Following Their Leaders: Political Preferences and Public Policy 1009323164, 9781009323161

Citation preview

Following Their Leaders Models of democratic decision-making tend to assume that voters have preferences and that candidates adjust their platforms to conform with those preferences; however, the direction of causation is largely the opposite. Political elites offer policy platforms to voters, and v­ oters adopt those policies – they follow their leaders. Following Their Leaders argues that policies are designed by the elite and the ­electorate has little say. Preferences for public policy tend to be anchored in a political identity associated with a candidate, party, or ideology; v­ oters’ preferences on most issues are derived from their anchor preferences. Holcombe argues that because citizens adopt the policies offered by the elite, democratic institutions are ineffective constraints on the exercise of political power. This volume explores political institutions that help control the elite who exercise political power and discusses the implications political preferences have on democracies. Randall G. Holcombe  is DeVoe Moore Professor of Economics at Florida State University. He served on Florida Governor Jeb Bush’s Council of Economic Advisors from 2000 to 2006, and is past ­president of the Public Choice Society and the Society for the Development of Austrian Economics.

Published online by Cambridge University Press

Cambridge Studies in Economics, Choice, and Society Founding Editors Timur Kuran, Duke University Peter J. Boettke, George Mason University This interdisciplinary series promotes original theoretical and empirical research as well as integrative syntheses involving links between individual choice, institutions, and social outcomes. Contributions are welcome from across the social sciences, particularly in the areas where economic analysis is joined with other disciplines such as comparative political economy, new institutional economics, and behavioral economics. Books in the Series: Terry l. Anderson and Gary D. Libecap, Environmental Markets: A Property Rights Approach Morris B. Hoffman The Punisher’s Brain: The Evolution of Judge and Jury Peter T. Leeson: Anarchy Unbound: Why Self- Governance Works Better Than You Think Benjamin Powell Out of Poverty: Sweatshops in the Global Economy Cass R. Sunstein The Ethics of Influence: Government in the Age of Behavioral Science Jared Rubin Rulers, Religion, and Riches: Why the West Got Rich and the Middle East Did Not Jean-Philippe Platteau Islam Instrumentalized: Religion and Politics in Historical Perspective Taizu Zhang The Laws and Economics of Confucianism: Kinship and Property in Preindustrial China and England Roger Koppl Expert Failure Michael C. Munger Tomorrow 3.0: Transaction Costs and the Sharing Economy Carolyn M. Warner, Ramazan Kilinç, Christopher W. Hale and Adam B. Cohen Generating Generosity in Catholicism and Islam: Beliefs, Institutions, and Public Goods Provision Randall G. Holcombe Political Capitalism: How Political Influence is Made and Maintained Paul Dragos Aligica Public Entrepreneurship, Citizenship, and Self-Governance Vernon L. Smith and Bart J. Wilson Humanomics: Moral Sentiments and the Wealth of Nations for the Twenty-First Century Andreas Thiel, William A. Blomquist and Dustin E. Garrick Governing ­Complexity: Analyzing and Applying Polycentricity Alex Nowrasteh and Benjamin Powell Wretched Refuse?: The Political Economy of Immigration and Institutions Shelby Grossman The Politics of Order in Informal Markets: How the State Shapes Private Governance Taisu Zhang The Ideological Foundations of Qing Taxation: Belief Systems, ­Politics, and Institutions

Published online by Cambridge University Press

Following Their Leaders Political Preferences and Public Policy

RANDALL G. HOLCOMBE Florida State University

Published online by Cambridge University Press

Shaftesbury Road, Cambridge CB2 8EA, United Kingdom One Liberty Plaza, 20th Floor, New York, N Y 10006, USA 477 Williamstown Road, Port Melbourne, VIC 3207, Australia 314–321, 3rd Floor, Plot 3, Splendor Forum, Jasola District Centre, New Delhi – 110025, India 103 Penang Road, #05–06/07, Visioncrest Commercial, Singapore 238467 Cambridge University Press is part of Cambridge University Press & Assessment, a department of the University of Cambridge. We share the University’s mission to contribute to society through the pursuit of education, learning and research at the highest international levels of excellence. www.cambridge.org Information on this title: www.cambridge.org/9781009323161 DOI: 10.1017/9781009323178 © Randall G. Holcombe 2023 This publication is in copyright. Subject to statutory exception and to the provisions of relevant collective licensing agreements, no reproduction of any part may take place without the written permission of Cambridge University Press & Assessment. First published 2023 A catalogue record for this publication is available from the British Library ISBN 978-1-009-32316-1 Hardback ISBN 978-1-009-32319-2 Paperback Cambridge University Press & Assessment has no responsibility for the persistence or accuracy of URLs for external or third-party internet websites referred to in this publication and does not guarantee that any content on such websites is, or will remain, accurate or appropriate.

Published online by Cambridge University Press

Following Their Leaders Models of democratic decision-making tend to assume that voters have preferences and that candidates adjust their platforms to conform with those preferences; however, the direction of causation is largely the opposite. Political elites offer policy platforms to voters, and v­ oters adopt those policies – they follow their leaders. Following Their Leaders argues that policies are designed by the elite and the ­electorate has little say. Preferences for public policy tend to be anchored in a political identity associated with a candidate, party, or ideology; v­ oters’ preferences on most issues are derived from their anchor preferences. Holcombe argues that because citizens adopt the policies offered by the elite, democratic institutions are ineffective constraints on the exercise of political power. This volume explores political institutions that help control the elite who exercise political power and discusses the implications political preferences have on democracies. Randall G. Holcombe  is DeVoe Moore Professor of Economics at Florida State University. He served on Florida Governor Jeb Bush’s Council of Economic Advisors from 2000 to 2006, and is past ­president of the Public Choice Society and the Society for the Development of Austrian Economics.

Published online by Cambridge University Press

Cambridge Studies in Economics, Choice, and Society Founding Editors Timur Kuran, Duke University Peter J. Boettke, George Mason University This interdisciplinary series promotes original theoretical and empirical research as well as integrative syntheses involving links between individual choice, institutions, and social outcomes. Contributions are welcome from across the social sciences, particularly in the areas where economic analysis is joined with other disciplines such as comparative political economy, new institutional economics, and behavioral economics. Books in the Series: Terry l. Anderson and Gary D. Libecap, Environmental Markets: A Property Rights Approach Morris B. Hoffman The Punisher’s Brain: The Evolution of Judge and Jury Peter T. Leeson: Anarchy Unbound: Why Self- Governance Works Better Than You Think Benjamin Powell Out of Poverty: Sweatshops in the Global Economy Cass R. Sunstein The Ethics of Influence: Government in the Age of Behavioral Science Jared Rubin Rulers, Religion, and Riches: Why the West Got Rich and the Middle East Did Not Jean-Philippe Platteau Islam Instrumentalized: Religion and Politics in Historical Perspective Taizu Zhang The Laws and Economics of Confucianism: Kinship and Property in Preindustrial China and England Roger Koppl Expert Failure Michael C. Munger Tomorrow 3.0: Transaction Costs and the Sharing Economy Carolyn M. Warner, Ramazan Kilinç, Christopher W. Hale and Adam B. Cohen Generating Generosity in Catholicism and Islam: Beliefs, Institutions, and Public Goods Provision Randall G. Holcombe Political Capitalism: How Political Influence is Made and Maintained Paul Dragos Aligica Public Entrepreneurship, Citizenship, and Self-Governance Vernon L. Smith and Bart J. Wilson Humanomics: Moral Sentiments and the Wealth of Nations for the Twenty-First Century Andreas Thiel, William A. Blomquist and Dustin E. Garrick Governing ­Complexity: Analyzing and Applying Polycentricity Alex Nowrasteh and Benjamin Powell Wretched Refuse?: The Political Economy of Immigration and Institutions Shelby Grossman The Politics of Order in Informal Markets: How the State Shapes Private Governance Taisu Zhang The Ideological Foundations of Qing Taxation: Belief Systems, ­Politics, and Institutions

Published online by Cambridge University Press

Following Their Leaders Political Preferences and Public Policy

RANDALL G. HOLCOMBE Florida State University

Published online by Cambridge University Press

Shaftesbury Road, Cambridge CB2 8EA, United Kingdom One Liberty Plaza, 20th Floor, New York, N Y 10006, USA 477 Williamstown Road, Port Melbourne, VIC 3207, Australia 314–321, 3rd Floor, Plot 3, Splendor Forum, Jasola District Centre, New Delhi – 110025, India 103 Penang Road, #05–06/07, Visioncrest Commercial, Singapore 238467 Cambridge University Press is part of Cambridge University Press & Assessment, a department of the University of Cambridge. We share the University’s mission to contribute to society through the pursuit of education, learning and research at the highest international levels of excellence. www.cambridge.org Information on this title: www.cambridge.org/9781009323161 DOI: 10.1017/9781009323178 © Randall G. Holcombe 2023 This publication is in copyright. Subject to statutory exception and to the provisions of relevant collective licensing agreements, no reproduction of any part may take place without the written permission of Cambridge University Press & Assessment. First published 2023 A catalogue record for this publication is available from the British Library ISBN 978-1-009-32316-1 Hardback ISBN 978-1-009-32319-2 Paperback Cambridge University Press & Assessment has no responsibility for the persistence or accuracy of URLs for external or third-party internet websites referred to in this publication and does not guarantee that any content on such websites is, or will remain, accurate or appropriate.

Published online by Cambridge University Press

Following Their Leaders Models of democratic decision-making tend to assume that voters have preferences and that candidates adjust their platforms to conform with those preferences; however, the direction of causation is largely the opposite. Political elites offer policy platforms to voters, and v­ oters adopt those policies – they follow their leaders. Following Their Leaders argues that policies are designed by the elite and the ­electorate has little say. Preferences for public policy tend to be anchored in a political identity associated with a candidate, party, or ideology; v­ oters’ preferences on most issues are derived from their anchor preferences. Holcombe argues that because citizens adopt the policies offered by the elite, democratic institutions are ineffective constraints on the exercise of political power. This volume explores political institutions that help control the elite who exercise political power and discusses the implications political preferences have on democracies. Randall G. Holcombe  is DeVoe Moore Professor of Economics at Florida State University. He served on Florida Governor Jeb Bush’s Council of Economic Advisors from 2000 to 2006, and is past ­president of the Public Choice Society and the Society for the Development of Austrian Economics.

Published online by Cambridge University Press

Cambridge Studies in Economics, Choice, and Society Founding Editors Timur Kuran, Duke University Peter J. Boettke, George Mason University This interdisciplinary series promotes original theoretical and empirical research as well as integrative syntheses involving links between individual choice, institutions, and social outcomes. Contributions are welcome from across the social sciences, particularly in the areas where economic analysis is joined with other disciplines such as comparative political economy, new institutional economics, and behavioral economics. Books in the Series: Terry l. Anderson and Gary D. Libecap, Environmental Markets: A Property Rights Approach Morris B. Hoffman The Punisher’s Brain: The Evolution of Judge and Jury Peter T. Leeson: Anarchy Unbound: Why Self- Governance Works Better Than You Think Benjamin Powell Out of Poverty: Sweatshops in the Global Economy Cass R. Sunstein The Ethics of Influence: Government in the Age of Behavioral Science Jared Rubin Rulers, Religion, and Riches: Why the West Got Rich and the Middle East Did Not Jean-Philippe Platteau Islam Instrumentalized: Religion and Politics in Historical Perspective Taizu Zhang The Laws and Economics of Confucianism: Kinship and Property in Preindustrial China and England Roger Koppl Expert Failure Michael C. Munger Tomorrow 3.0: Transaction Costs and the Sharing Economy Carolyn M. Warner, Ramazan Kilinç, Christopher W. Hale and Adam B. Cohen Generating Generosity in Catholicism and Islam: Beliefs, Institutions, and Public Goods Provision Randall G. Holcombe Political Capitalism: How Political Influence is Made and Maintained Paul Dragos Aligica Public Entrepreneurship, Citizenship, and Self-Governance Vernon L. Smith and Bart J. Wilson Humanomics: Moral Sentiments and the Wealth of Nations for the Twenty-First Century Andreas Thiel, William A. Blomquist and Dustin E. Garrick Governing ­Complexity: Analyzing and Applying Polycentricity Alex Nowrasteh and Benjamin Powell Wretched Refuse?: The Political Economy of Immigration and Institutions Shelby Grossman The Politics of Order in Informal Markets: How the State Shapes Private Governance Taisu Zhang The Ideological Foundations of Qing Taxation: Belief Systems, ­Politics, and Institutions

Published online by Cambridge University Press

Following Their Leaders Political Preferences and Public Policy

RANDALL G. HOLCOMBE Florida State University

Published online by Cambridge University Press

Shaftesbury Road, Cambridge CB2 8EA, United Kingdom One Liberty Plaza, 20th Floor, New York, N Y 10006, USA 477 Williamstown Road, Port Melbourne, VIC 3207, Australia 314–321, 3rd Floor, Plot 3, Splendor Forum, Jasola District Centre, New Delhi – 110025, India 103 Penang Road, #05–06/07, Visioncrest Commercial, Singapore 238467 Cambridge University Press is part of Cambridge University Press & Assessment, a department of the University of Cambridge. We share the University’s mission to contribute to society through the pursuit of education, learning and research at the highest international levels of excellence. www.cambridge.org Information on this title: www.cambridge.org/9781009323161 DOI: 10.1017/9781009323178 © Randall G. Holcombe 2023 This publication is in copyright. Subject to statutory exception and to the provisions of relevant collective licensing agreements, no reproduction of any part may take place without the written permission of Cambridge University Press & Assessment. First published 2023 A catalogue record for this publication is available from the British Library ISBN 978-1-009-32316-1 Hardback ISBN 978-1-009-32319-2 Paperback Cambridge University Press & Assessment has no responsibility for the persistence or accuracy of URLs for external or third-party internet websites referred to in this publication and does not guarantee that any content on such websites is, or will remain, accurate or appropriate.

Published online by Cambridge University Press

Following Their Leaders Models of democratic decision-making tend to assume that voters have preferences and that candidates adjust their platforms to conform with those preferences; however, the direction of causation is largely the opposite. Political elites offer policy platforms to voters, and v­ oters adopt those policies – they follow their leaders. Following Their Leaders argues that policies are designed by the elite and the ­electorate has little say. Preferences for public policy tend to be anchored in a political identity associated with a candidate, party, or ideology; v­ oters’ preferences on most issues are derived from their anchor preferences. Holcombe argues that because citizens adopt the policies offered by the elite, democratic institutions are ineffective constraints on the exercise of political power. This volume explores political institutions that help control the elite who exercise political power and discusses the implications political preferences have on democracies. Randall G. Holcombe  is DeVoe Moore Professor of Economics at Florida State University. He served on Florida Governor Jeb Bush’s Council of Economic Advisors from 2000 to 2006, and is past ­president of the Public Choice Society and the Society for the Development of Austrian Economics.

Published online by Cambridge University Press

Cambridge Studies in Economics, Choice, and Society Founding Editors Timur Kuran, Duke University Peter J. Boettke, George Mason University This interdisciplinary series promotes original theoretical and empirical research as well as integrative syntheses involving links between individual choice, institutions, and social outcomes. Contributions are welcome from across the social sciences, particularly in the areas where economic analysis is joined with other disciplines such as comparative political economy, new institutional economics, and behavioral economics. Books in the Series: Terry l. Anderson and Gary D. Libecap, Environmental Markets: A Property Rights Approach Morris B. Hoffman The Punisher’s Brain: The Evolution of Judge and Jury Peter T. Leeson: Anarchy Unbound: Why Self- Governance Works Better Than You Think Benjamin Powell Out of Poverty: Sweatshops in the Global Economy Cass R. Sunstein The Ethics of Influence: Government in the Age of Behavioral Science Jared Rubin Rulers, Religion, and Riches: Why the West Got Rich and the Middle East Did Not Jean-Philippe Platteau Islam Instrumentalized: Religion and Politics in Historical Perspective Taizu Zhang The Laws and Economics of Confucianism: Kinship and Property in Preindustrial China and England Roger Koppl Expert Failure Michael C. Munger Tomorrow 3.0: Transaction Costs and the Sharing Economy Carolyn M. Warner, Ramazan Kilinç, Christopher W. Hale and Adam B. Cohen Generating Generosity in Catholicism and Islam: Beliefs, Institutions, and Public Goods Provision Randall G. Holcombe Political Capitalism: How Political Influence is Made and Maintained Paul Dragos Aligica Public Entrepreneurship, Citizenship, and Self-Governance Vernon L. Smith and Bart J. Wilson Humanomics: Moral Sentiments and the Wealth of Nations for the Twenty-First Century Andreas Thiel, William A. Blomquist and Dustin E. Garrick Governing ­Complexity: Analyzing and Applying Polycentricity Alex Nowrasteh and Benjamin Powell Wretched Refuse?: The Political Economy of Immigration and Institutions Shelby Grossman The Politics of Order in Informal Markets: How the State Shapes Private Governance Taisu Zhang The Ideological Foundations of Qing Taxation: Belief Systems, ­Politics, and Institutions

Published online by Cambridge University Press

Following Their Leaders Political Preferences and Public Policy

RANDALL G. HOLCOMBE Florida State University

Published online by Cambridge University Press

Shaftesbury Road, Cambridge CB2 8EA, United Kingdom One Liberty Plaza, 20th Floor, New York, N Y 10006, USA 477 Williamstown Road, Port Melbourne, VIC 3207, Australia 314–321, 3rd Floor, Plot 3, Splendor Forum, Jasola District Centre, New Delhi – 110025, India 103 Penang Road, #05–06/07, Visioncrest Commercial, Singapore 238467 Cambridge University Press is part of Cambridge University Press & Assessment, a department of the University of Cambridge. We share the University’s mission to contribute to society through the pursuit of education, learning and research at the highest international levels of excellence. www.cambridge.org Information on this title: www.cambridge.org/9781009323161 DOI: 10.1017/9781009323178 © Randall G. Holcombe 2023 This publication is in copyright. Subject to statutory exception and to the provisions of relevant collective licensing agreements, no reproduction of any part may take place without the written permission of Cambridge University Press & Assessment. First published 2023 A catalogue record for this publication is available from the British Library ISBN 978-1-009-32316-1 Hardback ISBN 978-1-009-32319-2 Paperback Cambridge University Press & Assessment has no responsibility for the persistence or accuracy of URLs for external or third-party internet websites referred to in this publication and does not guarantee that any content on such websites is, or will remain, accurate or appropriate.

