging Frequency and
Probability,” Cognitive Psychology, 5, 207–233.
Kahneman, Daniel, Paul Slovic, and Amos Tversky, eds. (1982) Judgment under Uncertainty: Heuristics and Biases.
New York; Cambridge University Press.
Kelly, James Terence. (2012) Framing Democracy. Princeton; Princeton University Press.
Kinder, Donald and Nathan Kalmoe. (2017) Neither Liberal nor Conservative: Ideological Innocence in the American
Public. Chicago; University of Chicago Press.
Landemore, Hélène. (2012) Democratic Reason. Princeton; Princeton University Press.
Leiter, Brian. (2019) “A ‘Mathematical Proof’ That ‘Diversity Trumps Ability’ Turns Out to Be Just More
Diversity Blather.” Leiter Reports: A Philosophy Blog. leiterreports.typepad.com/blog/2019/11/a-
mathematical-proof-that-diversity-trumps-ability-turns-out-to-be-just-more-diversity-blather.html.
Lewis-Beck, Michael, William Jacoby, Helmut Norpoth, and Herbert Weisberg. (2008) The American Voter
Revisited. Ann Arbor; University of Michigan Press.
Lodge, Milton, and Charles Taber. (2013) The Rationalizing Voter. New York; Cambridge University Press.
Mason, Lilliana. (2018) “Ideologues without Issues: The Polarizing Consequences of Ideological Identities,”
Public Opinion Quarterly, 82, 280–301.
McCann, James A. (1997) “Electoral Choices and Core Value Change: The 1992 Presidential Campaign,”
American Journal of Political Science, 41, 564–583.
Mutz, Diana. (2006) Hearing the Other Side. Cambridge; Cambridge University Press.
Quirk, Paul. (2014) “Making It Up on Volume: Are Large Groups Really Smarter?,” Critical Review, 26,
129–150.
Rasinki, Kenneth A. (1989) “The Effect on Question Wording on Public Support for Government Spending,”
Public Opinion Quarterly, 53, 388–394.
Somin, Ilya. (2013) Democracy and Political Ignorance. Stanford; Stanford University Press.
Taber, Charles, and Everett Young. (2013) “Political Information Processing,” in Huddy, Sears, and Levy 2013,
525–558.
Tajfel, Henry. (1981) Human Groups and Social Categories. New York; Cambridge University Press.
Tajfel, Henri. (1982) Social Identity and Intergroup Relations. Cambridge; Cambridge University Press.
Tajfel, Henry, and J. C. Turner. (1979) “An Integrative Theory of Intergroup Conflict,” in W. G. Austin and
S. Worchel eds. The Social Psychology of Intergroup Relations. Monterey, CA; Brooks-Cole.
Thompson, Abigail. (2014) “Does Diversity Trump Ability?,” Notices of the American Mathematical Society,
61, 1–24.
Westen, Drew. (2008) The Political Brain. New York; Perseus Books.
Westen, Drew, Pavel S. Blagov, Keith Harenski, Clint Kilts, and Stephan Hamann. (2006) “The Neural Basis of
Motivated Reasoning: An fMRI Study of Emotional Constraints on Political Judgment During the U.S.
Presidential Election of 2004,” The Journal of Cognitive Neuroscience, 18, 1947–1958.
Zaller, John. (1992) The Nature and Origins of Mass Opinion. New York; Cambridge University Press.
Zechmeister, Elizabeth. (2006) “What’s Left and Who’s Right? A Q-Method Study of Individual and
Contextual Influences on the Meaning of Ideological Labels,” Political Behavior, 28, 151–173.
400
32
VOTING RULES
Itai Sher
1 Introduction
This article addresses a basic democratic question: what voting rule should be used to decide an
election? Also, what criteria should guide the choice of voting rule?
There is an extensive academic literature on voting rules going back to Borda (1784) and
Condorcet (1785) and even earlier. In this article, I will focus on formal axiomatic approaches from
social choice theory. My approach will be to focus on majority rule as a central case. In elections
between two candidates, what is the justification for majority rule? How should majority rule be
extended to the case of multiple candidates? How does majority rule compare to alternatives?
The aim of the essay is not to be exhaustive – there is no space for that – but rather to give a
flavor for the nature of axiomatic analysis of voting rules. At the same time, I have attempted to keep
formalities to a minimum, and make the article as accessible as possible without sacrificing faith-
fulness to the underlying ideas.
Implicitly, the paper evaluates voting rules from a procedural rather than an instrumental per-
spective. I view the axioms as being justified in terms of what would be procedurally best from a
democratic standpoint rather than in terms of what would lead to the best outcomes. Note however
that the procedural content of democracy is by no means exhausted by the nature of its voting rules,
as other aspects of institutions, culture, and deliberation are very important as well.
This article will focus on single-winner voting methods, that is, voting methods for electing a
single candidate for a position. These methods can also be used to decide a policy question.
In what follows, I deal with binary elections (Section 2), multi-candidate elections (Section 3),
strategic issues (Section 4), and voting power (Section 5). Following the conclusion of the article,
I provide further reading.
2 Binary Elections
I start by considering binary elections, that is, elections with two alternatives, x and y. These al-
ternatives could be candidates for office or policies.
There is a set of n voters: voter 1, voter 2, …, voter n. Each voter may vote for x or y. A vote
profile is a list of votes, one for each voter. For example, with three voters – voter 1, voter 2, and
voter 3 – the voter profile (x, y, x ) represents the situation in which voter 1 votes for x, voter
2 votes for y, and voter 3 votes for x.
DOI: 10.4324/9780367808983-39 401

Itai Sher
A voting rule is a function f that takes as an input a vote profile v and gives as an output a winner of
the election, which is either x or y. Who the winner is depends on how people vote. For example,
voting rule f might specify that if the vote profile is (x, y, x ), then candidate x wins and if the vote
profile is (y, y, x ), then candidate y wins.
We also allow that there may be a tie. In case of a tie, we may flip a coin to determine the
winner.
There are many different voting rules, for example:
• Majority rule. The candidate with the most votes wins.

• 2/3 Supermajority with status quo x. The status quo x wins unless at least 2/3 of voters vote for y.
• Two-tier voting. Each voter is assigned a jurisdiction, and each jurisdiction is assigned a certain
number of electoral votes. A majority rule election is conducted within each jurisdiction, and
the winner of a jurisdiction’s majority election receives all of the jurisdiction’s electoral votes.
The winner of the overall election is the candidate who receives the most electoral votes.
The US Electoral College is an example of a two-tier system used in the US Presidential election. In
contrast, National Popular Vote is the name given to the proposal that the US Presidential election
will be decided by national majority rule without regard for the states in which voters reside.
2.1 Axioms
But how do we decide which voting rule we should use? One approach is to impose some in-
tuitively desirable requirements on the voting rule, known as axioms. Then we can look for a rule
that satisfies these requirements.
The particular axioms that I explore in this section are taken from May (1952).
The first axiom is anonymity. This axiom says that all voters are to be treated equally. How do we
represent this formally? Suppose the three voters are Ann (voter 1), Bob (voter 2), and Carol (voter 3).
Suppose the vote profile is v = (x, y , x ), so that Ann votes for x, Bob votes for y, and Carol votes
for z. Now suppose Ann and Bob exchange votes, so that Ann votes for y and Bob votes for x, leading
to vote profile v = (y, x, x ). If v results from v in this way, via an interchange of the votes of
two voters, then we say that v is a voter transposition of v. If v can be derived from v via a series of
transpositions, then we say that v is a voter permutation of v.
I now state the first axiom formally.
Axiom: Anonymity. If v is a voter transposition of v, then the outcome of the election is

the same in v as in v .
The anonymity axiom says that all voters are treated equally: if we interchange the votes of two
voters, then the outcome won’t change.
Clearly, majority voting is anonymous, as interchanging the votes of two voters won’t alter the
vote totals.
In contrast, the electoral college is not anonymous. Recall the 2000 Presidential election between
Al Gore and George W. Bush. Florida was pivotal, meaning that whichever candidate won Florida
would win the whole election. Modify the facts so that Nader, the Green Party candidate, had not
run, and that the only two candidates were Gore and Bush, and imagine that Florida, which was
pivotal, had been decided for Bush by just one vote. Let Ann be a Florida voter who voted for Bush.
Bob is a California voter who voted for Gore; Gore won California easily. If we had exchanged the
votes of Ann and Bob, so that Ann had voted for Gore and Bob had voted for Bush, then the
outcome in California would not have changed. The outcome in Florida, however, would have
402
Voting Rules
changed from a Bush win by one vote to a Gore win by one vote, and hence the outcome of the
entire election would have changed. This implies that the electoral college is not anonymous. If we
think that anonymity is a good axiom, then this forms a basis for criticizing the electoral college and
preferring national popular vote (i.e., majority rule).
It is natural to think that the anonymity axiom is compelling because all voters should be treated
in the same way. Notice however that this rationale is context dependent. The argument in favor of
anonymity depended on there being no significant difference between Ann and Bob that could
justify treating their votes differently. In some contexts, there might be an adequate justification for
different treatment. For example, non-citizens or children may not be eligible to vote, which could
be modeled formally as their votes having no effect on the outcome of the voting rule. Alternatively,
in some contexts, we might think that the votes of people most affected by a policy should be
weighed more heavily. In any context in which we think there are morally relevant distinctions –
citizenship status, age, stake in the policy – that justify different treatment of different voters (or
potential voters), we may reject anonymity.1
Let us now consider the 2/3 supermajority rule. It is easy to see that it is in fact anonymous:
Interchanging which voter votes for which candidate does not affect whether candidate y gets at
least 2/3 of the vote. However, at least in some contexts, the 2/3 supermajority rule is problematic
because it treats candidates asymmetrically.
Anonymity concerned equal treatment of voters. We now consider equal treatment of candidates.
Consider vote profile v = (x, y , x ). Let v = (y , x, y ). We arrived at v from v by changing which
candidate each voter voted for. If a voter voted for x in v, that voter votes for y in v and vice versa. We
refer to v as the candidate transposition of v. The same definition applies with a larger number of voters.
Axiom: Neutrality. Let v be the candidate transposition of v. Then if one candidate wins
in v, the other wins in v . If there is a tie in v, there is also a tie in v .
Majority voting satisfies neutrality: If we flip everyone’s vote, we thereby flip the winner. The
electoral college also satisfies neutrality. If we flip everyone’s vote, we flip the winner in each state.
We thereby flip the electoral scores of the two candidates and so flip the outcome of the election.
The 2/3 supermajority rule violates neutrality. Suppose 60% vote for the status quo x and 40%
vote for the alternative y. Then the status quo wins. If we flip everyone’s votes, we have 40% for x
and 60% for y. Now y is in the lead, but the support for y is not sufficient to surpass that 2/3
threshold. So the status quo x still wins. This violates neutrality, which requires that winner to flip
when we flip the votes. So the neutrality axiom is a basis for rejecting the 2/3 supermajority rule.
Neutrality is a good axiom if we want to base election outcomes only on the votes of individuals
and not on any inherent differences among the alternatives. However, in some circumstances, it may
make sense to privilege the status quo, especially if the status quo encodes basic rights and liberties or
the democratic rules of the game. Constitutions often do this; in that case, we may have good reason
to treat alternatives asymmetrically and so violate neutrality.
Next consider minority rule, according to which whichever candidate gets fewer votes wins.
Minority rule satisfies both anonymity and neutrality, but nonetheless seems like a bad rule.2 The
problem is that votes “point in the wrong direction.” We need an axiom that ensures that votes
point in the right direction.
Axiom: Positive responsiveness. Suppose that v differs from v only in that there is a single
voter i, such that, in v, i votes for x, and, in v , i votes for y, then
1. If v leads to a tie, then y wins in v .

2. If y wins in v, then y still wins in v .
403
Itai Sher
The positive responsiveness axiom ensures that a vote for y really is a vote for y. It also says that if we
are at a tie, a single additional vote for one candidate is enough to break it.
Positive responsiveness may initially seem asymmetric with respect to its treatment of x and y, but
it is easy to see that the above axiom is logically equivalent to the same statement with the roles of x
and y reversed.
2.2 May’s Theorem

May (1952) proved the following theorem.
Theorem: May’s theorem. Majority rule is the unique binary voting rule that satisfies
anonymity, neutrality, and positive responsiveness.
The proof is as follows. First, it is easy to see that majority rule satisfies the axioms. But we must
also show that majority rule is the unique rule satisfying these axioms. Assume an even number of
voters for simplicity. (One can still prove the theorem with an odd number.) Suppose our voting
rule satisfies the three axioms. One can show that anonymity implies that the outcome of the
election depends only on the number of votes for each candidate, and not on who votes which way.
If an equal number of voters vote for each candidate, neutrality implies there must be a tie. Starting
from a tie and switching one vote from one candidate to the other, positive responsiveness implies
that the latter must win. And positive responsiveness also implies that as we keep switching votes to
the latter, that candidate continues to win. However, what we have just described is majority rule.
So any voting rule that satisfies all three axioms must be majority rule.
May’s theorem provides a justification for majority voting in binary elections. If the question is,
“Why should we decide elections via majority voting?,” an answer might be that anonymity,
neutrality, and positive responsiveness are compelling normative axioms, and majority voting is the
unique binary voting rule that satisfies them.
2.3 Weak Anonymity

Consider a symmetric electoral college in which each state (or jurisdiction) has the same population and
the same number of electoral votes. In some sense, a symmetric electoral college seems to treat
voters equally. Each voter votes only within a certain jurisdiction, but that jurisdiction contains the
same number of voters and has the same number of electoral votes as the jurisdiction that contains
any other voter. Nevertheless, a symmetric electoral college does not satisfy the anonymity axiom.3
The argument is the same as the one given above for an asymmetric electoral college; that argument
did not depend on the states having different populations or electoral votes.
A symmetric electoral college does have the property that if we transpose the votes of two (equal-
sized) jurisdictions, as opposed to the votes of two voters, the outcome does not change. So if Ann
resides in Florida and Bob in California, and Florida and California are assumed to have equal po-
pulations, we can always interchange Ann and Bob’s votes without changing the electoral outcome
provided we appropriately interchange the votes of other pairs of voters in Florida and California. This
formalizes a sense in which all voters play the same role in a symmetric electoral college.
Bartholdi et al., 2021 present an axiom – which they call equity – generalizing the type of
symmetry found in a symmetric electoral college. I prefer the term weak anonymity, as I am unsure
the axiom makes a voting rule equitable. Bartholdi et al. (2020) characterize the voting rules that
satisfy equity, neutrality, and positive responsiveness, and show that this class is broader than just
majority rule.4
404
Voting Rules
3 Multi-Candidate Elections
3.1 Plurality Rule

A simple voting rule for multiple candidates is plurality rule: a vote is just a vote for a certain
candidate (as opposed to, e.g., a ranking of candidates), and the candidate who receives the most
votes wins. Unlike majority rule with two candidates, in which the winner always receives at least
50% of the votes cast, under plurality rule, the winner may receive less than 50% of the votes, since
the vote may be split multiple ways. This latter feature can be problematic.
Recall again the 2000 Presidential election. The election within Florida for Florida’s electoral
votes was conducted by plurality rule. In addition to the major party candidates, Bush and Gore,
Ralph Nader, the Green Party candidate, was on the ballot. Whereas Bush defeated Gore by a
margin of 537 votes, Ralph Nader received 97,488 votes. Nader’s vote was less than 2% of the total,
but it was enough to swing the election. If Nader had not been on the ballot, it is likely that some of
his voters would have still voted and that more of them would have voted for Gore than Bush. Since
the margin of victory in the election was less than 1% of Nader’s vote, it is likely that, had Nader not
run, Gore would have won.5 The problem with plurality rule is that a candidate with little support –
a so-called spoiler – can change the outcome of the election.
Consider next an election with seven left-wing candidates and one right-wing candidate.
Suppose 70% of voters are left-wing and 30% are right-wing. The left-wing voters equally split their
vote among slightly differing left-wing candidates so that those candidates get about 10% of the vote
each. The right-wing candidate gets 30%, the full right-wing vote, and wins. Any of the left-wing
candidates would have beaten the right-wing candidate in a head-to-head election by a wide
margin: 70–30%. It seems clear that, given the population’s preferences, in this case, a left-wing
candidate should have won.
The examples show that plurality rule suffers from a number of problems:
1. Spoiler effect: adding a candidate who does not win to the ballot – indeed a candidate with only
weak support – may change the winner.
2. The winner of a plurality vote election may lose by a wide margin in a head-to-head contest
against other candidates.
These problems arise uniquely in the multi-candidate setting and do not seem well addressed by the
axioms previously introduced, adapted to the multi-candidate setting. Plurality rule seems to be both
anonymous and neutral, in that it treats voters and candidates symmetrically. Moreover, the problem
does not seem to be that votes don’t “point in the right direction.” Rather it seems that the pro-
blems have to do with the perverse effects of adding new candidates to the ballot and with the fact
that candidates with low support can win.
3.2 Ranked Voting Methods

In binary elections, it is straightforward what a vote is: a vote is just a vote for a single candidate. In
multi-candidate elections, there are many more possibilities. A vote could be a vote for a single
candidate as in plurality voting; a vote could indicate which alternatives are acceptable, as in a system
known as approval voting (see Brams and Fishburn, 2007); a vote could give a grade or score to each
candidate (see, e.g., Balinski and Laraki, 2007).
In ranked voting systems, a vote is a ranking of candidates: if the candidates are x, y, and z, then a
ranking might be y first, x second, z third. A voter who submits this ranking is stating that they
prefer y most and z least. In this system, we may or may not allow rankings with ties (e.g., x and y
can be ranked on the same level).
405
Itai Sher
A ranked voting rule takes (i) a ballot, which specifies the set of candidates running, and (ii) a list of
votes in the form of rankings of candidates on the ballot, one for each voter, as inputs; it gives as an
output a winner (or multiple winners, in the case of ties).
We can reformulate our axioms for ranked voting rules allowing multiple candidates. Anonymity
again means that voters are treated equally in the sense that if we interchange the votes of different
voters, the outcome won’t change. Neutrality again has a similar meaning. Consider the situation in
which there are three voters, Ann, Bob, and Carol, and candidates x, y, and z. Suppose that votes are
as in Table 32.1. In Table 32.1, Ann ranks x first, y second, and z third, and other voters’ votes are
interpreted similarly.
Now consider the vote profile that results by interchanging the role of x and z in each ranking as
in Table 32.2. For example, in Table 32.1, Ann ranked x first and z last, and, in Table 32.2, this is
reversed. For ranked voting rules, neutrality says that if x wins the election in Table 32.1, then z
wins the election in Table 32.2 (and vice versa).
Next, consider positive responsiveness. Table 32.3 differs from Table 32.1 only in Ann’s vote.
Ann’s vote differs in Table 32.3 as opposed to Table 32.1 only in that y moved up one spot and
overtook x. In this context, let positive responsiveness mean that
• If y wins in Table 32.1, then y also wins in Table 32.3.

• If y ties for victory with either x or z or both in Table 32.1, then y is the sole winner in
Table 32.3.
Table 32.1
Ann Bob Carol
x y z
y x x
z z y
Table 32.2
Ann Bob Carol
z y x
y z z
x x y
Table 32.3
Ann Bob Carol
y y z
x x x
z z y
406
Voting Rules
More generally, positive responsiveness means that if any alternative y moves up in someone’s ranking,
while the ranking is otherwise unchanged, that can only cause y to win if it has any effect at all; it cannot
cause y to lose or another alternative to win. Moreover, such a change will break ties in y’s favor.6
I will illustrate these axioms with the Borda count (Borda, 1784), a voting rule that assigns a score
to each candidate based on voter rankings. If there are three candidates, then for each voter’s
ranking, x is assigned a score of 2 if x is the top choice, a score of 1 if x is the second choice, and a
score of 0 if x is the last choice. Then we add up the scores given to x by all voters. The candidate
with the highest total score wins.
Let us apply the Borda count to Table 32.1. On Ann’s ballot, x is assigned a score of 2; on both
Bob’s and Carol’s ballot, x is assigned a score of 1. The total score of x is 4 = 2 + 1 + 1. Similarly,
the total score of y is 3, and the score of z is 2. So, by the Borda count, x wins.
More generally, with n candidates, the Borda count assigns a score of n 1 to the top choice,
n 2 to the second choice, …, and 0 to the bottom choice. Equivalently, a candidate’s Borda score
is the total number of votes the candidate gets across all pairwise elections with the other candidates.
It is easy to see that the Borda count satisfies the neutrality and anonymity axioms. Borda also
satisfies positive responsiveness because moving an alternative up in some voter’s ranking only in-
creases its own score without increasing the score of any other alternative.
However, the Borda count does suffer from problems similar to plurality rule.
Suppose for the moment that Table 32.4 represents not the votes actually cast in an election, but
rather the preferences of voters. For example, Ann prefers Gore to Bush to Nader.
Suppose only Gore and Bush were on the ballot, and that the election were conducted using the
Borda count. Then each voter would rank just Bush and Gore, basing their ranking on the which
one they rank higher in Table 32.4. The resulting ballot is as in Table 32.5.
The Borda scores are:
Gore: 1 + 1 + 1 + 0 + 0 = 3
Bush: 0 + 0 + 0 + 1 + 1 = 2
So Gore wins. Notice that Gore is also the majority rule winner. This is not a coincidence; the
Borda count and majority rule coincide for binary elections. In fact, the Borda count is one way of
generalizing majority rule to multi-candidate elections.
Now suppose Nader joins the race, and that voters vote as in Table 32.4. With all three can-
didates in competition, the Borda scores are:
Table 32.4
Ann Bob Carol Dan Edna
Gore Gore Gore Bush Bush

Bush Bush Bush Nader Nader
Nader Nader Nader Gore Gore
Table 32.5
Ann Bob Carol Dan Edna
Gore Gore Gore Bush Bush

Bush Bush Bush Gore Gore
407
Itai Sher
Bush: 1 + 1 + 1 + 2 + 2 = 7
Gore: 2 + 2 + 2 + 0 + 0 = 6
Nader: 0 + 0 + 0 + 1 + 1 = 2
So, if Nader enters the race, Bush wins, just as in the case of plurality rule discussed above. As in the
example of plurality rule, Nader has little support, but his presence or absence determines the outcome.
This observation motivates the following axiom:
Axiom: Nash Independence of Irrelevant Alternatives (NIIA). If we remove a losing candidate

from the ballot, this should not change the winner.
This axiom is called the independence of irrelevant alternatives because it says that removing an
irrelevant alternative (e.g., Nader in the above example) – irrelevant in the sense that it is not chosen –
should not affect the outcome of the election. The axiom is named after Nash because it resembles an
axiom that appeared in Nash (1950) and to distinguish it from a similar axiom due to Arrow (1950,
1951).7 One might think that the term “irrelevant” is question-begging and might better be replaced
by “unchosen.”
3.3 Impossibility Theorems

Proposition. There is no ranked voting rule satisfying anonymity, neutrality, positive
responsiveness, and Nash independence.
To see this, consider a three-candidate election where voters have one of three ideologies – Left,
Center, or Right – with rankings as in Table 32.6.
Suppose there are three voters with each ideology. Then, as the situation is symmetric, by neutrality
and anonymity, there must be a three-way tie between x, y, and z. Now suppose that one voter switches
from a Left to a Center ideology. Then there are two Left voters, four Center voters, and three Right
voters. Because this change amounts to one voter moving up z from last to first place in their ranking,
positive responsiveness implies that the three-way tie is broken in favor of z. So z is the winner.
But now suppose candidate x is removed. Then the new voting table becomes Table 32.7.
By NIIA, the winner must still be z. But May’s theorem – which only uses axioms we have
assumed for this proposition – implies that the winner of a binary election should be the majority
Table 32.6
Left Center Right
x z y
y x z
z y x
Table 32.7
Left (2 Voters) Center (4 Voters) Right (3 Voters)
y z y
z y z
408
Voting Rules
winner. Here, y beats z, 5–4. So the winner must be y. So we have a contradiction since we have
argued from the axioms both that the unique winner must be z and that the unique winner must be y.
It follows that there is no voting rule satisfying all four axioms, establishing the proposition.
The above proposition is an impossibility theorem. An impossibility theorem proves that several
desirable properties of a voting rule cannot be jointly satisfied. The most famous impossibility
theorem is Arrow’s impossibility theorem. Arrow’s axioms are different but similar in spirit to the
above. Arrow appeals to a different version of the independence of irrelevant alternatives axiom.
(Arrow, 1950, 1951)
There are several possible responses to impossibility theorems. One is to examine the axioms
more carefully in order to reject at least one. Another is to conclude that all the axioms represent
desirable properties but they cannot be satisfied simultaneously, so there is a trade-off between
different desirable properties. A third is to restrict the domain of application by restricting the set of
admissible rankings, for example, by assuming that rankings are “single peaked” in the sense that
they are all organized around a single left-right axis and differ only in their bliss point (Black, 1948);
a voting rule may satisfy the axioms on this restricted domain even if it does not satisfy them for all
possible rankings.
It is sometimes said that Arrow’s theorem undermines the idea of democracy or that it proves that
there is no good voting rule. But neither interpretation is correct. What the various impossibility
theorems show is that there is no voting rule that satisfies every desirable property that one might
think of, but that is consistent with the possibility that some voting rules are much better than others.
3.4 The Condorcet Criterion

This section considers a way of extending majority rule to multi-candidate elections. If voters submit
ranked lists, we can determine the outcome of all binary elections between pairs of candidates on the
ballot. For example, in Table 32.1, we can see that x would beat y head-to-head because Ann and
Carol prefer x to y while only Bob prefers y to x. Similarly, y would beat z, and x would beat z.
Notice that x beats both y and z in a majority vote. We call x a Condorcet winner. More
generally, a Condorcet winner is a candidate that beats every other candidate in a pairwise majority
vote (Condorcet, 1785).
Axiom: Condorcet criterion. If x is a Condorcet winner, then x wins the election.
A voting rule that satisfies the Condorcet criterion is called a Condorcet rule.
Consider the Borda count. With votes as in Table 32.4, Bush wins according to the Borda count.
But, in Table 32.4, Gore is the Condorcet winner. So the Borda count is not a Condorcet rule.
It is tempting to argue for the Condorcet criterion as follows.
1. May’s theorem establishes that we should decide binary elections by majority rule.
2. Therefore, a candidate x who beats all other candidates in a pairwise majority vote should be
the winner of the multi-candidate election.
However, observe that the Borda rule satisfies anonymity, neutrality, and positive responsiveness –
the axioms in May’s Theorem – in a multicandidate context, but is not a Condorcet rule. It follows
that May’s axioms do not by themselves imply the Condorcet criterion, and some additional
principle – beyond May’s axioms – is required in step 2 of the above argument.
A Condorcet winner does not always exist. Consider Table 32.6 and suppose again that there are
2 Left voters, 4 Center voters, and 3 Right voters. Then:
409
Itai Sher
• x beats y, 6 votes to 3; so x wins by a margin of 3 votes.