Published online by Cambridge University Press

For Lora, Ross, Emily, Mark, Bailey, Connor, and Becca

Published online by Cambridge University Press

Published online by Cambridge University Press

Contents

List of Figures page x List of Tables xi Preface xiii 1

Introduction

1

1.1 The Impact of Enlightenment Ideas 3 1.2 Democracy as a Mechanism for Aggregating Preferences 6 1.3 Making Social Choices 8 1.4 Democracy as Ideology 12 1.5 Democratic Institutions and the General Will 12 1.6 Democracy as Freedom from Government Oppression 15 1.7 The Ideology of Democracy Cuts Two Ways 17 1.8 There Is No Public Interest 18 1.9 Conclusion 19

2 Instrumental and Expressive Voting 2.1 2.2 2.3 2.4 2.5 2.6 2.7 2.8 2.9 2.10 2.11

20

Individual Choice in Markets and in Voting 21 Instrumental Voting 23 Strategic Voting 23 Are Expressive Votes Wasted? 24 Rational Ignorance 26 Feedback in Markets and in Voting 28 Examples of Expressive Voting 29 Costs and Benefits of Expressive Voting 32 Expressive Behavior of Voters and Sports Fans 34 Ethical Voting 35 Conclusion 37

3 Influences over Preference Formation 39 3.1 Market Choices 43 vii

Published online by Cambridge University Press

Contents

viii

3.2 Advertising 45 3.3 Created Desires 48 3.4 Political Advertising 49 3.5 Hope and Change 51 3.6 Negative Advertising 52 3.7 Conclusion 54

4 Preference Aggregation through Voting 56 4.1 Voting Models and Public Policy 57 4.2 Economic Models of Voting 61 4.3 The Logic of Collective Choice Models 61 4.4 Differing Methods of Aggregation 62 4.5 Expressive versus Instrumental Preferences 64 4.6 Aggregating Expressive Preferences 65 4.7 Social Choice and Social Welfare 66 4.8 The Link between Preferences and Outcomes 67 4.9 Elites and Masses 68 4.10 Conclusion 70 5 The Formation of Political Preferences 72 5.1 Preferences and Choices 75 5.2 Instrumental Preferences 76 5.3 The Status Quo 78 5.4 The Endowment Effect 80 5.5 Peer Pressure 81 5.6 Cognitive Dissonance 84 5.7 The Bandwagon Effect 86 5.8 Political Advertising 87 5.9 Candidate Charisma 89 5.10 Current Conditions 91 5.11 Mass Media 93 5.12 Conclusion 94 6 Anchor Preferences and Derivative Preferences 96 6.1 Public Policy Preferences 96 6.2 Anchor Preferences 99 6.3 Principled Preferences 102 6.4 Abstention 103 6.5 Derivative Preferences 105 6.6 Most Political Preferences Are Derivative 108 6.7 Defining the Issue Space 111 6.8 Conclusion 114 7 Preferences of Elites and Masses 117 7.1 The Quest for Power 118 7.2 Economic Power and Political Power 121 7.3 Economic Power Can Buy Political Power 123 7.4 Political Power Can Extort Economic Power 125 7.5 The Discontinuity in Political Power 127

Published online by Cambridge University Press

Contents

7.6 7.7 7.8 7.9

Transaction Costs and Public Policy 129 Controlling the Agenda 132 Elite Influence on the Preferences of the Masses 134 Elite Objectives 137

8 Policies That Maximize Political Power 8.1 8.2 8.3 8.4 8.5 8.6 8.7 8.8 8.9 8.10 8.11 8.12 8.13

ix

139

Principles and Politics 141 The Exercise of Power in Markets and in Politics 142 Politics and Power 144 The Adversarial Nature of Politics 146 Elites Want to Have Power, Not Use It 147 The Power of Populism 148 Elites Design Public Policy 150 Popular Policies 151 Power Leads to More Power 154 Perpetual Crises 155 Rule of Law and Political Power 156 Creating Dependence on the Political Elite 158 Conclusion 160

9 Patriotism, Propaganda, and the Public Interest

161

10 Implications for Democracy

179

9.1 The Social Contract 163 9.2 Patriotism and Propaganda 167 9.3 Thank You for Your Service 169 9.4 Choosing Adversaries 170 9.5 Can’t We All Just Get Along? 171 9.6 The Authority of the Elite 173 9.7 The Escape from Anarchy 175 9.8 Conclusion 177 10.1 10.2 10.3 10.4 10.5 10.6 10.7 10.8 10.9 10.10

Academic Analysis 181 Democracy Is Divisive 183 The American Founders Did Not Create a Democracy 184 Constitutional Limits on Popular Opinion 187 Eroding Constitutional Constraints 188 Checks and Balances 189 Constraining Leviathan 192 Competition among Elites 193 Political Preferences and Public Policy 194 The Quest for Power 196

References 199 Index 209

Published online by Cambridge University Press

Figures

9.1 Prisoners’ dilemma game 9.2 Masses and the elite 9.3 Escape from Hobbesian anarchy

x

Published online by Cambridge University Press

page 167 174 176

Tables

1.1 Preferences that produce a cyclical majority 1.2 Value of preferences that produce a cyclical majority

xi

Published online by Cambridge University Press

page 9 10

Published online by Cambridge University Press

Preface

The idea on which this book is built is that people adopt the public policy preferences of the candidates and parties with whom they identify. In the era of Donald Trump, this idea has a ring of plausibility to much of the general public, who see Trump supporters as people who blindly accept what President Trump has told them. At the other end of the political spectrum, a different set of people see Bernie Sanders supporters’ the same way. Nobody in the general public thinks this is true of them, personally. They will tell you they have sound reasons for the public policy preferences they express. They just think that other people uncritically accept the policy preferences espoused by candidates and parties. So you, the reader of this book, should read it as if its arguments apply to other people, not to you. There may be an element of truth in that because readers of this book are likely to be better informed about public policy than most of the general public. However, the academic literature on the subject does suggest that well-informed citizens may be more susceptible to political persuasion than is the average citizen. This idea is more difficult to sell to an academic audience than to the general public. Academicians studying the political process consistently begin with the assumption that citizens and voters have preferences, and that candidates and parties adopt platforms that correspond with those citizen preferences. Rarely does an academic analysis of preference aggregation through democratic processes take a step back to consider how those citizen and voter preferences are formed. There has been research on political preference formation, but it is rarely taken into account when social scientists model the process by which individual preferences are aggregated to make collective choices. xiii

https://doi.org/10.1017/9781009323178.001 Published online by Cambridge University Press

xiv

Preface

Academics have taken a partial step toward addressing this issue by noting that when voters make choices at the ballot box the choices they make, as individuals, do not affect the outcome they get. In market transactions, people get what they choose. If a diner orders a salad rather than pizza, the diner gets a salad. At the ballot box, the same candidates and parties will be elected regardless of how any individual voter votes. Because of this, voters may express preferences at the ballot box that are different from what they would choose if the choice were theirs alone. A voter may vote for candidate A even if the voter would prefer B to win the election. Voters can do this at no cost to themselves because they know their one vote will not change the election outcome. Voting is an expressive act that has no instrumental effect on election outcomes. This idea, which has met with some resistance from academics, is considered in the chapters that follow, but even with this idea established, that academic literature has not taken the next step to analyze why citizens and voters form the policy preferences they express. The preferences they express at the ballot box may be at odds with policies they would choose if the choice were theirs, but if this is the case, where do those preferences originate? If they are influenced by friends or family, this just pushes the question one step further back. Where do those friends and family get their political preferences? This volume explains that people adopt the political preferences of the candidates and parties with whom they identify. If the public policy preferences of citizens come from the political elite, one implication is that citizens in democratic countries have less control over their governments than the ideology of democracy would suggest. Supposedly, democratic governments are accountable to their citizens, but that accountability breaks down if citizens adopt the political preferences of the political elite. Academic models of voting conclude that candidates and parties design their platforms to correspond with the political preferences of voters, but the direction of causation goes the other way. Voters adjust their preferences to conform with the platforms put forward by the political elite. This volume offers a pessimistic view of democratic government, but because there are competing members of the elite, that competition among elites offers some hope that that abuses of government power can be limited. The checks and balances built into political institutions offer stronger constraints on the abuse of power than democratic oversight by citizens and voters.

https://doi.org/10.1017/9781009323178.001 Published online by Cambridge University Press

Preface

xv

The usual disclaimer that any shortcomings in this volume are the responsibility of the author is especially applicable here. If citizens adopt the policy preferences offered them by political elites, this calls into question some long-held conclusions about preference aggregation in the academic areas of public choice and social choice. One might expect that researchers would be cautious about accepting ideas that undermine long-held conclusions in those areas. I am under no illusion that this volume represents the last word on the subject. My hope is that it will prompt others to explore the merits – and demerits! – of the ideas in this volume. I appreciate comments from Roger Congleton, Judit Kalman Joseph Newhard, and David Skarbek. I presented material from the manuscript at Texas Tech University and at the Public Choice Society annual meetings, where I received many helpful comments. A hazard in thanking readers for comments is that I have neglected to recognize some helpful readers, and that is surely the case here. I am deeply grateful for the support of my family, and dedicate the book to my wife, Lora, my three sons, and my three daughters-in-law.

https://doi.org/10.1017/9781009323178.001 Published online by Cambridge University Press

https://doi.org/10.1017/9781009323178.001 Published online by Cambridge University Press

1 Introduction

The idea that the preferences of citizens should be taken into account to determine a government’s public policies is a relatively new one in human history, only a few hundred years old. Until Enlightenment ideas became influential in the late seventeenth century, citizens were viewed – and viewed themselves – as subjects of their governments. The population was divided into two classes – the rulers and the ruled – and one duty of the ruled was to serve the interests of the state, as determined by its rulers. Enlightenment ideas offered an ideology that reversed that relationship. Rather than citizens serving the state, the state should serve its citizens. Before the state can act to further the interests of its citizens, it must know what those interests are, and that is often viewed as one of the main functions of a democratic government. Democratic government works, first, to place the control of government in the hands of its citizens. But democracy is often also viewed as a mechanism for revealing and aggregating the preferences of its citizens so that government can carry out the will of the people. When democracy is viewed that way, as a mechanism that reveals the preferences of its citizens, the actions of democratic governments gain a legitimacy that can lead to an abuse of power. The way that social scientists have analyzed how political institutions aggregate citizen preferences took a major step forward in the 1960s, which marked the beginning of the public choice revolution. Forerunners can be found, to be sure,1 but a major premise upon which this revolution 1

See Appendix 2 in James M. Buchanan and Gordon Tullock, The Calculus of Consent (Ann Arbor: University of Michigan Press, 1962). This book could be viewed as initiating the public choice revolution, and Appendix 2 is titled “Theoretical Forerunners.”

1

https://doi.org/10.1017/9781009323178.002 Published online by Cambridge University Press

2

Introduction

in political thought was based is that there is little difference between the motivations of people in the private sector and people in the public sector. In contrast to the view (still widely held) that people in the private sector act to further their private interests while those in the public sector work to further the public interest, the public choice approach recognizes that while most people do want to do what is best for others, everyone responds to the incentive to do what is best for themselves. Elected officials often take actions that help them to maintain their offices, or move up the political hierarchy, even when those actions may not further the public interest. Government bureaucrats will act to further their careers, and will avoid actions that threaten their job security, even when those actions may not further the public interest. The public choice assumption is that people in the public sector are no better and no worse than those in the private sector. The public choice approach to politics looks at the incentives people face when they make decisions and the information they have available to them when they make those decisions. This approach looks at the way political processes actually work, not how we might hope that they would work. When analyzing politics, objective analysis should replace wishful thinking. James Buchanan, one of the founders of the public choice movement, referred to this approach as “politics without romance.”2 Thinking about democratic institutions as a way of aggregating the policy preferences of individual citizens into some vision of the public interest requires an understanding of how those institutions aggregate individual preferences, which has been done extensively in the public choice analysis undertaken by political scientists and economists. Efforts along these lines will be discussed below. But it also requires an understanding of how citizens form the preferences they express through democratic institutions, and this has seen much less development. Political preferences are often assumed as given and exogenous, and the primary interest of this volume is to examine in more detail how those preferences are formed, and as a result, the implications for public policy. Social scientists typically assume citizen preferences to be given, and examine how politicians and other government officials design their political platforms to correspond to the preferences of their citizens.3 There are 2 3

James M. Buchanan, “Public Choice: Politics without Romance,” Policy 19, no. 3 (Spring 2003), pp. 13–18. Anthony Downs, An Economic Theory of Democracy (New York: Harper & Row, 1957), is a good example. Downs concludes that politicians design their platforms to conform to the preferences of the median voter.

https://doi.org/10.1017/9781009323178.002 Published online by Cambridge University Press

1.1  The Impact of Enlightenment Ideas

3

good reasons to believe that the direction of causation goes mainly in the other direction: that citizens adopt the policy preferences of the political elite, rather than the elite adjusting their platforms to conform to the preferences of their constituents. This idea is far enough away from the mainstream views in political science that much of this volume is devoted to explaining how this happens. If policy preferences ultimately are determined by the elite, this raises the further question of what determines the preferences of the elite. Answering those questions links political preferences and public policy.

1.1  The Impact of Enlightenment Ideas For most of human history, societies were divided into the rulers and the ruled. Citizens were subjects of their governments and were obligated to obey the orders of their rulers. Thomas Hobbes, writing in 1651, argued that doing so was in the best interest of the citizens.4 Without a government to enforce order, Hobbes argued that life in anarchy would be solitary, poor, nasty, brutish, and short, and would be a war of all against all. Everybody would always be at risk from predation, both from others in their society and from outside invaders. Productivity would be low because people would have no incentive to produce things that would likely be stolen from them in a lawless society. The solution, Hobbes said, was for everyone to agree to abide by the government’s rules. This would allow an escape from anarchy and enable an orderly society. Obeying the rules of the sovereign was a social contract that would be – and must be – agreed to by all members of society, according to Hobbes. Several features of Hobbes’s social contract point toward the preEnlightenment view of government. Hobbes argued that the sovereign had the right to put to death anyone who violated the rules of government. That is one way to ensure that everyone agrees to abide by the rules: kill those who do not! Hobbes saw no alternative to abiding by all of the sovereign’s rules. People could not pick and choose which rules they wanted to follow; to do so would lead right back to anarchy. What if some of the sovereign’s rules were in some way flawed? It did not matter. One could hope to live under a government that acted in the best interests of its subjects, but allowing people the discretion to decide that some rules are unjust or otherwise flawed would undermine the order created by a government’s rules.

4

Thomas Hobbes, Leviathan (New York: E.P. Dutton, 1950 [orig. 1651]).

https://doi.org/10.1017/9781009323178.002 Published online by Cambridge University Press

4

Introduction

Hobbes’s vision of the social contract divided people into two classes: the rulers and the ruled. The rulers made the rules, and the ruled were required to follow them. Government’s subjects had the obligation to act in the interest of their government, that is, in the interest of the ruling class, because as Hobbes described it, to do so was also in the interest of the government’s subjects. It allowed them to live in an orderly society and escape from an anarchy that would be a war of all against all. According to Hobbes, people got their rights from government. The ruling class made the rules and their subjects were obligated to follow them. This idea is difficult to comprehend in the twenty-first century because Enlightenment ideas have reinforced the notion that everyone should be equal in the eyes of the law and that in some cases governments violate people’s human rights. When Hobbes wrote, even the world’s most civilized nations assigned people to different classes based on their birth. Some people were born into royalty and enjoyed privileges associated with it. Commoners, simply as a result of the families they were born into, could never enjoy the privileges of royalty. Slavery was common throughout the world, and people accepted these class divisions with Hobbesian reasoning. Government makes the rules and its subjects are obligated to follow them. If some people were born into higher status than others, the social contract obligated everyone to recognize the distinction. Slavery, rejected as immoral in the twenty-first century, was accepted in the seventeenth as a part of the orderly society enforced by the mandates of government. Only a few decades later, John Locke published his Two Treatise of Government in 1690, with a very different view of the social contract.5 In contrast with Hobbes, who argued that people got their rights from government, Locke argued that people naturally have rights and that the role of government is to protect those rights. Locke begins with the idea that people own themselves. They have a right to their bodies, and therefore they have a right to their labor. Therefore, they have a right to own what they produce with their labor. Locke thus develops a theory of self-ownership that leads to the right to own property, and the social contract, as Locke saw it, is that people are obligated to not violate the rights of others. Locke saw a problem similar to the one Hobbes saw, which is that opportunistic individuals might violate that social contract and infringe 5

John Locke, Two Treatises of Government (Cambridge: Cambridge University Press, 1960 [orig. 1690]).

https://doi.org/10.1017/9781009323178.002 Published online by Cambridge University Press

1.1  The Impact of Enlightenment Ideas

5

the rights of others. The role of government, Locke argued, is to protect people’s rights. Here, Locke and Hobbes are in agreement. But Locke’s vision of the social contract differed from Hobbes’s vision in (at least) two important ways. First, Locke viewed that people naturally have rights, in contrast with Hobbes who said that the government makes the rules and determines what rights people have. Second, government is a party to the social contract as Locke describes it, whereas Hobbes’s social contract is among government’s subjects, who are obligated to abide by government’s rules. Locke put forward the revolutionary idea that if government fails to uphold its obligations under the social contract, citizens have a right to overthrow and replace their government. This idea was, quite literally, revolutionary. Prior to the American Revolution, pamphleteers were arguing that the king of England was violating the rights of the colonists, so the colonists had the right to replace that government with one that was dedicated to protecting their rights. While most Americans at the time would not have read Locke, pamphleteers advocating independence from Britain were referring to Locke’s ideas, so the colonists would have been familiar with Locke’s ideas.6 Throughout the eighteenth century, Enlightenment ideas changed the way that citizens viewed their relationship to government. The view that citizens were subjects of their governments and obligated to serve their governments was reversed, so people increasingly thought that government should serve its citizens rather than the other way around. A series of newspaper columns published in the London Journal under the name of Cato from 1720 to 1723 advocating Enlightenment principles of liberty was very influential, and the columns were ultimately compiled into a book, first published in 1755, titled Cato’s Letters.7 By the time of the American Revolution in 1776, the idea that government should serve the interests of its citizens was well established, changing the dominant view from a century prior that citizens should serve their governments. If governments should serve the interests of their citizens, what are those interests? One thought is that democratic decision-making processes can reveal those interests. The institutions of democracy serve as 6

7

The role of Locke’s ideas in the American Revolution is discussed by Bernard Bailyn, The Ideological Origins of the American Revolution (Cambridge, MA: Harvard University Press, 1967). The authors of the letters were Thomas Gordon and John Trenchard. The letters can be found in Essays on Liberty, Civil and Religious, and Other Important Subjects, Ronald Hamowy, ed. (Indianapolis: Liberty Fund, 1995).

https://doi.org/10.1017/9781009323178.002 Published online by Cambridge University Press

6

Introduction

a preference revelation mechanism that identifies the interests of its citizens. Democracy is widely viewed that way, and politicians promote the idea with claims, after winning elections, that they have a mandate to implement policies on which they campaigned. This vision points to a direct link between political preferences and public policy. As appealing as that idea might at first appear, thinking of democracy that way is problematic.

1.2  Democracy as a Mechanism for Aggregating Preferences James Buchanan and Gordon Tullock depict democracy as a mechanism for revealing the collective preferences of groups of individuals. They say, “Collective action is viewed as the action of individuals when they choose to accomplish purposes collectively rather than individually, and the government is seen as nothing more than the set of processes, the machine, which allows such collective action to take place.”8 They recognize, however, that groups have no preferences; only individuals have preferences. So, to say that some policy is in the interest of a group of people can mean nothing more than that it is in the interest of the individual members of the group. Democracy can aggregate preferences to make a collective choice, but it is misleading to say that the group, as such, has expressed a preference. Buchanan and Tullock use agreement as a benchmark to judge whether collective action is in the best interest of members of a group, recognizing that individuals cannot expect that every collective action a group takes will benefit every single individual. Rather, the expectation of group members is that everyone is better off with collective action undertaken by government than without it. If this is not the case, then the ideas of Locke and other Enlightenment thinkers would lead citizens to work to replace their government with one that does improve their welfare. Consider, for example, a system of traffic lights that regulates the flow of traffic. In some cases, an individual may come to a red light and have to stop even though there is no other traffic at the intersection. The individual is worse off for having to stop, and nobody is better off because there is no conflicting traffic, so in this specific instance, stopping at that intersection imposes a cost on the driver, but nobody benefits. Social 8

James M. Buchanan and Gordon Tullock, The Calculus of Consent (Ann Arbor: ­University of Michigan Press, 1962), p. 13.

https://doi.org/10.1017/9781009323178.002 Published online by Cambridge University Press

1.2  Democracy as a Mechanism for Aggregating Preferences

7

welfare is reduced. But overall, drivers are better off with the system of traffic lights than without that system, even if in specific cases the system imposes some costs in excess of their benefits. In the aggregate, everybody benefits from an orderly flow of traffic. The example extends to collective decision-making more generally. Ideally, democracy produces a set of institutions that improves the welfare of everyone.9 There is no guarantee that this is the case, and when public policies are enacted that benefit some, even if it is a large majority, but harm others, there is no way to compare the gains for some against the losses of others, so any suggestion that the outcome of democratic decisionmaking reveals the public interest is problematic. Even if somehow such a determination could be made, one would still be hard-pressed to say the outcome is in the public interest. If one person were to gain more utility from owning a slave than another would lose from being enslaved, could we say that enslaving the second person would be in the public interest? Ultimately, the hope is that even if some individual policies go against the interest of some individuals, everyone agrees to the process by which those policies are made. They agree to the rules and institutions, even if they do not always agree with the outcomes those rules and institutions produce. Democracy does not always work that way. John Stuart Mill refers to a “tyranny of the majority” in which “society is itself the tyrant – society collectively, over the separate individuals who compose it,”10 which suggests the possibility that democracy and freedom can be at odds with each other. People do benefit from collective action to produce goods that are collectively consumed, such as roads, municipal water supplies, wastewater treatment, and more. The challenge is to design a system of collective decision-making that enables people to cooperate to produce those goods, without enabling some people to use that same system to benefit themselves at the expense of others. Daron Acemoglu and James Robinson describe such a system as a narrow corridor between a government so weak that it fails to protect people’s rights and a Leviathan government that abuses its power to violate people’s rights.11 Similarly, James Buchanan describes the limits 9 10 11

This example is taken from James M. Buchanan, “The Relevance of Pareto Optimality,” Journal of Conflict Resolution 6, no. 4 (December 1962), pp. 341–354. John Stuart Mill, On Liberty, People’s Edition (London: Longmans, Green, and Co., 1913), p. 8. Daron Acemoglu and James A. Robinson, The Narrow Corridor: States, Societies, and the Fate of Liberty (New York: Penguin Press, 2019).

https://doi.org/10.1017/9781009323178.002 Published online by Cambridge University Press

8

Introduction

of liberty as lying between anarchy and Leviathan.12 It should be obvious that if citizens’ preferences are aggregated through any mechanism in which some people – a majority, or even a powerful minority – can impose their decisions on others, public policies may not reflect the political preferences of the citizens they affect. Without unanimous support, it is obvious that they do not reflect the political preferences of everyone. The risk is that democratic governments might slip out of that narrow corridor, beyond the limits of liberty, to become an oppressive Leviathan.

1.3  Making Social Choices Democratic decision-making is more than just majority rule voting. Democratic institutions vary substantially from one government to another, and those differences can make nontrivial differences in the scope and nature of government activity. Differences among democratic governing institutions include presidential versus parliamentary systems of government and plurality versus proportional voting mechanisms, but many smaller nuances also differentiate various democratic governments.13 Presidential systems, as in the United States, elect the executive and legislative branches of government separately, whereas the parliamentary systems that are commonly used in Europe elect a parliament which then chooses government ministers who lead the executive branch. Plurality voting selects the candidate who receives the most votes as the single winner in an election, whereas proportional voting elects party members in proportion to their total votes. In plurality voting, a candidate who gets 20 percent of the vote loses the election. In proportional voting, a party that receives 20 percent of the votes gets 20 percent of the seats in the parliament. These nuances are important, but an analysis of the basic framework underlying democratic decision-making reveals issues that call it into question as a mechanism for revealing collective preferences. One commonly used framework for describing the results of democratic elections is the median voter model, which concludes that when 12 13

James M. Buchanan, The Limits of Liberty: Between Anarchy and Leviathan (Chicago: University of Chicago Press, 1975). Torsten Persson and Guido Tabellini, Political Economics: Explaining Economic Policy (Cambridge: Cambridge University Press, 2000), discuss the different outcomes that tend to result from presidential versus parliamentary democracy, and from plurality versus proportional voting.

https://doi.org/10.1017/9781009323178.002 Published online by Cambridge University Press

1.3  Making Social Choices

9

Table 1.1  Preferences that produce a cyclical majority Voters 1

2

3

A B C

B C A

C A B

preferences are aggregated by majority rule, the aggregated preference of the group is the preference of the median voter.14 The model assumes that voter preferences can be arrayed on a single-dimensioned continuum, and concludes that the outcome preferred by the median voter can defeat all other alternatives by simple majority rule. Looked at in this way, majority rule voting is a system of preference aggregation. The group wants to make a collective decision, and if the model is descriptive, aggregating the individual preferences of group members through majority rule voting means that the collective choice will be the outcome that the median voter prefers. The model is likely to be descriptive in many but not all situations. Voter preferences are often viewed as existing on a left-to-right continuum, and the model concludes that in representative democracies, candidates and parties tend to design platforms that appeal to the median voter.15 However, under some circumstances there may be no option that can win the support of a majority over all others. Table 1.1 gives a well-known example in which preferences are aligned so that there is a cyclical majority. The table represents the rank-order preferences of three voters for three different alternatives, A, B, or C. For example, voter 1 prefers A to B and B to C. In a majority rule vote pitting A against B, voters 1 and 3 would vote for A, so A defeats B. But if A runs against C, C gets the votes 14

15

This model was introduced and developed by Howard R. Bowen, “The Interpretation of Voting in the Allocation of Economic Resources,” Quarterly Journal of Economics 58, no. 1 (November 1943), pp. 27–48; Duncan Black, The Theory of Committees and Elections (Cambridge: Cambridge University Press, 1958); Anthony Downs, An Economic Theory of Democracy (New York: Harper & Row, 1957). I discuss the model in more detail in Randall G. Holcombe, Advanced Introduction to Public Choice (Cheltenham, UK: Edward Elgar, 2016), but this simple description is sufficient to give the flavor of the conclusions the model draws.

https://doi.org/10.1017/9781009323178.002 Published online by Cambridge University Press

Introduction

10

Table 1.2  Value of preferences that produce a cyclical majority Voters 1 Alternatives A $1000 B $2 C $1

2

3

$1 $3 $2

$2 $1 $3

of voters 2 and 3, so C defeats A. And if B runs against C, B gets the votes of 1 and 2, so B defeats C. B defeats C, A defeats B, and C defeats A, so no alternative can defeat all others in a majority rule vote.16 This example shows that under some circumstances there will be no equilibrium outcome under majority rule voting. If democratic decisionmaking is used to select among policies A, B, or C, none of those alternatives is preferred by a majority over all others. That does not imply that none of the alternatives is better than the others. One possibility, illustrated in Table 1.2, is that voter 1 places a high value on alternative A ($1000), and voters 2 and 3 place values of only $3 on alternatives B and C, respectively. The rank order of preferences in Table 1.2 is the same as in Table 1.1, so the same cyclical majority exists, but with the values in Table 1.2, A would clearly be the highest valued choice. Majority rule voting will not reveal that. The democratic choice is as likely to be the lower-valued alternatives B or C as the higher-valued alternative A. Furthermore, once an alternative is chosen, there will always be another one that a majority of voters prefer to the chosen alternative. Even if A is chosen, a majority would prefer C over A. This suggests the possibility of instability in democratic government. If preferences resemble those in Table 1.1, regardless of the status quo, a majority would always prefer something different. In light of this possibility, it is remarkable that democratic governments (often) appear to be so stable.17 Reasons for this remarkable

16

17

This example is given early in Kenneth J. Arrow, Social Choice and Individual Values (New Haven, CT: Yale University Press, 1953), and is sometimes referred to as the Arrow paradox. R. D. McKelvey, “Intransitivities in Multi-Dimensional Voting Models and Some Implications for Agenda Control,” Journal of Economic Theory 12, no. 3 (June 1976), pp. 472–482, suggests that instability in democratic government may be a common

https://doi.org/10.1017/9781009323178.002 Published online by Cambridge University Press

1.3  Making Social Choices

11

stability will be explored later in this volume, but one implication from this discussion is that democratic decision-making may not be a good method for identifying outcomes that best represent the interests of citizens. If a group has very homogeneous preferences, then it will not matter much what type of collective decision-making mechanism is used. Every­ body wants about the same thing, so almost any collective ­decision-making mechanism will produce what the people in the group want. When prefere­nces differ among group members, democratic decision-mak­ing tends to break down, as the cyclical majority example illustrates, and results may not reflect the preferences of the members of the group. Demo­ cracy works worst when an effective collective decision-making mechanism is needed the most. This, along with the possibility of the tyranny of the majority, suggests that democratic decision-making is not a robust system for identifying the preferences of a group. Ultimately, groups do not have preferences and groups make no choices. Individuals have preferences and individuals make choices, so any reference to group preferences must be shorthand for references to the preferences of the individuals in the group, and any references to group choices must be references to an outcome of a preference aggregation mechanism that aggregates the preferences of the individuals in the group. In general, there is no mechanism that can identify some unique “group preference” that can be discovered by aggregating votes. Democratic decision-making offers a mechanism for making collective choices, but those choices are not necessarily the ones that are best for the members of the decision-making group. This is worth noting in light of the fact that elections are often viewed as reflecting the policy preferences of the voters. However, for precisely this reason, democratic governments are always subject to constitutional constraints on allowable actions. In the American case, the Constitution of the United States gives the government limited and enumerated powers rather than determining the scope of government through democratic processes. Increasingly, the popular view of government is moving away from the idea of constitutionally limited powers toward the idea that government should do what its citizens want, as revealed through democratic processes. To see the effects of that shift, one must understand the relationship between political preferences and public policy. condition. However, Gordon Tullock, “Why So Much Stability?” Public Choice 37, no. 2 (1982), pp. 189–202, observes that democratic governments appear very stable, questioning McKelvey’s conclusion.

https://doi.org/10.1017/9781009323178.002 Published online by Cambridge University Press

12

Introduction

1.4  Democracy as Ideology Evaluated as a mechanism for making collective decisions, the preceding discussion shows that democracy falls well short of a method to identify the public interest. When there is widespread consensus, democracy can identify the consensus view, but then, so can any other method of choosing public policy. When preferences among individuals differ, democracy may be no better than choosing policies at random, and might be worse because of the possibility that a majority can use democratic institutions to exploit minorities. But democracy is more than just a mechanism for making collective decisions. It has a strong ideological component. Democracy as an ideology cuts two ways. When viewed as a method for identifying the general will, it runs the risk of legitimizing government actions that exploit minorities, or worse, actions that allow an elite few to exploit the masses. A different ideological vision of democracy is that government should be responsive to the interests of its citizens, and that democratic institutions give citizens the right to replace their governments when they are not, as, for example, the American colonists did with their Declaration of Independence.

1.5  Democratic Institutions and the General Will Jean-Jacques Rousseau, in 1762, gave a clear explanation of the view that the democratic decision-making process reveals the will of the people. He said, The citizen gives his consent to all the laws, including those which are passed in spite of his opposition, and even those which punish him when he dares break any of them … When in the popular assembly a law is proposed, what the people is asked is not exactly whether it approves or rejects the proposal, but whether it is in conformity with the general will, which is their will. When therefore the opinion that is contrary to my own prevails, this proves neither more nor less that that I was mistaken, and what I thought to be the general will was not so.18

This view of democracy legitimizes the decisions of democratic governments by asserting that the outcome of a democratic decision-making process conforms with the will of the people. 18

Jean-Jacques Rousseau, The Social Contract, or Principles of Political Right, translated by G. D. H. Cole. www.constitution.org/jr/socon.htm. 1762, Book IV, Ch. 1, no. 2. While this is a translation from the original French, note that Rousseau uses people as a singular term, further reinforcing the idea that there is a general will that transcends the individual wills of those who compose the people.

https://doi.org/10.1017/9781009323178.002 Published online by Cambridge University Press

1.5  Democratic Institutions and the General Will

13

Rousseau’s vision of democracy has been reinforced in the twentieth and into the twenty-first centuries in the ideology of progressive democracy. For the first century of its existence, citizens of the United States perceived their government as founded on an ideology of liberty, which viewed the role of government as protecting individual rights, along the line of reasoning followed by Locke. Toward the end of the nineteenth century and into the twentieth, the ideology of progressivism began displacing the ideology of liberty. The progressive ideology incorporated an expanded vision of the role of government: Government’s role was not only to protect individual rights but also to look out for people’s economic well-being.19 Progressivism began largely as a reaction against increasingly concentrated economic power. As the nation industrialized, industrialists and financiers such as Rockefeller, Vanderbilt, Carnegie, and Morgan amassed wealth more rapidly than anyone before in human history, and many believed that they were using their economic power to take advantage of those who had less economic power. Progressive policies included regulation – especially of the railroads – antitrust laws, and even court decisions Terry Anderson and P. J. Hill cite the 1877 Supreme Court decision of Munn v. Illinois as a major advance in progressive public policy.20 In response to the complaints of farmers that grain elevator operators were using their monopoly power to undercompensate them for their crops, the state of Illinois began regulating grain elevator prices. When those regulations were challenged, the Supreme Court ruled in favor of Illinois. This was the first time the courts ruled that government has the right to set the terms of exchange, including prices, for private transactions. The Interstate Commerce Commission was established in 1887 to regulate railroads, and the Sherman Antitrust Act was passed in 1890, which allowed the federal government to break up firms that were accused of monopolizing industries. Progressivism was from its beginning redistributive in nature. It justified imposing costs on some for the benefit of others. In its earliest 19

20

I discuss the implications of this ideological shift in Randall G. Holcombe, Liberty in Peril: Democracy and Power in American History (Oakland, CA: Independent Institute, 2019). See also Robert Higgs, Crisis and Leviathan: Critical Episodes in the Growth of American Government, Twenty-Fifth Anniversary Edition (Oakland, CA: Independent Institute, 2012), for a discussion of the implications of this ideological shift. See Terry L. Anderson and P. J. Hill, The Birth of a Transfer Society (Stanford, CA: Hoover Institution Press, 1980), for a discussion of the implications of Munn v. Illinois.

https://doi.org/10.1017/9781009323178.002 Published online by Cambridge University Press

14

Introduction

days, the costs of progressive policies were borne primarily by the very wealthy  – those who were perceived as abusing their substantial economic power. As progressivism developed through the twentieth century, the ideology of progressivism increasingly justified imposing costs on some not because they were taking advantage of others but as a means to the end of providing economic benefits to some other group. The progressive welfare state imposes costs on a broad range of citizens to enhance the economic well-being of others. Progressive regulation is designed to limit the power of business for the benefit of workers and consumers. Progressive taxation is designed to transfer income from those with higher incomes. The twenty-first-century ideology of progressivism, embodied in the welfare state and the regulatory state, justifies imposing costs on some for the benefit of others. Combining the ideologies of progressivism and democracy, the ideology of progressive democracy legitimizes the activities of democratic government. The ideology of progressivism justifies imposing costs on some for the benefit of others, and the ideology of democracy says that when this is done by a democratic government, it is carrying out the will of the people. If one could be confident that democratic institutions revealed the will of the people, as Rousseau suggested, this would not be a concern, but as noted earlier, democratic institutions fall short as a method of aggregating individual preferences to identify the preferences of a group. The language used to describe government actions influences how people perceive those actions. Progressivism conveys the impression of policies that serve the public interest rather than narrow individual interests, and democracy conveys the impression of a government that responds to the demands of its citizens. What some might call the tyranny of the majority is what others call progressive democracy. But the institutions of democratic governments do not always benefit a majority at the expense of a minority. Often it is a minority that benefits at the expense of a majority – narrow special interests can benefit at the expense of the general public. The ideology of progressive democracy legitimizes all actions undertaken by democratic governments. The actions of a democratic government are justified because they embody the preferences of the governed, according to the ideology of progressive democracy. One cannot even disagree with this statement, according to Rousseau, who says that anyone who disagrees is mistaken. Those who have political power in a democracy can use it to benefit themselves at the expense of others, and the ideology of progressive democracy says that when they do so, they are acting in the public interest. This

https://doi.org/10.1017/9781009323178.002 Published online by Cambridge University Press

1.6  Democracy as Freedom from Government Oppression

15

places a heavy burden on any links between political preferences and public policy. If the link is weak, the ideology of progressive democracy offers what amounts to a blank check to the political elite who exercise political power.