• y beats z, 5 votes to 4; so y wins by a margin of 1 vote.
• z beats x, 7 votes to 2; so z wins by a margin of 5 votes.
So we have a cycle – x beats y beats z beats x – and no Condorcet winner. This situation is known as
the Condorcet paradox, although it is not a paradox in the strict logical sense.
The existence of Condorcet cycles implies that there are many different Condorcet rules, which
select different winners when there is a cycle.
In the above example, let us rank the three pairwise elections by their margin of victory. The z vs x
election has the biggest margin of victory of 5. So let us conclude that z is better than x. The x vs y
election has the next biggest margin of 3. So we conclude that x is better than y. Since z is better than x
and x is better than y, transitivity implies that z is better than y. Finally, we examine the y vs z election,
having the smallest margin of victory. That election tells us that y is better than z. However, this
conflicts with our previous conclusion that z is better than y. Therefore, we discard the last election
with the smallest margin, and conclude that z is better than x and x is better than y, so that z wins.
This general method of ranking elections by margin of victory, and prioritizing results of elec-
tions with a larger margin, and discarding results that conflict with conclusions arrived at, via
transitivity, from elections with bigger margins, is called ranked pairs (Tideman, 1987). Ranked pairs
always selects a Condorcet winner when there is one. Therefore, ranked pairs is a Condorcet rule.
Other Condorcet rules deal with the absence of a Condorcet winner in different ways.
Dasgupta and Maskin (2008) provide a justification for Condorcet rules. They define majority rule
in a multi-candidate context to be a rule that selects the Condorcet winner when it exists and selects
no winner otherwise. This departs a little from our formal framework in allowing for the possibility of
no winner. Notice also that this is a broader use of the term “majority rule” than I have used above;
I shall instead refer to it as Condorcet.
Say that candidate x is dominated by candidate y if everyone ranks y above x. In addition to the
above axioms, Dasgupta and Maskin present the following:
Axiom: Unanimity. If x is dominated by some other candidate, then x is not the winner.
This seems quite compelling. Why would we pick a candidate who is dominated by some other
candidate? Formally, Dasgupta and Maskin work in a setting with a continuum of voters; given the
anonymity axiom and a finite number of candidates, this can be interpreted roughly as a setting in
which a voting rule takes as an input the percentages of voters with each ranking. Dasgupta and
Maskin also present a generic decisiveness axiom, which says roughly that for almost all vote profiles,
the voting rule produces a unique winner. A domain D is a subset of the set of all possible rankings.
We can consider a voting rule in a setting where only rankings in D are permissible as inputs.
Moreover, we can ask whether a voting rule satisfies an axiom on D – that is, when rankings are
restricted to D. Call the combination of anonymity, neutrality, NIIA, unanimity, and generic de-
cisiveness the DM axioms.
Dasgupta and Maskin show that if there exists a domain D on which some voting rule f satisfies
the DM axioms, then Condorcet satisfies the DM axioms on D also, and if f differs from Condorcet
on D, then there exists another domain D , such that Condorcet satisfies the DM axioms on D but f
does not. So, in a sense, Condorcet satisfies the DM axioms more often than any other rule; this
provides a foundation for Condorcet.
An alternative foundation is provided by Horan, Osborne, and Sanver (2019), who show that
Condorcet rules uniquely satisfy the May axioms – anonymity, neutrality, and positive responsiveness –
together with NIIA on vote profiles for which there is a Condorcet winner. See also Alemante,
Campbell, and Kelly (2016).
410
Voting Rules
3.5 Instant Runoff Voting

One commonly advocated multi-candidate voting rule is instant runoff voting (IRV). (Sometimes this
is called ranked choice voting, but many rules use rankings!) Under IRV, voters submit a ranking. For
each voter, a vote is entered for their top ranked alternative. If some candidate wins a majority –
more than 50% of vote – then that candidate wins. Otherwise, the candidate getting the smallest
number of votes is disqualified. If a voter voted for the disqualified candidate, the voter’s vote is
reallocated to their second choice. We then again check whether any candidate receives a majority,
in which case that candidate is the winner, and if not, we repeat the process until we arrive at a
candidate that receives a majority. As we progress, some voters may have their votes count for their
second, third, fourth, and so on choices, as their higher ranked choices are eliminated.
In the 2000 Florida election, in the first round, no candidate would have gotten a majority. So
Nader would have been eliminated. So the winner would have been whichever of Bush and Gore
would have won head-to-head. So IRV deals well with the simple examples of spoilers that plague
plurality rule.
Consider Table 32.8, where we have specified the percentage of voters in each group. Under
IRV, x, which has the lowest top choice support, is eliminated in the first round. Then Group 1’s
voters are reallocated to y. So y beats z in the second round.
Notice however that
• x beats y, 65–35%.
• x beats z, 65–35%.
So x is the Condorcet winner and beats both alternatives by a wide margin, and yet y wins according
to IRV. This seems problematic. Moreover, IRV violates NIIA, because removing the loser z from
the ballot would cause x to win instead of y.
While IRV is imperfect, it handles spoilers with low support like Nader well. It is also important
that voters understand the voting method and IRV seems easier to understand than Condorcet rules
such as ranked pairs. This makes IRV an attractive candidate to replace the more commonly used
plurality rule.
4 Strategic Incentives
Another problem with plurality rule is that it creates strategic incentives. For example, if there are
two left-wing candidates and one right-wing candidate, then left-wing voters may have to guess
which of the left-wing candidates will be stronger and a failure to coordinate on one of them may
cause the right-wing candidate to win. IRV is supposed to address this by making it safe for voters to
express their true preferences without worrying about the electoral consequences. For example, if
the 2000 Florida election had been conducted via IRV, Nader voters who preferred Gore to Bush
could safely have ranked Nader first and Gore second without worrying about throwing the election
to Bush.
Table 32.8
Group 1 (30%) Group 2 (35%) Group 3 (35%)
x y z
y x x
z z y
411
Itai Sher
Unfortunately, IRV is also subject to strategic incentives. Consider Table 32.8. Under IRV,
voters in Group 3 had an incentive to pretend to be in Group 1 in the hope of helping x make it into
the second round, because if x is eliminated in the first round, then this will cause y, Group 3’s least
preferred choice, to win in the second.
A natural question is whether there exists a voting rule that makes it always safe to submit one’s
true preference ranking. Say that a ranked voting rule is strategyproof if, no matter how others vote,
no voter could secure an outcome that she prefers more by misrepresenting her preferences than by
reporting her true preference ranking. One reason strategyproofness is desirable is that it makes it
more likely that the information a voting rule takes as input reflects true preferences. Gibbard (1973)
and Satterthwaite (1975) proved the following theorem.
Theorem: Gibbard-Satterthwaite Theorem. Suppose there are at least three alternatives.

Then there does not exist a ranked voting rule that satisfies the unanimity axiom and is also
strategyproof.
This result implies that any reasonable voting rule will sometimes create incentives for voters to
misrepresent their preferences. However, some voting rules are better than others with regard to
minimizing strategic incentives to misrepresent, for example, IRV is superior to plurality rule.
Using a framework similar to Dasgupta and Maskin (2008), Dasgupta and Maskin (2020) show
that if a voting rule satisfies strategyproofness, unanimity, anonymity, neutrality, NIIA, and generic
decisiveness on a domain, the rule must coincide with Condorcet on that domain. In that sense, if
one accepts the other axioms, Condorcet performs best with regard to strategyproofness.
5 Voting Power
One way to assess voting rules is in terms of voting power, that is, the influence that a voter has on the
outcome of an election.
For simplicity, in this section, we consider binary elections.
A voter is pivotal if changing her vote would change the outcome of the election. Whether a
voter is pivotal depends on how others vote. A voter’s pivotality probability is the probability the voter
will be pivotal. We may equate a voter’s pivotality probability with their voting power, or their
capacity to influence the election. This suggests a criterion for evaluating voting rules: we may
search for rules that equitably and maximally allocate voting power.
Voting power depends not just on the voting rule but on the probabilities of how others will
vote. This in turn depends on the probabilities they will have various preferences, and the incentives
generated by the rule.
Consider random dictatorship: each of n voters is selected with probability 1/n and that voter may
select the outcome. The pivotality probability of each voter is 1/n. Can we do better? Suppose there
are three voters and we use majority rule. Suppose each voter is equally likely to prefer x and y, and
these probabilities are independent across voters. Each voter is pivotal exactly if the other two voters
vote for different candidates, which happens with probability 1/2. So, under majority rule, each voter
is pivotal with probability 1/2, in contrast to 1/3 for random dictatorship. So, in this respect, majority
rule is superior to random dictatorship. Observe however that this result depends on the independence
of voter preferences. At the other extreme, we could assume that voter preferences were perfectly
correlated. Suppose for example that all three voters base their decision in the same way on a publicly
observed event. Then, under majority voting, none is ever pivotal since the others vote in the same
way. In that case, random dictatorship gives voters more voting power than majority voting.
Let us call the situation in which all voters are equally likely to prefer either candidate and these
probabilities are independent across voters the random voting model. In the random voting model,
412
Voting Rules
voting power defined as pivotality probability is known as Banzhaf power (Banzhaf, 1964).8 Dubey
and Shapley (1979) show that, in the random voting model, majority rule is optimal in the sense that
it maximizes average voting power; it also distributes voting power equally.
Gelman, Katz, and Bafumi (2004) have criticized the random voting model. In reality, different
voters have different probabilities of voting for each candidate and these probabilities can be cor-
related because of, e.g., local or national events that move voter sentiment.
Laruelle and Valenciano (2005) distinguish decisiveness from success. Decisiveness is just pivotality,
that is, the probability that a voter’s vote will determine the outcome of an election. Success is the
probability that a voter’s vote will agree with the outcome. We saw above that whether majority rule
is optimal with regard to decisiveness depends on the probability model. In contrast, majority voting
always maximizes the average probability of success regardless of the probability model. For ex-
ample, the average probability of success across voters is always at least as large under majority rule as
it is under random dictatorship or the electoral college. So success provides a more robust foun-
dation for majority rule than does decisiveness.
We have discussed voting power in relation to binary elections. Bolger (1986) and Sher (2020)
provide approaches to voting power in multi-candidate elections.
6 Conclusion
This paper has used the axiomatic method of social choice to analyze voting rules. The axioms
correspond to qualitative criteria by which to evaluate voting rules – some rules satisfy each axiom
and some do not. The axioms in turn must be critically evaluated. For binary elections, May’s
theorem shows that majority rule performs uniquely well with regard to compelling axioms.
Dasgupta and Maskin (2008) and Horan, Osborne, and Sanver (2019) provide a similar justification
for Condorcet rules, which generalize majority rule. These results are less decisive than in the binary
case because no multi-candidate rule satisfies all the axioms one might desire.
There may be practical considerations that are not well captured by the formal axioms.
Condorcet rules are rarely used, and the reason may concern the difficulty of explaining them to
voters. It is important for voting rules to be transparent to participants. For this reason, we might
prefer instant runoff voting as an alternative to plurality rule even though it is axiomatically inferior
to Condorcet.
Notes
1 I am not taking a stand on whether the examples that I mention actually justify a departure from anonymity.
2 One would think that voters would respond to minority rule by voting for their least favorite candidate,
effectively turning minority rule into majority rule. For such strategic responses, see Section 4. For now,
imagine that voters vote sincerely.
3 An exception is the degenerate case with one voter per state in which case the electoral college collapses into
majority rule.
4 A symmetric electoral college does not, in general, satisfy positive responsiveness.
5 See Herron and Lewis (2007).
6 More precisely, it can break preexisting ties in y ’s favor. If previously x had been winning, the change might
create a tie between x and y .
7 Arrow’s IIA is formulated in a model of voting rules in which the inputs are profiles of rankings and the
outputs are social rankings, and not just winner from the overall ballot, as in this paper. It says that the social
ranking of any two alternatives should depend only on the individual rankings of those two alternatives and
not how other alternatives are individually ranked.
8 The underlying concept was discovered previously by Penrose (1946). An alternative voting power index is
the Shapley-Shubik index (Shapley and Shubik, 1954).
413
Itai Sher
Further Reading
Arrow (1951) is a classic foundational work of social choice theory and voting introducing Arrow’s famous
impossibility theorem. Chapter 2 of Gura and Maschler (2008) presents an accessible treatment of Arrow’s
impossibility theorem. Sen (2018) is an updated and expanded edition of Sen’s classic 1970 work of social
choice and voting. Saari (1995) presents a geometric approach to voting.
Dietrich and Spiekermann (2013) provide a critical discussion of the Condorcet jury theorem, which purports
to provide an epistemic foundation for majority rule, and proposes alternative results on epistemic de-
mocracy which are more credible.
Felsenthal and Machover (1998) provide a detailed treatment of voting power.
Gaertner (2009) is a book-length introduction to social choice. Pacuit (2019) provides a survey of voting rules
covering some topics that I did not have space to cover. List (2013) provides a survey of social choice theory,
including both voting theory and topics going beyond voting theory.
Posner and Weyl (2015) present quadratic voting, a method involving the purchase of votes aimed at making
voting outcomes efficient. Laurence and Sher (2017) provide a critique of quadratic voting, and more
generally voting methods that involve the purchase of votes for money.
References
Alemante, F., Campbell, D. E., and Kelly, J. S. (2016) “Characterizing the Resolute Part of Monotonic Social
Choice Correspondences,” Economic Theory, 62, 4, 765–783.
Arrow, K. J. (1950) “A Difficulty in the Concept of Social Welfare,” Journal of Political Economy, 58(4), 328–346.
Arrow, K. J. (1951) Social Choice and Individual Values, New York; Wiley.
Banzhaf, J. F. (1964) “Weighted Voting Doesn’t Work: A Mathematical Analysis,” Rutgers Law Review, 19,
317–343.
Balinski, M. and R. Laraki (2007) Majority Judgement: Measuring, Ranking and Electing, Boston; MIT Press.
Bartholdi, L., Hann-Caruthers, W., Josyula, M., Tamuz, O., & Yariv, L. (2021) “Equitable Voting Rules,”
Econometrica, 89, 2, 563–589.
Black, D. (1948) “On the Rationale of Group Decision-Making,” Journal of Political Economy, 56, 23–34.
Bolger, E. M. (1986) “Power Indices for Multicandidate Voting Games,” International Journal of Game Theory,
15, 3, 175–186.
Borda, J.-C. de (1784) “Mémoire sur les élections au scrutin par M. de Borda” in Mémoires de l’Académie Royale
des Sciences année 1781, Paris; l’Imprimerie Royale, pp. 657–665.
Brams, S. and P. Fishburn (2007) Approval Voting, New York; Springer.
Condorcet, Nicolas de (1785) Essay sur l’Application de l’Analyse à la Probabilité des Décisions Rendue à la Pluralité
des Voix, Paris; De l Imprimerie Royale.
Dasgupta, P., and Maskin, E. (2008) “On the Robustness of Majority Rule,” Journal of the European Economic
Association, 6, 5, 949–973.
Dasgupta, P., and Maskin, E. (2020) “Strategy-Proofness, Independence of Irrelevant Alternatives, and Majority
Rule,” American Economic Review: Insights, 2, 4, 459–474.
Dietrich, F., and Spiekermann, K. (2013) “Epistemic Democracy with Defensible Premises,” Economics &
Philosophy, 29, 1, 87–120.
Dubey, P., and Shapley, L. S. (1979) “Mathematical Properties of the Banzhaf Power Index,” Mathematics of
Operations Research, 4, 2, 99–131.
Felsenthal, D. S., and Machover, M. (1998) The Measurement of Voting Power, Cheltenham, UK; Edward Elgar.
Gaertner, W. (2009) A Primer in Social Choice Theory: Revised Edition, Oxford; Oxford University Press.
Gelman, A., Katz, J. N., and Bafumi, J. (2004) “Standard Voting Power Indexes Do Not Work: An Empirical
Analysis,” British Journal of Political Science, 34, 4, 657–674.
Gibbard, A. (1973) “Manipulation of Voting Schemes: A General Result,” Econometrica, 41, 4, 587–601.
Gura, E. Y., and Maschler, M. (2008) Insights into Game Theory: An Alternative Mathematical Experience,
Herron, M., and Lewis, J. (2007) “Did Ralph Nader Spoil a Gore Presidency? A Ballot-Level Study of Green
and Reform Party Voters in the 2000 Presidential Election,” Quarterly Journal of Political Science, 2, 3,
205–226.
Horan, S., Osborne, M. J., and Sanver, M. R. (2019) “Positively Responsive Collective Choice Rules and
Majority Rule: A Generalization of May’s Theorem to Many Alternatives,” International Economic Review,
60, 4, 1489–1504.
414
Voting Rules
Laruelle, A., and Valenciano, F. (2005) “Assessing Success and Decisiveness in Voting Situations,” Social Choice
and Welfare, 24, 1, 171–197.
Laurence, B., and Sher, I. (2017) “Ethical Considerations on Quadratic Voting,” Public Choice, 172, 1, 195–222.
List, C. (2013) “Social Choice Theory,” The Stanford Encyclopedia of Philosophy (Winter 2013 Edition), Edward
N. Zalta (ed.), https://plato.stanford.edu/archives/win2013/entries/social-choice/.
May, K. O. (1952) “A Set of Independent Necessary and Sufficient Conditions for Simple Majority Decision,”
Econometrica, 20, 4, 680–684.
Nash Jr, J. F. (1950) “The Bargaining Problem,” Econometrica, 18, 2, 155–162.
Pacuit, E. (2019) “Voting Methods,” The Stanford Encyclopedia of Philosophy (Fall 2019 Edition), Edward N.
Zalta (ed.), https://plato.stanford.edu/archives/fall2019/entries/voting-methods/.
Penrose, L. S. (1946) “The Elementary Statistics of Majority Voting,” Journal of the Royal Statistical Society, 109,
1, 53–57.
Posner, E. A., and Weyl, E. G. (2015) “Voting Squared: Quadratic Voting in Democratic Politics,” Vanderbilt
Law Review, 68, 441–500.
Saari, D. (1995) Basic Geometry of Voting, Berlin; Springer.
Satterthwaite, M. (1975) “Strategy-Proofness and Arrow’s Conditions: Existence and Correspondence
Theorems for Voting Procedures and Social Welfare Functions,” Journal of Economic Theory, 10, 2, 187–217.
Sen, A. (2018) Collective Choice and Social Welfare, Cambridge, MA; Harvard University Press.
Shapley, L. S., and Shubik, M. (1954) “A Method for Evaluating the Distribution of Power in a Committee
System,” American Political Science Review, 48, 3, 787–792.
Sher, I. (2020) “Freedom and Voting Power.” Unpublished Manuscript.
Tideman, T. N. (1987) “Independence of Clones as a Criterion for Voting Rules,” Social Choice and Welfare, 4,
3, 185–206.
415
33
ENABLING INFORMED AND
EQUAL PARTICIPATION
Thomas Christiano
1 Basic Rights to Participate

Democracy is a system of collective decision-making in which adults have the right to participate as
equals in some essential stage of the decision-making. In a modern representative democracy, people
participate as equals in voting for representatives in the legislature and the executive and sometimes
in the judiciary. The most basic framework of rights under which people participate as equals in a
democracy consists of the right to vote, the right to run for office, the rights to associate with others
in political parties and other political associations, the rights to express one’s opinions freely in
support of candidates, parties, policies, or the basic aims of policy. The right to vote is subject to an
egalitarian principle; the principle is one person one vote. This implies that persons have equal votes
in each election they may participate in, it requires that the decision rule is majoritarian, and in many
circumstances, it requires that legislative districts are of equal size so that each person’s vote counts
for the same in terms of its effect on the legislature.
When the basic rights listed above are properly protected, and the outcome of the vote de-
termines which candidates are elected, then the system of elections is described as free and fair. And
when legislatures and executives make law in a way that reflects the outcomes of the vote and the
society is genuinely regulated by the rule of law, then we have a genuine democracy. We might also
call a modern representative democracy in which the above rights are well protected a minimally
egalitarian democracy.
These are complex rights. They are negative: the rights to vote, to free speech, and to free
association require that the state and people not interfere with the exercise of the rights. But they are
also positive: the government has a duty to provide services to help realize the various rights. Just to
give a quick sense of this, the government of the United Kingdom spent nearly 200 million dollars
on its snap parliamentary election in 2017 (Dallison, 2017). In the United States, the cost is mostly
unknown since it is shared by Federal, State, County, and Municipal governments. We do know
that the Federal Government allocated 3 billion dollars in 2002 to upgrade voting machines. (Hubler
and Underhill, 2018). Any serious curtailing of this process will result in protests in the name of legal
and moral rights to participation.
The right to participate in democracy has made a very substantial difference to all those to whom it
has been extended. The rise of universal white manhood suffrage in America precedes the rise of the
progressive era of reform in American politics; the rise of women’s suffrage concludes this develop-
ment in the United States with the rise of the welfare state in the New Deal. African-American men
416 DOI: 10.4324/9780367808983-40

Enabling Informed and Equal Participation
initially acquired the right to vote in the United States with the ratification of the 15th Amendment to
the US Constitution in 1870. They made great gains in southern states till the 1890s at which point the
Southern white elites figured out how to deprive them of the right to vote, and then deprive them of
the gains. In 1965, African-Americans were able once again to make governments work in their
interests after having recovered the vote (Wright, 2013).
Minimally egalitarian democracy has brought great benefits to the societies in which it has been
achieved. They have radically improved the protection of basic human rights, stimulated economic
growth, avoided going to war with each other, promoted public goods in the society and many
other goods (Christiano, 2011; Acemoglu et al., 2019). We need to remember both the egalitarian
and the common good features of minimally egalitarian democracies as we proceed to discuss issues
of participation.
The guiding ideal of democracy that I recommend to you here is the idea of democracy as
political equality in which there is an egalitarian distribution of political power over collective
decision-making. This distribution of political power is not achieved just by the implementation of
minimally egalitarian democracy. To understand this, we need to get a grasp of the different di-
mensions of political power in a democracy. The first dimension of political power in a democracy is
that of minimally egalitarian democracy. This distributes a kind of brute power over the collective
decision-making in the sense that it does not yet involve the ability to persuade others.
2 Informational Power
There is a second dimension that is essential to understanding democracy and that is the dimension
of informational power. The question is, what do people do with the rights that they have? This
depends on the information they have about society, about policy, and about politicians’ dispositions
towards policies. By “information” I mean the content of beliefs and values they have.
To grasp the idea of informational power it is worth trying to grasp the idea of a minimally
egalitarian democracy in which every person has very well-developed values and full information
about the society they live in, the effects of policies on that society, the relations between these
effects and their values and the dispositions of politicians towards those policies. In such a world, a
minimally egalitarian democracy would be a fully egalitarian democracy. In such a society, an
election would simply consist of people announcing that they will run for office. With that in-
formation, each person would know exactly who to vote for, given that they have well-developed
values, an understanding of society, how policy affects it, and what politicians will do. The outcomes
of elections would be perfectly determinate, and politicians would be expected to do as they said
they would. If we suppose that people do not initially know each other’s preferences, then there
would be some period of adjustment in which politicians would put themselves up for election by
proposing platforms and they would adjust them when they get little support.
We understand the issue of informational power when we see that human beings are very far
from perfectly or fully informed about society, policy, and politicians and their values and con-
ceptions of their interests are only modestly reflectively formed. We are at best only modestly
informed. We start out usually quite ill-informed and sometimes make some progress towards
becoming better informed. What I call informational power has two different components. One
component is the ability politically to understand and make use of transmitted information. Another
component is the power to transmit information.
When we are far from full information, being well-informed in politics involves having access to
a coherent and defensible, though not necessarily correct, picture of the relations between one’s
experience, one’s values and interests, the larger society, the available alternatives for policy and the
relations of politicians and parties to those policy alternatives. This conception enables a person to
navigate the constant and chaotic stream of information they experience in modern societies. It is
417
Thomas Christiano
what makes the political system practically intelligible to a person to some degree. We might call it a
political intelligibility framework. It provides a framework with which a particular citizen can in-
terpret what is happening in the society and with which a person can make decisions about which
representatives and parties to support. It does not have to be inflexible, indeed it must not be. To be
clear, as I will articulate more later, all the elements of the picture or conception cannot and need
not be held in any single person. It can be distributed across reasonably like-minded persons who are
properly connected in a network.
A political intelligibility framework can be more or less sophisticated. For example, I receive
information from a major newspaper that the proposed minimum wage increase will cause un-
employment. Already, of course, I have to know what minimum wage is and I have to know what
unemployment is. But do I know that the situation is significantly more complex than is suggested in
the newspaper article, for example, that most economists think small increases do not cause much
unemployment and that the poverty-reducing benefits may be greater than the losses? Do I know
that the current proposal is not only to raise the minimum wage but also to implement policies
lowering unemployment? These are just some of the things a sophisticated receiver of political
information will think through.
We can see that a lot of what happens in a democracy is an attempt to overcome, and, in some
cases, take advantage of, the mix of paucity of information for many and chaotic information from
many sources. Election campaigns attempt to inform voters about what the candidates stand for and
what the implications of the policy platforms are for society. They attempt to make the system
intelligible and argue for practical implications. Of course, election campaigns are also designed to
motivate citizens who are already expected to favor the candidates (we will discuss this issue more
below). Newspapers, political parties, political groups, associations like the chambers of commerce,
churches, and unions attempt to inform citizens. And of course, ordinary citizens talk to each other
and discuss politics as well. What is key here is that the development of bodies of information and
the transmission of information are costly activities.
Another dimension of the process of transmission of information is information transmitted to
politicians after an election, which they can use to craft legislation. Politicians are also only modestly
informed at best. They need a lot of help in putting legislation together. Political parties, lobbying
firms, and secondary associations play a large role in helping legislators craft legislation. Secondary
associations play a large role in aggregating information among persons, for instance, unions play a
large role in collecting information about the conditions of workers and then disseminating it.
3 The Distribution of Informational Power

In a world of highly imperfect informedness, the distributions of the capacities to transmit in-
formation and to receive information in a sophisticated way are essential constituents of the dis-
tribution of political power. Once we have established a minimally egalitarian democracy, the
distribution of informational power is the key element of political power.
To be sure, the distribution of wealth and the distribution of economic power have a great
influence on the distribution of political power, but this is because they have great influence over the
capacities to transmit information and receive information in a sophisticated way. The reason for this
is that the transmission of information and the sophisticated reception of information are very costly.
Taking each in turn, the cost of transmission of information about politics in the United States
can be seen in the cost of election campaigns. The costs of campaigns tend to be much more
expensive than the government costs of putting on an election, which we noted above. This cost
makes it the case that normally no one can run for office without a substantial fund to finance the
campaign. I include in this the expenditures of groups that give independent support to political
campaigns. This substantial fund comes primarily from the affluent part of society or the top 10% of
418
the income distribution (Sorauf, 1994). And very big donations come from the very top of the
income distribution. This has an impact on who runs for office. Presumably, only those who can
appeal to the affluent can run. Either the affluent choose those like-minded persons to run or
politicians adjust their platforms to enable them to secure the support of the affluent (Christiano,
2012). This does not imply that the person who raises the most money wins, but it does imply that
one must raise a lot of money to be in the game. And this is not a trivial fact because preferences are
not distributed randomly throughout the population. The affluent tend to favor, for example, less
redistribution than the less well off (Erickson, 2015). And so, this tends to put a damper on the
willingness of the state to engage in redistribution. To be sure, there are many disagreements among
the affluent, so there is still a lot of pluralism in the political system even when the affluent are able to
play a disproportionate role.
To be clear, there are some associations that buck this bias in favor of the affluent in financing
election campaigns. Unions have traditionally raised money from their members to support parties
and candidates’ campaigns and to support lobbying efforts. And these members tend to be lower
middle class. Here we see the function of unions in aggregating information about workers and
informing each of them as well as funding the broadcast of information about worker’s needs and
demands to the larger society as well as to the lawmaking process. Political parties in urban en-
vironments are also sometimes able to organize lower-income people in political campaigns as in
door-to-door campaigns or telephone banks. Furthermore, churches also are able to raise money to
support campaigns and they are able to aggregate information about parishioners. But with unions in
decline, the presence of the non-affluent in financing election campaigns is diminished (Rosenfeld, 2014).
Let us look at the side of citizens’ sophisticated reception of information. Here the key element is
the sophistication of voters in discriminating among candidates in terms of policy concerns. The
level of sophistication of voters is an essential part of the possession of political power. There are two
reasons for this. One reason for the importance of sophistication to participation is that it enables
people to develop conceptions of their interests and what justice and the common good require. It
helps them see the place of their interests in the larger society. And it also helps them connect the
interests which they experience in their everyday lives with the legal and political system in such a
way that they can see how it can advance those interests. It helps them see who else shares these
interests and thereby enables them to come to a better understanding of those interests and the social
conditions under which they thrive. People who have more sophisticated frameworks with which to
make the system intelligible are less prone to being moved by superficially plausible information or
simple emotional appeals. One way to think of this is to imagine a person who possesses an au-
tomobile but who does not know how to drive it and has no sense of where to take it. They have a
kind of external power, but their lack of sophistication implies that they are powerless.
The second reason for this is that politicians are more responsive to sophisticated voters than they
are to unsophisticated voters. The explanation for this is that unsophisticated voters are not paying
very close attention to politicians. Politicians have much more freedom regarding unsophisticated
voters than they do with sophisticated voters. Unsophisticated voters tend to focus more on per-
sonality and on party identification of politicians than do sophisticated voters (Erickson, 2015). And
so, politicians can select policies that the more sophisticated are in favor of as they can rely on the less
sophisticated to vote for them on other grounds. I do not mean to suggest that party identification is
not a valuable cue for people. It is very important, but it is not as refined as what sophisticated voters
are concerned with.
To be sure, sophistication comes in degrees. People who know very little about politics except
that one political party tends generally to advance their concerns while the other does not, already
have important sophistication. But it is not as effective as persons who have more refined under-
standings of their interests and the common good as well as the connections of these to society and
policy.
419
Thomas Christiano
4 What Is Rational Participation?