1.6  Democracy as Freedom from Government Oppression The ideology of democracy does not always conform with the ideology of progressivism. People often view democracy more broadly as a form of government that holds those with political power accountable to their citizens. Democratic institutions facilitate that accountability, but features of democracy, such as elections, do not guarantee it. In this broader vision of democracy, voting is of secondary importance; the primary characteristic of a democratic government is that it is accountable to its citizens. Democratic elections enable citizens to have some say over who holds the power of government, but the scope of government, in this vision of democracy, is limited to that narrow corridor between anarchy and Leviathan. Democracy in this sense means that citizens control their governments rather than governments controlling their citizens. This idea goes back at least to John Locke and is embodied in the Declaration of Independence that created the United States. The Declaration of Independence consists mostly of a list of grievances against the king of England. The colonists asserted that the British government was violating their rights, and therefore they had the right to establish their own government dedicated to protecting their liberty. Democratic institutions are a means to an end that allows citizens to control their governments. In this view, democracy means government that is accountable to, and controlled by, its citizens. This ideology that views democracy as a component of a government that preserves freedom and protects individual rights remains strong, if often underappreciated. When people are generally in agreement with the activates of their governments, they tend to view democracy as a set of institutions – elections, legislatures, and so forth – rather than viewing it as a guarantor of freedom. When citizens view democracy as a system of government that is controlled by and accountable to its citizens, they are inclined to rise up in protest when government compromises their freedom. The influence of Locke’s ideas on the American Revolution has already been noted, but more contemporary examples show that people still hold Locke’s ideas, even if they are not expressed that way.

https://doi.org/10.1017/9781009323178.002 Published online by Cambridge University Press

16

Introduction

When the Berlin Wall fell in 1989, citizens of the Eastern bloc countries rose up and rallied for democracy to replace their dictatorships, but as they looked across the wall to the West, the right to vote was a small part of what they sought. They wanted the freedom that they saw in the democratic West relative to the oppression that they experienced themselves. While in many ways’ democracy is an impediment to freedom, in this powerful way the ideology of democracy supports freedom because popular opinion equates the two.21 Martin Luther King spoke against government discrimination that compromised individual freedom in the democratic United States. In addition to the civil rights movement that gained prominence in the 1960s, student protests against the Vietnam war offer another example. College students had personal reasons for opposing the war because they were subject to the military draft that compromised the freedom of those who were drafted, but like Martin Luther King, they were protesting a nominally democratic government that was violating this broader ideology of democracy by compromising the freedom of individuals. The Occupy Wall Street movement that began in 2011 offers another example. After the financial crisis that began in 2008, the bursting of the housing bubble along with an accompanying recession meant that many people could not pay their mortgages, and their homes were foreclosed. They perceived a government policy that bailed out the financial firms that owned depreciated mortgage-backed securities, leaving foreclosed homeowners to fend for themselves. In the language of the protesters, the government was supporting the 1 percent rather than the 99 percent – the vast majority of citizens. Government policy supported the insiders and cronies rather than doing what was in the broader public interest. The Occupy Wall Street protesters were objecting to a government that, as they viewed it, subverted democracy by favoring the few at the expense of the many. One can debate the claims of the protesters and some will argue that the government’s policy response was appropriate to the situation, but the point here is that the protesters were objecting to policies that, in their views, were antithetical to the ideology of democracy because they were 21

Bo Rothstein, “Epistemic Democracy and the Quality of Government,” European Politics and Society 20, no. 1 (2019), pp. 16–31, notes that citizens will rebel against their nominally democratic governments when they believe they are acting against the interests of their citizens.

https://doi.org/10.1017/9781009323178.002 Published online by Cambridge University Press

1.7  The Ideology of Democracy Cuts Two Ways

17

not in the best interest of most citizens. Rather than accept Rousseau’s view that policies constructed through democratic institutions express the will of the people, they were objecting to policies that they viewed as being designed by a group of cronies for their own benefit. They were explicitly rejecting the idea that their democratic government was furthering the general will. Believing that a democratic government should act to benefit the masses, the protesters perceived that in this case the policies created through a democratic decision-making process worked against the public interest. In 2019 there were widespread protests against the governments in Hong Kong and in Iran precisely because the protesters viewed their governments as compromising their freedoms. In both cases, some protesters lost their lives to government violence during the protests, indicating a willingness of the protesters to risk their lives to rise up against oppressive governments and demand their freedom. People do not just identify democratic government with citizen rights to vote. Viewed in this way, the ideology of democracy can create a link between political preferences and public policy. When citizens view that link as broken, even when they have the right to vote, history reveals many instances in which they take to the streets to protest the actions of their governments.

1.7  The Ideology of Democracy Cuts Two Ways One ideological view of democracy is that it is a form of government that aggregates citizen preferences to identify a collective preference that represents the general will. Rousseau clearly expressed this view of democracy. Another ideological view is that democracy is a system of government that acts in the interests of its citizens. While similar on the surface, a major difference between them is that the first view of democracy legitimizes the actions of government, whereas the second offers a route to call them into question. The political elite – the people who make the rules that the masses must obey – have a strong interest in promoting that first view because it legitimizes any actions they take. The chapters that follow suggest that the political elite have been fairly successful at promoting this vision of democracy – but not always successful. In general, the political elite have been able to lead citizens to think that the policies they enact embody the political preferences of the masses. When forming their political preferences, citizens are often following their leaders.

https://doi.org/10.1017/9781009323178.002 Published online by Cambridge University Press

18

Introduction

1.8  There Is No Public Interest One complication in thinking about the relationship between political preferences and public policy is that groups do not have interests; only individuals have interests. When one thinks about the interests of a nation, that can mean nothing more than the interests of the individuals who make up that nation. Surely there are some broadly shared interests that can be thought of as national interests, but still, it is the individuals in a nation who have those interests, not some aggregate “nation” that exists beyond the individuals who compose it. People do have broadly shared values, mostly associated with the protection of individual rights, which might constitute a public interest. There is widespread agreement that people should not assault each other or kill each other, and that people should not take the property of others (although there may be disagreement about what constitutes just ownership of property). Public policy can garner widespread agreement in addressing these issues. Issues can be referred to as being in the public interest as a shorthand way of saying that they are in the interests of all of the individuals who make up the public. Issues like these are a small subset of public policy in the twenty-first century. Consider a policy such as requiring that motor fuels contain ethanol, as is the case in the United States. Proponents of this mandate claim it is in the public interest because it lessens the nation’s dependence on foreign oil, and because it is more sustainable and environmentally sound. Meanwhile, it produces financial gains to the producers of ethanol and imposes costs on motorists who must accept ethanol in their fuel regardless of whether they want it. (It is easy to conclude that they do not want it because if they did, there would be no reason for a mandate.) This mandate would appear to be a policy that furthers the interests of a narrow special interest group by imposing widely dispersed costs on the general public. However, supporters of this policy (and every other public policy) argue that the policy is in the public interest. For the policy to be enacted, the public interest argument must at least be plausible enough that the politicians who support it can claim to be acting in the public interest. Many citizens may not care enough about issues like this to become informed, especially if a plausible public interest argument is offered by the policy’s supporters. This raises the question of how interest groups are able to shift political preferences sufficiently that policies that benefit their interests are supported, or at least not opposed, by the political preferences of the masses.

https://doi.org/10.1017/9781009323178.002 Published online by Cambridge University Press

1.9 Conclusion

19

Political preferences are subject to outside influences, and the next several chapters discuss how this happens. There is no such thing as the public interest beyond the individuals who make up the public, but there are good reasons to think that narrow special interests and the political elite can influence the political preferences of the masses.

1.9 Conclusion Prior to the Enlightenment, the ruling class dictated public policy and the masses viewed themselves as subjects of their governments, obligated to act to further the interests of their governments. Enlightenment ideas reversed that view of the relationship between citizens and their governments. Rather than citizens serving their governments, governments should serve the interests of their citizens. This Enlightenment ideology implies that public policies reflect the political preferences of a nation’s citizens. One question, that has been dealt with extensively by scholars over the ages is what correspondence (if any) exists between political preferences and public policy. A second question, which has not been dealt with as extensively, is what determines people’s political preferences. Social scientists often take preferences as given and begin their analysis from there, but one major conclusion in the following chapters is that the masses get their political preferences largely from the political elite. They are following their leaders. This conclusion is enough of a deviation from most analyses in the social sciences that it merits explaining in some detail.

https://doi.org/10.1017/9781009323178.002 Published online by Cambridge University Press

2 Instrumental and Expressive Voting

Analysis of individual behavior in the social sciences typically begins with the assumption that individuals have preferences (utility functions) and that they make choices to maximize their well-being (their utility). Faced with choices, they choose the alternatives that best further their ends. This utility-maximizing framework puts little attention toward understanding why people aim toward the ends they do.1 The value people put on certain ends is always subjective. Some people prefer chocolate ice cream, while others prefer vanilla. Some individuals will work overtime to earn more money to buy a boat, while others do not view a boat as worth giving up enough leisure time to work for it. In either case, one cannot say that one person is right and the other is wrong. They subjectively value the alternatives differently, and make different choices as a result. The preferences people act on in their marketplace decisions are different from those they express as voters. The idea that there are fundamental differences in the nature of the choices people face when they are consumers as compared to when they are voters is well established if not universally accepted by social scientists. This chapter explains that idea. Social scientists have not made much progress in taking the next step beyond recognizing that difference, which is to explain how

1

George J. Stigler and Gary S. Becker, “De Gustibus Non Est Disputandum,” American Economic Review 67, no. 2 (March 1977), pp. 76–90, depict those ends as the result of the relative prices and budget constraints people have faced in the past. A discussion of the way preferences are influenced is undertaken in Chapter 3, and Chapter 5 looks specifically at the formation of political preferences.

20

https://doi.org/10.1017/9781009323178.003 Published online by Cambridge University Press

2.1  Individual Choice in Markets and in Voting

21

people form their political preferences. One of the major goals of this volume is to understand how people form preferences regarding public policies. With an idea of how those preferences are formed, the volume then turns to what types of political preferences people tend to express. The first step is to understand why preferences people express when they vote can differ from the preferences they act on when they make market choices.

2.1  Individual Choice in Markets and in Voting In markets, the choices people make determine what they get as a result, so they have an incentive to choose the alternatives that give them the most utility. When voting, the choices people make have no effect on the outcomes they get. Because their one vote will have no effect on the outcome, there is no link between the choices they make at the ballot box and the outcomes they get as a result of elections. When engaging in market activity, people have an incentive to choose the outcomes they most prefer, because they get what they choose. There is no such incentive when people vote. Because their political choices have no effect on political outcomes, people may express political preferences for outcomes they would not choose if the choices were theirs alone. Vilfredo Pareto (1848–1923), Italian economist and sociologist, made a distinction between what he labeled logical actions and nonlogical actions.2 Logical actions, as Pareto framed them, directly affect outcomes. A diner at a restaurant orders a pizza rather than a salad, and the diner gets a pizza rather than a salad. Nonlogical actions are actions that have no effect on determining an outcome. In an election with a large number of voters, regardless of whether a voter votes for candidate A or candidate B, or chooses not to vote at all, the same person will be elected. Ordering lunch in a restaurant is a logical action; voting is a nonlogical action, to use Pareto’s terminology. Pareto is not saying that nonlogical actions are irrational; just that the choice a person makes has no effect on the outcome the person gets. Pareto made that distinction more than a century ago, so this is not a new idea, but it has more recently been applied, with different terminology, to voting. Geoffrey Brennan and Loren Lomasky call Pareto’s 2

Vilfredo Pareto, The Mind and Society: A Treatise on General Sociology (New York: Dover Publications, 1963).

https://doi.org/10.1017/9781009323178.003 Published online by Cambridge University Press

22

Instrumental and Expressive Voting

logical action instrumental action and nonlogical action expressive action.3 As Pareto explained, instrumental actions determine outcomes, whereas expressive actions do not. In elections, except when the number of voters is very small, the likelihood of a voter casting a decisive vote is vanishingly small, so voting is not an instrumental action. The choice a voter makes has no effect on the outcome of an  election. Voters must realize this, even if they are reluctant to admit it. Anyone who has voted in a few elections must have been able to see that the election outcome would have been the same if they had voted differently, or not voted. They know that their one vote will not change an election outcome. Although most voters will not be familiar with this terminology, they must recognize that voting is an expressive, or nonlogical, act. Voters vote to express their views, even knowing that the election outcome will be the same regardless of how they vote, or even whether they vote. The utility people get from having political preferences and voting comes solely from the satisfaction they get from having those preferences and expressing them. They might get more utility from expressing a preference for an outcome that differs from the outcome they would most prefer to see occur. Academic analysis of voting often depicts voters as choosing among social alternatives, but because a voter’s one vote will not be decisive, voters are not choosing among social alternatives, and it is misleading to depict them as doing so. They are acting expressively, not instrumentally, and as individuals they are not choosing an outcome, they are expressing a preference. There are many reasons to think that the preferences they express at the ballot box may differ from outcomes they would prefer if the choice among social alternatives were actually theirs to make. When a voter votes for A over B, one can conclude that the voter would rather express a preference for A over B, but one cannot conclude that the voter would rather see A win the election over B, because that voter has no ability to determine which alternative wins. Expressing a preference for A over B is not the same as actually preferring A over B, because the voter’s expressed preference has no effect on the election outcome. Voters may have good reasons for expressing a preference for an outcome that is not the voter’s most preferred outcome. One cannot determine whether a voter prefers A or B based on how the voter votes. 3

Geoffrey Brennan and Loren Lomasky, Democracy and Decision: The Pure Theory of Electoral Preference (Cambridge: Cambridge University Press, 1993).

https://doi.org/10.1017/9781009323178.003 Published online by Cambridge University Press

2.3  Strategic Voting

23

2.2  Instrumental Voting Social scientists have developed models of voting in which they assume that voters vote instrumentally, as if their one vote can affect the election outcome. Because voters must realize that one vote will not affect an election outcome (unless there are very few voters), a “rational” voter will not be voting instrumentally, as social scientists define rationality. A rational voter would weigh the costs and benefits of voting and, determining that casting a vote will have only a vanishingly small chance to affect the election outcome, would choose to abstain rather than to incur the cost, no matter how small, of casting a vote. A rational voter would know that there is almost no chance that their one vote will affect the election outcome. For voters who are not “rational” in this sense, it would seem to be a leap of faith to say that they will vote for the outcome they most prefer. To make this leap would require questionable assumptions about rational behavior, although social scientists who model voting systems make those assumptions when they assume voters vote instrumentally.

2.3  Strategic Voting In most cases, instrumental voters vote for the outcomes they would choose if the choice were theirs alone. Candidates A, B, and C are running for office, and the voter would prefer that candidate B be elected, so the voter votes for candidate B. In some cases, instrumental voting may be strategic voting. A voter prefers candidate B, but candidates A and C are far more popular, and the instrumental voter sees little possibility that preferred candidate B could win the election. The voter does not want to cast a wasted vote and votes for candidate A because, even though the voter most prefers B, B has no chance of winning, and the voter prefers A over C. A possible example in which strategic voting might have made a difference is the US presidential election in 2000, when George Bush won over Al Gore. The electoral votes nationwide were close enough that whichever candidate won Florida would win the election, and George Bush won Florida by 537 votes. In that election, third-party candidate Ralph Nader received about 97,000 votes in Florida, and it is likely that if those voters had not voted for Nader, most of them would have voted for Gore, giving him the election. A strategic Nader voter who preferred Gore over Bush would have voted for Gore, knowing that a vote for Nader was taking a vote from Gore, and that Nader could not win.

https://doi.org/10.1017/9781009323178.003 Published online by Cambridge University Press

24

Instrumental and Expressive Voting

If all Nader voters had voted strategically, Gore almost certainly would have been elected to the presidency in 2000. But Gore did not win, and those 97,000 Nader voters cast votes for a candidate they knew had no chance. They were voting expressively, to support Nader for some reason, realizing as they voted that their votes would have no effect on the outcome. And they were correct that their votes had no effect. If all Nader voters had voted for Gore, Gore would have won the election. But each individual voter had only one vote to cast, and as close as the election was, one more vote for Gore would not have changed the election outcome. Nader still would have received about 97,000 votes, and Bush would have won by 536 votes. It was a close election, but still, Bush still would have won had any individual voter decided to vote differently.