One way to characterize rational participation is to say that people maximize utility in their actions.
Utility, here, does not have simply to mean a person’s well-being or interests. It merely refers to the
preferences of the agent, which in the context of politics can be for the common good, justice, or
the particular interests of a group of persons. The benefit of information can be understood in terms
of how much the information advances the satisfaction of the preferences of the agent and the cost is
understood in these terms as well. In politics, the benefit of becoming informed is to be understood
in terms of one’s estimate of the extent to which one can advance one’s aims with the help of the
extra information. The cost is normally understood in terms of the resources, time, and effort that
must be expended to acquire a piece of information as well as the foregone opportunities.
There is a piece of complexity here that one must grasp. The process of becoming informed
contributes to one’s understanding of how to satisfy one’s preferences. It also contributes to critical
reflection on the preferences and sometimes changes of preference. People’s sense of what is just or
fair or what their interests are can and does change in this process of reflection. In economic terms,
this must be represented as satisfying a higher-order preference to make one’s preferences genuinely
be for the good and the right (Sen, 1977).
Here there are some major sources of disagreement that can lead to vastly different expectations
about political participation and different views of what one is required to do in politics. On the one
hand, many people think that rational persons advance their aims individualistically. They simply
ask, what can I do to make things better? They treat other people’s actions merely as part of the
environment in which they act and then try to do the best they can. Such a person determines how
likely their vote is to be decisive. The chances of being decisive are extremely small (See Barnett,
2020, for a discussion of how to estimate this). They then determine how valuable the outcome is
and decide whether voting is worthwhile by comparing this small, expected benefit with the costs of
voting and informing oneself. It is often thought that voting and participation generally are not
especially valuable. Such thinkers also think that democracies tend to be ruled by ignorant persons
(Somin, 2014). In any case, they do not put much store in the idea that democracy realizes intrinsic
values of equality or freedom (Schumpeter, 1950).
Others think that people act in many contexts in collective ways. They think of themselves as
acting with other agents and sharing projects (Maskivker, 2019). They assess what the group should
be doing and then ask how they can do their fair share of this, even if their individual action, taken
by itself, would make little difference. Many people think that recycling is of this kind. Each act of
recycling makes an infinitesimally small difference to the quality of the environment, but the col-
lection of the acts of many people makes a big difference. So, each person contributes by doing their
bit. There is ample evidence for this mode of reasoning among people (Fehr and Fischbacker, 2005).
And there have been attempts to vindicate it as a legitimate independent form of reasoning. In the
history of philosophy, Cicero (1991) and the natural law tradition (Grotius, 2012) as well as
Rousseau (1987) and Kant (2012) defend variants of this kind of reasoning. More recently, econ-
omists and philosophers have attempted to vindicate it (Roemer, 2019; Bajaj, 2021). On this view,
politics is typically not a context in which one advances one’s aims in an individualistic way. One
advances one’s aims with other people. In general, the aims are usually connected with the common
good and justice and the way in which one pursues the aims is in collaboration with other like-
minded people. My vote is just one among many. I vote with many other persons in the hope of
achieving, with their aid, desirable outcomes. I participate in organizing people only on the con-
dition that many others are doing the same. I can recognize that my own contribution does not
make a great deal of difference while also thinking that I am willing to do my bit if enough others
are willing to do so as well (as in public goods games, see Bowles, 2016).
420
There is another set of disagreements that are close to this and that are less well articulated
philosophically. To put it in stark terms, we might distinguish between purposive and identity
approaches to politics (Achen and Bartels, 2016). The purposive conception is that citizens parti-
cipate to achieve an outcome, either as individuals or as members of groups. An extreme version of
the identity approach to participation suggests that the way people participate is determined by their
group identities. One proposed mechanism, which contrasts with the purposive approach, is that
people mimic each other’s performances. Another might be the collective sharing of emotion as we
see in fan participation in sports events. A conciliatory approach would give the purposive aspect of
participation priority in characterizing rational participation, but it would emphasize the importance
of groups in advancing these purposes both because of the mode of practical reasoning (each doing
her bit in the larger social project) and in the division of labor within groups, as I will try to show
below. The conciliatory approach would also concede that the rational pursuit of ends is often
sidetracked in politics by powerful emotions, polarization, and other pathologies. In part, the proper
design of institutions of participation would be guided by the need to diminish the pathological
elements in politics.
5 The Social Conditions of Sophistication

Social conditions can enhance or set back the interest in sophistication. Insofar as some people have
better access to sophistication than others, they have more political power. And given the nature of
the interest in sophistication and the conditions that enhance it, persons have positive rights to the
social conditions that give greater access to sophistication. I proceed to this argument now.
To understand the sources of sophistication, we need to look at how people acquire information.
It is useful to employ a basic economic picture here. The most developed approach is that of
Anthony Downs (1957). He proceeds from the basic economic principle that one collects more
information up to the point where the benefit of an extra increment of information is equal to the
cost of collecting that new piece of information. Up to that point, each new piece of information
brings more benefit than the cost of acquiring it. This maximizes the utility of the person collecting
information. For example, imagine looking for an apartment in some town you are moving to. You
look at a number of places and, if you are lucky, you have some friends you can consult with. At
some point, you stop looking, satisfied that the place you have found is good. You have decided at
that point that looking at more places will not help you find a significantly more valuable place and is
not worth the extra time, effort, and resources you would need to spend to look further. You decide
that if you search more, the marginal cost becomes more than the marginal benefit.
How does one know that one needs more information, or where to look for it? The absence of
information is not easily overcome because one doesn’t know what one doesn’t know. If one is
extremely unsophisticated, one might not know that one needs to become more sophisticated, at
least in a particular area.
This is the place in which the idea of “free information” becomes significant. Information and
tools to analyze it are free when one acquires essentially them as byproducts of the pursuit of other
purposes, which information then becomes useful for a new task. For example, to take up our
apartment search. During that search, I might find a person who I like very much and who I
wouldn’t have found otherwise. The information about that person is free information.
One main source of sophistication for the initially unsophisticated is education. It seems clear that
children get educations not because they know the value of the information they are receiving, but
because they are trying to please their elders. But the normal consequence of their education is their
increased sophistication. The better educated are more sophisticated than the worse educated. They
have more information and more tools with which to search for and analyze new information and
put it to use.
421
Thomas Christiano
Another important source of sophistication is work. Many people receive a great deal of free
information at their jobs because of the nature of their work and through informal discussions with
their colleagues. For example, a lawyer receives a great deal of information about the law while she
works. She must not only know what the law is in the area of her work, but she must also have some
idea of how it will evolve and thus the political processes in which it evolves. She receives
the information in one context for the purpose of doing her job and then can apply the information
in the context of thinking about politics. Furthermore, if she is a member of a law firm, she receives
information in discussions with the other partners in developing policy for the firm and in informal
discussions with colleagues who also develop sophistication in their work. We can see the same
processes at work in businesspersons, government workers, and other professionals. In addition,
noting that the family and friends of persons will tend to be in similar professions or work, the
informal conversations that people have with friends and family will also give them sophisticated
information (Downs, 1957). And of course, many, though not all, persons who have these kinds of
work tend to receive high-quality education as preparation and so they receive high-quality free
information from education, work, colleagues, family, and friends.
We can contrast this situation with the work of a typical non-unionized dishwasher, waiter, or
any other kind of unskilled labor. These persons typically receive very little in the way of in-
formation about law or politics at work. It is simply unnecessary to their jobs. And to the extent that
their co-workers are similarly placed, they receive little from these informal conversations as well.
Furthermore, if the members of their families and their friends are similarly situated, they will tend
normally to learn little about politics from these conversations beyond the bluntest information.
Because of this, these citizens are more vulnerable to manipulative and emotional appeals regarding
the political system than other citizens.
Four remarks help clarify the above claims. One, these ideas do not imply that unskilled laborers
will never be well informed or sophisticated or that every professional will be well informed. They
only suggest broad tendencies in sophistication. Two, these ideas do not suggest in any way that the
less sophisticated are less capable of becoming sophisticated. The effects I am discussing are the result
of persons occupying different positions in the division of labor in society. Economists since Adam
Smith have observed that different positions in the division of labor can have large effects on the
intellectual activities of persons, without any suggestion that this results from differential natural
talents (Smith, 2003). Indeed, we will consider some changes to the division of labor that might alter
these broad tendencies. Third, these workers are likely to have an appreciation of their needs, the
above claim concerns the relationship between their appreciation of their needs and the political
system. Finally, these observations do not imply that the unsophisticated cannot advance their in-
terests in the political system or advance their concerns. They can still advance their concerns by
voting on the basis of party identification and other cues. But these kinds of cues are more blunt than
more refined knowledge of policy.
There is another way that differences in wealth can influence sophistication. The cost of ac-
quiring information is lower for the wealthy than for the poor because wealth has diminishing
marginal utility and so the same expenditure of wealth on more information is more costly for a poor
person than an affluent person.
Empirical evidence does suggest that the wealthy are better informed and more likely to parti-
cipate than the worse off (Erickson, 2015). And there is empirical evidence that politicians are more
responsive to the wealthy than other citizens (Bartels, 2008; Gilens, 2012). So, the model I am
suggesting for how people become sophisticated and the effects of sophistication is partly confirmed.
But there is other evidence that also supports this picture. There is substantial evidence that
members of unions tend to be significantly better informed about politics than persons in the same
jobs who are not members of unions (MacDonald, 2019; Kim and Margalit, 2017). And there is
422
evidence that politicians tend to be more responsive to lower middle class and working-class voters
in districts that have high union density (Becher and Stegmuller, 2020).
This is supporting evidence because unionization involves a change in the division of labor.
Workers have power over workplace conditions, wages, and other aspects of company policy. This
division of labor requires workers to be more aware of both the law governing labor and the political
forces shaping that law. For example, the presence of unions enhances the implementation of oc-
cupational health and safety law in a firm because the union aggregates the information, and each is
more knowledgeable and able to invoke it (Donado and Walde, 2012). They must also be more
aware of the legal environment in which the firm operates and of the political forces that shape
that law.
6 The Division of Labor Among Citizens

The proposed solutions to citizen ignorance are meant to be compatible with the idea that citizens
themselves still make use of a division of labor regarding knowledge. It is not necessary that each
person know all there is to know about candidates, policy, and society or even more than a fraction
of this. Indeed, it is not desirable, since each person needs to take care of the other aspects of their
lives. If citizens find themselves within reasonably well-functioning networks of persons, the
knowledge can be possessed in a way that is distributed across citizens. For example, I can have some
informed views about a candidate’s foreign policy while my friends or co-workers have informed
views about economic policy. If they share the basic values with me and I have some reason to trust
the networks from which they are getting information, I can mostly defer to them on these matters.
In a well-functioning set of political parties and secondary associations, knowledge can be distributed
across many people and at the same time serve the aims and interests of each individual (Christiano,
2013). This is a common feature of modern life (Hutchins, 1995). In economic life, knowledge
about our individual lives is distributed among doctors, mechanics, tax accountants, government
officials, and many others. The big question is whether the institutions that structure these relations
do a good job.
7 Equality in the Transmission of Information

So far, we have discussed the conditions under which citizens can acquire the kind of sophistication
that makes the political system responsive to them. Democracy involves, in addition, large-scale
deliberation among citizens. Election campaigns, secondary associations such as unions, churches,
social movements, and chambers of commerce are all devoted to informing citizens and persuading
them of the importance of their aims and interests. Sophistication is essential to this process.
Sophisticated citizens can make their interests and concerns heard by other citizens and they are able
to listen to the concerns of other citizens and reflect on them. But sophistication is not enough.
Citizens must be guided by the principle of equality in democracy to take seriously the expressed
concerns of other citizens, no matter how much they disagree. They must be willing to take each
other seriously and respond to each other’s arguments. They must be able to fit the interests of
fellow citizens into a larger picture of the common good and justice. Some of this can be learned in
the kinds of workplace contexts, for example in democratic unions, in which people have learned to
discuss and respond to contrary views on matters of common concern (Pateman, 1970). But an ethos
that enables people to listen to each other from very different walks of life is more difficult to
develop and maintain.
Another dimension of equality is in the distribution of opportunities to persuade and influence
fellow citizens in the collective decision-making process. The basic principle of justice here is that
the only things that ought to give some people more influence than others are the greater
423
Thomas Christiano
persuasiveness of their arguments and the greater willingness to participate politically. This principle
combines the need for robust debate and deliberation in democracy with the demand for equality.
One implication of this principle is that differences in wealth and income ought not to have a
significant impact on the processes of influence. And differences in gender, race, or religion ought
not to have significant impacts.
To the extent that election campaigns are privately financed, the affluent play a dominant role in
the processes of elections, out of proportion to their position in society. This is not trivial, as I
mentioned above, because the affluent have distinct preferences from the rest of society. They thus
benefit from unequal opportunities to influence the processes of democratic decision-making. And
to the extent that we can expect their interests to influence the kinds of information transmission
they subsidize, we have a reasonably clear way in which the electoral system seems publicly to favor
their interests. No doubt similar mechanisms favor white people over black people in the political
system as well as other minorities. The usual mechanisms for rectifying these imbalances are public
financing of elections or alternatively a ban on television advertising, which absorbs most of the cost
of campaigning (Christiano, 2012).
More broadly, one might wonder about the longer-term impact of inequality of wealth on the
transmission of information and the processes of persuasion. The major media are privately owned.
And it is generally much easier for the wealthy to communicate their views than it is for others. This
must have a significant impact on the long-run determination of opinion. Here too, one might
advocate for a strengthening of the kinds of associations that tend to support the interests of worse
off persons in the society. Again, unions have often been mentioned in this context (O’Neill and
White, 2018). Another proposal is to give vouchers for speech to every individual to spend ex-
clusively on the associations of their choice for broadcasting opinion (Ackerman and Ayres, 2004).
This goes some way towards disconnecting the ability to express opinions from the distribution of
wealth. It does not limit freedom of expression because it supplements the process of expression.
8 The Interest in Free Information and the Right to Participation

The foregoing discussion suggests that there are important social conditions for an effective right to
political participation. We discussed the importance of education, workplace conditions, and the
wider network of persons in which a person finds herself for enabling sophisticated participation.
Here I want to discuss the normative implications of these findings. The questions to be answered
are whether and how the right to participate in politics is linked up with other rights that might
support the right to participation.
The right to participate as an equal in politics contains two distinct elements: brute power rights
to vote, express oneself, associate with others and run for office in free and fair elections; and
informational power rights which include rights to the capacity to become a sophisticated voter and
the capacity to transmit information. Both of these are essential to equal participation.
The realization of informational power rights depends on social conditions such as education,
workplace conditions, broader civil society associations as well as the networks of persons which
might depend on these.
There is already a positive right to education, whose realization is very poorly distributed in the
United States. There are only modestly protected negative rights to join unions and there is some
fluctuating positive support for this in the United States and the United Kingdom. There are ne-
gative rights of association in liberal democratic countries, but there may be a question as to whether
these should be bolstered with some positive rights.
In those cases, such as the right to education, I think there is reason to think that the conception
of the right to participate adds some additional weight to the need to protect and promote the right
to education in an egalitarian way. Education clearly supports informational sophistication in
424
politics. Additionally, an enhanced role for persons in the economic division of labor also provides
key support to informational sophistication.
There are already many arguments for enhanced rights of participation in economic life. One,
union membership can play a significant role in decreasing inequalities of wealth and income as well
as the concomitant effect of inequality of opportunity in society (Stansbury and Summers, 2020).
Two, union membership can lessen the degree of inequality of power in the economy more directly
by lessening the differences in bargaining power between workers and employers (Freeman and
Madoff, 1984). These give us reasons for a positive right to unionization. The political sophistication
argument gives us an additional reason for thinking there is a right here.
The basic argument for this being a right is analogous with the right to education. The argument
is a kind of bootstrapping argument. The basic problem of information is that, at a certain level of
information, you don’t know what you don’t know. Without a certain level of free information, our
capacity to receive information and our ability to pursue further information are highly vulnerable to
manipulation, deception, and fruitless search for further information or alternatively we become
simply disconnected from political issues either completely or in significant degree.
The right to education is partly grounded in this kind of bootstrapping concern. We don’t expect
children to want to be educated; they don’t know what they are missing. The information provided
gives them a base from which to expand their understanding. The argument for enhancing power in
the workplace and in economic life more generally has a similar structure. It provides sophisticated
information about politics and law, which people may not initially think of as important. It provides
a site in which discussion and argument about important collective matters can happen and so
provides a sense that there is something to be learned from others who disagree. Without this base,
many citizens are vulnerable to manipulation, deception, or disconnection from politics to some
degree. The problem is that without political sophistication one is vulnerable to manipulation by
clever political entrepreneurs, or one is likely simply to treat the information as not very meaningful.
This is a recipe for creating a highly unresponsive representative system for those persons. And hence
it is a recipe for disempowering a person in a representative system. Hence, I think we can argue that
there is a deep interest in economic arrangements in which free information about politics and law is
quite pervasive. This deep interest gives an additional grounding for a right to participate in the
running of economic firms that complements the other grounds.
But there is more to be said about these rights to free information. Though they ought to be
conceived in part as negative rights, they must also be thought of as positive rights. The boot-
strapping character of free information implies that a person does not procure it for herself. Her
reception of free information is an external effect of her relations with other people, in the sense that
she does not seek it out. It provides a foundation on which to seek out further information and
refine one’s understanding. If the networks are not very good, the likelihood is that she does not
receive the free information that will enable her to advance her interests or her sense of justice.
This bootstrapping argument defeats a potential argument for keeping the right to free in-
formation a merely negative right. One argument for a merely negative right to some good is that
the person is responsible for acquiring that good. And we might think that a person must be held
responsible for the acquisition of free information. But the bootstrapping argument shows that one
cannot hold persons in general responsible for the failure to acquire high-quality free information
when it is normally an unsought byproduct of the network of relations one finds oneself in. Hence
the responsibility argument in favor of a negative right only cannot succeed, at least with respect to
the most basic right to free information.
Yet, the interest is profound and central to the right to participate as an equal in the political
society. Hence, the interest that is central to the right to participate as an equal seems sufficient to
ground duties in the rest of the society to assist in providing the social conditions that create the free
information.
425
Thomas Christiano
Someone might object that this conception of the right to participation breaches the proper
autonomy that the economy and civil society ought to have from the political system. To be sure,
the modern welfare state intervenes quite substantially in the economy for the sake of economic
growth and equality and so breaches the autonomy of the economy quite regularly. I am proposing a
further breach that involves intervention in the economy for the sake of greater political equality,
which imposes requirements on the economic division of labor. This view allows for substantial
autonomy; it is not a proposal for state control of the economy. I say we can justify the limitation on
civil society on the already noted grounds that there is independent justification for enhancing the
power of workers in the economy and because such enhancement enables citizens to participate as
equals. So, though there is some breach, it is one that converges with an already widely accepted
form of state intervention.
References
Acemoglu, D., Naidu, S., Restrepo, P. and Robinson, J. (2019) “Democracy Does Cause Growth,” Journal of
Political Economy, 127, 1.
Achen, C. and Bartels, L. (2016) Democracy for Realists: Why Elections Do Not Produce Responsive Government,
Princeton; Princeton University Press.
Ackerman, B. and Ayres, I. (2004) Voting with Dollars: A New Paradigm for Campaign Finance, New Haven, CT;
Yale University Press.
Bajaj, S. (2021) “The Weight of Fairness,” Politics, Philosophy and Economics, 18, 4, 2019.
Barnett, Z. (2020) “Why You Should Vote to Change the Outcome,” Philosophy and Public Affairs, 48, 4.
Bartels, L. (2008) Unequal Democracy, Princeton; Princeton University Press.
Becher, M. and Stegmuller, D. (2020) “Reducing Unequal Representation: The Impact of Labor Unions on
Legislative Responsiveness in the US Congress,” Perspectives on Politics, July.
Bowles, Samuel. (2016) The Moral Economy: Why Good Incentives Are No Substitute for Citizens, New Haven, CT;
Yale University Press.
Christiano, T. (2011) “An Instrumental Argument for a Human Right to Democracy,” Philosophy and Public
Affairs, Spring.
Christiano, T. (2012) “Money in Politics,” David Estlund Ed. Oxford Handbook of Political Philosophy, Oxford;
Christiano, T. (2013) “Rational Deliberation among Experts and Citizens.” In Jane Mansbridge and John
Parkinson Ed. Deliberative Systems: Deliberative Democracy at the Large Scale, Cambridge; Cambridge
University Press.
Cicero (1991) On Duties, Cambridge; Cambridge University Press.
Dallison, P. (2017) “UK election cost 140 million pounds,” POLITICO, September 17.
Donado, Alejandro and Walde, Klaus (2012) “How Trade Unions Increase Welfare,” The Economic Journal,
122, 563.
Downs, A. (1957) An Economic Theory of Democracy, New York; Harper and Row.
Erickson, R. S. (2015) “Income Inequality and Policy Responsiveness,” Annual Reviews of Political Science, 18,
11–29.
Fehr, E. and Fischbacker, U. (2005) “The Economics of Strong Reciprocity.” In Hebert Gintis, Samuel
Bowles, Robert Boyd and Ernest Fehr Ed. Moral Sentiments and Material Interests: The Foundations of
Cooperation in Economic Life, Cambridge, MA; MIT Press.
Freeman, R. and Madoff (1984) What Do Unions Do?, New York; Basic Books.
Gilens, M. (2012) Affluence and Influence, Princeton; Princeton University Press.
Grotius, H. (2012) Hugo Grotius on the Law of War and Peace. Student Edition. Stephen C. Neff. Cambridge;
Hubler, K. O. and Underhil, W. (2018) “Election Costs: Who Pays and with Which Funds?” National
Conference of State Legislatures, 26, 1 March 2018. https://www.ncsl.org/research/elections-and-campaigns/
election-costs-who-pays-and-with-which-funds.aspx.
Hutchins, E. (1995) Cognition in the Wild, Cambridge, MA; MIT Press.
Kant, I. (2012) Groundwork of the Metaphysics of Morals. Mary Gregor Ed. Cambridge; Cambridge University
Press.
426
Kim, S. E. and Margalit, Y. (2017) “Informed Preferences? The Impact of Unions on worker Policy Views,”
American Journal of Political Science, July.
MacDonald, D. (2019) “How Labor Unions Increase Political Knowledge: Evidence from the United States,”
Political Behavior, April.
Maskivker, J. (2019) The Duty to Vote, Oxford; Oxford University Press.
O’Neill, M. and White, S. (2018) “Trade Unions and Political Equality.” In Collins, Hugh, Gillian Lester,
Victoria Montavalou Ed. Philosophical Foundations of Labour Law, Oxford; Oxford University Press.
Pateman, C. (1970) Participation and Democratic Theory, Cambridge; Cambridge University Press.
Roemer, J. (2019) How We Cooperate: A Theory of Kantian Optimization, New Haven, CT; Yale University
Press.
Rosenfeld, J. (2014) What Unions No Longer Do, Princeton; Princeton University Press.
Rousseau, J-J. (1987) On the Social Contract. Donald Cress Ed. Indianapolis, IN; Hackett Pub.
Schumpeter, J. (1950) Capitalism, Socialism and Democracy, New York; Harper and Row.
Sen, A. (1977) “Rational Fools: A Critique of the Behavioral Foundations of Economic Theory,” Philosophy
and Public Affairs, Summer.
Smith, A. (2003) [1776]) The Wealth of Nations, New York; Bantam Dell.
Somin, I. (2014) Democracy and Political Ignorance: Why Smaller Government Is Smarter, Princeton; Princeton
University Press.
Sorauf, F. (1994) Inside Campaign Finance: Myths and Realities. New Haven, CT; Yale University Press.
Stansbury, A. and Summers, L. (2020) “The Declining Worker Power Hypothesis: An Explanation for the
Recent Evolution of the American Economy,” NBER Working Paper Series, 27193.
Wright, G. (2013) Sharing the Prize: The Economics of the Civil Rights Revolution in the American South, Cambridge,
MA; Harvard University Press.
427
34
WHAT, IF ANYTHING, CAN
JUSTIFY LIMITING WORKERS’
VOICE?
Lisa Herzog
1 Introduction
In most countries, citizens take it for granted that in the workplace, they have no right to a voice. They
accept that workplace relations are hierarchical: bosses tell employees what to do, and all that employees
can do if they disagree is to leave. However, for many individuals leaving is not a realistic option: they
need an income and looking for a new job would cost more time and money than they have. So instead
of speaking up, they comply with the conditions of employment they are offered. But is this justifiable?
Why do we think that power needs to be held accountable in politics, but not in the economy? Why
do the citizens of so many countries accept that workers, by and large, have no voice?1
Take the following thought experiment: imagine that you live in the year 2500, in a world in
which work has been fully democratized.2 Might these future generations look back upon the current
hierarchical structures of workplaces with the same kind of moral horror with which we, today, look
back upon slavery or feudalism? After all, these practices also seemed natural, and without alternatives,
to their contemporaries. Later generations, in hindsight, understood how being deeply unjust they
were. Contemporaries had sophisticated theories that justified these practices, even though, in
hindsight, it is clear that these were mere ideological smokescreens. What about the theories that
justify hierarchical workplace relations today, might they be similar kinds of smokescreens?
The debate about how a just society should arrange workplaces has picked up speed in recent
years, but it has a much longer history, going back at least to the 19th century, for example, in the
work of John Stuart Mill (2008 [1848/1873]). Since the beginnings of the Industrial Revolution, the
question about a right to voice for workers, inside and outside of firms, has been debated by theorists
and activists. After intense discussions in the 1960s and 1970s, the debate died down a bit, but it saw
a revival after the 2008 Financial Crisis (Pausch, 2013). Many capitalist countries have gone through
long periods of stagnating wages and skyrocketing material inequality, which raises serious questions
about the ability of the current economic system to serve “the many” instead of “the few” (cf. the
title of Reich, 2015). There are thus serious questions about the feasibility of alternative models, and
about how the economic system could be brought better in line with democratic principles (see also
Kelly and Howard, 2019). Considerations about democratizing the workplace, and giving workers
the right to a voice, are among such proposals.3
Various lines of arguments in favor of giving workers a voice can be distinguished (see Frega
et al., 2019, for an overview): arguments about the analogy between states and firms, and hence the
applicability of arguments in favor of democracy to firms (e.g., Dahl, 1985; McMahon, 1994, 2013;
428 DOI: 10.4324/9780367808983-41

What Can Justify Limiting Workers’ Voice?
Landemore and Ferreras, 2016); arguments about meaningful work (e.g., Schwartz, 1982; Yeoman,
2014); arguments about the avoidance of domination in the workplace (e.g., Anderson, 2017;
Breen, 2015; González‐Ricoy, 2014; Néron, 2015; Gourevitch, 2015); or arguments about positive
spillover effects towards political democracy (e.g., Pateman, 1970; Gould, 1988; Mason, 1982). In
addition, the digital transformation of might exacerbate current problems, for example, by making
even more work precarious, but it might also offer opportunities for democratization, for example
by lowering the costs for participatory decision-making through online tools (see Herzog, 2019).
In this chapter, I approach the question of workplace relation from a reversed perspective: instead
of asking what might speak in favor of structuring workplaces differently than is currently the case, I
try to disentangle which of the arguments that speak against such proposals are worth taking ser-
iously, and what this means for the institutional arrangements we should aim at.4 I am less interested
in defending an abstract moral right to workplace democracy (see, e.g., Dahl, 1985; Mayer, 2000),
but rather in an all-things-considered, political judgment about the kind of economic system we
ought to have. This is a typical “PPE” question, in the sense that philosophical, political, and
economic considerations need to be brought together. It is also a topic for which both normative
and empirical arguments are relevant, and an additional aim of this chapter is to create a clearer sense
of their relation; one of the challenges being, of course, that absent fully democratic economies, we
can only use empirical evidence drawn from certain partial models.
The three regimes that will serve as my points of comparison are a regulated market economy with
capitalist firms, workplace republicanism, and workplace democracy. By the first category, I mean a
system in which labor markets are regulated by democratic politics from the outside, but firms are
internally controlled by capital owners alone.5 By “workplace republicanism,” a term coined by Hsieh
(see in particular Hsieh, 2005), I mean a system in which workers are protected from the exercise of
arbitrary power and have some rights to contest decisions, but no democratic rights to control firms.
Co-determination, as it exists in some European countries, is a real-life instantiation of workplace
republicanism (see, e.g., Page, 2011, on Germany). By workplace democracy, I mean models in which
employees have ultimate control over companies (see similarly Dow, 2003, 5), even though capital
owners may also have some rights. Workplace democracy can take on different forms, for example, as
cooperatives (where workers are also owners) or in the form of bicameral models in which both
employees and capital owners elect representatives into two governing bodies (Ferreras, 2017).
I use a regulated market economy, instead of the current version of relatively unregulated ca-
pitalism, in order to compare these regimes on a fair basis. It is a danger in many PPE discussions that
one compares proposals with different degrees of idealization, for example, current forms of ca-
pitalism with an idealized version of workplace democracy, or non-ideal versions of workplace
democracy with idealized models of capitalism. For an honest discussion, we need to shift all regimes
to a comparable level of idealization.
I divide my discussion into arguments from rights (2) and arguments from consequences (3),
thereby following a distinction that Sen (1985) has introduced for thinking about markets. For
reasons of space, I bracket the question of transition, that is, how countries could move towards
different systems. I conclude by briefly reflecting on what one should not expect from republican or
democratic firms, emphasizing that it should not be understood as a freestanding proposal, but as
proceeding in lockstep with other reforms (IV).
2 Arguments From Rights
2.1 Arguments From Property Rights

A first candidate for how to justify capitalist firms, in contradistinction to republican or democratic
workplaces, is to draw on property rights. The argument is simple: it holds that the owners of a company,
429
Lisa Herzog
qua owners, have the right to control what is going on within the company.6 Note, first, that if this
argument were valid, it would present no problems for cooperatives, because these are owned by
workers themselves. Some defenders of workplace democracy have indeed favored cooperatives for that
very reason (e.g., Gonzalez-Ricoy, 2020). Other models of republican or democratic workplaces, in
contrast, could be justifiably rejected if this simple argument from ownership were valid.
But is this argument valid? No! Ownership provides control over things, it does not automatically
include the right to tell others what to do with that property (see in particular McMahon, 1994, 15,
and passim). Owners may have a right to certain returns, and this may best be realized by giving them
some say about what happens in a company, but that does not imply that they have an exclusive right to
control (see also Dow, 2003, 4). According to a widely shared theory of ownership, first proposed by
Honoré (1961), our traditional notion of ownership consists of a set of rights, which are not always
bundled together in the same way – think, for example, about ownership of a listed building, where
the right to modification is strictly limited. Ownership rights in companies could be cut out differently
than they currently are, to give workers more rights (though investors would also retain some rights). If
a critic objects to that line of argument by holding that company owners must have a right to ownership
that implies full control, she is simply begging the question. An independent argument is needed to show
why that bundle of rights for owners, rather than a different one, would be justified.
Another point is worth noting in this context. The argument from property rights may sound
most convincing when we imagine a small business owner who is personally liable when the
company goes bankrupt. It may seem plausible to say that in such a constellation, there are strong
reasons to give the owner some say in what happens to the company. Without such control, the
economic liberties of small business owners might indeed be restricted too much. But note that such
a right is compatible with workers having some rights of contestation, along the lines suggested by
defenders of workplace republicanism, and maybe also with some forms of co-determination in
which power is shared.
However, many businesses are not these kinds of small family businesses. Instead, they are legal
constructs with limited liability, in which the losses investors might incur are strictly limited to their
shares, and do not touch their private fortunes. This construction is only possible by a legal privilege:
the act of incorporation of the business as a legal person. Critics have in fact argued that the rights
given to corporations, as legal persons, have gone too far, sacrificing the common good for the sake
of private interests (see, e.g., Ciepley, 2013). While I cannot here enter the debate about the nature
of the corporation, it is not clear why the rights of financial investors – which are not prototypical
ownership rights anyway – should not be constructed in a way that allocates more rights to workers.
2.2 Arguments From Contractual Rights

A more plausible candidate for rights-based limitations on workers’ voice – and indeed the most
important element of the current legal justifications of hierarchical workplaces – are contractual rights.
Workers, this argument holds, should have the freedom to enter into contracts with employers, in
which they agree to accept hierarchical working conditions in exchange for receiving a salary. This
looks like a standard liberal argument: liberal states treat their citizens as grownups, who can make
their own choices and conclude contracts as they wish. Some critics (in particular Ellerman, see, e.g.,
1992) hold that any contract in which individuals hire out their work to non-democratic firms turns
them into non-persons, and should therefore be considered illegitimate. But that claim seems too
strong: why shouldn’t individuals agree to rent out their work to others, as long as there are adequate
protections against violations of their interests and their dignity? Do they necessarily lose their equal
moral standing when doing so? This argument seems to get the phenomenology of labor markets
wrong (at least when these are part of somewhat regulated economies, maybe with some elements of
workplace republicanism): employees do not turn into non-persons simply by becoming employees.
430
They continue to think for themselves, and often retain considerable amounts of autonomy (for
discussions see e.g., Dow, 2003, 33, or Mayer, 2000, 315–8).
The key question with regard to the contractual argument, however, is whether individuals
actually do freely consent to labor contracts, or whether they do so for lack of alternatives. This free
consent, not the mere fact of someone signing a contract, carries normative weight. This is of
particular importance for employment contracts, which differ from other market contracts.
Economists describe them as “incomplete” contracts: instead of anticipating all tasks that the em-
ployee will undertake, he or she agrees to obey orders, without knowing exactly what he or she will
be told to do (Williamson, 1975). Why would individuals consent to such contracts? Many critics of
capitalist firms suspect that they only do so because they need an income and have no reasonable
alternative to signing up for wage labor.7
Many defenders of capitalist companies, in contrast, insist that such alternatives do exist, for
example in the form of self-employment or by working for another company. But are these claims
about exit options realistic? Modern economies rely on complex forms of divided labor that can only
function when these are integrated into collective workplaces; in many occupations, self-
employment is therefore simply not an option.8 For most individuals, it is, therefore, more realistic
to move to a different job. But this can carry considerable (material and immaterial) costs (see, e.g.,
González‐Ricoy, 2014; Frega et al., 2019). Moreover, individuals would often not have a choice to
move to a republican or democratic company, but only to a different capitalist one. It seems that
they should at least have the option to work in republican or democratic structures (see also Jacob and
Neuhäuser, 2018). If they had such an option, one could see their choice to work for a hierarchical
company, for example, in exchange for a higher wage, as a genuine expression of free choice – if
not, this is a problematic assumption.
While the argument for free consent seems strong, it is important to note that it would require
massive changes compared to the status quo to make sure that workers really have a free choice
between differently structured workplaces. One strategy, which has been suggested in particular by
Taylor (2017), would be to strengthen workers’ exit rights. He mentions policies such as “relocation
vouchers,” “demogrants,” or an unconditional basic income (2017, 53–4; see also Widerquist, 2013
on the latter). There are no empirical cases of countries having installed such mechanisms on a large
scale, so we do not know how exactly these mechanisms would work, and what kind of macro-
economic dynamics they would create. Given that companies exist for individuals to work together,
one wonders whether putting so much emphasis on exit rights is a viable strategy. At some point,
most workers would presumably have to re-enter a workplace, and then the question of a positive
vision of how work should be organized, rather than an argument about the possibility to leave
certain organizations, comes up again. But of course, exit rights can take effect long before workers
exercise them because they have an anticipatory effect on how companies treat workers. So this line
of argument is not without promise.
If we could assume that the choice to work for a capitalist company was indeed freely consented
to, that is, that workers give up their voice voluntarily, this would indeed be a weighty argument.
But would it be the end of the discussion? Two lines are worth exploring further. One is the
argument that individuals should not be allowed to enter all kinds of contracts, especially not
contracts that put their basic interests at risk. For example, contracts to sell oneself into slavery are
invalid in most countries, and rightly so. Such an argument can be understood as being based on the
inalienability of certain rights, or it can be understood as protecting individuals, in a paternalistic
fashion, from violating their own interests. This type of argument may seem too far-fetched when it
comes to selling one’s work to non-democratic firms in general. But it is quite plausible when it
comes to contracts that do not include any basic protection for workers’ interests, together with
some rights to contestation to ensure that the rights will indeed be protected. Thus, one might say
that ceteris paribus, workers should be free to choose between republican and democratic companies
431
Lisa Herzog
but should be paternalistically protected from contracts in which they give up all rights to con-
testation because this creates too great a risk that their basic interests will be violated.9
A second line of argument is more complicated and involves empirical considerations concerning
the effects of the interactions between individuals. One could imagine a scenario in which the
exercise of the right of some to work in non-democratic firms would jeopardize the right of others to
freely choose between democratic and non-democratic firms. Such a scenario could happen if, for
example, there are interdependencies between firms and there are competitive pressures – in
economies in which some but not all firms are capitalist – on other firms to turn capitalist as well
(see, e.g., Jacob and Neuhäuser, 2018). Here, however, more empirical input would be needed to
better understand whether such a scenario is likely. If so, the rights of different individuals would
mutually affect each other, and it is not clear a priori how to adjudicate between them. On the other
hand, if a peaceful co-existence of different workplace regimes was feasible, then the central premise
of this argument would disappear.
To summarize: of the various rights-based arguments that might justify limiting workers’ voice
that has been put forward in the debate, the most promising one is an argument from contractual
rights under conditions of meaningful consent. It is far more plausible for republican workplaces – in
which individuals’ basic interests are protected – than for capitalist firms, and it would justify ex-
emptions from the imperative to make all workplaces fully democratic. But it would require creating
conditions in which employees really have meaningful alternatives so that their consent to a capitalist
labor contract is truly free.
3 Arguments From Consequences

In addition to arguments from rights, arguments for limiting workers’ right to voice might come
from consequences. Here, we face the challenge that the empirical evidence on which we can base
our predictions is rather incomplete. There are many possible effects of republican and democratic
workplaces, which economists have described in theoretical models, but it is less clear what the actual
effects are (see similarly Dow, 2003, 8–9). In what follows, I will, where available, draw on empirical
results from companies with co-determination and worker representation in Europe, and from the
cooperative sector in various countries. They provide the best estimates of what to expect if workers
were given full (democratic) or partial (republican) voice.
3.1 Arguments From Consequences for Workers

The first set of consequences concerns the direct effects on the welfare of workers employed in
capitalist, republican, or democratic companies. Here, we can distinguish various dimensions. One is
the protection of workers’ interests within companies. In a well-regulated market economy –
which, as mentioned earlier, should be our point of comparison – these are secured by law, for
example, by occupational health and safety standards. However, such laws need to be enforced, and
outside enforcement can be costly and intrusive. Hence, republican or democratic structures might
be helpful on the ground, simply for making sure that companies stick to existing laws. For that
aspect, however, fully democratic structures might not be necessary, because the control function of
workplace republicanism might suffice. In any case, it seems that these kinds of workers’ interests are
better secured in companies in which workers have more voice, so finding an argument about limiting
workers’ voice here would be difficult.
A second, much-debated question is whether workers would have to face lower wages because of
the alleged lower productivity of non-capitalist firms. Many theoretical arguments have been
brought forward concerning, for example, the lower inclination towards capital-intense innovations
in democratic firms (Bowles and Gintis, 1993). But it is not clear whether this phenomenon can be
432
observed in reality; below, I will discuss the question of productivity in more detail. There are also
some studies that look directly at wages. Studies about mixed economic systems, in which some
companies have work councils and others do not, found higher wages in firms with work councils
10
(Addison et al., 2001; Hübler and Jirjahn, 2003). Burdin and Pérotin (2019), in a study that took
advantage of the quasi-experiment of the introduction of employee representation in several
European countries, also found positive effects on non-material dimensions of work, such as
working-time flexibility. Another point worth mentioning is job security: evidence about co-
operatives shows that they have a higher tendency than conventional firms to keep workers em-
ployed in recessions (see, e.g., Olsen, 2013, 95; Péroti, 2016). Thus, this would be another effect on
the welfare of workers that speaks in favor of non-capitalist firms.
In terms of direct negative consequences for individual welfare, one might think about two other
possible effects. One concerns the additional burdens of democratic self-governance, for example,
the time needed for deliberation, or the emotional turmoil if two candidates campaign for leadership
in a company. However, whether or not this is an additional burden – compared to the burdens that
are part of all organizational life – depends a lot on how the republican or democratic structures are
organized and managed. Badly designed republican or democratic companies might indeed create
such problems, but it is not clear that this would be a necessary effect. Moreover, digital tools might
help to lower transaction costs and to secure transparency, so that the additional burden, if there is
any, could be further reduced. Last, even if there were some additional burdens, these might be
justified by the overall advantages of republican and democratic firms.
Second, Taylor (2017, 7, 58–9) sees a risk that giving workers more voice might increase
domination.11 He worries that bodies such as works councils or unions “can and inevitably will”
misuse their power “in the pursuit of private ends, be they financial (e.g., bribery), tribal (e.g.,
bureaucratic-class interests), or ideological” (2017, 7). However, he provides no argument for why
this should “inevitabl[y]” be the case. To be sure, control mechanisms are needed – that holds for
any collective body that has any form of power in a democratic society. Republican and democratic
firms would have the same protection of basic rights as capitalist firms in a well-regulated market
economy, and these would hold both against domination by capital owners and by other institu-
tions. One can also acknowledge that combining voice and the possibility of exit – which Taylor
emphasizes so much – is often the best policy; this would imply that one should make sure that the
labor market remains sufficiently flexible in a system of democratic work. But Taylor seems to
commit the error mentioned in the introduction, of contrasting an idealized model of “exit” with a
very non-ideal version of “voice,” which leads to unjustified conclusions.
So, the consequences for individual workers seem, overall, positive – no argument for limiting
workers’ voice can be derived from them. But there might be other reasons for caution. As
Dow rightly notes, “the interests of society as a whole may demand that other goals take priority”
(2003, 42). Therefore, we need to explore further whether there are any broader consequences of
workplace republicanism or workplace democracy that would justify limiting workers’ voices.
3.2 Arguments for Consequences for Society

One candidate for such a societal interest might be overall productivity: what if republican or
democratic firms would exhibit systematically lower productivity? Many theoretical considerations
turned around this question, suggesting mechanisms such as lower incentives for monitoring em-
ployee performance (Alchian and Demsetz, 1972), higher transaction costs within firms and higher
costs of capital (Hansman, 1996), or a lower willingness to take risks (e.g., Bowles and Gintis, 1996;
for discussions, see Dow, 2003). It is, however, not clear whether these factors affect only demo-
cratic workplaces, or affect them to a greater degree than capitalist ones. One can just as well adduce
theoretical considerations that would suggest higher degrees of productivity in democratic firms. For
433
Lisa Herzog
example, employees might be more motivated and more willing to comply with the decisions,
because they see them as more legitimate. Companies that allow for more voice might also make
better use of workers’ knowledge (Gerlsbeck and Herzog, 2020). Such effects might compensate for,
or even outweigh, possible negative effects. From a theoretical perspective, it, therefore, remains
unclear whether non-capitalist enterprises really are less productive.
The empirical evidence does not support this claim either. For example, once established, co-
operatives seem to have lower failure rates than conventional firms, which suggests that they suffer
no productivity disadvantage (Dow, 2003, 184; Penceval, 2012; Olsen, 2013).12 As one recent
review of the literature puts it, cooperatives are more productive than conventional firms, with “staff
working ‘better and smarter’ and production organized more efficiently” (Pérotin, 2016, 3). Similar
results can be found in studies of companies with worker councils. In a summary of research about
German workers’ councils, Addison et al. (2004) distinguish three phases of research. In the first
phase, some “rather alarming adverse consequences” of worker councils on productivity were
discovered, but in the second and third phase, inconclusive and even positive effects of work
councils on productivity were found (2004, 276). Addison et al. warn that there are methodological
challenges in separating the effects of workers’ councils from other effects, but they acknowledge
that the initial negative findings were probably overblown, or alternatively, might have been due to
initial problems that were overcome through learning processes within companies (2004, 276).
However, let us assume, for the sake of argument, that the productivity of republican or de-
mocratic firms would indeed be somewhat lower, and that there might thus be a negative effect on
productivity if more and more companies turned republican or democratic. Would this actually
be a problem? Thinking about this question leads to the question of who would benefit from higher
productivity – which is, after all, standardly measured in relation to capital input. If, counterfactually,
capital ownership were spread evenly in society, for example, along the lines of a “property-owning
democracy” (e.g., Thomas, 2016), then arguments about declining productivity would have
plausibility, because all members of society would suffer from it. Similarly, this argument would be
worth taking seriously if there were indirect benefits of high capital productivity to society as a
whole, for example, via the taxation of capital income or via more innovative products.
Thus, the question really is who benefits from “higher productivity” – society as a whole, or only
capital owners? The more capital is concentrated in the hands of the few, and the less it contributes
to society through taxation and other positive effects, the more arguments about productivity turn
into thinly veil capitalist ideology: all they boil down to is the fact that a few capital owners want to
keep their privileges.
If one assumes, however, that the market economy is well-regulated, such that productivity gains
do indeed benefit the society as a whole, directly or indirectly, and if one assumes (against the
evidence!) that capitalist firms were more productive – would it then be worthwhile to keep ca-
pitalist companies in place in order to achieve higher overall productivity? The answer to this
question depends to some extent on the overall level of welfare a society has already achieved – if
basic needs are not met, then increasing productivity is of great importance.
One worry in particular deserves attention, especially in poorer countries: the incentives for
investment and for starting new companies. New companies, which replace older ones with out-
dated technologies or organizational structures, are often taken to be a key driver of the overall
productivity of an economy. Now, it is not clear whether republican or democratic companies really
do invest at lower rates, as predicted by theoretical models. Addison et al. (2007) found neither a
positive nor negative effect of works councils on investment in German companies (see also Pérotin,
2016, 17–8, on cooperatives not necessarily being less capital intensive than conventional firms). But
what seems to be true is that the rate of newly started democratic firms is rather low, a phenomenon
for which various possible reasons have been adduced, such as credit constraints or collective action
problems (Olsen, 2013, 99–101; Dow, 2003, 210–2). Shouldn’t one allow resourceful individuals to
434
start new businesses, and give them exclusive control rights, at least for the first few years, so that
they can realize their ideas?
Maybe entrepreneurs do indeed need exclusive control rights, and exclusive access to profits, as a
motive for starting new companies (or maybe this is a cultural prejudice of our capitalist era, and
entrepreneurs would, in a different kind of society, just as well start companies that are republican or
democratic from the start, motivated by the drive to realize their ideas or to receive public re-
cognition). If so, this might be a reason to allow the establishment of non-democratic companies.
Maybe they could be turned republican, or democratic, after having grown to a certain size – after
all, the structures of companies need to become more formal when they become larger. Note,
however, that this argument only holds if the companies that are thus started really deliver societal
benefit; it is, after all, an argument from consequences, not from rights. If these capitalist start-ups
were established in an environment in which there are sufficient numbers of alternative, democratic
or at least republican, job companies, and hence reasonable exit options for the employees of ca-
pitalist startups, that might be a defensible compromise.
For such a mixed model to function, the question of transition would be crucial: at what point
should companies adopt republican or democratic structures? Should this be a voluntary or a man-
datory move, or something in between (e.g., with tax incentives for a transformation)? As Olsen notes
(2013, 101–2), more research is needed in order to understand why we do not already now see more
conversions of conventional companies into democratic ones. If this puzzle were solved, one could
think about how to address the obstacles to voluntary conversions. If more and more companies
became republican or democratic, more evidence about the effects, obstacles, and challenges, would
become available, and other companies could learn from that evidence. Social norms and expectations
would probably also change over time, with democratic work becoming the new normal. This would
be the kind of transition that John Stuart Mill had hoped for in the 19th century. It has not happened
so far, but that does not mean that might not happen in the future, if the preconditions for it were
better understood and if there were more political will to support workers’ voice.
4 Conclusion: What Not to Expect From Non-Capitalist Workplaces