2.4  Are Expressive Votes Wasted? In US presidential elections, people often make the argument that they will not vote for a third-party candidate rather than a Democrat or Republican because a third-party candidate cannot win and they do not want to waste their vote. Voting for a third-party candidate could take votes away from the major party candidate they prefer, as apparently was the case for many voters in the 2000 presidential election. If all those Nader voters had voted for Gore, Gore would have won by a comfortable margin. The problem with this reasoning is that each voter only has one vote, and the election was decided by more than one vote.4 The argument that voters waste their votes if they vote for third-party candidates is an interesting one that shows how voters deceive themselves as they rationalize how they will cast their votes. It is clear that voting for a third-party candidate is an expressive act, because the voter knows that the third-party candidate cannot win. The vote is wasted in the sense that it cannot affect the election outcome. But the same is true of voting for a major party candidate. Each voter has only one vote. In the 2000 election, if a voter had voted for Bush, Bush would have been elected to the presidency. If the voter had voted for Gore, Bush still would have been elected. If the voter had voted for Nader, or not voted at all, Bush still would have been elected. Voting is not an instrumental act, and in that 4

Even this may overstate the influence of individual votes. The election was tied up in court cases for more than a month before the Supreme Court of the United States ordered recounts to be stopped, giving Bush the victory. One might argue that in close elections, the courts rather than the voters ultimately decide the winner.

https://doi.org/10.1017/9781009323178.003 Published online by Cambridge University Press

2.4  Are Expressive Votes Wasted?

25

sense every vote is wasted, if by wasted one means that it had no impact on the election outcome. A vote for the winning candidate is wasted unless that candidate wins by only one vote. The winner would have won without that one vote. A vote for a losing candidate is clearly a wasted vote, because the candidate did not win. Following this reasoning, all of the individual votes for Gore in the 2000 election were wasted, because those voters voted for Gore but Bush won the election. Because no one vote will be decisive, even though all the votes in the aggregate determine the election outcome, every vote is wasted in that sense. Except when the number of voters is very small, the wasted vote argument does not reflect the reality of electoral institutions. Even when the number of voters is small, the logic that points toward expressive voting remains valid. When millions of votes are cast in an election, the likelihood of casting a decisive vote is not perceptively different from zero. Even if there are a few dozen voters, so the likelihood of casting a decisive vote is higher, it is still far from certainty. With market choices, the chooser’s choice is decisive. Choosers get what they choose. Even in small elections, one vote is unlikely to be decisive, so expressive voting is likely. People may claim to be voting instrumentally because they want to feel that their act of voting is worthwhile. They frequently argue that their vote along with all others who voted the same way will make a difference. While that is true, each individual still gets only one vote, so the fact that all votes taken together determine an election outcome is not a reason to vote instrumentally. A rational instrumental voter, someone who weighs the costs and benefits of voting based on the likelihood of casting a decisive vote, will abstain, because the time cost alone of casting a vote will impose a greater cost on the voter than the expected benefit of casting a decisive ballot. Even if the cost of voting is very small, the likely benefit of casting a vote that affects an election outcome is even smaller. Potential voters who decide how to vote solely on the utility gain expected from affecting an election outcome rationally decide to abstain. Voters must realize this, but many people still choose to vote even though their votes have no instrumental consequences. They must be voting because they get utility from having and expressing their political preferences through voting. If this is so, then one can understand what they choose to vote for only by understanding what motivates them to make the choice to vote. They know their individual votes will not affect

https://doi.org/10.1017/9781009323178.003 Published online by Cambridge University Press

26

Instrumental and Expressive Voting

the election outcome but they still choose to vote, for some reason. The same reasons that draw them to the polls will determine how they vote. Political preferences are expressive, not instrumental.

2.5  Rational Ignorance Anthony Downs, in an influential book on democratic decision-­making, observed that because voters must realize that their votes have only a vanishingly small likelihood of affecting an election outcome, they will tend to be poorly informed.5 There is little payoff to gathering information to become informed about a choice that has no instrumental consequences. Voters tend to be ignorant of the choices they make at the polls, but rationally ignorant. They have little incentive, as voters have to be informed about a choice that has no effect on the election outcome. People have an incentive to make informed choices when they engage in market transactions, because market transactions have instrumental consequences. People get what they choose in markets, unlike when they vote. When buying an automobile, people have an incentive to collect information about the differences in the many cars on the market so that they can choose the one that best suits their intended purposes. If someone makes a bad choice when buying an automobile, that person suffers the consequences. In voting, a voter suffers no consequences from making a bad (or good) choice. The outcome of the election would have been the same had the voter chosen differently. Some voters are informed, because they are interested and collect information for their own interest. Some people enjoy being informed about politics in the same way that others enjoy being informed about sports. People follow their favorite sports teams and gather statistics on athletes and teams because they are interested, and the same is true for others who gather information about politics. But in both cases, there is no instrumental benefit to gathering the information. Being more informed about a sports team, or being a bigger fan, will not help that team win, just as gathering more political information will have no instrumental effect on an election outcome. Voters have little incentive to cast informed votes, and little incentive to change their expressive preferences at the ballot box even when the preferences they express when 5

Anthony Downs, An Economic Theory of Democracy (New York: Harper & Row, 1957).

https://doi.org/10.1017/9781009323178.003 Published online by Cambridge University Press

2.5  Rational Ignorance

27

voting conflict with their instrumental preferences. People may vote for alternatives they would not choose if the choice were theirs alone. It costs them nothing to do so. Even this line of reasoning may be overly optimistic about voter preferences. Bryan Caplan has make a good case that voters are rationally irrational.6 He notes that people tend to have political biases that, if acted upon, work against their personal interests. Caplan looks at surveys of popular opinions on issues and compares them with “expert” opinions, which on economic issues means the opinions of economists. He finds that members of the general public tend to have an anti-immigrant bias, an anti–free trade bias, a make-work bias, and in general, a skepticism that markets produce results that enhance people’s well-being. The specific issues are of secondary importance. Immigration tends to increase productivity because low-skilled immigrants often take jobs that domestic citizens find undesirable while high-skilled immigrants add to human capital. Free trade benefits everyone because people engage in trade only when there are mutual gains. People often favor makework projects, viewing jobs as a benefit, but on net, jobs are a cost and what workers produce is the benefit. (Of course, workers benefit from the income, but make-work projects impose this cost on everyone else.) But aside from any specific issues, people have little incentive to examine and adjust their views on public policy because their political preferences have no instrumental effects. They can continue holding irrational political beliefs – beliefs that if acted on would make them worse off – with no feedback mechanism to impose costs on them if they vote based on those irrational preferences at the ballot box. Since the beginning of the Industrial Revolution, for example, people have tended to view labor-saving innovations as negative because they cost people their jobs. Economists see this as efficiency gains that allow more goods to be produced at lower cost. A look at the economic history of the past several hundred years shows that even as some jobs are displaced by labor-saving innovations, other jobs are created so that economies remain close to full employment. One would think that, after centuries of labor-saving machinery being introduced into the world economy, nobody would have a job if labor-saving innovations created unemployment. Still, people are critical of automation because they say it costs jobs. 6

Bryan Caplan The Myth of the Rational Voter: Why Democracies Choose Bad Policies (Princeton: Princeton University Press, 2007).

https://doi.org/10.1017/9781009323178.003 Published online by Cambridge University Press

28

Instrumental and Expressive Voting

While economists might be wrong on these issues, Caplan argues that they are more likely to be right than laypeople who have no incentive to study or understand these issues. The result is that voting based on irrational beliefs often leads to bad public policy. In an experimental setting, Andrea Robbett and Peter Hans Matthews found that when subjects were asked to vote on factual questions, when information was costly to acquire they collected less information and voted in a more partisan way than when they were rewarded personally for correct responses.7 Their experimental design uncovered the same type of rational ignorance in voting that Downs described. When choices are instrumental, people will collect more information and will be more likely to make choices in their own interests than when choices are purely expressive. The fact that voters have no instrumental incentive to acquire informed political preferences is an important element in understanding the political preferences they actually express. People have an incentive to make informed choices in their market decisions, because they bear the consequences of those decisions. The political preferences they hold have no effect on political outcomes, so there is no instrumental benefit to citizens and voters from becoming informed.

2.6  Feedback in Markets and in Voting Market transactions offer consumers a feedback mechanism that provides an incentive to make utility-maximizing choices. If someone goes to a restaurant and gets a bad meal, that person has an instrumental incentive to avoid visiting that restaurant again, to avoid getting another bad meal. Voting offers no similar feedback mechanism, because there is no instrumental consequence to making a bad choice. Thus, there is no incentive for voters to make informed choices. Gathering information about differences in automobiles, or differences among restaurants, can make a difference to people’s quality of life. Gathering information about political parties and candidates will make no difference to one’s quality of life, because the same people will be elected regardless of how an individual votes. So, voters are rationally ignorant. 7

Andrea Robbett and Peter Hans Matthews, “Partisan Bias and Expressive Voting,” Journal of Public Economics 157, no. 1 (January 2018), pp. 107–120. Jean-Rovert Tyran and Alexander K. Wagner, “Experimental Evidence on Expressive Voting,” Oxford Handbook of Public Choice, vol. 2 (Oxford: Oxford University Press, 2019), ch. 45, survey additional experimental studies that support the existence of expressive voting.

https://doi.org/10.1017/9781009323178.003 Published online by Cambridge University Press

2.7  Examples of Expressive Voting

29

Not only is voting an expressive choice, feedback to voters is reduced when compared to markets because voters can never observe what a different election outcome would have produced. Buyers of automobiles can see the alternatives they passed up driving on the road. People who eat in restaurants can experience different restaurants to make a comparison. In politics, when one collective choice is made, one cannot observe what would have been the result if the choice had been different. If Al Gore had been elected president in 2000, how would things have been different? One can conjecture, but there is no way to actually observe the differences, because the collective choice of Bush prevented voters from observing the outcome they did not choose. The market offers a variety of choices to consumers so there are good ways to compare real-world alternatives, and to discover what one has passed up by making market choices. The same is not true in politics, where campaign promises are not always fulfilled, and where voters can only conjecture about what would have happened had different parties and candidates been elected. This reduces feedback from political choices even more, compounding the noninstrumental nature of voting. In that sense, voters can never be sure about whether they made a good (or bad) choice when they cast their ballots. The collective choice of one alternative eliminates the possibility of experiencing what would have happened had another alternative been selected. If candidate A is elected, they can experience the result, but they can only conjecture about what would have happened had candidate B been elected. Human nature likely pushes voters who voted for A to think that things would have been worse had B been elected, while voters who voted for B would conjecture that things would have been better had B been elected. There are no facts on which to judge, because in politics, the choice of one option precludes the other.

2.7  Ex amples of Expressive Voting Gordon Tullock provides an interesting example to consider the instrumental effects of expressive voting in an article titled “The Charity of the Uncharitable.”8 Tullock notes that people who have little inclination to engage in charitable activity themselves nevertheless may vote for parties and candidates who favor policies that target redistribution to the less 8

Gordon Tullock, “The Charity of the Uncharitable,” Western Economic Journal 9, no. 4 (December 1971), pp. 379–392.

https://doi.org/10.1017/9781009323178.003 Published online by Cambridge University Press

30

Instrumental and Expressive Voting

fortunate, because giving to charity is an instrumental act whereas voting for it is expressive. Individuals who decide to give money to charity will not be able to spend that money on themselves. They bear a direct instrumental cost from their charitable giving. However, if those same individuals vote for candidates, parties, or polices that promote redistribution, their individual votes will have no effect on the election outcome or on public policy. It is costly to give money to charity, but it imposes no cost on a voter to vote for a party or candidate who promotes charitable policies. By casting a charitable vote, voters can get a good feeling about doing something to help the less fortunate, but at no personal cost to themselves. They feel good about expressing charitable views without having to give up anything, because voting is a nonlogical action.9 Voters could vote for government redistribution programs as expressive acts, even though if the decision were theirs alone, they would choose not to fund those programs. People, as voters, can express support for programs and candidates they would not choose if the choices were theirs alone. Tullock does recognize another possibility, which is that individuals have an incentive to free ride on charitable activities because their individual contributions will be so small that they will not make a noticeable difference in overall charitable activity. However, they would agree to contribute themselves, if everyone else also agreed to give. Such an agreement would be difficult to design through private voluntary agreement, but government can enforce such an agreement by taxing the well-off to transfer resources to the poor.10 Without government coercion, people who want the less fortunate to be helped have an incentive to free ride on the charitable activities of others. Government redistribution overcomes this free rider problem. One problem with this alternative explanation is that there is no way to tell whether it is true. This explanation makes government’s coercive taxation appear to be the result of agreement when no actual agreement took place. If a rich person, taxed to provide transfers to the poor, were asked if he (or she) were in favor of such transfers and answered no, the response prompted by the argument in the previous paragraph would be that the rich person was just trying to free ride on the charitable 9

10

Francesca Gino, Michael I. Norton, and Roberto A. Weber. “Motivated Bayesians: Feeling Moral While Acting Egotistically,” Journal of Economic Perspectives 30, no. 3 (­Summer 2016), pp. 189–212. This is the thesis of Harold M. Hochman and James D. Rodgers, “Pareto Optimal Redistribution,” American Economic Review 59, no. 4, Part 1 (September 1969), pp. 542–557.

https://doi.org/10.1017/9781009323178.003 Published online by Cambridge University Press

2.7  Examples of Expressive Voting

31

contributions of others. This argument makes coercion appear to be the product of consent, when there was no actual consent.11 The argument attempts to justify government intervention, but does not demonstrate that the intervention actually is consistent with the instrumental preferences of those who are coerced by it. Another problem, which Tullock notes, is that most government redistribution does not go to the least well-off. It is targeted to students, to farmers, and to the elderly without regard to their economic status. People tend to use the political process to get government transfers for themselves, rather than working to benefit the least well-off. People who are voting for redistribution are, in large part, not voting to help the needy, but voting to help the politically well-connected. The poor have relatively little political clout compared with the elderly, the education establishment, and even supporters of the arts. One should not be surprised to note that government supports symphonies and operas – entertainment favored by the well-connected – but not rap music. But facts like this may play a minimal role in influencing the political preferences of rationally ignorant voters. Expressive voters are likely to be rationally ignorant of what they actually are financing when they vote for redistributive programs. They believe they are voting for charitable activity, but the political process tends to favor well-organized interest groups over the poorly organized least fortunate individuals in a society. Because voting is an expressive action, the diversion of redistributive funds away from the needy may not affect the way expressive voters vote. They are generating a virtuous signal by voting for redistribution, and may prefer to express their virtue regardless of the actual public policy outcome. In any event, their one vote will not change the outcome. Tullock’s uncharitable voters can rationalize their lack of private charitable activity by telling themselves that redistribution is not their personal responsibility; it is one of the responsibilities of government. The view that redistribution is government’s responsibility, not their own, can minimize any cognitive dissonance that arises from uncharitable people voting for redistribution. They cast expressive votes for charitable activities and feel good about themselves while giving little or nothing to charity from their 11

Along these lines, See Leland B. Yeager, “Rights, Contract, and Utility in Policy Espousal,” Cato Journal 5, no. 1 (Summer 1985), pp. 259–294, and Yeager, Ethics as a Social Science: The Moral Philosophy of Social Cooperation (Cheltenham, UK: Edward Elgar, 2001).

https://doi.org/10.1017/9781009323178.003 Published online by Cambridge University Press

32

Instrumental and Expressive Voting

personal incomes. Many people will not want to think of themselves as uncharitable, even if they are, so they vote expressively for redistributive candidates and parties while spending their own money on themselves. Another possible example of expressive voting is the election of Jesse Ventura as governor of Minnesota in 1998. Ventura, a retired professional wrestler who was an entertaining candidate and had good name recognition thanks to his former profession, ran as a third-party candidate against Democrat Hubert Humphrey III and Republican Norm Coleman. Both major party candidates spent more on their campaigns than Ventura, but apparently failed to excite their bases, leading to Ventura’s election. While one cannot climb inside the minds of Ventura voters, it is easy to imagine how expressive voting led to Ventura’s election. A Democrat going to the polls, unexcited about Humphrey, would certainly not vote for the Republican, but Ventura offered an alternative for that voter to express displeasure with the Democratic candidate. Similarly, Republican voters, unenthusiastic about Coleman, would not vote for the Democrat, but Ventura was available as an expressive outlet. So, Republicans and Democrats, unexcited about their party’s own candidates, could cast an expressive vote in protest to the underfunded but well-recognized thirdparty candidate Ventura. There is no instrumental cost to casting such a vote, because each individual’s one vote will not decide the election, and indeed, Ventura won by more than one vote. Expressive voting offers a plausible answer to how political novice and third-party candidate Ventura was able to win that election. Imagine an alternative in which, after arriving at the polls, a Ventura voter is pulled aside and told that for this election, Minnesota’s governor was going to be chosen by a different process. One voter would be chosen at random to decide who would be the governor, and the voter is told, “That voter is you. Who do you choose to be Minnesota’s next governor?” It is easy to conjecture that almost every Ventura voter would have chosen either the Democrat or the Republican, because under that alternative scenario, the choice would be instrumental. One individual would choose the governor. But because voting is an expressive rather than instrumental activity, Ventura got more votes than either major party candidate.

2.8  Costs and Benefits of Expressive Voting Brennan and Lomasky offer a simple numerical example to illustrate why people might vote for policies they would not choose if the choice

https://doi.org/10.1017/9781009323178.003 Published online by Cambridge University Press

2.8  Costs and Benefits of Expressive Voting

33

were theirs alone. Assume that if a particular policy were passed, an individual voter would suffer a loss in utility equal to $500. The individual would not choose to implement that policy if it were up to that individual to make the choice. However, assume that the individual would experience a utility gain of $0.05 for expressing a preference for that policy. Because an individual’s one vote will not determine whether the policy is implemented, there are no instrumental consequences from voting for the policy, so the individual votes for the policy to gain that $0.05 in utility. That expressive benefit might come from the good feeling the voter gets from voting for a charitable activity, as in Tullock’s example, or it might come from a feeling of camaraderie from friends or family who also express support for the policy. The instrumental cost the policy would place on the voter is not relevant to the choice of whether to vote for it, because the election outcome will not be affected by that voter’s vote. Voting carries with it no instrumental cost, so voters vote expressively. In one sense, expressive voting has an instrumental component, in that the reason a voter would choose to vote is to accomplish some end. The expressive nature of voting comes because the end the voter can hope to accomplish is not to affect the election outcome,12 but to accomplish some other end that is solely the result of being able to express a preference, even though it will not affect the election outcome. Arye Hillman refers to examples like this as policy traps.13 People’s instrumental preferences would lead them, as individuals, to favor one candidate or policy, yet the expressive value of voting leads them to, in the aggregate, vote for a different candidate or policy that is not in their interest and results in lower utility for everyone. Consider the hypothetical example that starts this section, and assume that those hypothetical numbers apply to all voters. Each individual would be $500 better off if the policy in the example were not passed, but votes for the policy because no one vote is decisive, and each individual gains $0.05 in utility from expressing a preference for the policy. The result is that each individual gains $0.05 from voting for the policy and loses $500 because

12

13

This point is made by Alan Hamlin and Colin Jennings, “Expressive Political Behaviour: Foundations, Scope, and Implications,” British Journal of Political Science 41, no. 3 July (2011), pp. 645–670. Arye L. Hillman, “Expressive Behavior in Economics and Politics,” European Journal of Political Economy 26, no. 4 (December 2010), pp. 403–418.

https://doi.org/10.1017/9781009323178.003 Published online by Cambridge University Press

34

Instrumental and Expressive Voting

the policy is approved, for a net loss of $499.95. The numbers are hypothetical but the principle is not. Expressive voting has the potential to produce collective choices that lower the utility of every individual in the voting group.