As I have tried to show, when considered in detail, and in the light of existing empirical research, the
justifications for limiting workers’ voice boil down to much less than is often assumed. If given
genuine choice, and if no negative effects on other people’s rights ensue, workers should be allowed
to choose (well-regulated) capitalist over republican or democratic workplaces, unless paternalistic
reasons or arguments from the interdependence of choice come in. But this “if” requires massive
changes in the structures of our economic systems. Moreover, exceptions for innovators and en-
trepreneurs who start genuinely socially useful companies and thereby help increase productivity
seem justifiable. What this implies is that a system in which the majority of workplaces are de-
mocratic, or at least republican, is harder to reject than many people think. The best system would
probably be one in which there is quite some variety of company constitutions, with the possibility
of experimentation and social learning processes about what works best for what type of company.
But this does not mean that workplace democracy should be understood as a panacea that would,
once and for all, address all problems of our economic system (see e.g., Schweickart, 2011, for
broader reform proposals). Let me briefly mention some areas in which we should not expect
workplace democracy to help. Obviously, democratic workplaces would still be populated by
human beings whose ideas, worldviews, desires, and ambitions sometimes clash. Democratic or-
ganizations will need mechanisms for resolving conflicts, just as much as non-democratic organi-
zations do, and workplace republicanism or workplace democracy will not make that annoying
colleague who gets on everyone’s nerves disappear (but at least he or she will be less likely to become
your boss). A second set of issues that workplace democracy does not, by itself, address, are questions
435
Lisa Herzog
about the boundaries between work and nonwork and justice with regard to unpaid work; what it
can do, however, is give greater weight to workers’ interests in finding ways to combine “work”
and “life” (including unpaid work). Nor will it dissolve problems of discrimination and injustices
towards women and minorities at the workplace. To address these, additional mechanisms are
needed, but workplace republicanism and workplace democracy might help enforce anti-
discrimination legislation on the ground.
Depending on which form it takes, workplace republicanism or workplace democracy may have
larger or smaller effects on overall distributive justice and the reduction of domination in society.
One could imagine a scenario in which capital ownership in a country is highly concentrated in the
hands of a few super-rich families, but these capitalists have social-democratic leanings (or are fearful
of being dispossessed) and therefore agree to give workers more voice. This would obviously be
unsatisfactory from a perspective of distributive justice and non-domination. Workplace democracy,
on its own, might help to raise the incomes of workers, but would probably not do enough, on its
own, to create a “society of equals” (Anderson, 1999). Nor would it, by definition, answer questions
about justice towards those who, for various reasons, cannot participate in the workforce.
Moreover, there continues to be a need for political regulation and enforcement in countries
with democratic economic systems, to deal with unintended consequences or evasive reactions of
workplace republicanism or workplace democracy. For example, the experiences in European
countries with relatively well-protected workforces have shown the risk of creating a bifurcated
labor market, in which there are permanent, well-paid jobs with many benefits for some, and
precarious, low-paid jobs without any benefits for others. Obviously, such developments need to be
counteracted. If the introduction of democratic workplaces would necessarily lead to such a bi-
furcation, then one would have to seriously consider whether it is a step worth taking, from a
perspective of justice, because it would likely harm the most vulnerable members of society. But it is
not clear that this is a necessary development that could not be counteracted by other means, for
example, a job guarantee offered by the state as “employer of last resort” (e.g., Tcherneva, 2018).
All in all, workplace republicanism and workplace democracy should be seen as one element of a
whole set of progressive reforms that would re-embed the current form of cut-throat capitalism,
and, step by step, transform it into an economic system that truly serves all members of society. In
different countries, the urgency of different reforms varies, depending on what had already been
achieved and on how their labor market is structured. I take it, however, that strengthening workers’
voice is a crucial desideratum in many countries, and I venture the guess that it would also help spur
other reforms, because workers would be able to use their voice, both within companies and in the
broader public, to demand further changes. Whether or not such reforms will come forth is a
question that philosophers are not competent to answer. What they can, and should, do, however, is
to examine the arguments, and to sort the wheat from the chaff, when it comes to the assumptions
about our social and economic order that keep us trapped in the status quo.
Notes
1 There are some exceptions: some European countries have developed forms of co-determination (which are
mandatory at least for certain types of companies) that can be classified as instances of “workplace repub-
licanism,” but not fully developed “workplace democracy.”
2 Such a world could also include democratic rights for, say, the users of internet platforms. For reasons of
scope, however, I focus on workers.
3 For reasons of space, I focus on voice inside firms; a fully democratic economic system would also require
other elements, such as unions, collective bargaining, and the “primacy of politics” on the political level.
Again for reasons of space, I here cannot discuss the interrelations between these different levels.
4 These arguments, if valid, are arguments against mandating workplace democracy – not necessarily against
privileging it (e.g., by subsidies or tax reductions) as long as other options are also available.
436
5 This is compatible with granting workers certain legal rights (e.g., through safety regulations) and with
unions playing an important role in the economic system, but it would not include a right to contestation.
6 I here abstract from the additional questions that arise because of the separation, especially in large cor-
porations, between ownership and control (with the latter being in the hands of management); I assume that
owners have at least an indirect say over control because they decide who controls a corporation.
7 One possible reason why individuals might prefer not to work in democratic firms is that if these come in the
form of cooperatives, they have to bundle risks in ways that may appear suboptimal: they have to invest
(some of) their capital in the same company in which they invest their labor power, rather than being able to
diversify their portfolio. One reply in the literature has been that shares in cooperatives could be freely
tradeable, see, for example, Thomas, 2016, 240–5, for a summary of the debate. Note, however, that the
problem only arises if workplace democracy takes on certain forms; it does not arise at all for workplace
republicanism.
8 If one wanted to insist on self-employment for everyone, one would have to bite the bullet of massive
welfare losses.
9 Maybe it could also be justifiable to let people enter into contracts with capitalist firms if these are strictly
regulated from the outside, to secure workers’ basic rights. Then, a key question (which would have to be
answered empirically) is whether this external regulation is sufficiently effective to really secure their rights.
10 As Pérotin (2016, 3), notes with regard to cooperatives, the gap between executive and non-executive pay
is also smaller.
11 He thereby positions himself against various writers (e.g., Gourevitch, 2015) who have defended workplace
democracy as reducing domination.
12 It might be said, though, that cooperatives are only successful in certain industries, in which the additional
costs for running firms democratically might not be so high. It is true that there are certain industries in
which few cooperatives or otherwise democratically organized firms can be found; Hansman’s (1996)
theoretical arguments would hold that this has to do with cost structures in these industries. But of course
cost structures are in turn a result of the regulatory environment, and could potentially be changed (e.g., by
taxes or subsidies).
References
Addison, J. T., Schank, T., Schnabel, C. and Wagner, J.(2007) “Do Work Councils Inhibit Investment?,”
Industrial and Labor Relations Review, 60, 29, 187–203.
Addison, J. T., Schnabel, C. and Wagner, J.(2001) “Works Councils in Germany: Their Effects on
Establishment Performance,” Oxford Economic Papers, 53, 4, 659–694.
Addison, J. T., Schnabel, C. and Wagner, J.(2004) “The Course of Research into the Economic Consequences
of German Work Councils,” British Journal of Industrial Relations, 42, 2, 255–281.
Alchian, A. A., and Demsetz, H.(1972) “Production, Information Costs, and Economic Organization,”
American Economic Review, 62, 777–795.
Anderson, E. (1999) “What Is the Point of Equality?,” Ethics, 109, 2, 287–337.
Anderson, E. (2017) Private Government: How Employers Rule Our Lives (and Why We Don’t Talk About It),
Bowles, S., and Gintis, H. (1993) “A Political and Economic Case for the Democratic Enterprise,” Economics
and Philosophy, 9, 75–100.
Bowles, S., and Gintis, H. (1996) “The distribution of wealth and the viability of the democratic firm.” In U. Pagano
and R. Rowthorn eds. Democracy and Efficiency in the Economic Enterprise, 82–97, New York, NY; Routledge.
Breen, K. (2015) “Freedom, republicanism, and workplace democracy,” Critical Review of International Social and
Political Philosophy, 18, 4, 470–485.
Burdin, G., and Pérotin, V. (2019) “Employee Representation and Flexible Working Time,” Labour Economics,
61, 101755.
Ciepley, D. (2013) “Beyond Public and Private: Toward a Political Theory of the Corporation,” American
Political Science Review, 107, 1, 139–158.
Dahl, R. (1985) A Preface to Economic Democracy, Berkeley, CA; University of California Press.
Dow, G. K. (2003) Governing the Firm: Workers Control in Theory and Practice, Cambridge; Cambridge University Press.
Ellerman, D. (1992) Property and Contract in Economics, London; Blackwell.
Ferreras, I. (2017) Firms as Political Entities. Saving Democracy through Economic Bicameralism, New York;
Frega, R., Neuhäuser, C., and Herzog, L. (2019) “Workplace Democracy – The Recent Debate,” Philosophy
Compass, 14, 4, online.
437
Lisa Herzog
Gerlsbeck, F., and Herzog, L. (2020) “The Epistemic Potentials of Workplace Democracy,” Review of Social
Economy, 78, 3, 307–330.
González‐Ricoy, I. (2014) “The Republican Case for Workplace Democracy,” Social Theory and Practice, 40, 232–254.
Gonzalez-Ricoy, I. (2020) “Ownership and Control Rights in Democratic Firms – A Republican Approach,”
Review of Social Economy, 78, 3, 411–430.
Gould, C. (1988) Rethinking Democracy, Cambridge; Cambridge University Press.
Gourevitch, A. (2015) From Slavery to the Cooperative Commonwealth: Labor and Republican Liberty in the Nineteenth
Century, Cambridge; Cambridge University Press.
Hansman, H. (1996) The Ownership of Enterprise, Cambridge, MA; Harvard University Press.
Herzog, L. (2019) Die Rettung der Arbeit. Ein politischer Aufruf, Berlin; Hanser.
Honoré, A. M. (1961) “Ownership.” In Guest, A. G. ed. Oxford Essays in Jurisprudence, 107–147, Oxford;
Hsieh, N. (2005) “Rawlsian Justice and Workplace Republicanism,” Social Theory and Practice, 31, 115–142.
Hübler, O., and Jirjahn, U. (2003) “Works Councils and Collective Bargaining in Germany: The Impact on
Productivity and Wages,” Scottish Journal of Political Economy, 50, 4, 471–491.
Jacob, D., and Neuhäuser, C. (2018) “Workplace Democracy, Market Competition and Republican
Self‐Respect,” Ethical Theory and Moral Practice, 21, 4, 927–944.
Kelly, M., and Howard, T. (2019) The Making of a Democratic Economy. Building Prosperity for the Many, Not Just
the Few, Oakland, CA; Berrett-Koehler Publishers.
Landemore, H., and Ferreras, I. (2016) “In Defense of Workplace Democracy: Towards a Justification of the
Firm–State Analogy,” Political Theory, 44, 53–81.
Mason, R. (1982) Participatory and Workplace Democracy, Carbondale, IL; Southern Illinois University Press.
Mayer, R. (2000) “Is There a Moral Right to Workplace Democracy?,” Social Theory and Practice, 26, 301–325.
McMahon, C. (1994) Authority and Democracy: A General Theory of Government and Management, Princeton, NJ;
McMahon, C. (2013) Public Capitalism: The Political Authority of Corporate Executives, Philadelphia; University of
Pennsylvania Press.
Mill, J. S. (2008) [1848/73] Principles of Political Economy and Chapters on Socialism, Oxford; Oxford University Press.
Néron, P.-Y. (2015) “Rethinking the Very Idea of Egalitarian Markets and Corporations: Why Relationships
Might Matter More Than Distribution,” Business Ethics Quarterly, 25, 93–124.
Olsen, E. (2013) “The Relative Survival of Worker Cooperatives and Barriers to Their Creation.” In Sharing
Ownership, Profits, and Decision-Making in the 21st Century (Advances in the Economic Analysis of Participatory &
Labor-Managed Firms, Vol. 14), Bingley; Emerald, 83–107.
Page, Rebecca (2011) “Co-determination in Germany –A Beginner’s Guide,” Working Paper 313 (November),
Düsseldorf; The Hans Böckler Foundation.
Pateman, C. (1970) Participation and Democratic Theory, Cambridge; Cambridge University Press.
Pausch, Markus (2013) “Workplace Democracy: From a Democratic Ideal to a Managerial Tool and Back,”
The Innovation Journal, 19, 1, 1–19.
Penceval, J. (2012) “Worker Cooperatives and Democratic Governance.” SIEPR Discussion Paper No. 12‐003.
Pérotin, V. (2016) “What Do We Really Know About Worker Co‐operatives?” https://www.uk.coop/sites/
default/files/uploads/attachments/worker_co‐op_report.pdf.
Reich, R. (2015) Saving Capitalism. For the Many, Not the Few, New York; Alfred Knopf.
Schwartz, A. (1982) “Meaningful Work,” Ethics, 92, 4, 634–646.
Schweickart, D. (2011) After Capitalism. Second Edition, Lanham, MD; Rowman & Littlefield.
Sen, A. (1985) “The Moral Standing of the Market,” Social Philosophy and Policy, 3, 1–19.
Taylor, R. S. (2017) Exit Left: Markets and Mobility in Republican Thought, Oxford; Oxford University Press.
Tcherneva, P. (2018) “The Job Guarantee: Design, Jobs, and Implementation.” Levy Institute Working Paper
No. 902.
Thomas, A. (2016) Republic of Equals: Predistribution and Property‐Owning Democracy, New York; Oxford
University Press.
Widerquist, K. (2013) Independence, Propertylessness, and Basic Income: A Theory of Freedom as the Power to Say No,
New York, NY; Palgrave Macmillan.
Williamson, O. E. (1975) Markets and Hierarchies: Analysis and Antitrust Implications, New York; The Free Press.
Yeoman, R. (2014) Meaningful Work and Workplace Democracy: A Philosophy of Work and a Politics of
Meaningfulness, Houndmills, Basingstoke, Hampshire; Palgrave Macmillan.
438
35
SOCIAL TRUST
1 Introduction
Social trust, when it exists, facilitates cooperation and contributes to social order. It generally refers
to the extent to which people view those they do not know as trustworthy. Robinson and Jackson
(2001: 117) state that social trust reflects “a positive expectation about the trustworthiness of the
generalized, abstract other, and a person’s level of social trust is thus a standard estimate of the
trustworthiness of an unknown other.” It also refers to the degree to which there is general trust in
institutions such as government, business, the military, or religious organizations. In societies with
high generalized social trust, institutions function fairly smoothly, the economy is often robust, and
individuals are willing to engage with those they do not know to transact and often to accomplish
collective tasks.
In research that documents variations in social trust across countries, it is not surprising that the
Nordic countries like Denmark, Sweden, and Norway rank near the top while countries in the
former Soviet Union such as Bulgaria and Romania as well as some countries in South Africa and
South America (e.g., Peru, Columbia, and Brazil) rank quite low. These differences can be linked to
several factors but among the most important are the level of trust in government agencies and
officials, the level of corruption, and the degree of inequality in society. General social trust has also
been associated with ethnic diversity specifically in the neighborhoods in which people reside,
hinging on the degree and sometimes the nature of the contact between ethnic groups (Putnam,
2007; Stolle et al., 2011; Van der Meer and Tolsma, 2014; Dinensen and Sonderskov, 2015), though
the evidence is mixed. In this chapter, we review some of the findings regarding the significance of
social trust in people as well as institutions and we examine evidence concerning the factors asso-
ciated with high and low social trust cross-culturally.
Research on general social trust has been conducted at both the interpersonal and the societal
levels dealing with the effects of trust on a variety of outcomes ranging from collaboration and
cooperation among individuals or teams in work settings to social cohesion and economic devel-
opment at the community and societal level. Typically generalized trust is linked most directly to
aggregate-level outcomes (see Schilke, Reimann, and Cook, 2021) while relational or particularized
trust focuses on personal or group-based trust. We focus primarily on the aggregate level, identifying
ways in which generalized social trust affects and is affected by various institutions in society.
This review begins with an examination of the relationship between general social trust and
economic development. We then add the role of governing institutions and the relationship
DOI: 10.4324/9780367808983-42 439

between general social trust and trust in institutions. The second section considers factors that
mediate both general social trust and trust in institutions, in particular the role of inequality and
corruption. Our review underscores how conceptually rich – but also perplexing – the field has
become in recent years. We conclude by (1) clarifying concepts, (2) highlighting notable pathways,
and (3) proposing an expansion of comparative research that investigates these pathways across a
greater variety of cultures and institutional settings.
2 Trust, Economy, and Governance
2.1 Does Social Trust Support Collective Action?

Cooperation is central to society. It facilitates collective action and participation in governance,
contributing to the production of social order. It is often argued that trust leads to cooperation, but
cooperation is typically plagued with free-riding problems at the heart of social dilemmas in which
individual interests are pitted against collective interests. Such social dilemmas typically occur in
complex social systems in which trust alone is unlikely to offer a solution (Cook and State, 2017). It
might offer the initial grounds for cooperation at the micro-level providing a basis for limited risk-
taking that may lead to longer-term cooperation and the joint production of collective action, but
the evolution of some form of institutional support for continued risk-taking is likely required over
time. Ostrom (1990), in work that earned her a Nobel Prize, provides evidence of the evolution of
such institutions for the support of collective actions, including the enforcement of local norms and
the use of relevant sanctions in fieldwork around the globe (e.g., Kenya, Guatemala, Nepal,
Switzerland, etc.).
In her fieldwork, Ostrom identified a number of mechanisms used by various communities to
manage common resources producing empirical evidence of the ways in which people avoid the
“tragedy of the commons” or the failure to provide for the common good. These include, but are
not limited to, specifying group boundaries, setting rules for engagement and access, monitoring
behavior, applying relevant sanctions, and developing agreed-upon ways to resolve disputes and
revise rules. In essence, her work with her collaborators proved that communities did not always fall
prey to the “tragedy” of the commons. Certainly, general social trust within the community fa-
cilitated the process of managing common resources effectively. But as Ostrom and Cox (2010)
argue, there is no silver bullet. Social dilemmas often occur in very complex situations so reliance on
trust alone is unlikely to be a universal panacea. Cook and State (2017: 25) conclude that “where it
does exist it extends cooperation, reduces reliance on sanctioning and other mechanisms and pro-
vides for increased social cohesion … in an increasingly global context in which conflicts have
widespread consequences.” But as Cook, Hardin, and Levi (2005) argue, there are many other
mechanisms that facilitate cooperation within organizations, communities, and societies that do not
rely specifically on general trust although they may enhance it in the long run.
Along with facilitating the solutions to social dilemmas, social trust helps to create the grounds for
economic development as argued by Fukuyama (1995) and others. Nobelist, Kenneth Arrow (1972,
1974) argued much earlier that trust is fundamental to robust economic activity in any society.
Ostrom’s work suggests how trust within communities can contribute to the emergence of in-
stitutional supports for collective action often serving as the basis for the efficient use of common
resources and related economic activity.
2.2 Does Social Trust Enhance Economic Development?

In Trust: The Social Virtues and the Creation of Prosperity (1995) Francis Fukuyama argues that the
global economy requires more flexible business organizations that are facilitated by social trust. He
440
Social Trust
pursues this thesis with a sweeping examination of the role of sociability, social capital, and cultural
values in political stability and economic development over time, especially at the dawn of the 21st
century. As Fukuyama (1995: 27) notes: “Widespread distrust in a society … imposes a kind of tax
on all forms of economic activity, a tax that high-trust societies do not have to pay.” This tax is the
cost of negotiation, regulation, and enforcement mechanisms that are required to secure transactions
when there is low social trust, and the risks of defection are high.
By Fukuyama’s (1995) account, those societies with low social trust have relied on relatively small
family firms and businesses. As a result, they had much less capacity to develop large, private business
organizations or firms that require some level of general social trust that moves beyond the close
radius of personal trust relations. The examples noted by Fukuyama are Taiwan, Hong Kong,
France, and Italy (mainly southern Italy) where the “reluctance of non-kin to trust one another
delayed and in some cases prevented the emergence of modern, professionally managed corpora-
tions” (Fukuyama, 1995: 30). These factors also inhibit the flexibility of the society to adopt new
organizational forms related to technological innovations and ever-changing markets. The examples
of higher trust societies that meet these challenges are Germany, Japan, and, to some extent, the
United States (a more complicated case) in Fukuyama’s analysis.
In a recent review of the empirical literature on the linkages between social trust and economic
growth Bjornskov (2018) identifies the various arguments about the direct and indirect mechanisms
that help to explain these linkages. Among the direct effects are such things as the reduction in
transaction costs that higher levels of social or organizational trust allow given that monitoring and
sanctioning efforts and related costs are lower. In societies without generalized trust anonymized
transactions are difficult, especially when enforcing institutions are weak.
As Bjornskov notes (2018: 538) more generally: “trust greases the wheels of innovation and
makes economies more productive.” Indirect effects include those that operate through informal
social norms, for example, that lead people to abide by internalized constraints that keep them from
exploiting others and engaging in self-dealing. Trust also increases investment which improves a
number of factors (e.g., those linked to increased innovation and productivity) that result in greater
economic growth (Zak and Knack, 2001). Other factors that are important mediators or those that
condition these effects are education and the extent to which democratic political institutions exist
in the country (or region) being studied (see Bjornskov, 2018).
The strong relationship between confidence or trust in democratic institutions and generalized
trust demonstrates how important the role of institutions can be in regulating interactions between
those who are not directly known to one another. When trust in institutions is low, people may
have to rely on those they know or on third-party reputation systems in order to obtain what they
need. This is the case in settings in which black markets emerge or informal systems of barter come
to exist outside the range of weak institutional regulation. Examples of these informal barter systems
abound in the Eastern European countries, especially following the transition from authoritarian rule
to more democratic institutions, a trend now in reversal in some locations several decades later.
More generally Alesina and La Farrara (2002) conclude that strong economic performance in society
requires the type of trust or confidence in institutions that reduce transactions costs and support
economic growth. We return to this topic in our discussion of factors that yield trust in government.
2.3 Does Trust Affect Government Performance?

Trust in governmental institutions (as distinct from generalized interpersonal trust) has been linked
to economic growth, as noted above. Political theorists have written extensively about the role of
trust in government and democratic institutions. Some view trust as essential to the smooth func-
tioning of government. Without the willing consent and compliance of the governed, those who
attempt to govern often have to revert to more coercive mechanisms of rule. Often political theorists
441
such as Dunn (1988), for example, place the focus on trust in government as the key foundation of
governing in what is referred to as “social contract theory.” But exactly by what mechanism social
trust works its magic is less clear in much of this work.
Some argue that it is the trust of citizens of each other that lays the groundwork for a more
productive and cooperative society and thus more effective governance and increased social order.
For Putnam, for example, higher levels of trust among citizens lead to greater civic engagement and
a more vibrant civil society, and this has a stabilizing effect leading to an increase in the governability
of the citizens and social order (1995).
Some of the literature on trust in government thus grows out of Putnam’s arguments about the
role of social capital in society. In 1992, Putnam wrote about the differences in economic growth
and development in northern and southern Italy arguing that what he called social capital (i.e., the
engagement of the citizens in civic duties and the networks they formed as a result) improved
government performance and thus economic productivity, more in the north of Italy than in
southern Italy. This argument set off several decades of research across the globe examining the
effects of social capital on government performance and economic development. Mansbridge
(1997) built on Putnam’s empirical work in Italy on the role of social capital but, instead of
attempting to explain variation in government performance, she investigated trust in government
institutions.
In successive writings on the effects of social capital on trust in government, Putnam’s concept of
civic engagement was combined with Coleman’s concept of generalized interpersonal trust (1988).
These authors distinguished and attempted to measure both concepts as a two-dimensional construct
of social capital (Brehm and Rahn, 1997; Keele, 2007). They also concurred that between the two
components, the generalized trust component was more predictive of trust in government than was
civic engagement.
Others believe that it is trust in governmental institutions that is more significant than citizens
trusting one another. They view the trust in institutions of governance as the glue that binds society
together and allows for stability since it provides legal enforcement and support in the face of failed
obligations. These factors produce a backdrop against which it is possible for citizens thus to trust
one another and to find those they do not know to be more trustworthy. In studies of taxation and
conscription, Levi (1997) and Levi and Stoker (2000) argue that only under certain circumstances do
citizens trust their government and comply with its requirements. In their work, it is the perceived
legitimacy of government based on its capacity to govern effectively and the extent to which citizens
view it as abiding by principles of procedural justice that determine the extent of compliance with
government dictates.
Another camp, also responding to Putnam’s work, reversed the causal direction of Putnam’s
argument and replaced his definition of social capital with Coleman’s. The earliest of these rebuttals
studied the effects of trust in government on social capital, which they defined as generalized in-
terpersonal trust (Tarrow, 1996; Levi, 1996, 1998). Later, some authors reintroduced the concept of
civic engagement but put it on the causal side of the equation, along with trust in government, in
order to predict generalized interpersonal trust. Moreover, they continued to define generalized
trust as social capital. The result is that both camps now view social capital as synonymous with
generalized interpersonal trust and treat Putnam’s original concept of civic engagement either as a
predictor or as a covariate of social capital. From the perspective of trust, this literature is quite
consistent. One camp argues that generalized trust predicts trust in government. The other argues in
the opposite direction, that trust in government predicts generalized trust. Future research will be
needed to sort out the exact effects of these two important sources of trust in society – interpersonal
general trust and institutional confidence – and their interrelationship.
Within this literature, the concept of trust in government has also expanded over time, but in a
more straightforward manner. As mentioned above, whereas Putnam (1992) originally studied the
442
Social Trust
effects of social capital on regional government performance, several authors in the effects camp
switched the dependent variable to trust in government institutions (Mansbridge, 1997; Brehm
and Rahn, 1997). Keele (2005, 2007) went a step further by distinguishing the two concepts
and using government performance as a predictor of trust in government. The opposite prog
ression occurred in the causes camp with the earliest authors emphasizing trust in institutions
(Tarrow, 1996; Levi, 1996, 1998). Research on transitional societies emphasizes the importance of
government performance, particularly in terms of levels of corruption and macro-economic de-
velopment, in fostering general trust in political institutions (Cleary and Stokes, 2006). Modern
welfare states, for example, often act as “third-party” enforcers and thus create grounds for trust and
increase the possibility of the provision of public goods.
In general, the two different sources of social trust that are significant in society, primarily with
respect to economic growth and general social order, are trust in those we do not know and
confidence or trust in our key institutions, particularly governmental institutions. Both sources of
trust have been demonstrated to be important in economic activity as well as government perfor-
mance. However, the causal directions are not clear and seem to point in both directions. That is,
strong institutions promote confidence in them and support broader economic engagement in
innovation as well as productive activities and such engagement builds confidence or trust in these
institutions. In a similar way, trusting others, particularly those not yet known to be trustworthy,
facilitates engagement in a broader range of economic activity often yielding positive outcomes, and
this activity, when it does produce positive outcomes, reinforces the act of trusting others (even
anonymous others). In addition, it is clear that general social trust and trust in institutions are linked
and it is to that complex problem that we turn in our consideration of inequality and corruption in
relation to trust, interpersonal and institutional. Several general factors associated with social trust
include the effects of immigration and increasing ethnic fragmentation (or diversity) on interpersonal
trust, as well as issues of fairness and procedural justice at the level of government performance and
its link to institutional confidence.
3 Trust in Society and Institutions
3.1 Does Diversity Lower Social Trust and Social Cohesion?

A number of researchers have studied the links between general social trust and social cohesion.
With widespread distrust in society, interactions tend to be limited to those who occupy a closed
network with a small trust radius. People are unwilling to take a risk on those they do not know or
about whom they have little reliable reputational information. As we note, such circumstances also
restrict or constrain economic development because transactions are not likely to occur between
strangers or those outside one’s trust radius given the risk of exploitation. A separate argument, made
popular by Robert Putnam (2007), was that increased ethnic diversity lowers social trust and reduces
social cohesion, claims we examine more fully below.
In the literature on social capital, generalized trust (as a measure of social capital) was initially found
to be higher in ethnically homogeneous populations (Knack and Keefer, 1997). In addition, social
capital, when defined in terms of participation in civic activities, was found to be lower in ethnically
fragmented communities (Alesina and La Farrara, 2002). Putnam (2007) popularized these findings in
sociology and political science proposing that individuals in ethnically diverse communities hunker
down, and that this harms interpersonal trust and reduces within-group ties. In the research programs
which followed, this hunkering down argument was also referred to as the constrict claim.
Following Putnam’s 2007 publication, a veritable cottage industry of empirical research emerged
to test and clarify the relationship between ethnic diversity and social cohesion, but this research
yielded mixed results in part due to conceptual and methodological issues. Using a structural-
443
equation model with geographic scope, Letki (2008), for example, found that it was low neigh-
borhood status, and not racial diversity, which undermined social cohesion. By low neighborhood
status they mean communities in which most of the occupants have low socio-economic status. In a
cross-national study, Kesler and Bloemraad (2010) found that immigration-generated diversity only
partially accompanied reductions in generalized trust, group membership, and civic participation.
Stolle and Harell (2013), in contrast, found that while neighborhood heterogeneity negatively af-
fected generalized and out-group trust, these effects were also reduced to some extent by interethnic
neighborhood contact.
In a more recent publication commenting on the literature on ethnic diversity and social trust,
Dinensen and Sonderskov (2015) argue that “residential exposure to ethnic diversity reduces social
trust,” providing some support for Putnam’s original argument concerning the negative effects on
trust of more integrated communities or neighborhoods. The data Dinensen and Sonderskov (2015)
analyze are uniquely suited to test such a claim since they come from the Danish registry which they
link to relevant social survey data in order to get measures of trust.
Conceptually, Portes and Vickstrom (2011) agree that ethnic diversity may well reduce ties and
participation at the neighborhood level; however, they contend that it is norm-based organic solidarity
that constitutes social cohesion in modern societies and not Putnam’s nostalgic view of an older version
of communitarianism. In their words: “The ‘discovery’ that immigration reduces cultural homogeneity
and communitarianism is perfectly reasonable. The alarm following that discovery that migration would
lead to social disorganization and breakdown is not” (Portes and Vickstrom, 2011: 474).
In related work on the topic of social trust and social cohesion, criticisms have focused on the
various conceptualizations and measures of social cohesion. In the literature investigating Putnam’s
(2007) “hunkering down” claim, social cohesion is typically defined in terms of one of four di-
mensions: formality, mode, target, and geographic scope according to Van der Meer and Tolsma
(2014: 461). Formality indicates a distinction between formal and informal relations (Pichler and
Wallace, 2007); mode indicates a distinction between attitudinal and behavioral measures (Hooghe
et al., 2009); target indicates distinctions between the in-group, the out-group, or the general
population and geographic scope indicates neighborhood, municipalities, and/or country (Lundasen
and Wollebaek, 2013). Failing to draw these distinctions in many studies leads to effects of ethnic
diversity on social cohesion that are inconsistent and often contradictory, as discussed more fully by
Van der Meer and Tolsma (2014).
However, once the definition of social cohesion and its measure(s) are specified, a number of
findings are fairly consistent. These include the fact that ethnic diversity is only one of many factors that
reduce inter-ethnic group trust. For this reason, immigration and heterogeneous environments alone
should not be viewed as the primary reasons for decreases in interethnic social cohesion (Stolle and
Harell, 2013). A second significant finding that is relatively consistent across studies is the conclusion
that Putnam’s constrict claim (i.e., the hunkering down hypothesis) does not universally apply to all
geographic scopes, but specifically to social cohesion which is spatially bounded by neighborhood.
Just as Portes and Vickstrom (2011) argued, ethnic diversity may decrease intra-neighborhood
cohesion but does not necessarily decrease other forms of cohesion (Marschall and Stolle, 2004),
most notably the organic solidarity provided by norms and institutions in modern societies. In
addition, it is clear that support for Putnam’s claims is more common in the United States than
elsewhere suggesting that Putnam’s concepts and arguments may confound culturally specific at-
tributes of the American experience with other factors that may vary cross-culturally in distinct ways
(see Van der Meer and Tolsma, 2014, who refer to this fact as “American exceptionalism”).
In their comprehensive review, Van der Meer and Tolsma (2014) also identify various me-
chanisms that might underly the linkage between ethnic heterogeneity and social cohesion (often
measured in terms of general social trust). The mechanisms they examine include homophily,
anomie, group threat, and social disorganization (or disorder). Homophily refers to a preference for
444
Social Trust
those who are similar or “like me.” Anomie is a sense of anxiety or alienation resulting from a lack of
clear social and ethical standards, often arising during times of rapid social change. In the authors’
overview, they argue that there is little support for homophily and anomie, but some support for
perceived group threat derived from increased interethnic contact given greater opportunity for such
contact in diverse neighborhoods. In fact, they suggest that the negative effect of increased ethnic
diversity is limited to neighborhoods (consistent with Dinensen and Sonderskov, 2015) and no other
geographic boundaries. They conclude that perceived threat may lead to distrust of others but that it
“certainly does not always originate from living in ethnic heterogeneous environments” (Van der
Meer and Tolsma, 2014: 471). They also point to the significant moderating effects of the level of
segregation between groups and degrees of inequality.
In a study focusing on the effects of immigration on civic activity (one form of social capital) over
time, Kesler and Bloemraad (2010: 339) find that cross-national changes in ethnic diversity are not
everywhere accompanied by reductions in collective mindedness. Policies relating to multiculturalism
and the mitigation of inequality make a difference. Studies have been conducted in various cultures
focusing on ethnic fractionalization or diversity and its effects on social trust and cooperation. In one of
those field studies in Africa (Kenya in this case), Buchan and Rolfe (2017: 234) conclude that: “In both
Western and non-Western societies, when ethnic fractionalization exacerbates income inequality –
often prompted by globalization – enmities and tensions among ethnicities is increased and co-
operation is lessened.” Further studies are needed across groups of the mitigation effects of inequality,
fairness, and other factors (Hofman, Bulte, and Voors, 2017) on cooperation and trust. As we have
noted, social trust and the resulting collective orientation within neighborhoods and communities are
often associated with engagement not only in civic activities but also in a variety of forms of collective
action often involving the production of important public goods.
3.2 Does Inequality Decrease Social Trust?