2.9  Expressive Behavior of Voters and Sports Fans The analogy between voters and sports fans is relevant to understanding what motivates voters, as it was when considering rational ignorance. Sports fans will attend games and cheer for their teams even though their participation as fans will make no difference to the outcome of the sporting event. Even if in the aggregate the cheering of fans increases a team’s likelihood of winning (and home court and home field advantage suggests this is the case), each individual fan’s contribution to the crowd’s noise will be inconsequential. Just as no one vote determines an election outcome but in the aggregate all the votes do, no one fan’s support will affect the outcome of the contest even if the support of all fans in the aggregate do. Fan support, like voting, is an expressive rather than instrumental activity. When sports fans cheer for their favorite teams, they know that their cheering will not help the team win; rather, they are expressing their loyalty to the team. The same is true of expressive political behavior. When people tell others about their support for candidates and parties, and when they vote for those candidates and parties, they are expressing their loyalty, so that they can associate themselves with the candidates and parties they support. They do it to reinforce their own political identity, and to convey that identity to others. In the case of Tullock’s charity of the uncharitable, they get a good feeling for themselves by casting the charitable vote, and they can leverage their support to express to others their political leanings. Their expressed political preferences make them feel virtuous, in their own eyes and, they hope, in the eyes of others. The value of the expressive action comes solely from the action itself rather than from consequences that result from the action. One can see, in the case of sports fans, that for almost everyone the outcome of a sports contest will have no instrumental value. People’s lives will be no better or worse, materially, whether their teams win or lose, even though those people may be happier if their teams win. There is no instrumental reason for someone to be the fan of one team rather than another. The same is not true for political choices, which in the

https://doi.org/10.1017/9781009323178.003 Published online by Cambridge University Press

2.10  Ethical Voting

35

aggregate do have instrumental consequences. The material well-being of sports fans is independent of which teams win sporting contests,14 whereas people’s material well-being is directly affected by the which candidates and parties win elections. While readers should be wary of making too much of the analogy between sports and politics, the material well-being of those who are on sports teams does depend on the success of their teams, and on their individual performance. Coaches who lose too many contests also lose their jobs, and players who perform well get paid more while those who perform poorly also lose their jobs. The players and coaches, in contrast to the fans, make instrumental decisions that determine whether their teams win. These people can be thought of as the sports elite, to differentiate them from sports fans. Similarly, while the masses make no instrumental choices when they vote, the political elite make the instrumental decisions that determine public policy, and affect their ability to keep their positions. They can lose elections, or if not in elected positions, can lose their jobs by making unpopular choices. The political elite and the sports elite act instrumentally, whereas voters and sports fans act expressively. The sports elite have the clear objective of winning, and the same is true of the political elite, who want to win elections and keep their jobs. However, keeping in mind Caplan’s idea of rational irrationality in voters, one might question whether policies that win elections are also those that are in the best interest of voters.

2.10  Ethical Voting One possible reason people may choose to express their political preferences through voting is that they feel they have an ethical responsibility to others, not only to vote, but as Bart Engelen conjectures,15 to vote in the public interest. Thus, although people vote expressively, the preferences they express are those they believe are in the best interests of everyone. The fact that voting has no instrumental consequences for the individual voter may even facilitate other-regarding behavior.

14

15

Sports bettors might gain materially if the teams they bet on win, but serious bettors bet based on their estimates of teams winning or beating the spread, rather than blindly as fans for their team. Bart Engelen, “Solving the Paradox: The Expressive Rationality of the Decision to Vote,” Rationality and Society 18, no. 4 (2006), pp. 419–441.

https://doi.org/10.1017/9781009323178.003 Published online by Cambridge University Press

36

Instrumental and Expressive Voting

Consider again Tullock’s example of the charity of the uncharitable – the people who do not engage in private charitable activity because it imposes a personal cost on them, but nonetheless vote for government redistribution. One way to view this is that the expressive nature of voting lowers the cost of engaging in other-regarding behavior. People vote to collectively act in the public interest, even though they are very selfinterested in their private actions. People maximize their own utility in their private actions, but choose to further the public interest when they vote because there is no instrumental cost to doing so. This line of reasoning is plausible, because the utility in casting an expressive vote lies only in having and expressing those preferences. There is no instrumental effect, so people might feel good about casting a vote in what they perceive is the public interest. They get utility from casting the charitable vote, without bearing any personal cost. Still, one must recognize that the act of voting is expressive, not instrumental, and so carries with it all of the above-discussed issues with expressive voting. Setting aside agency problems (once in office, parties and candidates do not always do what they promised), voters tend to be rationally ignorant about the alternatives in front of them. They have no instrumental incentive to understand the platforms on which parties and candidates are running. This raises the question, addressed in the following chapter, of how voter preferences are actually determined. Voters may have good intentions when voting, but many voters are expressing preferences based on limited information. Good intentions do not overcome rational ignorance. This points toward the question of whether voters have an ethical obligation to vote. The argument is often made (by politicians) that people have a patriotic duty and an ethical responsibility to vote. But if voters are poorly informed, or do not want to vote for other reasons, why would they have an ethical obligation to vote and dilute the voting power of those citizens who have a positive preference for casting a ballot? Politicians like high voter turnout because it makes the winners appear that they have a mandate and the support of the citizens, but this creates no ethical obligation to vote. Indeed, one might make an argument that if individuals are uninformed about the political choices they face – and there is an incentive to be rationally ignorant – that those individuals have an ethical obligation to abstain and allow collective decisions to be made by those who are knowledgeable. There is the possibility that people are more likely to vote in what they perceive is the public interest because doing so brings with it no

https://doi.org/10.1017/9781009323178.003 Published online by Cambridge University Press

2.11 Conclusion

37

instrumental costs. Nevertheless, the decision to do so is an expressive choice, not an instrumental one. This points toward the question of how people determine their expressive preferences.

2.11 Conclusion Voters who engage in fully rational behavior, as economists describe it, would not vote. When deciding whether to vote, “rational” voters would weigh the expected instrumental costs and expected instrumental benefits, and the costs would always outweigh the benefits. The likelihood that a voter’s individual vote will affect the outcome of an election with anything other than a very small number of voters is vanishingly small, so there is no instrumental benefit from voting. If there is even a tiny cost that comes with taking the trouble to cast a ballot, a weighing of the costs and benefits tells the instrumental voter to abstain. But many people do vote, even though they must realize that their individual votes have no effect on an election’s outcome. The utility they get from casting a vote must come solely from the utility they get from having a preference and expressing that preference at the ballot box. They vote because they gain utility from expressing a preference. They do not vote because they intend to affect an election’s outcome. Voters are not choosing an outcome when they vote; they are choosing to express a preference that has no effect on an outcome. Thus, one cannot conclude that expressing a choice for A over B at the ballot box means that the voter would prefer that A be collectively chosen over B. Before one can understand why voters vote for one alternative over another, one must understand why voters vote at all. Whatever motivates them to go to the polls will also be what motivates them to express the preferences they do once they are there. The idea that individuals choose differently when the choices they face are expressive rather than instrumental has a long history in the social sciences, dating back at least a century to Pareto’s clearly formed distinction. The idea has been recognized more recently by social scientists who have applied it specifically to voting, but it has not been fully incorporated into voting theory. Models of voting mechanisms do not make any distinction between instrumental and expressive preferences. One possible reason that the distinction has rarely been taken into account is that while the literature on expressive voting notes that expressive preferences are likely to differ from instrumental preferences, that literature has not explained how expressive preferences are formed.

https://doi.org/10.1017/9781009323178.003 Published online by Cambridge University Press

38

Instrumental and Expressive Voting

When looking at instrumental preferences, economists often assume that individuals want to maximize their incomes and wealth, and that they get utility from consuming goods and services. These instrumental outcomes are not relevant to expressive choices, because expressive choices have no effect on instrumental outcomes. This raises the question, largely unaddressed by social scientists, about what factors shape the expressive political preferences of citizens and voters. To integrate expressive voting into an understanding of democratic decision-making requires some understanding of how individual preferences are determined when they make expressive choices. The task that lies ahead is to further that understanding.

https://doi.org/10.1017/9781009323178.003 Published online by Cambridge University Press

3 Influences over Preference Formation

Chapter 5 looks specifically at factors that influence people’s political preferences. This chapter discusses preference formation more generally. Economic analysis typically takes individual preferences as given and begins its analysis from there. Nobel laureates George Stigler and Gary Becker make a strong case for this methodology, saying that once we start attributing different choices people make to differences in their utility functions – their preferences – that amounts to saying we do not know why they are making the choices they do.1 The analysis that follows is not at odds with Stigler and Becker’s recommended methodology. Stigler and Becker simply say that an economic explanation for the different preferences people exhibit must rest on differences in the relative prices and budget constraints they face. People who face different opportunities will make different choices. The ideas discussed in this chapter carry over into Chapter 5, which looks at factors that can affect the utility people get from holding and expressing particular political preferences. Given the economist’s predilection for assuming preferences are given, they might express skepticism about the arguments in Chapter 5 about factors that influence political preferences, and in particular, factors that cause people to change their political preferences. This chapter takes a step back from political preferences to look at factors that affect preferences more generally. Consider the individual who grew up in a household whose parents listened to classical music recordings, perhaps played such music themselves, and had their children take classical music 1

George J. Stigler and Gary S. Becker. “De Gustibus Non Est Disputandum,” American Economic Review 67, no. 2 (March 1977), pp. 76–90.

39

https://doi.org/10.1017/9781009323178.004 Published online by Cambridge University Press

40

Influences over Preference Formation

lessons. These factors lower the cost of acquiring a taste for classical music, whereas someone who grew up in a household that listened to country music would be more likely to have a taste for that type of music. Preferences in music differ, but they can be explained by past exposure that has lowered the cost individuals face in acquiring a taste for certain types of music. Food and fashion are perhaps more obvious examples. People’s preferences for different types of food, fashion, and music originate in experiences they have had in the past, and those with whom they have associated. Everyone understands the influence that people’s cultural background has on their preferences, or that when people are exposed to others with different preferences, their own preferences might shift toward those of their peer group. One might “develop a taste” for new foods, new fashions, and new forms of entertainment after being exposed to them. The same might be true of political preferences, but that idea is explored in Chapter 5, after considering preference formation more generally in this chapter. Individuals can make investments themselves to acquire and refine their preferences, but outside influences also play a role, and this chapter deals specifically with those outside influences that can alter people’s preferences. Schools offer an obvious example. With regard to political preferences, critical race theory offers a controversial example from the 2020s, questioning the way that race has been considered in American history and in contemporary politics and political institutions.2 Others have argued that it presents a biased, divisive, and misleading analysis of racial issues.3 Without taking sides on critical race theory, it is apparent that people on both sides of the issue intend to influence the political preferences of students and the population more generally. The debate on critical race theory is more along the lines of the subject matter in Chapter 5, which discusses how people’s expressive preferences are influenced. The present chapter makes the case that people’s instrumental preferences are also subject to influence. If the choices people make when those choices have direct consequences are subject to outside influence, it is much more likely that expressive choices, which have no instrumental consequences, can be shaped by outside influences. 2 3

See, for example, Richard Delgado and Jean Stefancic, Critical Race Theory: An Introduction, 2nd ed. (New York: New York University Press, 2012). For example, James Lindsay and Helen Pluckrose, Cynical Theories: How Activist Scholarship Made Everything about Race, Gender, and Identity (Durham, NC: Pluckstone Publishing, 2020).

https://doi.org/10.1017/9781009323178.004 Published online by Cambridge University Press

Influences over Preference Formation

41

James Buchanan says that economists should focus their attention on the study of exchange rather than the study of choice.4 Building on definitions of economics that go back to Lionel Robbins, Buchanan says that economists tend to view the profession as studying how individuals and societies choose to allocate scarce resources to satisfy competing ends.5 This focus on choice takes place when looking at individual behavior, characterizing individuals as choosing combinations of goods to maximize their utility functions subject to constraints. It takes place at the aggregate level when nations are characterized as collectively choosing how to allocate scarce resources. A common example, dating back to World War II, is analyzing how nations choose to allocate resources between guns and butter. Should we give up some consumption opportunities to produce more security? Buchanan first notes that aggregates, such as nations, do not choose; only individuals choose. This observation is especially relevant to an analysis of political preferences, in that the electorate does not choose anything in the process of an election. Rather, individual voters make their own choices on how, or whether, to vote. Then votes are aggregated to make a collective decision, but it is not the collective that decides anything. It depends both on how individuals vote and on how votes are aggregated. As noted in the previous chapter, this can clearly be seen in voting for the US president, which is done though the electoral college. From time to time, one candidate receives a higher number of popular votes while another receives a higher number of electoral votes and wins the election. If votes were aggregated by totaling up the votes for each candidate, a different candidate might win than if they are aggregated through the electoral college.6 In voting, all choices are individual choices. The aggregate outcome is the result of a particular method of aggregating those individual choices. 4 5 6

James M. Buchanan, “What Should Economists Do?” Southern Economic Journal 30, no. 3 (January 1964), pp. 213–222. Lionel Robbins, The Nature and Significance of Economic Science (London: Macmillan, 1932). With electoral college voting, one cannot be sure that a candidate who receives the most popular votes would also receive the most popular votes if the election were determined by popular vote totals, because candidates would campaign differently and voters would vote differently under different electoral institutions. In early twenty-first century presidential elections, for example, there was little doubt that Democrats would win California, so Republican presidential candidates had no reason to campaign there, and Republican voters would be more likely to abstain from voting. If the popular vote total were what counted, rather than the state’s electoral votes, Republican presidential candidates surely would have received more popular votes in California than they actually did under the electoral college.

https://doi.org/10.1017/9781009323178.004 Published online by Cambridge University Press

42

Influences over Preference Formation

Buchanan also is critical of economic analysis for focusing on the way that individuals choose. The neoclassical framework, promoted by John R. Hicks and Paul A. Samuelson, depicts individual choice as maximization subject to constraints.7 Individuals have utility functions and faced with constraints, they choose the utility-maximizing consumption bundle. This line of reasoning depicts individuals as acting like computers, not actually making choices but following a pre-determined algorithm that determines their actions. Buchanan views the economy as a set of institutions through which individuals interact. Individuals make their own decisions about what is important to them, and interact with other individuals so that they can improve their well-being. Given what individuals hope to achieve, the economy is a set of institutions that allows people to interact to further their goals. Government fits into this institutional framework when people have goals that require collective action. In addition to food, clothing, and shelter, people want law and order, to be protected from potential aggression from others. They want infrastructure that will be shared and consumed collectively. Government institutions open the opportunity for people to interact with others to move toward these collectively shared goals. Economic analysis, as Buchanan sees it, should focus less on how individuals decide what they want, but given what they want, how they interact with others to get it. Even when economists analyze choice, however, they tend to take individual preferences as given. People have preferences, and economic analysis examines how the economy works given those preferences. Paul Samuelson’s revealed preference theory is an example.8 Samuelson presents a mechanism by which people’s preferences can be revealed – assuming those preferences are given and constant. This is the standard method of economic analysis. People’s preferences are taken as given, and analysis begins from there. The chapters that ­follow build on the idea that people’s political preferences are ­subject to change, and look at the factors that influence them. Assumptions that preferences are given misrepresent the way that democratic (and other) governments make decisions and produce  public  policy. 7

8

John R. Hicks, Value and Capital: An Inquiry into Some Fundamental Principles of Economic Theory (Oxford: Oxford University Press, 1939) and Paul Anthony Samuelson, Foundations of Economic Analysis (Cambridge, MA: Harvard University Press, 1947). For an excellent critique of this line of analysis, see Meir Kohn, “Value and Exchange,” Cato Journal 24, no. 3 (Fall 2004), pp. 303–339. Paul A. Samuelson, “A Note on the Pure Theory of Consumer Preference,” Economica n.s. 5, no. 17, pp. 61–71.

https://doi.org/10.1017/9781009323178.004 Published online by Cambridge University Press

3.1  Market Choices

43

Because  the  ways that political preferences are formed, and the ways that they are subject to being influenced and changed, are key elements in the analysis that follows, this chapter takes a critical look at ways that preferences are influenced. For some purposes, assuming people’s preferences to be given and unchanging works well, but analyzing how political preferences affect public policy is not one of them.