Among the most important factors identified consistently as those associated with low levels of social
trust (as well as trust in institutions) is the degree of inequality in the society. It helps to partially explain,
for example, what is referred to as the “Nordic exceptionalism” with respect to such high levels of
general trust. Early case studies such as the famous work of Banfield (1958) demonstrate how poverty
led to social distrust in rural Italy (along with strong in-group bonds), which was widespread and, to
some extent, the basis for the subsequent claims of Putnam (1992) about major differences between
southern and northern Italy with respect to economic development over time. Using information
derived from national and international panel data, Uslaner (2002; Uslaner and Brown, 2005) shows
that economic inequality is one of the strongest predictors of weakened generalized trust.
A number of mechanisms have been proposed linking economic inequality to weakened gen-
eralized trust. Social ties, inference on social relationships, conflicts over resources, and the op-
portunity cost of time are those identified by Jordahl (2009). Coleman (1988) and others have
argued that people are more likely to form social ties and trust those like themselves. Inequality
undermines tie formation because it exacerbates the differences across socio-economic groups.
Social relationships are not only pipes of exchange, but prisms of perception. Inequality across social
relationships can lead to exploitation, or at least to feelings of envy and perceptions of unfairness
(Fischer and Torgler, 2006). Moving beyond social relationships, inequality has also been thought to
raise incentives for deceitful behavior among classes, particularly from the poor to the rich
(Rothstein and Uslaner, 2005). Lastly, low-income earners are likely to be more sensitive to changes
in wages and more likely to verify the trustworthiness of others, resulting in a higher opportunity
cost of time (Zak and Knack, 2001). Jordahl (2009) reviewed cross-country studies, within-country
studies, and experiments and found the strongest evidence that the first two mechanisms, social ties,
and inference on social relationships, link inequality to weakened generalized trust.
445
3.3 Does Low Social Trust Increase Corruption?

In addition to inequality, low levels of social trust (and trust in institutions) have been linked with
corruption. Several mechanisms have been identified that underly this association. Theories of social
capital equate strong binding ties and within-group cohesion to weak generalized trust. These
factors also cause groups to impose strong constraints on group members that tend to enforce what
Portes (1998) refers to as downward leveling norms. The conditions of weak generalized trust which
give rise to corruption are not only the effect of strong within-group cohesion but also of weak
between group cohesion and weak social institutions. As Gambetta (1993) makes clear in his work
on the Mafia, its roots took hold in southern Italy because of the combination of weak governmental
institutions and strong within-group ties (social bonds). This combination was subsequently argued
to be the basis for particularized trust whereas strong between-group bridging ties and social co-
hesion were linked to generalized trust (e.g., Graeff, 2009).
Corruption is typically learned over time in societies in which particularized trust comes to
dominate general social trust. With strong generalized trust, corruption norms are weaker than
universalistic laws and norms and those engaged in corruption tend to have strong specific ties with
one another (to limit exposure). Early indications of this relationship came from cross-national
studies showing that capital markets were smaller in countries having weaker laws protecting in-
vestors and weaker enforcement of those laws (La Porta et al., 1997). Cross-national studies of
mature and newly established democracies yielded mixed results with respect to performance and
within-group cohesion. While corruption predicted negative evaluations of performance, this
evaluation was significantly attenuated among those having close informal ties to incumbent political
authorities (Anderson and Tverdova, 2003).
Normative links are difficult to empirically test at the macro-level, but some research focuses
attention on corruption in particular institutional contexts such as judicial corruption or in gov-
ernmental bureaucracies. Within-country studies, particularly of Japan, have attributed norms of
bureaucratic corruption to decreased general trust in public officials. Low levels of public trust in
Japan were blamed on decades of scandals (Rose-Ackerman, 1999). This distrust, Pharr (2000)
argued, was not only due to the scandals themselves but also due to their relentless coverage in the
media. The corruption of Japanese public officials is taken for granted as bribery scandals involving
government officials and public works are everyday stories in Japan’s mass media (Inoguchi, 2002).
In the United States, Richey (2010) used state-level panel data to demonstrate that an increase in
judicial corruption led to decreased social trust.
The above two bodies of research have linked inequality and corruption to decreased social
trust, respectively, but these two pathways may also be linked in a single, more complex process.
Wealth disparity has been shown to predict corruption (You and Khagram, 2005). Uslaner
(2005) argued that this connection is in fact mediated by a decrease in generalized trust. If
increased corruption, in turn, perpetuates inequality, then the entire process is actually a re-
inforcing cycle. Not easily escaped, this cycle was later coined the inequality trap (Uslaner, 2002).
Subsequent studies have followed suit, attempting to demonstrate this mechanism not only on a
normative, societal level but also in specific institutional settings, such as unfair legal systems
(Uslaner, 2009).
4 Conclusion
General social trust often refers not only to the trust of individuals one does not know, the standard
focus but also of governing institutions. General trust in individuals has been associated with eco-
nomic growth due to reduced transaction and monitoring costs, as well as with increased co-
operation and various forms of collective action. Similarly, general trust in our institutions such as
446
Social Trust
government, the military, and religious organizations, among others, is central to the production of
cooperation, social cohesion, and social order in society.
Unfair norms and practices have the potential to undermine each type of general trust. Much ink
has been spilt arguing the detrimental effects of diversity on cohesion and trust; however, this
research agenda ultimately only found narrowly applicable effects. By contrast, inequality appears to
undermine general trust in individuals more than any other factor. As a manifestation of unfairness in
socio-economic matters, inequality across social relationships and classes can generate powerful
negative feelings and perceptions that undermine our mutual sense of homophily and trust-
worthiness. Corruption exerts a similarly detrimental effect on general trust in institutions. As a form
of unfairness in public and political spaces, corrupt norms and practices undermine our beliefs
that justice is blind, that rules be enforced impartially, and that leaders have our best interests at heart.
The distinctions between trust in individuals and institutions, and between the unfairness of in-
equality and corruption, offer clarity to a field of research that has been clouded by catch-all concepts
and inconsistent measurements. At the same time, we recognize that in the real world, forms of trust
and unfairness are equally as intertwined as are economy and governance. While undermining general
trust in institutions, corruption can perpetuate socio-economic inequalities and compel over-reliance
on particular trust in known individuals (within-group ties) at the expense of general trust in unknown
individuals (between-group ties). Pursing the specific interests of a few over the collective interests of
the whole leads again to corruption and inequality, and the cycle continues.
The state plays a significant role in either propagating or mitigating unfairness and general distrust
of individuals and institutions. When the state permits inequality and corruption to persist, we come
to distrust governing institutions as well as the motives of others. Conversely, where the state
mitigates inequality and corruption, it restores faith in institutional efficacy and expands our civic
commitment beyond the narrow cliques of those we know to our community and society at large.
Democratic governance which promotes fairness and general social trust will necessarily reap the
benefits of economic development and institutional efficacy.
While social trust may be important in every society, it is subject to the social and political
organization of each particular society. This review has laid out a number of pathways involving
general social trust; however, further research is needed regarding (1) the specific mechanisms
linking fairness, trust, and performance, as well as (2) the interactions between these pathways,
particularly those spanning governance and the economy. To what extent can civil institutions
compensate for a weak state or unequal economy? Are the effects of inequality and corruption
additive or multiplicative? How is general trust affected in highly unequal societies with low levels of
corruption or highly corrupt societies with low levels of inequality? Is there is zero-sum relationship
between the particular trust of known individuals and the general trust of unknown members of
society? We know that social trust matters, but only a purposeful expansion of empirical research
across multiple cultures and institutional settings will begin grappling with the breadth of what we
have yet to learn.
References
Alesina, A. and La Farrara, E. (2002) “Who Trusts Others?”, Journal of Public Economics, 85, 2, 207–234.
Anderson, C. J. and Tverdova, Y. V. (2003) “Corruption, Political Allegiances, and Attitudes toward
Government in Contemporary Democracies”, American Journal of Political Science, 47, 1, 91–109.
Arrow, Kenneth. (1972) “Gifts and Exchanges”, Philosophy and Public Affairs, I, 343–362.
Arrow, Kenneth. (1974) The Limits of Organization, New York; Norton.
Banfield, Edward C. (1958) The Moral Basis of a Backward Society, Glencoe, IL; The Free Press.
Bjornskov, Christian. (2018) “Social Trust and Economic Growth”, In Eric Uslaner (ed.) The Oxford Handbook
of Social and Political Trust (pp. 535–555), Oxford, UK; The Oxford University Press.
447
Brehm, John and Rahn, Wendy. (1997) “Individual-Level Evidence for the Causes and Consequences of Social
Capital”, American Journal of Political Science, 41, 3, 999–1023.
Buchan, Nancy R. and Rolfe, Robert. (2017) “The Influence of Globalization and Ethnic Fractionalization on
Cooperation and Trust in Kenya”, In Paul A.M. van Lange, Bettina Rockenbach, and Toshio Yamagishi
(eds.) Trust in Social Dilemmas (pp. 213–236), Oxford, UK; Oxford University Press.
Cleary, Matthew R. and Stokes, Susan C. (2006) Democracy and the Culture of Skepticism: Political Trust in
Argentina and Mexico, New York, NY; Russell Sage Foundation.
Coleman, James S. (1988) “Social Capital in the Creation of Human Capital”, American Journal of Sociology, 94,
S95–S120.
Cook, Karen S., Hardin, Russell, and Levi, Margaret. (2005) Cooperation without Trust?, New York City;
Russell Sage Foundation.
Cook, Karen S. and State, Bogdan. (2017) “Trust and Social Dilemmas: A Selected Review of Evidence and
Applications” In P. A. M. Van Lange, B. Rockenbach, and T. Yamagishi (eds.) Trust and Social Dilemmas
(pp. 9–30), Oxford, UK; Oxford University Press.
Delhey, Jan. (2005) “Predicting Cross-National Levels of Social Trust: Global Pattern or Nordic
Exceptionalism”, European Sociological Review, 21, 4, 311–327.
Dinensen, P. T. and Sonderskov, K. M. (2015) “Ethnic Diversity and Social Trust: Evidence from the Micro-
Context”, American Sociological Review, 80, 550–573.
Dunn, J. (1988) “Trust and Political Agency”, In Diego Gambetts (ed.) Trust: The Making and the Breaking of the
Cooperative Bond (pp. 73–93), Oxford, UK; Basil Blackwell.
Fischer, J. A., and Torgler, B. (2006) “The Effect of Relative Income Position on Social Capital”, Economics
Bulletin, 26, 4, 1–20.
Fukuyama, Francis. (1995) Trust: The Social Virtues and the Creation of Prosperity, New York; Free Press.
Gambetta, Diego. (1993) The Sicilian Mafia: The Business of Private Protection, Cambridge, MA; Harvard
University Press.
Graeff, P. (2009) “Social Capital: The Dark Side”, In G. Svendsen and S. Gunnar Lind Hase (eds.) The
Handbook of Social Capital (pp. 143–161), Cheltenham, UK; Edward Elgar.
Hetherington, M. (1998) “The Political Relevance of Trust”, American Political Science Review, 92, 791–808.
Hofman, Paul, Bulte, Erwin, and Voors, Maarten. (2017) “Trust in African Villages: Experimental Evidence
from Rural Sierra Leone”, In P. A. M. van Lange, B. Rockenbach and T. Yamagishi (eds.) Trust and Social
Dilemmas (pp. 263–276), Oxford, UK; Oxford University Press.
Hooghe, M., Reeskens, T., Stolle, D., and Trappers, A. (2009) “Ethnic Diversity and Generalized Trust in
Europe: A Cross-National Multilevel Study”, Comparative Political Studies, 42, 198–223.
Inoguchi, T. (2002) “Broadening the Basis of Social Capital in Japan”, In R. Putnam (ed.) Democracies in Flux:
The Evolution of Social Capital in Contemporary Society (pp. 359–392), Oxford, UK; Oxford University Press.
Jordahl, Henrik. (2009) “Economic Inequality”, In G. T. Svendsen and G. L. H. Svendsen (eds.) Handbook of
Social Capital. Cheltenham, UK; Edward Elgar.
Keele, Luke J. (2005) “The Authorities Really Do Matter: Party Control and Trust in Government”, Journal of
Politics, 67 (August), 445–456.
Keele, Luke J. (2007) “Social Capital and the Dynamics of Trust in Government”, American Journal of Political
Science, 51, 2, 241–254.
Kesler, Christel and Bloemraad, Irene. (2010) “Does Immigration Erode Social Capital? The Conditional
Effects of Immigration-Generated Diversity on Trust, Membership and Participation Across 19 Countries
1981–2000”, Canadian Journal of Political Science, 43, 2, 319–347.
Knack, Stephen and Keefer, Philip. (1997) “Does Social Capital Have an Economic Pay-off?”, Quarterly Journal
of Economics, 112, 1251–1288.
La Porta, R., Lopez‐de‐Silanes, F., Shleifer, A., and Vishny, R. W. (1997) “Legal Determinants of External
Finance”, The Journal of Finance, 52, 3, 1131–1150.
Letki, Natalia. (2008) “Does Diversity Erode Social Cohesion? Social Capital and Race in British
Neighborhoods”, Political Studies, 56, 1, 99–136.
Levi, Margaret. (1996) “Social and Unsocial Capital: A Review Essay of Robert Putnam’s Making Democracy
Work”, Politics and Society, 24, 1, 45–55.
Levi, Margaret. (1997) Consent, Dissent, and Patriotism, New York, NY; Cambridge University Press.
Levi, Margaret. (1998) “A State of Trust”, In Trust and Governance, Valerie Braithwaite and Margaret Levi (eds.)
New York; Russell Sage Foundation.
Levi, Margaret and Stoker, Laura. (2000) “Political Trust and Trustworthiness”, Annual Review of Political
Science, 3, 1, 475–507.
448
Social Trust
Lundasen, Susanne W. and Wollebaek, Dag. (2013) “Diversity and Community Trust in Swedish Local
Communities”, Journal of Elections, Public Opinion and Parties, 23, 3, 299–321.
Mansbridge, Jane J. (1997) “Social and Cultural Causes of Dissatisfaction with U.S. Government”, In Joseph
Nye, Jr. Philip D. Zelikow, and David C. King (eds.) Why People Don’t Trust Government (pp. 133–153),
Cambridge, MA; Harvard University Press.
Marschall, Melissa J. and Stolle, Dietlind. (2004) “Race and the City: Neighborhood Context and the
Development of Generalized Trust”, Political Behavior, 26, 2, 126–153.
Ostrom, Elinor. (1990) Governing the Commons: The Evolution of Institutions for Collective Action, New York;
Ostrom, Elinor and Ahn, T. K. (2009) “The Meaning of Social Capital and Its Link to Collective Action”, In
Gert Tinggaard Svendsen and Gunnar Lind Haase Svendsen (eds.) The Handbook of Social Capital,
Cheltenham, UK; Edward Elgar.
Ostrom, Elinor and Cox, M. (2010) “Moving Beyond Panaceas: A Multi-Tiered Diagnostic Approach for
Social-Ecological Analysis”, Environmental Conservation, 37, 4, 451–463.
Paxton, Pamela. (1999) “Is Social Capital Declining in the United States? A Multiple Indicator Assessment”,
American Journal of Sociology, 105, 1, 88–127.
Pharr, S. J. (2000) “Official’s Misconduct and Public Distrust: Japan and the Trilateral Democracies”, In S. J.
Pharr and R. D. Putnam (eds.) Disaffected Democracies: What’s Troubling the Trilateral Countries (pp. 173–201),
Pichler, Florian and Wallace, Claire (2007) “Patterns of Formal and Informal Social Capital in Europe”,
European Sociological Review, 23, 4, 423–435.
Portes, Alejandro. (1998) “Social Capital: Its Origins and Applications in Modern Sociology”, Annual Review of
Sociology, 24, 1–24.
Portes, Alejandro and Vickstrom, Erik. (2011) “Diversity, Social Capital and Cohesion”, Annual Review of
Sociology, 37, 461–479.
Putnam, R. D. (1992) Making Democracy Work: Civic Traditions in Modern Italy, Princeton, NJ; Princeton
University Press.
Putnam, R. D. (1995) “Bowling Alone: America’s Declining Social Capital”, Journal of Democracy, 6, 65–78.
Putnam, R. D. (2007) “E Pluribus Unum: Diversity and Community in the Twenty-first Century”,
Scandinavian Political Studies, 30, 2, 137–174.
Richey, S. (2010) “The Impact of Corruption on Social Trust”, American Politics Research, 38, 4, 676–690.
Robinson, Robert V., and Jackson, Elton F. (2001) “Is Trust in Others Declining in America? An
Age–Period–Cohort Analysis”, Social Science Research, 30, 1, 117–145.
Rose-Ackerman, S. (1999) Corruption and Government: Causes, Consequences, and Reform, New York; Cambridge
University Press.
Rothstein, B., and Uslaner, E. M. (2005) “All for All: Equality, Corruption, and Social Trust”, World Politics,
58, 41.
Schilke, Oliver, Reimann, Martin, and Cook, Karen S. (2021) “Trust in Social Relations”, Annual Review of
Sociology, 47, 239–259.
Stolle, Dietlind, and Harell, A. (2013) “Social Capital and Ethno-Racial Diversity: Learning to Trust in an
Immigrant Society”, Political Studies, 61, 1, 42–66.
Stolle, Dietlind, Petermann, Soeren, Schoenwaelder, Karen, and Schmitt, Thomas. (2011) “Consequences of
Immigration-related Diversity on Social Integration and Social Cohesion – Bringing Contact Back In”,
Paper Presented at ECPR Conference in Reykjavik 2011, Section 45.
Stolle, Dietlind, Soroka, Stuart, and Johnston, Richard. (2008) “When Does Diversity Erode Trust?
Neighborhood Diversity, Interpersonal Trust and the Mediating Effect of Social Interactions”, Political
Studies, 56, 1, 57–75.
Sturgis, Patrick, Brunton-Smith, Ian, Read, Sanna, and Allum, Nick (2011) “Does Ethnic Diversity Erode
Trust? Putnam’s ‘Hunkering-Down’ Thesis Reconsidered”, British Journal of Political Science, 41, 1, 57–82.
Tarrow, Sidney. (1996) “Making Social Science Work Across Space and Time: A Critical Reflection on
Robert Putnam’s Making Democracy Work”, American Political Science Review, 90, 2, 389–397.
Torcal, M. and J.R. Montero. (2006) Political Disaffection in Contemporary Democracies: Social Capital, Institutions,
and Politics, London; Routledge.
Uslaner, Eric. (2002) The Moral Foundations of Trust, Cambridge, UK; Cambridge University Press.
Uslaner, Eric M. (2005). “The Bulging Pocket and the Rule of Law: Corruption, Inequality, and Trust”, In
Conference on the Quality of Government: What it is, How to Get it, Why it Matters (pp. 17–19).
Uslaner, Eric M. (2009) “Corruption, Inequality, and Trust”, In G. Svendsen and S. Gunnar Lind Hase (eds.)
The Handbook of Social Capital, Cheltenham, UK; Edward Elgar.
449
Uslaner, Eric M. (2012) Segregation and Mistrust, New York; Cambridge University Press.
Uslaner, Eric M. and Brown, Mitchell. (2005) “Inequality, Trust and Civic Engagement”, American Politics
Research, 33, 6, 868–894.
Van der Meer, Tom and Jochem Tolsma. (2014) “Ethnic Diversity and Its Effects on Social Cohesion”, Annual
Review of Sociology, 40, 459–478.
Van der Meer, Tom W. G. (2018) “Economic Performance and Political Trust”, In Eric M. Uslaner (ed.) The
Oxford Handbook of Social and Political Trust (pp. 600–615), New York; The Oxford University Press.
You, J. S., and Khagram, S. (2005) “A Comparative Study of Inequality and Corruption”, American Sociological
Review, 70, 1, 136–157.
Zak, Paul and Knack, Stephen. (2001) “Trust and Growth”, Economic Journal, 111, 295–321.
450
INDEX
15th Amendment, 1870 416 Anderson, P. 38, 127

Andreou, Chrisoula 91–92
A androcentric norms 56
Abeler, J. 111 Anglo-American political philosophy 367
Absolute Idea 45–46, 49 anomie 445
accountability principle 194 anonymity 402
Acemoglu, Daron 165–166 “antimarket principle” of communal reciprocity”
Achen, Christopher 393 286n2
Ackerly, Brooke 384 Appiah, Anthony 393
Ackerman, Bruce 289 Aquinas, Thomas 321
adaptive preferences and internalized oppression arguments from rights: from contractual rights
59–61 430–432; from property rights 429–430
Addison, J. T. 434 Arnold, Samuel 197n2
agent-based modeling 29 Arrow, Kenneth 9–10, 18, 20–21, 25n10, 77, 440
Ain’t I a Woman? (Hooks, 1987) 370 Arrow’s IIA 413n7
air pollution 195 Arthur, W. B. 30–32, 38
Akerlof, G. A. 20, 113 Ashby, W. R. 29, 31
Alaska Permanent Fund 294 Astill, S. 35
Albert, Michael 281, 284, 286n12 Atkinson, Anthony 297
Alemante, F. 410 Attenuation of Rights 338
Alesina, A. 124, 441 Aumann, Robert J. 110
Aligica, Paul Dragos 195 Austen-Smith, David 17
allocation of productive resources 342, 347–348 Austin v. Michigan Chamber of Commerce 177–178
“the allocation problem” 281 Axelrod, Robert 72
allocative vs. the distributive role of markets 348 axioms, binary elections: anonymity 402–403;
Alstot, Anne 289 neutrality 403; positive responsiveness 403–404
alternative distributivist principles 339n4 Axtell, R. 29
Althaus, Scott 396 Ayala-López, Saray 127, 130n5
Althusser, Louis 43 Ayers, Clarence 150
Amazon 278
American Community Survey 260n1 B
Analyzing Oppression (Cudd, 2006) 53 baby bonds 295–296
Anarchy, State, and Utopia (Nozick) 333, 336 “bad” institutions 318
anchoring bias 121 Bafumi, J. 413
Anderson, Elizabeth 227–229, 371–372 Bangladesh Fire and Building Safety Accord 373
451
Index
Banks, Jeffrey 17, 127 Bowles, Robert Boyd 420

Banzhaf power 413 boycotts 181
bargaining and cooperative games 76–77 Boyd, R. 29
bargaining models, Nash and Rubinstein 20 Braithwaite, Richard 77
Bar-Hillel, Maya 83 Brake, Elizabeth 57
Barrett, J. 37 Bratman, Michael 154, 157
Bartels, Larry 393 Braunstein, M. L. 99–100
Bartholdi, L. 404 Brennan, G. 29, 115–116, 138, 230–233, 235
basic rights and liberties 342–345 Brennan, Jason 229
“basic structure of society” 153 Breuer, Felix 24
basins of attraction 31–32 Brexit 35, 391
“bathroom bill” 176, 181 bribes 317–318
Battle of the Sexes 18 British East India Company 322n5
BEAST model 100 Broome, John 9–10, 15n15, 83
Becker, Gary S. 109 Brownstein, Michael 130n6
“Becoming” 46 “brute luck” 351
Beeghly, Erin 130n5 Buchan, Nancy R. 445
behavioral decision theory 94–95 Buchanan, James 152
behavioral economics 10, 95 Buchanan v. Warley 253–254
behavioral rule, social norms as 138–139 “bundles of resources” 366
“behavioral” rule-following 111 burdens of justice, evading 359–361
behavioral theories, prominent 97–98 Burdin, G. 433
“Being” 45–46 Burwell v. Hobby Lobby 179–180
Bénabou, R. 111 Busemeyer, J. R. 104
“the benefits of my pains” 168 Bush, George W. 402
Berges, Sandrine 60
Berle, Adolf 150 C
Berlin, Isaiah 21 Cairney, P. 35
Bernheim, B. D. 111 Calculus of Consent (Buchanan and Gordon, 1962) 152
Bertalanffy, L. V. 29 Calculus of Consent (Buchanan and Tullock, 1962; E.
Bertrand, Marianne 130n2 Ostrom, 2011) 187
Bessi, A. 36 California Judicial Council 260n2
Betham, Jeremy 341–342 Campbell, D. 29, 410
Beveridge Report 270 Capital (Marx) 43, 50
“Beyond Markets and States: Polycentric Governance of capital assets tax 282
Complex Economic Systems” (Ostrom, 2010a) 187 capitalism 215n3; capitalism as “socialism” 272–274;
Bezos, Jeff 278–279 crony capitalism is not capitalism 271–272; how
Bhatia, S. 104 capitalism works 269–271; introduction to
bias 121–122 267–269
Bicchierie, Cristina 76, 114–116, 138, 141–142, capitalism, how it works 269–271
144–145 capitalism and women’s oppression 61–62
binary elections: axioms 402–404; introduction to capitalism as “socialism” 272–274
401–402; May’s Theorem 404; weak “capitalist democracy” 21
anonymity 404 care, gender of 56–59
Binmore, Ken 77 caregiver suport policy 248–249
Bipartisan Campaign Reform Act (BCRA, aka caregiving, elevating the status of 247
McCain-Feingold), 2003 177 caregiving leave, paid 242–244
Birch, Jonathan 78 Carens, Joseph 285–286n1
birdcage metaphor 53 caring labor, gender of 56–59
Bjornskov, Christian 441 Carlana, Michela 124
Black Live Matter, 2020 181 Carlson, Erik 91
Black Lives Matter 365 Carnell Construction Corporation v. Danville
Bloemraad, Irene 443, 445 Redevelopment and Housing Authority 183n3
Bolshevism 302 Cartwright, Nancy 21–22, 25n6
Booth, Anthony 228 case for (contingent) market meaning 231–234
Borda, P. 401, 407 catallaxy 31, 34–35
Boulding, K. E. 29 categorical vs. comparative thinking 9–10
Bovens, Luc 88 “ceremonial behavior” 150
452
Index
“certification” service 8 The Communist Manifesto (Marx and Engles) 268