3.1  Market Choices Because political preferences only have expressive value, there is good reason to think that they are more subject to outside influence than preferences in markets, where the choices people make determine the outcomes they get. So, a good place to start to analyze preferences is to look at the preferences people have for alternatives when they make market choices. If those choices are subject to outside influence, then it is even more likely that political preferences will be also. An early and insightful observer of consumer choice was Thorstein Veblen, who explained that people’s consumption choices often are affected by the way they believe that they will be perceived by others.9 Consumer choices not only have instrumental value but also, in a social setting, expressive value. Veblen used the term conspicuous consumption to refer to consumer choices that people make to impress others. While there may be some instrumental value in buying jewelry or luxury automobiles, it is easy to see that a significant share of the utility from those goods comes from their ability to impress others. What additional value beyond showing it off to others is there in having a one-carat diamond in an engagement ring rather than a half-carat, or even a ring with no diamond? Similarly, a luxury automobile provides little if any additional instrumental value over a less-expensive automobile, but signals to others a person’s high status. Rare works of art provide perhaps a better example. Art critics might find few nuances in paintings that are valued at millions of dollars that are not present in paintings costing only hundreds, but it is unlikely that the well-heeled purchaser is buying the rare and expensive painting because of those nuances. Purchasers hire experts to help them avoid buying forgeries, indicating that individual purchasers are unable to tell the difference. When excellent reproductions of great art works are available, it would seem that the only reason to pay a premium to buy an original is to impress others. 9

Thorstein Veblen, The Theory of the Leisure Class (New York: Macmillan, 1899).

https://doi.org/10.1017/9781009323178.004 Published online by Cambridge University Press

44

Influences over Preference Formation

Economist Robert Frank has written extensively on the way that consumer preferences and consumer satisfaction are affected by the choices of others. He says that once people’s basic needs are taken care of, the utility they get from goods they purchase depends on how they compare with the goods of others around them.10 He offers an example of an outdoor grill, saying that a very basic grill can cook food just as effectively as high-end models that cost many times more, but people buy those highend models because they want to impress their friends and neighbors. The high-end grill is an indicator of status, Frank says, like the luxury car and the huge diamond in the engagement ring. Beyond the very basic models, goods are in this sense positional. The value of that big diamond to the person who wears it does not come from its absolute size, but rather from how big it is relative to the diamonds other people are wearing. Frank argues that people wastefully overspend on goods to engage in conspicuous consumption, and uses this as an argument for highly progressive income taxation. If an individual drives a Ferrari while his neighbor only has a Corvette, the Ferrari driver gets utility from the increased status associated with the Ferrari. But, Frank argues, if both individuals had lower incomes – perhaps because of a more progressive tax structure – and the first individual had a Corvette while the neighbor had only a Camaro, the difference in status would be preserved and the individual would get just as much utility from the Corvette (comparing it to the neighbor’s Camaro) as he would have received from the Ferrari (when compared to the neighbor’s Corvette). If this is the case, Frank offers some advice on pursuing happiness. Choose a peer group in which you can be a high-status individual. People are happier if – to use an old saying – they are the big fish in a small pond rather than the small fish in a big pond.11 The big fish in the small pond has the satisfaction of being a high-status individual within the person’s peer group. Rather than stretching one’s budget to buy a house in the most exclusive neighborhood, where all the neighbors will drive nicer cars and take more expensive vacations, a person will be happier in a more modest neighborhood in which that person’s cars and vacations look better than the neighbors’. 10 11

Robert Frank, Luxury Fever: Weighing the Cost of Excess (Princeton: Princeton University Press, 2000). Robert H. Frank, Choosing the Right Pond: Human Behavior and the Quest for Status (New York: Oxford University Press, 1985).

https://doi.org/10.1017/9781009323178.004 Published online by Cambridge University Press

3.2 Advertising

45

The “bandwagon effect” is another example of people’s consumption choices being based on the opinions of others. In this case, people make consumption choices because they appear to be popular choices. They do not choose what to consume to look better than their peers, as with conspicuous consumption, but to fit in and look the same as their peers. Fashion is an almost tautological example, in that things are fashionable because many people choose them, whether one is talking about clothing, musical tastes, or fad diets. People’s choices are influenced by the choices of others. The point is that people’s preferences are affected by the preferences of those around them – the bandwagon effect – and by their views on how others will perceive those preferences. People’s consumer choices have a substantial expressive component, even though they have instrumental effects, so one should expect that political preferences are even more likely to be affected by the preferences of others. Veblen also describes a snob effect, which leads people to buy goods that indicate high income and status – goods that those with lower incomes and status would not buy. A Rolex watch may be a good example. Inexpensive watches keep time just as well and can be just as durable, but they do not signal high status in the same way as a Rolex. People’s preferences are affected by the way they believe that others perceive them. If this is true over preferences for “snob” goods that reduce the purchaser’s overall purchasing power, it is likely more true for political preferences, which carry with them no instrumental costs.

3.2 Advertising It (almost) goes without saying that advertising is based on the premise that people can be persuaded to alter their consumption patterns, and their political preferences, with the right kind of sales pitch. Vance Packard, in 1957, wrote about techniques that he called the hidden persuaders, because they were based on psychological techniques that tended to pass unnoticed by those who were persuaded.12 The use of these hidden persuaders has become more refined over the many decades since Packard wrote about them. Packard says that “many of the nation’s leading public-relations experts have been indoctrinating themselves in the lore of psychiatry and the social sciences in order to increase their skill at ‘engineering’ our consent to their propositions.” He discusses firms that advertise to sell goods, but also notes that these methods are 12

Vance Packard, The Hidden Persuaders (London: Longmans, Green and Co.,1957).

https://doi.org/10.1017/9781009323178.004 Published online by Cambridge University Press

46

Influences over Preference Formation

“showing up nationally in the professional politicians’ intensive use of symbol manipulation and reiteration on the voter, who more and more is treated like Pavlov’s conditioned dog.”13 Even back in the 1950s when Packard was writing, he saw the potential for the masses to be manipulated, noting, “We move from the genial world of James Thurber into the chilling world of George Orwell and his Big Brother, however, as we explore some of the extreme attempts at probing and manipulating now going on.” He goes on to observe, “The national chairman of a political party indicated his merchandising approach to the election of 1956 by talking of his candidates as products to sell.”14 Twenty-first-century readers are unlikely to be surprised by this, indicating the degree to which people have come to accept the idea that people can be persuaded to change their views on what products to buy, and what political positions to support. Most individuals will be reluctant to admit that their own preferences have been manipulated this way, even as they perceive that the preferences of others have been. Packard goes on to explain how people buying products are buying an image as much as the physical product itself. Products do have instrumental value to consumers, but they also have expressive value, and Packard concludes that the more effective way to sell products is to advertise the image the product expresses rather than the instrumental utility it provides. People have the option of buying many different detergents and many different automobiles. Obviously, they will not shop for a detergent if their instrumental goal is to purchase transportation, but choosing among the many options that would satisfy the purchaser’s instrumental purposes, they choose the option that appeals to their expressive preferences. Conspicuous consumption, the bandwagon effect, and the snob effect observed by Veblen play a heavy role in advertising, and from the money they spend on advertising, it is apparent that advertisers believe it affects consumer preferences. Advertisers realize what drives a purchaser’s choices and direct their advertising toward those expressive desires. When applied to politics, the choices among parties and candidates are purely expressive, because the individual’s choice does not affect the political outcome. Political advertising is designed to persuade people that they would be happier by expressing a preference for the advertiser’s candidates. If advertisers can persuade consumers to buy one 13 14

Packard, Hidden Persuaders, p. 1. Packard, Hidden Persuaders, p. 2.

https://doi.org/10.1017/9781009323178.004 Published online by Cambridge University Press

3.2 Advertising

47

product over another, surely advertising would be more effective in the purely expressive realm of political preferences. A subfield of motivational research helps inform advertisers of the most effective ways to manipulate people’s preferences. The secret to selling a product, or a candidate, is to appeal to people’s feelings. Advertising alters people’s choices by addressing their subconscious desires. It can pass facts along to consumers (and voters), and economists tend to view advertising as a way to convey information, but its persuasive power comes from the subconscious appeal of the ad more than through the information it conveys. Advertisers, Packard says, try to “build images that would arise before our ‘inner eye’ at the mere mention of the product’s name, once we have been properly conditioned.”15 This is likely more true when the product is a political candidate than when it is a good offered for sale. Packard notes, “The manipulative approach to politics is of course not a discovery of the nineteen-fifties, or even the twentieth century. Napoleon set up a press bureau that he called, perhaps in a playful moment, his Bureau of Public Opinion. Its function was to manufacture political trends to order. Machiavelli, was another who made some original contributions to the thinking in this field.”16 Echoing Pareto, who he references, Packard cites “growing evidence that voters could not be depended upon to be rational. There seemed to be a strong illogical or nonlogical element in their behavior, both individually and in masses.”17 Packard quotes a campaign strategist who says, “I think of a man in a voting booth who hesitates between two levers as if he were pausing between competing tubes of tooth paste in a drugstore. The brand that has made the highest penetration on his brain will win his choice.”18 The political preferences people hold should be more manipulable than their preferences for goods and services in the market, because they are expressive rather than instrumental. Their value comes solely from the utility people get from having and expressing them, so it follows that people will adopt preferences that make them feel good about themselves, and advertising is a tool that the political elite can use to do that. Advertising appears very effective at influencing people’s choices in markets, where their preferences have a direct and instrumental impact 15 16 17 18

Packard, The Hidden Persuaders, p. 26. Packard, The Hidden Persuaders, p. 103. Packard, The Hidden Persuaders, p. 105. Packard, The Hidden Persuaders, p. 111.

https://doi.org/10.1017/9781009323178.004 Published online by Cambridge University Press

48

Influences over Preference Formation

on outcomes, and it is not unreasonable to believe that advertising is even more effective in forming political preferences, which, because they are expressive rather than instrumental, tend to be based more on emotion than on reason.

3.3  Created Desires One of the goals of advertising is to convince people they want things they may not have otherwise wanted. John Kenneth Galbraith goes a step further with this idea, noting that many of those things that advertising convinces people they desire are created wants, desired only because they were created. If they did not exist, people would not miss them. Once created, advertising makes them appear desirable. One can talk about people’s desires to have goods to sustain their lives and make them more comfortable, but people would have no desire for many goods had they never been created. Nobody would desire those goods that Veblen classified as demanded for conspicuous consumption had they not been produced to begin with. Galbraith says, “One cannot defend production as satisfying wants if that production creates the wants.”19 Galbraith goes on to say, “Consumer wants can have bizarre, frivolous, or even immoral origins, and an admirable case can still be made for the society that seeks to satisfy them. But the case cannot stand if it is the process of satisfying wants that creates the wants.”20 Consider again a diamond engagement ring. It has expressive value but no instrumental value in that people would be no worse off if diamond rings did not exist. What if there was no symbolism associated with a diamond ring? There would be little demand for them – or for rings in general for that matter. Rings are a good example of goods that provide utility only for their symbolic and expressive value. They have no other function. Adam Smith noted in The Wealth of Nations in 1776 that while diamonds are expensive, they have little value in use. Smith said, “Things which have the greatest value in use have frequently little or no value in exchange; and on the contrary, those which have the greatest value in exchange frequently have no value in use. Nothing is more useful than water; but … scarce anything can be had in exchange for it. A diamond, 19 20

John Kenneth Galbraith, American Capitalism: The Concept of Countervailing Power (Boston: Houghton Mifflin, 1952), p. 134. Galbraith, American Capitalism, p. 135.

https://doi.org/10.1017/9781009323178.004 Published online by Cambridge University Press

3.4  Political Advertising

49

on the contrary, has scarce any value in use; but a great quantity of other goods may frequently be had in exchange for it.”21 Diamond jewelry – and jewelry more generally – has no function beyond its expressive value. As Smith noted, diamonds have no value in use. Advertising has sold people the idea that they should desire diamond rings. The De Beers company, the preeminent diamond-mining company, initiated an advertising campaign in 1947 to tell people “A Diamond Is Forever.” Encouraged by advertisers, consumers pay large sums of money to buy diamonds, which they desire in part because they cost large sums of money. Once people’s basic needs for food, clothing, and shelter are satisfied, their demand for additional goods and services is driven largely by expressive concerns. Their desire for status, to impress others, and to feel good about themselves can even overshadow their desire for necessities. People will give up meals to spend money for expressive purposes. Advertising feeds these expressive desires. Galbraith says, “The even more direct link between production and wants is provided by the institutions of modern advertising and salesmanship. These cannot be reconciled with the notion of independently determined desires, for their central function is to create desires – to bring into being wants that previously did not exist.”22 Advertising manipulates people’s preferences to make them want things they previously had no idea they wanted. Galbraith refers to this as the dependence effect and notes “that wants are dependent on production.”23 This holds true for political advertising as much as or more than for advertising to sell goods and services. The purchase of goods and services entails an opportunity cost. Money spent on one thing cannot be used to buy something else. But when advertising acts to sell people political preferences, there are no instrumental costs to buying the message that is being sold.

3.4  Political Advertising Galbraith calls the creation of wants through advertising and salesmanship the dependence effect, and he sees little merit in it. “Wants thus 21 22 23

Adam Smith, The Wealth of Nations (New York: Modern Library, 1937 [orig. 1776]), p. 28. Galbraith, American Capitalism, p. 135. Galbraith, American Capitalism, p. 136.

https://doi.org/10.1017/9781009323178.004 Published online by Cambridge University Press

50

Influences over Preference Formation

come to depend on output. In technical terms it can no longer be assumed that welfare is greater and an all-around higher level of production than at a lower one.”24 People would never want those goods if they were not produced. Friedrich Hayek offers a thoughtful critique of the dependence effect, noting that many of the greatest achievements of a civilized society fall into exactly that category.25 Great works of art, music, architecture, and other products of culture would never have been demanded had they not been created. Nobody would want to view the Mona Lisa had Leonardo da Vinci not painted her. Nobody would want to listen to Beethoven’s Ninth Symphony had he not written it. Nobody would value the architecture of Frank Lloyd Wright had he not designed those structures. Galbraith’s lament that producers and advertisers create wants that would not exist had they not created them must be tempered a bit, but it is worthwhile to examine how this plays into the creation of political preferences. Advertisers need something to sell. One might hope that what they are trying to sell is a better product. As Hayek notes in his critique of the dependence effect, selling a product that provides more instrumental utility to purchasers than alternative products will be more profitable than trying to sell something that ultimately produces less utility. Firms that try to use advertising to sell inferior products ultimately are doomed to failure as consumers realize there is better value elsewhere. Sometimes, however, there is little difference between the products of one producer and another, and the advertiser’s job is to create the perception of a difference. Many products, such as gasoline or milk, are very homogeneous, and because the advertiser’s task is to push consumers to buy the advertised product, the advertiser must make one brand of gasoline, or one brand of milk, look more desirable. The only way to do this if products really are homogeneous is to create emotional appeal for the product being advertised. Sellers give much thought to making their packaging look more attractive, and to crafting messages associated with their products that make people feel good about buying one product relative to another. In politics, this can be done in a few ways. One is to make candidates appear more likeable so voters are more comfortable voting for them. 24 25

Galbraith, American Capitalism, p. 136. Friedrich A. Hayek, “The Non-Sequitur of the ‘Dependence Effect,’” Southern ­Economic Journal 27, no. 4 (April 1961), pp. 346–348.

https://doi.org/10.1017/9781009323178.004 Published online by Cambridge University Press

3.5  Hope and Change

51

Voters want nice people to win. Candidates can also be sold as more competent. Future challenges are uncertain, and people want political leaders who are able to meet unforeseen challenges. The message is that this candidate is a better person for the job than the alternatives. The second way is to advertise differences in actual policy preferences among candidates, promoting one candidate’s or party’s policies as superior to the other’s. This strategy carries with it some risk in that some voters may not like the policies promoted by that candidate or party. So, it tends to be more effective to take the first strategy and say this candidate is better than that candidate, rather than that these policies are better than the alternatives.

3.5  Hope and Change The title of this section is the slogan that Barack Obama used when campaigning for the presidency in 2008. What is the actual policy meaning of this slogan? Brief reflection reveals that there is no policy meaning. This is a feel-good slogan. President Obama is not unique in this regard. President Reagan won the presidency in 1980 running with the slogan “It is morning in America.” President Trump’s winning slogan in 2016 was “Make America great again.” Slogans like these give no indication of the policies that the candidates using them would work to enact. Essentially, what these slogans suggest is that things are not as good as they could be, and if voters elect this candidate, things will be better. This is the message voters are hoping to hear. Just as important, it is a message with which nobody will disagree. Everyone can think of things that could be better, and candidates are saying that the way to make them better is to elect them to do it. An alternative would be for candidates to suggest specific policies that would improve things, but the problem with this strategy is that even though everyone agrees that things could be better, not everyone will agree on the specific policies that could make them better. So, voters will more readily agree with the candidate who says changes could be made to improve things than with the candidate who proposes specific policies to address perceived problems. The way to sell the candidate is to promote the vague message that the candidate will make things better without detailing how. Anyone who pays attention to actual political campaigns must see how devoid of specific policy suggestions (most) campaigns are.

https://doi.org/10.1017/9781009323178.004 Published online by Cambridge University Press

52

Influences over Preference Formation

A good example of this type of campaigning comes from Hubert Humphrey’s presidential campaign in 1972.26 I attended a campaign speech of his in Gainesville, Florida, when I was a college student and economics major, and was most impressed by his ability to win over a tough crowd. When he took the stage to speak in front of a packed auditorium, he was being booed so loudly that I thought he might never get the chance to speak. But he quieted the crowd, telling them, “Hear what I have to say before you make up your mind, and if you don’t like my message, you should boo me.” As he began speaking, the crowd did quiet down, and Humphry was a skilled and charismatic speaker who eventually won the crowd over. As an economics major, I recall one particular component of his speech especially well. Addressing economic issues, Humphrey said, “The economy is like an eight cylinder automobile that is only running on six cylinders. We have to get those other two cylinders running again!” The crowd roared in approval. His entire talk was filled with statements along these lines. What, exactly, does Humphrey’s insight have to say about economic policy? It says the economy is not doing as well as it could, and if he is elected, he will make it better. This is just the kind of political message that sells. It appeals to the emotions rather than to the intellect, and conveys a message that would command widespread agreement. Effective political advertising is designed to make voters feel good about candidates and parties, not to sell specific policy initiatives. As with advertising in general, it is designed to appeal to people’s emotions rather than to persuade them with facts. Can people be influenced to change their political preferences in response to advertising? If advertising is effective in selling goods and services, it should be at least as effective in selling candidates and parties, because what is being sold is not the usefulness of a product but its attractiveness and desirability.