Chambers, Clare 242 communitarian response to liberalism 153
“chaos” theorem 23, 25n10 Communitarians 153
Chicken 18 “community control” 191
“child penalities” 240 Community Development Block Grants 255
childcare 240–241, 246–247 “community policing” 191–192
China’s Great Leap Forward 193 comparative vs. categorical thinking 9–10
choice behavior, theories of: brief history 94–95; Competence Principle 395
choice problem 95–96; key behavioral properties complex orders and their features 31–35
98–102; prominent behavioral theories 97–98; complexity: complex orders and their features 31–35;
standard theories 96–97 conditions for complexity 30–31; defining 30;
“choice of a social system” 291 implications of complexity for PPE 35–39;
choice problem 95–96 introduction to 28; three stages 28–30
choice processes 102 Complexity (1995) 29
Christiano, Tom 181 complexity, conditions for 30–31
Chudek, M. 113 complexity, defined 30–31
Cialdini, R. B. 112 Complexity, Governance & Networks (2014) 29
Cicero 420 “the complexity and opacity of politics” 36
“circumstances of justice” 290 complexity dynamics 29
Citizens United v. Federal Election Commission, Jan. complexity studies, institutionalization of 29
2010 176, 178–179, 181–182, 183n5 compressed economic inequality 300
Civil Rights Act 144, 252 computer simulation, use of 29
Clarke, Kevin 17 The Concept of Law (Hart) 153
class struggle 302 concepts vs. conceptions 19
class struggles, as history of society 48–49 condensation 20, 22
classical liberalism 19, 310n1 conditional preferences 115, 144
classical liberalism and the evolution of high liberal Condorcet, Nicolas 401
tradition 341–342 Condorcet Criterion 409–410
ClassicalI Iinstitutional Economics 150 Condorcet paradox 410
ClassicalI Iinstitutionism 151–152 Condorcet rules 413
Clayton, P. 30 Condorcet’s Jury Theorem 393–394
closed exchange, markets as 232 Condorcet-Voting-Paradox Style Cycle 84–86, 92
Coase, Ronald 151, 317 configural weight theories 102
coercion condition 53 conflict resolution principle 194
coercive state action 329 Connolly, T. 36–37
cognitive diversity 394 “consent of the governed” 322n6
Cohen, D. 30 consequences, moral advantages: enrichment 210;
Cohen, G. A. 153, 286n2, 349, 359–361 poverty and inequality 210–211; virtue 211–212
Colander, D. 36 consequences for society 433–435
Cole, Daniel 196 consequences for workers 432–433
Coleman, James 445 conspiracy theories 36
Coleman, Jules 154 Constant, Benjamin 341
collateral and monetary policy 294–295 constitutional rules 194
collective behavior, other kinds of 141–143 Constructivist Institutionalism 152
collective choice rules 193–194 consumers as bosses 282–283
“collective patterns of behavior” 114 contigency meanings 229
Collier, Paul 315 contractual rights 430–432
Collins, Patricia Hills 370 Contribution to the Critique of Hegel’s Philosophy of Right
“color blindness” 370 (Marx) 43
color theory 83 Convention (Schelling, 1969) 76
Committee for Economic Development 191 Cook, Karen 440
commodification 228 cooperation and social democracy 301
commodification of women’s body 61 cooperative system of labor 305
common beliefs and values 156 cooperatives 437n12
common law 168 coordination game 68–69
“common store of goods” 293 coordination problems 143
Commons, John 150 “core design principles” 196–197
Communist Manifesto (Marx and Engels) 48, 50 Corneo, Giacomo 286n7
453
Index
corporations 136 desirability of economic socialism 302

corporations and electoral politics 177–179 Detroit riots 252
corporations and religion 183n8 Dewey, John 23, 150
corporations in our polity: corporations and electoral dialectical idealism 45
politics 177–179; corporations and political protest Dick, David 236
181–182; corporations and religious freedom The Difference Between the Democritean and Epicurean
179–181; introduction to 176–177 Philosophy of Nature (Marx) 43
Correll, Joshua 125 difference principle 333, 350
corruption: defined 315–316; functional corruption Dinensen, P. T. 444
316–318; introduction to 314–316; limited access Dire Straits Proviso 338
offers 318–321 “direct preference aggregation” 18
corruption, defined 322n1 disciplines as certifiers 8–9
corruption and social trust 446 discounted utility 96
corvée-style scheme 172 discounted utility theory 103
cost-benefit analysis 229 discretion elimination 128–129
Coulthard, Glen 368–369 Discuriseve Institutionalism 152
COVID-19 36, 39, 58, 253, 373 dissolution of concentrations of capital 350
Cox, M. 197n2, 319–320, 440 distracted decision field theory 101
“creative destruction” 322n10 distributive justice 384
“the Creative Powers of a Free Civilization” (Hayek, distributive justice and social minimum 348–351
1960) 305 distributive justice, the mitigation of inequalities, and
criminal system biases 125–126 permissible inequalities 349–350
crisis for refugees 385 distributive paradigm of justice 365–367, 374
Critique of the German Ideology (Marx and Engels) 44 distributivist rule 333
crony capitalism is not capitalism 271–272 diversity, social trust and social cohesion 443–445
Crown-chartered municipal corporations 322n7 “diversity trumps ability” 394
Cudd, Ann 53 division of labor 153, 366
cultural imperialism 54–55 division of labor among citizens 423
cultural schemas 130n5 domestic violence literature 61n2
customs 141–143 Dow, Chales 227
Dowding, Keith 19
D Downhill Slide 89–90, 92n2
Dagger, Richard 292 Downs, Anthony 421
Dahl, Robert 280 DRIFT model 101
Dasgupta, P. 410 “drift to oligarchy” 290
David, P. A. 36 dual-systems model of affect and deliberation 101
de Beauvoir, Simone 54 dual-systems model of risk 101
decision centers 187 Dunn, J. 441
decision-making and implicit bias See implicit bias and Durlauf, S. N. 34
decision making; criminal system 125–126; duty of justice 354, 361
education 123–124; employment 124; medical Dweck, Carol S. 128
system 125 Dworkin, Ronald 349–351
decision-making power 366
deliberate norm creation and elimination 144 E
democratic control 259–260 Eberhardt, Jennifer L. 124
democratic firms See worker’s voices, justification for economic advantages: incentives 208; information
limiting 208–209
democratic firms and workers’ preferences 437n7 “economic” approach to politics 11
democratic planning 283 economic calculation, under socialism 208
democratic socialism 301, 307–308 economic citizens 280
democratically planned socialism 281 economic democracy, concept of 279–280
demogrants 289, 295–296, 431 “Economic Democracy” (ED for short) 277, 279,
Demsetz, Harold 29, 169 281–282
denisty 32 economic democracy in practice 280–282
Department of Education 253 economic democracy, reasons for 283–285
Department of Moral Merit Measurement 333 economic democracy without socialism 282–283
descriptive norms 112, 142 economic imperatives vs ethical imperatives 284
desert 213–214 economic liberties 344
454
Index
economic veto 287n18 excludability principle 194

The Economics of Control (Lerner, 1944) 271 “exclusionary zoning” 255
economists 3–4 exclusionary zoning and segregation 258–259
ED workers 286n15 “Existence” 46
Edgington, Dorothy 90 expanisve conception of public goods 342
Edmundson, William A. 286n4, 290 expected utility theory 96, 103, 215n146
education, right to 425 expenditure tax 292–293
education biases 123–124 “experimentalist governance” 36
effective action, relationship building, and self- exploitation 54–55, 366–367; introduction to
concept management 112 217–218; non-idealism and inter-transactional
“efficient rent-seeking” 320 parity 221–226; structuralism and transactionalism
egalitarian end-state theorist 332 218–221
Elements of the Philosophy of Right (Hegel) 45–46 exploitation of labor 61
Elster, Jon 20, 59–60 extensive form 68–69
Emergence: Complexity and Organization (1999) 29 externalities 166
emergent order, scientific community as 189
empirical evidence and norms 138 F
empirical expectations 138 fables, defined 22
employment biases 124 fables, models as 21
Employment Division v. Smith, 1990 179 Facebook and President Trump 182
“empower the base” 278 fair distribution of income and wealth 350
enabling social cooperation 306 fair distribution of work and its product 306–307
Encyclopedia of the Philosophical Sciences in Basic Outline Fair Housing Act, 1968 254
(Hegel) 46 fairness principle 194
“End of Laissez-Faire” Keynes (1926) 269 fake news stories and Russia 183n5
end-state and patterned principles 335 “Famine, Affluence, and Morality” (Singer,
end-state doctrines/patterned doctrines 331–333 1972) 384
enforced reparation payments 337–338 Fannie Mae (Federal National Mortgage
enforcement costs 151 Association) 294
Engles, Friedrich 43–44, 46–50, 252 Fanon, Frantz 367–368
enlightened preference voting 396–399 fat-tailed distributions 34
See epistocracy, defense of Federal Election Campaign Act (FECA) 177–178
enrichment 210 Federal Housing Authority (FHA) 254
“epistemic friction” 371 federal rent freeze 257
epistemic injustice 370–371 federalism 192–193
Epistemic Injustice (Fricker, 2007) 370 feminist theory: adaptive preferences and internalized
epistemic labor, disciplines and division of: disciplines oppression 59–61; capitalism and women’s
as certifiers 8–9; limits of specialization 7; origins of oppression 61–62; gender inequlaity and the
specilization in intellectual enquiry 5–7 oppression of women 53–56; gender of caring
The Epistemology of Resistance (Medina) 371 labor 56–59; introduction to 52
epistocracy, defense of: enlightened preference voting Ferguson, Adam 186–187
396–399; incompetent rule is unjust 395–396; Feuerbach, Ludwig 43, 50
introduction to 391; voter behavior 101, 391–393; Fichte, Johann 367
we don’t make it up in bulk 393–394 filtering 257
Epstein, J. M. 29, 38 Financial Crisis, 2008 428
Equal Protection and Due Process clauses of the “A Fine Is A Price” 110
Constitution 178 First Amendment’s Free Exercise clause 179
equality principle 242 First National Bank of Boston v. Bellotti 177–178
equilibria of rules in repeated games, institutions as 73 first theorem of welfare economics and its proof
“equilibrium selection” 110 215n11
Eriksson, L. 115–116 Fischels, William 255
Estlund, David 15n16 five complexities concerning property rights 337–339
Ethical Trading Initiative 373 “five faces” of oppression 54–55
Eubulides 90 Five Giants of Poverty, Ignorance, Squalor, Disease,
“everyone does it” morality 316–317 and Idleness 270
eviction 253 flexible family support policy 245
Eviction Lab 253 Fligstein, Neil 152
excessive conformity 119n3 Folbre, Nancy 57
455
Index
“folk theorems” 72, 110 gender group, defined 54

Ford, Thompson 127 gender inequlaity and the oppression of women
formal equality of opportunity 346 53–56
formal models 17, 22 gender socialization 241–242
formal vs. informal institutions 157 gendered division of labor, eroding 247–248
Forrester, Katrina 374 General Systems Theory of Bertalanffy (1968) 29
Forst, Rainer 359 generative entrenchment 34
Foster, J. 38, 125 “gentleman scholars” 6
Foucault, Michel 17, 19 “Geometry, the Majority Vote and the Power of
“foundational role” and institutions 363n8 Agenda Control” (Breuer) 24
four structures of intransitive preferences: Condorcet- George, Henry 174, 294
Voting-Paradox Style Cycle 84–86; introduction Georgist tradition, convergence with 293–294
to 81–82; Negligible-Value-Differences and The German Ideology (Marx and Engels) 43, 46–47
Missing-Values Cycles 82–84; Sen’s-Libertarian- Geroge, Henry 291
Paradox Style Cycles 86–88; Sorites Cycles 88–92 Gheaus, Anca 242
four-condition model 53–54 Gibbard-Satterthwaite theorem 18, 412
frame of justice 369 Giffen good 57, 61n1
Fraser, Nancy 243–244, 248–249, 369 Gigerenzer, G. 104
Fraser Institute 211 Gilabert, Pablo 285–286n1
“Fraud, corruption, vice, crime.becomes divorced Gilbert, Margaret 154, 157
from a coordinated institutional emphasis 322n3 Gilens, Martin 396
Freddie Mac (Federal Home Loan Mortgage Gimenez, Martha 61
Corporation) 294 Gini Index 215n14
Frederick, S. 104 Ginsburg, Ruth Bader (Justice) 183n1
Free, Equal, Cooperators 305 Gintis, Herbert 67
“free and equal moral persons” 351 Glaeser, Edward 255
“free information” 421 global climate change 195–197
free information and the right to participation, interest Global Financial Crisis (GFC) 35
in 424–426 global gender justice 383–384
Freyerabend, P. 31 global poverty 380
Fricker, Miranda 370–371 global warming 195
Friedman, Milton 214, 304–306 globalization 268, 384
Frye, Marilyn 53 Gochen, G. A. 285–286n1
Fukuyama, Francis 440 “the Goldilocks zone” 32
functional corruption 316–318 Goodin, R. E. 115–116, 221
functionalist school of social norms 138 Google 181
Google and Parler 182
G Gore, Al 402
Gaissmaier, R. 104 governance of metropolitan areas 189–192
Galison, P. 29 graduated sanctions 194
Gallie, W. B. 291 “gray” or “black” mrket 317–318
Gambetta, Diego 446 Great Depression 252
game theory 17, 193; bargaining and cooperative Great Enrichment 210
games 76–77; introduction to 67; rationality in “Great Enrichment” 274
games 69–76; strategy, problem of 67–69 “Great Enrichment” of the Industrial Revolution
games of conflict 70 268–269
game-theorectic models 21–23 “Great Escape” of Globalization 269
Gaus, G. F. 24n2 The Great Society 38
Gaus, Jerry 77 Greenwald, Anthony 128
Gauthier, David 71, 77 Greenwald, T. H. 341
Gehlbach, Hunter 128 Gyourko, Joseph 255
Gelbach, Scott 17
Gelman, A. 413 H
gender, as construct 54 Hagglund, M. 284, 287n19
gender and the division of labor: cost of changes Hall, Hall 260
248–249; introduction to 239–240; issues with Halperin, Eran 128
240–242; social policy solutions 242–244; what we Haranyi, John 76
may do about it 244–248 Hardin, G. 31, 91
456
Index
Hardin, Russell 440 Hooker, C. 29

harm condition 53 Hooks, Bell 370
Harré, H. 113 Horan, S. 410
Harrington, Michael 278 household labor and childcare 239
Harrison, Jonathan 91 housing, taking off the market 260
Harsanyi, John 74, 78n8 “housing affordability crisis” 255
Hart, H. L. A. 116, 119n4, 153–154 housing markets: democratic control 259–260;
Hartley, Christie 247 exclusionary zoning and segregation 258–259;
Hart’s rules 116 injustices in housing 253–254; introduction to 252;
Harvey, David 35, 260 price controls 257–258; taking housing off the
Haslanger, Sally 127, 130n5 market 260; upzoning: defeat NIMBYS 255–256;
Hausman, Daniel 20 upzoning: housing costs 256–257; upzoning:
Hayek, F. A. 19, 29, 31–32, 34–35, 37–38, 108, 116, removing regulations 254–255
173, 189–190, 208, 214, 273–274, 293, How Innovation Works (Ridley, 2021) 267, 271
304–306, 348 Huber, J. 260
He, L. 104 HUD Act, 1968 254
Head, B. 37 Human Action (Mises) 274
Hegel, G. W. F. 367 See Marx’s materialist humanity as “species being” 310n8
conception of history Hume, David 116, 169, 304–306, 341–342
Henrich, J. 113 Humphreys, Macartan 22–23
hermeneutical injustice 371 “hunkering down” 444
heuristic theories 100 Hurricane Katrina 208
high liberal conception of public goods 351–352 hybrid systems 286n5
high liberal substantive equality of opportunity 346 hyperbolic discounting 98, 103
“high liberalism” 310n1
high liberalism 341–342, 345–346, 353n1; basic right I
and liberties 342–345; classical liberalism and the IAT 129n1
evolution of high liberal tradition 341–342; “ideal theory is an impossible enterprise” 37
distributive justice and social minimum 348–351; “ideal theory” of distributive justice 366
high liberal conception of public goods 351–352; ideal vs. non-ideal theory 12–13
public nature of political power and political “identity economics” 112
justification through public reasons 352; regulated immigration policy 152
market economy 347–348; substantive equality of immunity of expropriation 165, 174n1
opportunities 346–347 imperfectly competitive markets 215n4, 215n5
“highest value,” speech for political causes 181 impersonality of traditional market exchanges 232
historical entitlement 331–333 implicit anti-fat biases 125
“historical entitlement” view of justice inholdings 329 implicit bias and decision making: bias 121–122;
Historical Institutionalism 152 defined 122–123; implicit bias and decision
historical materialism 44 making 123–126; importance of implicit bias
historical materialism or determinism 302 126–127; interventions on implicit bias 127–129
Hobbes, Thomas 304–306 implicit bias, defined 122–123
Hobby Lobby 182 implicit bias, importance of 126–127
Hockett, Robert 294 implicit bias, interventions on 127–129
Holcombe, Randall 271 implicit social bias 121–122
“hold up” taxes 320 impossibility theorem 9–10, 18, 20–21, 25n10, 77,
Holland, J. 29, 36 408–409
Holly Lobby 183n6 impure semiotic objections 228
Holt, Charles 78n5 incentive-compatibility, market characteristics 233
The Holy Family (Marx) 49 incentives vs. motives 10
homelessness, measurements 253 “incidental detail” 21
Homer-Dixon, T. 35 “incidents” of ownership 164
“homevoter hypothesis” 255 incompetent rule is unjust 395–396
Homo Economicus 10 increasing returns 34
homophily 444–445 Independence of Irrelevant Alternatives 77
Hong-Page Theorem 393–394 independent behaviors 141–142
Honneth, Axel 369 India, corruption in, example 315
Honoré, A. M. 430 indication 231
Honoré, Tony 164 individualized behaviors 141–142
457
Index
Industrial Revolution 428 intellectual curiosity 6–7

“Industrial Revolution” 274 intellectual enquiry 6
inequality, question of 169–174 intellectual ill 44
inequality and social trust 445 interactions between options 99–100
inequality trap 446 interactions within options 98–99
inevitable immorality of capitalism 302 interdependent behaviors 141–142
information, wages as 208–209 interdependent behaviors, social norms as 141
information and citizen knowledge 391–392 interdisciplinarity 14n1
informational power 417–418 interdisciplinary work, general case for 3–5
informational power, distribution of 418 interfaces 33
informed and equal participation, enabling: basic interiorization of rules and norms 112–113
rights to participate 416–417; distribution of internalized oppression 59–61
informational power 418–419; division of labor International Labour Office (ILO) 57
among citizens 423; equality in transmission of International Monetary Fund (IMF) 384
information 423–424; informational power internationalism 302
417–418; interest in free information and right to interpretive models as fables 21–22
participation 424–426; rational participation intertemporal choice 94, 100
420–421; social conditions of sophistication inter-transactional fairness 224
421–423 inter-transactional vs. intra-transactional 222
initial acquisition 329–331 interventionist economic policy 300
injunctive norm 112 intransitive preference relation 88
injustices in housing 253–254 “invisible hand” 269–270, 285, 349
An Inquiry into the Nature and Causes of the Wealth of Ipsos Mori 391
Nations (Smith, 1776) 15n6, 269, 271 Israel 181
instant runoff voting 411 ITCH model 101–102
“institutional constraint on self-determination” 366 “Iteration Facilitation Board” 286n12
Institutional Economics 150–151
institutional economics 150–151 J
“institutional fact” 154 Jackson, Elton F. 439
institutional fundamentalism 357–359 Jacoby-Senghor, Drew S. 123
institutional ideal 355–357 Jaggar, Alison 384
institutional inequalities 217 Janlert, L.-E. 36
institutional justice, primacy of 357 Jaworski, Peter 229–233, 235
institutional pathologies 316–318 Jervis, R. 35
institutional political science 151–152 Jim Crow South 371
institutional responses 363n10 Johnson, Cathleen 78n5, 78n6
institutional responsibilities 363n6 Johnson, E. J. 104
institutionalism: institutional economics 150–151; Johnston, Michael 318, 322n4
institutional political science 151–152; Jonker, Julian 231
institutionalism in political and moral philosophy Jordahl, Henrik 445
153; legal institutionalism 153–154 Journal of Complexity (1985) 29
institutionalism, injustice, and personal responsibility: judgment and decision-making 95
evading the burdens of justice 359–361; just initial acquisition 331
institutional fundamentalism 357–359; institutional Just Restitution 337–338
ideal 355–357; introduction to 354–355; justice 212
reparations and unjust institutions 361–362 justice, natural duty of 360
“institutionalists” 136 justice across borders 381–382; foundational
institutions, defining: choosing what an institution is arguments 378–380; global gender justice
159; common beliefs and values 156; different 383–384; from the Good Samaritan to institutional
conceptions of rules 156–157; formal vs. informal interconnection 380–381; introduction to
institutions 157–159; introduction to 154–155; 377–378; refugee crisis 384–386; structural
patterns of behavior vs. rules 155–156 injustice 381–383
institutions and institutionalism: defining institutions justice and merit 310n17, 355–357
154–159; institutionalism 150–154; introduction Justice and the Politics of Difference (Young) 54–55
to 149–150; reasonable consensus 159–161 Justice and the Politics of Difference (Young, 199) 365
instrumental perspective 205 Justice as Fairness: A Re-Statement (Rawls, 2001) 294
intellectual activity and specialization 15n5
458
Index
K “Legalize Love” campaign 181

Kalai, Ehud 77 Lerner, Abba 271
Kalai-Smorodinsky solution (Kalai and Smorodinsky, Lessig, Lawrence 144
1975) 77 Letki, Natalia 443
Kaldor, Nicholas 292–293 “letters patent 319
Kamm, Francis 83–84 Levi, Margaret 440, 442
Kant, Emmanuel 217, 341–342, 356, 363n5, 420 Levin, S. 34
Kardos, Péter 128 Levinson, Justin D. 126
Kates, Michael 363n2 Levontin, Liat 128
Katz, J.N. 413 Lewis, David 76, 154
Kauffman, S. A. 34 liberal egalitarians 285–286n1
Keele, Luke J. 443 liberal egalitarians/left liberals 277–278
Kelly, Daniel 127 liberal significance of politcal equality and integrity of
Kelly, J. S. 410 democracy 345–346
Kersch, Ken 152 liberalism 273
Kesler, Christel 443, 445 “liberties of the moderns”/”liberties of the ancients”
Kessler, Judd 122 341–342
key behavioral properties: choice processes 102; liberty and patterns 333–335
interactions between options 99–100; interactions Liberty Upsets Patterns, Nozick 334
within options 98–99; transformations of values “limited access order” 316
100–101; transformations of weights 101–102 limited access orders 318–321
Keynes, John Maynard 270–271, 275, 297, 304 limited access orders/open-access orders 319
Keynesianism “road not travelled” 297 limited liability 180
Kézdi,Gábor 128 Limited Liability Corporations, in housing markets
Khader, S. 59–61 260–261
Kickstarter 196 Lindblom, C. 29, 37
King, David C. 442 List, Christian 18
Kingdom of Necessity - Kingdom of Freedom 274 Livernois, Rebecca 91–92
Knight, Frank 214 Lloyd, S. A. 242, 246
Kogelmann, Brian 145 Loasby, B. J. 29, 33
Kok-Chor Tan 153 local fit principle 194–196
Koput, K. 36–37 Locke, John 165, 167–168, 170, 174n1, 174n3, 217,
Kranton 113 329–331, 336, 338, 341–342
Kuhlmann, D. 256 Locke on just initial acquisition 329–331
Kuhn, T. S. 24n1, 30 Logic of Collective Action (Olson, 1965) 194
Kuhnian philosophy of science 34 logic of strategic choice 77
Kundera, Milan 24n4, 24n5 logic of strategic rationality 69
Kuran, Timur 140–141 Lomasky, Loren 213
Kymlicka, W. 217 “long jump” exploratory modulation 34
Low, Connie 122
L “luck egalitarianism” 371
La Farrara, E. 441
labor incentives 292 M
labor market, assumptions of 247–248 MacPhillamy, Paul 83
labor-mixing theory of just transfers 331 macro level wrong, exploitation as 218
laissez faire 215n12 macro-economic modelling 38
“laissez faire activism” 36–37 Madva, Alexa 127
Landermore, Hélène 394 Maina, Ivy W. 124
Landsburg, S. E. 109 Majone, G. 36
Lane, R. 28 majority rule 85, 402
Laruelle, A. 413 majority rule, general indeterminacy 20
Leave and Remain voters 391 Malleson, T. 277, 279–280, 285, 286n10, 286n16
Leboeuf, V. 128 Malthus, Thomas Robert 269, 275
Lefebvre, Henri 259 mandating workplace democracy 436n4
left-liberals 309 Mansbridge, Jane J. 442
left-libertarianism 173–174, 213 March, J. G. 34
“legal conventionalism” 154 March, James 152
legal institutionalism 153–154 Margalit, Avishai 83
459
Index
marginalization 54–55, 366–367 measurement costs 151

market distribution 300 medical system 125
market failure 209 “medodology of game theory” 18
market freedom 182 Meirowitz, Adam 17
market meaning, case against 229–231 Melenovsky, C. M. 153
market norms 212 Menakem, Resmaa 128
market production for profit vs. planned production Mencken, H. L. 39
for use 279 mercantilism 271
market socialism 281, 285 Mercantilist system 269
markets 186 meritocracy and legal equality 61
markets, advantages of: economic advantages Mershon, Carol 17
207–209; introduction to 205; markets and market Merton, Robert K. 317–318, 322n3
economies 206–207; moral advantages 210–214 Meshelski, Kristina 293, 296
markets, meaning of: case against market meaning metaconstitutional rules/norms 194
228–231; case for (contingent) market meaning “methodological approaches” 18
231–234; semiotic complaints, overview of “methodological individualism” 13
227–228; signaling overlap in market exchange microeconomics, standard models in 17
234–235 Mill, John Stuart 29, 294, 314, 341–342, 348,
markets and inequality 211 351, 428
markets and market economies 206–207 Mills, Charles W. 370, 374
markets and profit-seeking enterprises 281 Minimal Libertarianism 88, 92n1
Markose, S. M. 38 Minimax 71
Marmor, Andrew 154 minority rule, voters 413n2
Marshall, John (Chief Justice) 180 Miracle of Aggregation 393
Martín, Annette 127, 130n6 “The Mirage of Social Justice” (Hayek) 293
For Marx (Althusser) 43 Mises, Ludwig von 190, 208, 274, 282–283
Marx, Karl 55, 61, 217–218, 227, 229, 233, 260, 267, Mishra, Satyananda 317
269, 274–276, 278, 284, 290, 292, 310n8 Mislin, A. 78n6
See Marx’s materialist conception of history missing-values structure 83
Marx and the proletarian standpoint 48–50 Mitchell, Wesley 150
Marxism 44, 302 mitigation of economic inequalities 349
Marxist feminist critique of capitalism 61n3 mixed strategies 70
Marxists 284 mobile phones, evolution of 34
Marx’s materialist conception of history: introduction model, defined 18–20
to 43–45; Marx and the proletarian standpoint models, conceptual purposes of 20
48–50; philosophy and history in Marx and Hegel models and their uses: defined 18–20; fiction, fables,
45–48 models 20–22; introduction to 17–18; moral of the
Maskin, E. 410 story 22–23; using models, public places 23–24
Massey, D. S. 258 models in public places, using 23–24
master/slave dialectic 368 modularity 32–33
material inquality 170 money and moral rights 171–172
“the materialist conception of history” 43 moral advantages: consequences 210–212;
materialist conception of history 49 introduction to 209–210; rights 212–214
Maughan, T. 36 moral and economic limits, of markets 205
May, Kenneth 85 moral error 380
May’s Theorem 404 moral norms 142–144
McCarty, Nolan 17 moral of the story 22–23
McCloskey, Deirdre 210, 274 moral preferences 81
McConnell v. FEC, 2003 177–178 morality 316–317
McCormick, Neil 154 Morgenstern, Oskar 68
McGinnis, Michael 197n1 Moriarty, Jeffery 229–230
McKelvey, B. 23–24, 25n10, 32 Mosley, Walter 20–21
McKelvey-Schofield models 20 motivation, worse case and best case assumptions
McKeown, M. 218 15n10
McLean, Ian 24n1 motivational assumptions 10–11
Meade, James 292, 297, 348 motives vs. incentives 10
meaning of markets 230–231 “Movement to Amend” 178
“means of production” 278
460
Index
Movement to Boycott, Divest and Sanction Israel No Way Up 89

(BDS) 181 noisy retrieval model 101
Mruphy, Liam 213 non-capitalist workplaces, what not to expect
Muldoon, Ryan 77, 144–145 435–436
Mullainathan, Sendhil 130n2 Non-Diminished Opportunity Proviso 338
multiattribute choice 94 non-ideal vs. ideal theory 12–13
multi-candidate elections: Condorcet Criterion non-idealism and inter-transactional parity 222–226
409–410; impossibility theorems 408–409; instant non-market criterion of distributive justice 342
runoff voting 411; Nash Independence of non-market system 209
Irrelevant Alternatives (NIIA) 407; plurality rule non-reductive explanation, rules-following 115–116
405; ranked voting methods 405–408 non-worseness claims (NWC) 220–222
multi-national corporations (MNCs) 373 “Nordic exceptionalism” 445
Munger, Michael 271, 317 norm dynamics 144–146
Murphy, Liam 153, 171–173, 357 norm elimination 144–145
My Grandmother’s Hands (Menakem 2017) 128 “Norm entrepreneurs” 145
Myerson, Roger 18, 22–24, 25n8, 25n9 “norm psychology” 113–114
normal form 68–69
N “normalizing” assumptions, macroeconomics 31
NAACP v. Alabama 183n9 normative considerrations 11–12
Nachbar, Thomas B. 322n7 “normative desirability” 11
Nader, Ralph 402 normative expectations 138, 140
Nagel, Thomas 153, 171–173, 213 normativity of law 154
Nash, John 20, 70–71, 76 norms 116–118, 119n2
Nash Equilibrium 70–71, 74–75, 110, 138 North, Douglas 151, 316–317, 319–320, 322n9
Nash Independence of Irrelevant Alternatives (NIIA) Norwegian Government’s Pension Fund 294
408, 410 novelty 33
Nash solution 76, 78n9 Nozick, Robert 167, 170, 212, 214, 289, 329, 331,
National Assocation for the Advancement of Colored 333–336, 338
People (NAACP) 254 Nussbaum, Martha 242, 246, 372
National Popular Vote 402 Nye, J. 322n1
“natural person” 183n4
natural right of person 329–331 O
natural right of property 336–337 Obama, Barack 255
Negligible-Value-Differences and Missing-Values Obergefell v. Hodges 181
Cycles 82–84, 92 Ocasio-Cortez, Alexandria 276
Neoclassical Economics 151 Okin, Susan Moller 58, 240–241, 384
“neo-institutionalist” method 152 Okonofua, Jason A. 124
Neo-Keynesianism 297 Olsen, Johan 152
nested enterprises principle 194 Olson, Mancur 194
Network Institutionalism 152 open-access order vs. closed access order 316
“network of practices” 130n5 operational rules 193
neuroeconomics 95 oppression 366
neutrality 403 “Oppression” (Frye, 1983) 53
“New Class” 273 oppression, explaining 126
New Deal 416 oppression of women 53
“New Economics” 271 “option luck” 351
New Institutionalism 151–152 order 31–32
“The New Institutionalism: Organizational Factors in Organization for Economic Cooperation and
Political Life” (Olsen, 1984) 152 Development (OECD) 240
“new left” social movements 365 Original Position 19
“new methodological tools” 31 Osborne, M. J. 410
New Right 296 Ostrom, Elinor 74, 151, 158, 187, 189, 191–193,
New York Magazine 276 196–197, 197n1, 440
New York Times analysis (gendered labor) 239–240 Ostrom, Vincent 189–190, 192–195
NIMBYS, defeating 255–256 Otsuka, Michael 174
Nine, Cara 260 ownership and control, workplaces 437n6
9/11 35 owning fruits of labors 174n3
Nixon, Richard 257 ownship, defined 167–169
461
Index
P political liberties 343

Page, S. E. 18–19, 31, 34 political protest and corporations 181–182
Paine, Thomas 289 political representation principle 194
Palestinians 181 political scientists 3–4
Panopticon, The 19 political theorist on conceptual analysis 24n2
paradigm institutions 154 political-economic inequality 24
“paradox of caring labor” 57 politically actionable injustice 246
Paradox of the Heap 90 politics of recognition 368
Parecon app 281, 286n11, 286n12 Polyanyi, Michael 188–190
Pareto efficiency for symmetry 74 polycentricity: brief history of concept 188–193;
Pareto optimal outcome 78n5 introduction to 186–188; practical theory of self-
Pareto optimal points 76–77 governance 193–197
Pareto optimality 71 polycentricity, brief history of: decentralized
Pareto-efficiency or dynamic efficiency 215n10 production of knowledge in marketsand science
Parijs, Philippe Van 295–297 188–189; federalism 192–193; governance of
Parijs, Philippe van 349 metropolitan areas 189–192
Parks, Rosa 371 Popper, Karl 20
Parsons, T. 138 “popular sovereignty model” 391–392
participation, basic rights to 416–417 Portes, Alejandro 444
Pateman, Carole 61 “positive network externalities” 34
path-dependency 32 positive responsiveness 403–404
patterns of behavior vs. rules 154–155 Postema, Gerald 154
Payne, J. W. 99–100 poverty and inequality 210–211
Paypal 181 power-law distribution 34
Peltzman, S. 109 powerlessness 54–55
perfect equality of opportunities for welfare 346 PPE: as “counter-cultural” activity 14; economics and
perfect equality of opportunity 346 social sciences 135; and global citizens 384;
periodic crises 302 implications and completity for 35–39;
permissionless innovation 320 interdisciplinary understanding 17
Pérotin, V. 433 PPE, implications of complexity for: and economics
Perry, Stephen 116 38; in policy-making 36–37; and political
personal identity and earing a living 213 philosophy 37–38; in political science 35–36
personal liberties 343 PPE as intellectual enterprise: disciplines and division
personal norm 112 of epistemic labor 5–9; interdisciplinary work,
“personal-cost limitation” 360 general case for 3–5; shadows and conflicts 9–14
persons and corporations 183n4 practical reasoning, game-theorectic models as 22
Pew Charitable Trust 240 preferences
Pharr, S. J. 446 “preferences that are adaptations to and for
Phelan, Julie E. 124 oppression” (PAOs) 60–61
philosophers 3–4 “preference-satisfying” 12
philosophy and history in Marx and Hegel 45–48 presumption of value pluralism 356
Piaget, Jean 119n2 price controls 257–258
Pierpaolo, A. 29 primacy of economics 302
Pierson, P. 36, 152 Primo, David 17
Pierson v Post 168 principle of basic needs 342–345
Piketty, Thomas 271, 289–290, 297 Prisoner’s Dilemma 18, 71–72, 74, 91–92, 193
pivotality probability 413 private property 273
Plato 321 private property, importance of 165–166
pluralism of good 356 private property system 163
pluralistic ignorance 140–141 Private Truths and Public Lies (Kuran, 1995) 141
plurality rule 405 “privatizing the air above” 260
Pogge, Thomas 380–381, 383–384 privilege condition 53
“point in time” count (PIT), unsheltered homeless, “The Problem of Social Cost” (Coase, 1944) 151
annual 253 process of thinking 46
Polanyi, Karl 284 producer democracy 280
political and moral philosophy, institutionalism 153 productivity thesis 229
Political Liberalism (Rawls) 334 professional reputation 8
political libertarianism 329 profoundly undemocratic nature, capitalism as 278
462
Index
Progressive Era 271 363n3, 363n4, 365, 371–372, 374n1, 379–380,

proletarian movement 49–50 383–385, 386n1
“propensity to truck, barter, and exchange” (Smith) 6 Reagan, Ronald 297
property: introduction to 163; private property, reasonable consensus 159–161
importance of 165–166; property rights and reciprocity norms 139
externalities 166–167; question of inequality redistribution, alternative to 295–297
169–174; what are property rights 163–165; what redistribution and property rights 335–336
do we own? 167–169 “Redistribution and Property Rights” (Nozick,
property rights 163–165, 213, 363n5, 429–430 167–74) 335–336
property rights and externalities 166–167 redistributive egalitarianism 289
property rights and justice in holdings: five redistributive taxation 174
complexities concerning property rights 337–339; Reed, M. H. 35
historical entitlement 331–333; introduction to reference networks 139
329; liberty and patterns 333–335; Locke on just refugee crisis 384–386
initial acquisition 329–331; natural right of regulated bargaining 300
property 336–337; redistribution and property regulated capitalism 300
rights 335–336 regulated market economy 347–348
property-owning democracy: alternative strategies: “Regulators,” North Carolina 319
collateral and monetary policy 294–295; alternative rejection of bourgeois politics 302
to redistribution 295–297; convergence with rejection of dogmas of early social democracy 303
Georgist Tradition 293–294; importance of Rawl’s relational egalitarianism 371–372
contribution 291–293; introduction to 289–291 relational matter, justice as 359
“proportionality principle” 320 religious freedom and corporations 179–181
prospect theory 95, 97–98, 103 Religious Freedom Restoration Act (RFRA) 179
prospective reference theory 101 relocation 431
public (collective) character of justice 356 rent control 257, 259
public lawgiving 356 rent-seeking 322n11
public nature of political power and political reparations and unjust institutions 361–362
justification through public reasons 352 Republic (Plato) 321
public political justification of laws 342 Rescher, N. 28, 31, 39
public political power 342 Respiratory Distress Syndrome (RDS), example 223
pure procedural justice 348 resume field studies 122–123
pure strategies 70, 78n4 retrospective voting 393
Putnam, Robert 441–443, 445 revolution, not reform 302
rewards 215n6
Q RFRA, senate votes 183n7
Quinn, Warren 90 Richerson, P. J. 29
Quinn’s Self-Torturer 91 Richey, S. 446
Ridley, Matt 267, 271
R right to capital 291–294
Race IAT data 125 right to ownership 430
Rachels, Stuart 90 “right to the city” 259
Rachel’s Ecstasy 92n2 right-liberals 304–305
racial biases 123–126 rights, property and contract 212–213
racism inpolicing and police murders 125 rights, labeling as conventional 172–173
Radin, Margaret 227 risky choice 94
Rana Plaza disaster, 2013 383 risky choice theories 101
random dictatorship 413 Rittel, H. 29
random voting model 412–413 The Road to Serfdom (Hayek) 274
ranked voting methods 408–409 Robbins, Lionel 273
Rational Choice Institutionalists 152 Roberts, W. C. 284
“rational choice” model 10 Robeyns,Ingrid 242
rational choice rule-following 109–110 Robinson, James 165–166
rational participation 420–421 Robinson, Joan 297
Rawls, John 17, 19, 21, 31, 77, 153–154, 214, Robinson, Robert V. 439
285–286n1, 289–291, 293, 295, 297, 308–309, Rodrik, Dani 17, 22, 24n3
311n28, 333–335, 348, 350–351, 354–355, 360, Roemer, John 218, 285–286n1, 349
“role-rule” model of social behavior 113
463
Index
Rolfe, Robert 445 self-authorship 213

Rose, Jacqueline 369 self-governance practical theory 193–195
Rothstein, R. 258 “self-governing enterprises” 280
Rothwell, J. 258 “self-organized criticality” 34
Rousseau, Jean-Jacques 269, 275, 304–306, Selten, Reinhard 74
363n5, 420 semiotic complaints, overview of 227–228
“roving bandits”/”stationary bandits” 318–319 semiotic costs 234
Roy, Ananya 260 semiotic critics 236
Rubinstein, Ariel 20, 77 Sen, Amartya 210, 357, 363n8, 372
Rudman, Laurie A. 124 Sen’s-Libertarian-Paradox Style Cycles 86–88, 92
rule 117–118 shadows and conflicts 11–12; categorical vs.
rule-following: introduction to 108–109; comparative thinking 9–10; discounting for time
psychological accounts of rule-following 111–114; 12; ideal vs. non-ideal theory 12–13; introduction
rules and norms in philosophy 114–116; three to 9; methodology and philosophy of science
economic perspectives on 109–111 13–14; motivational assumptions 10–11; motives
rule-following, evolutionary accounts of 113–114 vs. incentives 10
rule-following, psychological accounts: effective shared politcal values 352
action, relationship building, and self-concept shared public reasons 342
management 112; evolutionary accounts of rule- Shelley v. Kramer 254
following 111; interiorization of rules and norms Sheltonm, J. Nicole 123
112–113; introduction to 111–112; “role-rule” Sherif, M. 112
model of social behavior 113 Shils, E. A. 138
rule-following, three economic perspectives: Shvetsova, Olga 17
“behavioral” rule-following 111; rational choice Sidgwick, Henry 215n12
rule-following 109–110; rule-following in signal overlap 234
equilibrium 110–111 “signal wrapped in an incentive” 189
rule-following equilibrium 110–111 signaling overlap in market exchange 234–235
rules, different conceptions of 156–157 “signals wrapped in an incentive” 186
rules and norms in philosophy: conditional Simon, H. 29
preferences 114–115; non-reductive explanation Simonovits, Gábor 128
115–116 Sinclair, Stacey 123
“rules of justice” 116 Singer, Peter 227, 379–380, 383–385
“a single science of history” 50
S single-payer system of healthcare 181
Sample, R. 221 Skocpol, Theda 152
Samuelson, Paul 270 Skowronek, Stephen 152
Sandel, Michael 227 Skyrms, Brian 74
Sander, Bernie 276 Slantchev, Branislav 25n8
Santa Clara County v. Southern Pacific R. Co (1886) 178 Slovic, Paul 83
Santa Fe Institute 29 “slum clearance” initiatives 254
Sanver, M. R. 410 “smart phones” 34
Satz, Debra 227–228 Smith, Adam 5–9, 11, 15n4, 15n6, 15n7, 31, 116,
Scanlon, Thomas 296 211, 213, 215n146, 233, 269–271, 273, 284–285,
Schafran, Alex 260 304–306, 322n5, 341–342, 349
Scheffler, Samuel 153 Smith, Joan 371
Schelling, Thomas 18, 20–21, 23, 25n8, 29, 76, 78n8 Smith, Matthew 260
Schmidtz, David 78n5 Smith, Vernon 78n1
Schumm, George F. 82–84 Snyder, J. 221
Schumpeter, J. 38 social arrangments, gendered assumptions 247–248
Schweickart, D. 277, 281–282, 285 social choice framework 9–10
science, methodology and philosophy of 13–14 social choice theory 17
The Science of Logic (Hegel) 46 social control over investment 281–282
“scientific socialism” 49 “social democracies” 211
Searle, John 154 social democracy: historical sketch 302–303;
Second Treatise (Locke) 336 introduction to 300–301; remarks on method
Secord, P. F. 113 301–302; social democracy vs. left-liberalism
Section 8 program 254 308–310; social democracy vs.democratic socialism
segregation and single family housing 258 307–308; as theory of justice 303–307
464
Index
social democracy, properties of 300 state control of the economy 279

social democracy vs. democratic socialism 307–308 “State of Fashion 2019” (McKinsey) 373
social democracy vs. left-liberalism 308–310 statism 279
social democrats 304 “Statue of Monopolies” 319
social expectations 139 Steiner, H. 174
social group condition 53 Stigler, George 13
social justice: critique of the distributive paradigm Stiltz, Anna 260
365–367; epistemic injustice 370–371; recognition Stoker, Laura 442
367–369; relational egalitarianism 371–372; Stolterman, E. 36
structural injustice 372–374 strategic incentives, voting rules 411–412
social learning 143 strategy, problem of: equilibrium selection 74–76;
social minimum 349 introduction to 67–69; Prisoner’s Dilemma 71–72;
social norms: defining 138–141; introduction to rationality in games 69–71; repeated games and
137–138; norm dynamics 144–146; other kinds of backwards induction 71–74
collective behavior 141–143; social norms and structural injustice 372–374, 381–383
social learning 143–144 structural justice 363n3
social norms 127 structuralism and transactionalism 218–221
social norms and social learning 143–144 structuralists 217
social norms, structure of 138 sub-game perfect equilibrium (SPE) 72–73
social parasitism, exploitation as 218 subsidiarity principle 194
social policy costs and benefits 245 substantive or fair equal opportunities for all citizens
social rules 157 342, 346–347
social structures 130n5 “successive approximation” method of policy
social trust: introduction to 439–440; trust, economy, development 37
and governance 440–443; trust in society and Sullivan, Colin 122
institutions 443–446 supermajority 402
socialism, what it isn’t 279 2/3 supermajority with status quo x 402
socialism as economic system 278–279 surrogacy, example 223–225
socialism as political philosophy 277–278 Sutcliffe, Peter 370–371
Socialism: Utopian and Scientific (Engels) 49 symmetry assumption 78n8
socialisms: economic democracy, concept of
279–280; economic democracy in practice T
280–282; economic democracy without socialism Tallying heuristic 98
282–283; introduction to 276–277; socialism, what tallying heuristic 98, 103
it isn’t 279; socialism as economic system 278–279; Tarko, Vlad 197n1
socialism as public policy 277–278; why economic tax-and-transfer schemes 170
democracy 283–285 taxation 170–172
socialist liberal egalitarian 286n2 taxation under socialism 279
socially-caused injustice 372–373 Taylor, Charles 368, 431
society and institutions, trust in: corruption and low Taylor, Keeanga-Yamahtta 259
social trust 446; inequality and social trust 445; Temkin, Larry 90–91
introduction to 443–445 tensions between philosophy, politics, and
Sonderskov, K. M. 444 economics 8
sophistication, social conditions 421–423 Terlazzo, Rosa 60–61
Sorites Cycles 88–92 Thatcher, Margaret 295, 297
Southwood, N. 115–116 The Theory of Economic Policy in British Classical Political
Soysal, Yasemin 152 Economy (Robbins, 1952) 272–273
Sparks, Jacob 227–228 theory of justice 301
“spatial contract” 260 A Theory of Justice (Rawls) 31, 153
spatial fix 260 theory of justice, social democracy as: first principle:
specialization, costs of 7 enabling social cooperation 306; introduction to
specialization and division of labor 14n3 303–304; second principle: a fair distribution of
Spirit 46 work and it’s product 306–307; working together
Spivak, Gayatri 370 304–306
spoiler effect 405 theory of rational choice 81
stability 33 theory of social norms 136
standard theories 96–97 third wold women 384
Starmer, C. 104 Thompson, Abigail 394
465
Index
three stages (of complexity): introduction of idea 29; US Department of Housing and Urban Development
systematic thinking 29; theoretical modelling (HUD) 253
29–30 US Electoral College 402
Tiebout, Charles M. 187, 189–190, 195 “usefulness” of intellectual work 6
Tillman Act 177 Uslaner, Eric M. 445–446
time, discounting for 12 utility theory 95
tipping-points 34 Utopophobia (David Estlund) 15n16
Tirole, J. 111
Tit-for-Tat 72 V
Tolsma, Jochem 444–445 Valenciano, F. 413
Tomasi, John 213 Vallier, Kevin 145
trade-off model 102 value and facts, distinction between 13
Tragedy of the Commons 91 value pluralism 356
transaction costs 151 van den Bergh, Linda 124
transactional parity and non-idealism 222–226 Van der Meer, Tom 444–445
transactionalistm and structuralism 218–221 Vanderschraaf, Peter 76–77
transactionalists 217 “vanity of the philosopher” 15n7
transformations of values 100–101 Veblen, Thorsten 150
transformations of weights 101–102 veil of ignorance method (A Theory of Justice) 31
transmission of informtion, equality in 423–424 Vickstrom, Erik 444
trickle-up effects 355 Villamayor Tomás, Sergio 197n2
Trost, M. R. 112 violence 54–56
Trounstine, J. 258 virtue 211–212
Trump, Donald 177, 255, 276 Voluntary Joint-Ownership 338
trust, economy, and governance: collective action and von Neumann, John 68
social trust 440; economic development and social voter behavior 101, 391–393
trust 440; trust and government performance voter permutation 402
441–443 voter transposition 402
trust and worthiness 78n6 voting power 412
trust in commercial society 8 voting rules 402; binary elections 401–404;
Trust: The Social Virtues and the Creation of introduction to 401; multi-candidate elections
Prosperity (Fukuyama, 1995) 440–441 405–411; strategic incentives 411–413; voting
Truth, Sojourner 370 power 412–413
Tucker, Benjamin 215n7 Vrousalis, Nicholas 218
Tullock, Gordon 152, 194, 315, 320, 322n4
‘turbo capitalism” 298 W
Tversky, A. 89–90 Waite, Morrison (Chief Justice) 178
Twitter and President Trump 182 Wallace, David 394
Wallis, John 316
U Walzer, Michael 227
Ullman-Margalit, Edna 138 War on Drugs 191
Ultimatum Game 211–212 Warren, Elizabeth 255
unanimity 410 Warren, Josiah 215n7
unconditional basic income 431 Warren, Robert 189–190, 195
unconditional social security 300 Watson, Lori 247
unfairness of outcome 224 Watt riots 252
“uniformity assumption” 31 weak anonymity 404
Unions 419 Wealth of Nations (Smith) 5–8
unions and workers’ legal rights 437n5 wealth tax on assets 289
“universal breadwinner” model of gender equity 244 wealthy and poor countries, relationship between 380
universality 49 weathering hypothesis 130n3
unpaid care 60 See also gender of caring labor Weaver, W. 29
upzoning: defeat NIMBYS 255–256 Web of Science (WoS) publication data 29
upzoning: housing costs 256–257 Webb, Beatrice 274
upzoning: removing regulations 254–255 Webb, Sidney 274
Urken, Adam 24n1 Webber, M. 29
US Census Bureau 252 Weber, E. U. 104
Weber, Max 322n6
466
Index
Weick, K. 29 from consequences 432–435; arguments from

weighted additive model 96, 103 rights 429–432; introduction to 428–429; non-
Weingast, Barry 316, 319–320 capitalists workplaces, what not to expect 435–436
welfare state regime 249 working together, theory of justice 304–306
welfare-state capitalism 295–297, 309 workplace democracy 281, 436n1
Wertheimer, Alan 219–220 workplace republicanism 436n1
West Bank 181 World Bank 267, 384
“What is the Point of Equality?” (Anderson) 372 “worst-off” in society 371
What Nations Fail (Acemoglu and Robinson) 165 Wright, Erik Olin 276, 278–279, 286n5
“When Did Everyone Become a Socialist” 276
Whitaker, Gordon 191–192 Y
“whited out,” Black testimony against Whites 370 YIMBY 258
Wiener, N. 29 YIMBY act 255
Williams, A. 35 YIMBY movement (Yes in My Backyard) 255
Williamson, Oliver 151 Yorkshire Ripper 370–371
“wisdom of the crowd” 394 Young, H. Peyton 78n9, 110
“without too much cost” clause 360 Young, Iris Marion 54–55, 218, 363n3, 365–367,
Wolkenstein, F. 260 369, 372–373, 374n1, 381–384
Wollstonecraft, Mary 60
women, constraints on 54 Z
women’s bodies as commodities 61 Zelikow, Phillip D. 449
Wood, A. 284 zero or non-zero sum games, as cooperative games 76
work as main source of income 300 zero-sum game 71
workers’ basic rights, contracts with capitalist firms Zhao, W.J 104
437n8 Zoot Suit riots 252
worker’s voices, justification for limiting: arguments
467

The Routledge Handbook of Philosophy, Politics, and Economics (Taylor Francis Group)

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

The Routledge Handbook of Philosophy, Politics, and Economics (Taylor Francis Group)

Uploaded by

Copyright:

Available Formats

THE ROUTLEDGE HANDBOOK OF

PHILOSOPHY, POLITICS, AND

Edited by C.M. Melenovsky

ISBN: 978-0-367-40767-4 (hbk)

1 PPE as an Intellectual Enterprise 3

2 On Models and their Uses 17

4 PPE in Marx’s Materialist Conception of History 43

7 Four Structures of Intransitive Preferences 81

8 Theories of Choice Behavior 94

10 Implicit Bias and Decision-Making 121

11 Social Norms 137

12 Institutions and Institutionalism 149

14 Corporations in our Polity 176

16 The Advantages of Markets 205

18 The Meaning of Markets 227

19 Gender and the Division of Labor 239

20 Housing Markets 252

23 Property-Owning Democracy 289

24 Social Democracy 300

26 Property Rights and Justice in Holdings: A Libertarian Perspective 329

27 High Liberalism 341

28 Institutionalism, Injustice, and Personal Responsibility 354

29 Social Justice 365

30 Justice Across Borders 377

31 In Defense of Epistocracy: Enlightened Preference Voting 391

32 Voting Rules 401

33 Enabling Informed and Equal Participation 416

34 What, If Anything, Can Justify Limiting Workers’ Voice? 428

35 Social Trust 439

7.1 Schumm’s Christmas Ornaments 82

32.2 Table 32.2 406

Sudeep Bhatia is Assistant Professor of Psychology at the University of Pennsylvania, where he

Lacey J. Davidson is Assistant Professor of Philosophy at the University of Indianapolis. Her

Kristina Meshelski is Associate Professor of Philosophy at California State University, Northridge.

Jacob Reidhead is a post-doctoral researcher at National Chengchi University. He recently

Geoffrey Sayre-McCord is the Morehead-Cain Distinguished Professor of Philosophy at the

Kok-Chor Tan is Professor of Philosophy at the University of Pennsylvania. He is the author of

Vanessa Wills is Assistant Professor of Philosophy at The George Washington University in

Introduction to Part I: Frameworks

1 The General Case for (Some) Interdisciplinary Work

2 Disciplines and the Division of Epistemic Labor

2.1 Why the Origins of Specialization in Intellectual Enquiry Are Different?

2.2 The Limits of Specialization

2.3 Disciplines as Certifiers

3 Shadows and Conflicts

3.1 Categorical vs. Comparative Thinking

3.2 Motives vs. Incentives

3.3 Motivational Assumptions

3.4 Normative Considerations

3.5 Discounting for Time

3.6 Ideal vs. Non-Ideal Theory

3.7 Methodology and the Philosophy of Science

3 Fiction, Fables, Models

4 The Moral of the Story

5 Using Models in Public Places

WoS "complexity" mentions per decade

4 Conditions for Complexity

1. numerous and diverse agents that

4. of the behaviors of other (“local”) agents

5 Complex Orders and Their Features

structures that manage complexity include federal, multi-ministry, and checks-and-balances

6 Implications of Complexity for PPE

6.1 Complexity in Political Science