3.6  Negative Advertising People claim to dislike negative political advertising, yet it persists. Negative ads, rather than touting the virtues of a candidate, present the candidate’s opponent in a negative light. They are effective because of the influence they have on voter turnout. 26

Humphrey was running to get the Democratic Party nomination, which he ultimately lost to George McGovern.

https://doi.org/10.1017/9781009323178.004 Published online by Cambridge University Press

3.6  Negative Advertising

53

As discussed further in Chapter 5, people are resistant to changing their political preferences. Certainly, some voters may be undecided and some may revise their preferences when presented with evidence that their views are more aligned with a different party or candidate. Most people, however, make up their minds relatively early in a political campaign and stick to their original positions. One variable that often swings elections is turnout. The purpose of negative ads is to discourage supporters of one’s opponents from showing up to vote. Becoming a voter’s more preferred party or candidate is only half the battle in winning that voter’s vote. The other half is getting the voter to actually show up and vote, while discouraging the supporters of opponents from voting. Positive ads can serve the purpose of enticing undecided voters to vote for the advertised candidate, but they likely have a laager impact by encouraging one’s supporters to vote. The flip side of that is the negative advertising that discourages supporters of one’s opponents from voting. One can see the effect of turnout in US election results in presidential election years versus off-year elections. As an empirical regularity, the president’s party almost always gains seats in the House of Representatives during presidential election years, and loses seats in the off-year elections. During presidential election years, candidates for the House of Representatives can ride the presidential candidate’s coattails, giving an advantage to the members of the elected president’s party. During off years between presidential elections, the president is not running, so the president’s supporters are not as inclined to show up to vote, and the president’s party loses seats in the House of Representatives. From 1960 to 1986 this was the case in every election. During presidential election years the party that won the presidency always gained seats, and in the elections between presidential elections the president’s party lost seats. There have been a few exceptions since then. In 1988 when the first George Bush was elected, Bush’s Republican party lost seats, and in 1992 when President Clinton was elected, his Democratic party lost seats. Then in 1998, which was not a presidential election year, President Clintons’ Democratic party gained seats. The last two exceptions were the two times that President Trump ran. In 2016 Trump won but the Republicans lost House seats, and in 2020 when President Biden unseated Trump, the Democratic party lost seats. Elections are held every two years, so there have been thirty-five elections spanning from 1960 to 2020, and in only five cases did that general pattern not hold. One explanation for the predominant pattern is that the

https://doi.org/10.1017/9781009323178.004 Published online by Cambridge University Press

54

Influences over Preference Formation

party that wins the presidency does so by winning the turnout battle by getting more of its supporters to turn out and by discouraging supporters of the opposing party from turning out. During off-years, without a presidential candidate to pull in supporters, turnout for the president’s party suffers. The message underlying negative political advertising is that gaining political support is a two-step process. First, candidates and parties need to convince people to adopt their political preferences, and second, they need to convince those whose preferences align with theirs to turn out and vote for them. People claim to despise negative political ads, but they are effective because they discourage voters from turning out and voting for the candidates the ads place in an unfavorable light. The utility from voting comes only from the satisfaction of expressing a preference at the polls, and potential voters who see unfavorable characteristics in their preferred candidates are less likely to show up to vote for them.

3.7 Conclusion When thinking about the relationship between political preferences and public policy, there is a tendency to envision voters and citizens as having preferences for policies, and politicians designing their campaigns and platforms to satisfy those preferences so they can be elected to power. The chapters that follow will suggest that the direction of causation goes the other way. The political preferences of citizens mostly come from the political elite. Citizens and voters adopt the political preferences of the politicians and parties they support, rather than parties and candidates adopting the preferences of citizens and voters. Members of the elite tell citizens what their policy preferences should be, and citizens follow their leaders. If this is the case, one question that then arises is the degree to which the preferences citizens hold are subject to outside influences. This chapter has focused primarily on preferences people have for goods and services offered on the market, where the preferences they express have instrumental consequences. Money devoted to buying one good or service cannot be used to buy something else, so consumers have an incentive to be careful about how they allocate their spending. Even here, when there is an instrumental cost to the choices people make, a substantial literature shows that their preferences are subject to outside manipulation. Businesses expend large sums on advertising because they believe that advertising can affect consumer preferences.

https://doi.org/10.1017/9781009323178.004 Published online by Cambridge University Press

3.7 Conclusion

55

If this is the case for choices that have an instrumental cost, it likely is even more the case when the choices people make are purely expressive, such as when they vote. The chapter noted the links between advertising for goods and services and political advertising, but the larger point of the chapter is to examine the assumption frequently made by economists that preferences should be taken as given and unchanging in economic analysis. To understand the implications of choices people make – especially choices about public policy – requires an understanding of how their political preferences are formed and how they can be manipulated through outside influences. Galbraith argues that most people understand that this is the case: “The businessman and the lay reader will be puzzled over the emphasis which I give to a seemingly obvious point. The point is indeed obvious. But it is one which, to a singular degree, economists have resisted. They have sensed, as the layman does not, the damage to established ideas which lurks in these relationships. As a result, incredibly, they have closed their eyes (and ears) to the most obtrusive of all economic phenomena, namely modern want creation.”27 Most people likely have a better intuitive understanding of this with regard to market choices than they do for political choices. The ideology of democracy – the idea that citizens control their democratic governments – is widely accepted by the general public. When one looks at the way political preferences are formed, it becomes apparent that people commonly underestimate the degree to which governments control their citizens – both the actions of their citizens and their thoughts.

27

Galbraith, American Capitalism, p. 136.

https://doi.org/10.1017/9781009323178.004 Published online by Cambridge University Press

4 Preference Aggregation through Voting

One of the challenges facing democratic governments is how electoral ­institutions can be designed to aggregate individual voter preferences to make collective decisions. The method of aggregation can affect the collective outcome. For example, electing a president by counting the popular vote might give a different outcome from the electoral college system used in the United States. Parliamentary systems and proportional voting might produce different collective outcomes than presidential systems and plurality voting. Open primaries can produce different results from closed primaries. One of the advances in the social sciences in the last half of the twentieth century has been the development of models that describe how different voting mechanisms for aggregating voter preferences produce collective decisions. Voting models depict the aggregation of the individual preferences of voters to produce a collective decision as a result of the modeled aggregation method. Kenneth Arrow demonstrated that there is no method of ­preference aggregation that always chooses a socially optimal ­outcome.1 Different methods of aggregating votes will produce different c­ ollective choices, and no method can be demonstrated to always produce a result that more closely reflects the underlying preferences of voters than other methods. The challenge is to find a method of aggregating votes that best satisfies the criteria that members of the group think are most important. 1

Kenneth J. Arrow, Social Choice and Individual Values (New Haven, CT: Yale University Press, 1953). T. Nicolaus Tideman and Gordon Tullock, “A New and Superior Process for Making Social Choices,” Journal of Political Economy 84 (December 1976), pp. 1145–1169, claim to present a mechanism that overcomes the problems Arrow demonstrates, but their mechanism brings with it a new set of problems.

56

https://doi.org/10.1017/9781009323178.005 Published online by Cambridge University Press

4.1  Voting Models and Public Policy

57

In an important sense, this is an impossible task. Conceptually, the idea is to aggregate preferences of the members of a group to produce an outcome that is the preference of the group, but groups do not have preferences. Only individuals have preferences, as Chapter 1 discussed in some detail. Nevertheless, groups that undertake collective action do have mechanisms that aggregate individual preferences to make collective decisions. Ideally, those collective decisions further the interests of those in the group. In voluntary organizations like clubs, evidence that group members are better off as a result of the group’s collective actions is that individuals choose to be members. One would hope that organizations which force members to participate, as governments do, also benefit those within the organization. A justification for collective organizations like clubs and governments is that they enable individuals to accomplish collectively ends that they would be unable to attain on their own.2 While one could never conclude that the outcome of a collective decision-making process is the group’s preference, because groups do not have preferences, academics who study v­oting mechanisms look for aggregation procedures that produce collective ­outcomes that reflect the underlying preferences of the individual voters. Two different problems exist when trying to use voting to produce a collective decision that reflects the underlying preferences of the voters. One problem, widely recognized by those who study voting mechanisms, is that the method by which individual votes are aggregated may produce a collective choice that does not accurately represent the underlying individual preferences. Another problem, rarely considered, is the issue that preferences expressed by voters at the ballot box may not reflect their preferences for actual outcomes. The concept of expressive voting, explained in Chapter 2, is well known, but it has not been integrated into the models of voting developed by social scientists. This chapter looks at both of these problems, and emphasizes the significance of the second, underrecognized, problem of expressive voting.

4.1  Voting Models and Public Policy The study of voting mechanisms fell within the domain of political science until the mid-twentieth century, when economists and other social scientists became increasingly interested in them. Prior to the mid-twentieth 2

This is an underlying theme of James M. Buchanan and Gordon Tullock, The Calculus of Consent (Ann Arbor: University of Michigan Press, 1962).

https://doi.org/10.1017/9781009323178.005 Published online by Cambridge University Press

58

Preference Aggregation through Voting

century, economists viewed the analysis of political decision-making as outside the realm of economics. The role of economists, according to the mainstream view of economics at the time, was to derive optimal public policies.3 The role of political science was to analyze how governments could implement those policies. Optimal policies are those that best satisfy the preferences of those who are governed by them, and the challenge democratic institutions face is implementing preference aggregation mechanisms that produce collective choices that conform to the individual preferences of those whose preferences are being aggregated. As cross-disciplinary interest in studying voting mechanisms grew, the normative goals of political scientists, economists, and social scientists more generally were roughly the same. They were interested in discovering mechanisms that would reveal the public policies that would further the public interest – the interest of the voters. The first step in the search for policies that satisfy the preferences of citizens is to look for methods of vote aggregation that would best reveal what policies voters as a whole favored. An extreme version of this idea dates back to Rousseau, who was quoted in Chapter 1 as saying that the policies that were enacted through a democratically elected body were in the public interest, or to use Rousseau’s terminology, they expressed the general will. Rousseau made the strong argument that those who disagreed that those policies were an expression of the general will were wrong. But as Chapter 1 also noted, voter preferences might be such that voting mechanisms are unable to reveal which among several alternatives is preferred by the group. The challenge is to try to design voting mechanisms that do a plausibly good job of revealing and aggregating voter preferences under most circumstances, despite Nobel laureate Kenneth Arrow’s demonstration that a completely satisfactory mechanism cannot be designed.4 Until the later part of the twentieth century, economists tended to ignore the collective decision-making problem entirely. The increasing mathematization of economics allowed economists to identify with mathematical precision the allocation of resources that would maximize social welfare.5 This theoretical welfare maximum was used as a benchmark 3

4 5

See, for examples, Francis M. Bator, “The Simple Analytics of Welfare Maximization,” American Economic Review 47, no. 1 (March 1957), pp. 22–59, and J. de V. Graaf, Theoretical Welfare Economics (London: Cambridge University Press, 1957). Arrow, Social Choice and Individual Values. The methodology is summarized and explained by Bator, “The Simple Analytics of ­Welfare Maximization,” and Graaf, Theoretical Welfare Economics, referenced earlier.

https://doi.org/10.1017/9781009323178.005 Published online by Cambridge University Press

4.1  Voting Models and Public Policy

59

for judging whether government policy might be called for to improve resource allocation. If for any reason markets fell short of the theoretical benchmark of welfare maximization, that reason was labeled a market failure, and having identified the theoretical welfare maximum, government policy could then, in theory anyway, correct that market failure.6 Market failures can occur because of externalities, public goods, monopolies, macroeconomic instability, and imperfect information, among other reasons. In short, if markets did not allocate resources perfectly, there was a potential role for government to correct any market deficiencies. Economists demonstrated the conditions under which resources would be allocated efficiently, and it was the task of those in government to design the policies that would satisfy the economist’s optimality conditions. From an academic standpoint, the design of those policies fell under the heading of political science. This line of reasoning uses rigorous economic analysis to identify problems the market mechanism can face in allocating resources, but then substitutes wishful thinking for rigorous analysis when looking at what government might do about it. Welfare economics provides mathematical conditions for optimal resource allocation, but does not identify any mechanism that would allow government to design a policy to get to that optimum. James Buchanan was among the leaders of a movement that advocated the use of the same tools economists use to analyze the market allocation of resources to also analyze how governments allocate resources.7 While markets can fail to allocate resources perfectly efficiently, so can governments, Buchanan persuasively argued. Without an analysis of the way that government actually addresses market failures, there is no guarantee that should government take action, government failure would not be worse than market failure. Government failure – the failure of government policies to maximize welfare – can occur for two reasons: information and incentives. Those in government who want to implement the optimal policy may not have sufficient information to do so. For example, public goods can produce a market failure because people in the private sector may have an incentive to underprovide them. Should this be the case, however, nobody in 6

7

Francis M. Bator, “The Anatomy of Market Failure,” Quarterly Journal of Economics 72, no. 3 (August 1958), pp. 351–379. An  excellent critique of this line of reasoning is found in Meir Kohn, “Value and Exchange,” Cato Journal 24, no. 3 (Fall 2004), pp. 303–339. James M. Buchanan, “Public Finance and Public Choice,” National Tax Journal 28, no. 4 (December 1975), pp. 383–394.

https://doi.org/10.1017/9781009323178.005 Published online by Cambridge University Press

60

Preference Aggregation through Voting

government can acquire the information to identify the amount of a public good that is in theory optimal. If the existence of public goods results in a market failure, government provision will also lead to a government failure because government does not have access to the information required to identify the optimum. Voting models are intended to partially address the issue of government failure by analyzing how citizen preferences can be aggregated to reveal a collective preference. While no voting procedure is ideal in every circumstance, the hope is that voting can provide some indication of collective demands. The information about what would be the optimal output of a public good is not available to policymakers, but through voting, some information about the preferences of group members can be revealed as individual preferences are aggregated into a collective decision. Even if the information is available, those in government often do not have the incentive to allocate resources optimally. It is common knowledge that elected officials often propose legislation that is designed to buy them political support, even when what they are proposing is inefficient. Similarly, government bureaucrats look out for their own interests, just as businessmen do, and may act to increase their agency budgets or give themselves more job security when those goals conflict with the public interest.8 To understand how government actually allocates resources requires an understanding of the incentives they face, which will be discussed further beginning in Chapter 7. This chapter deals with the problem of information. If government policies are intended to carry out the public interest, how do those in government identify the public interest? One answer is that the public interest is revealed through democratic institutions. Since the mid-1950s the social sciences have advanced the understanding of the relationship between voter preferences and collective choices through the development of voting models that reveal the mechanics of aggregating individual votes to make collective choices. These models focus on the first problem discussed earlier – the problem of aggregating individual preferences to produce a collective choice that reflects the underlying preferences. They do not consider the possibility that the preferences voters express at the ballot box may not reflect their preferences for public policy outcomes. 8

The incentives of bureaucrats are discussed in more detail by William A. Niskanen, Bureaucracy and Representative Government (Chicago: Aldine-Atherton, 1971).

https://doi.org/10.1017/9781009323178.005 Published online by Cambridge University Press

4.3  The Logic of Collective Choice Models

61

These voting models, originally falling within the domain of political science, do make a contribution to the economic analysis of public policy. Rather than relying on wishful thinking to assume that somehow government decision-makers will implement optimal policies, they examine the relationship between individual preferences and collective choices. There is no guarantee that even if the information on the preferences of the public is revealed, the government will enact policies to further those preferences. This is yet another problem that has been widely studied. But the first step is to look at the information that is revealed to policymakers through voting.

4.2  Economic Models of Voting Economic models of voting begin with the assumption that individuals have a given set of preferences and then examine the ways that different methods of voting aggregate the preferences of the voters to produce a collective decision. Analogous to supply and demand models that depict the way that preferences are aggregated to produce market outcomes, voting models depict the aggregation of political preferences to produce collective decisions. That aggregation represents the demand of the group. An election is held in which A runs against B, and A wins the election. The collective choice of the group is for A to hold that political office. As noted earlier, one would be hard-pressed to argue that the outcome of any democratic decision-making process represents the preference of the group, because groups do not have preferences. Furthermore, the way that votes are aggregated can affect the collective choice outcome. Different electoral procedures can produce different election results. An election outcome is simply the collective choice that was produced by aggregating individual preferences under a specific set of democratic institutions. What voting models reveal is the relationship between the preferences expressed by individual voters and the collective choice that is produced by aggregating them in a particular way.

4.3  The Logic of Collective Choice Models One way to clearly see the main issue this chapter raises is to describe it in general mathematical notation. Models of voting mechanisms depict the process by which elections aggregate the preferences of individual voters to produce a collective decision, given the individual preferences of voters

https://doi.org/10.1017/9781009323178.005 Published online by Cambridge University Press

62

Preference Aggregation through Voting

and the algorithm that is used to aggregate them. The mechanism itself is an algorithm that describes how, if voters express certain preferences, those preferences are taken into account to produce an election result. In general form, given individual preferences for n voters P1, P2, P3, … Pn, individual preferences are aggregated by some function f to produce a collective choice C, so

C = f(P1, P2, P3, … Pn). (4.1)

The model says that if voters express preferences P1 … Pn when they vote, and those preferences are aggregated through a process represented by f, the collective choice of the group is C. As a matter of basic logic, there is nothing wrong with depicting election methods this way. Voters vote, the votes are aggregated through the algorithm represented by f, and a collective choice is made. The problem comes in interpreting the relationship between voter preferences and the collective choice. The preferences are (almost) always interpreted as the instrumental preferences of voters – the outcomes that voters would most prefer. Representing this idea in mathematical notation, the instrumental preferences of voter i can be represented by Di and the collective choice from aggregating these preferences through voting can be represented by CI. Assuming instrumental voting, voter i expresses preferences Di for the outcome the voter would most prefer. Equation 4.1, depicting voters as voting instrumentally, can be restated as

CI = f(D1, D2, D3, … Dn). (4.2)

The collective choice with instrumental voting is CI and equation 4.2 shows the relationship between voters’ instrumental preferences and the election outcome. From the standpoint of welfare economics, the challenge is to find a vote aggregation mechanism f that produces a c­ ollective outcome CI that comes as close as possible to maximizing social welfare.

4.4  Differing Methods of Aggregation Much of the analysis of voting that takes place within voting models is devoted to showing the differences and similarities in results when different aggregation methods are used. In the previous section, the aggregation process was summarized as f, but there are many different aggregation procedures that can be used, and not all of them produce the same outcomes. Consider again the example of using the electoral college system of electing a president as is done in the United States

https://doi.org/10.1017/9781009323178.005 Published online by Cambridge University Press

4.4  Differing Methods of Aggregation

63

versus awarding the presidency to the candidate with the highest number of individual votes. Not all vote aggregation mechanisms produce the same outcome, so given the preferences Pi in equation 4.1, the primary issue of interest in voting models is what outcome C is produced by aggregating through function f. For example, the median voter model is a well-known model for understanding the relationship between voter preferences and electoral outcomes. The model depicts voters as having preferences that stretch along a single-dimensioned continuum from political left to political right, and candidates compete for votes by trying to choose a platform along that continuum to win majority support from the voters and therefore win the election. In this framework, the candidate who wins the vote of the median voter will win that vote and all votes to one side of the median, giving that candidate a majority of the votes and an electoral victory. Thus, when the aggregation procedure f is to award the election to the candidate with the most votes when voter preferences line up on a singledimensioned continuum, C = Pm, where Pm is the outcome preferred by the median voter. Anthony Downs uses this model to explain why winner-take-all elections tend to have two candidates, and why candidate platforms converge on the median voter’s preference.9 The conclusion is that when votes are aggregated through a representative democracy, the preferences of the median voter are the preferences expressed by the group. Different voting mechanisms can produce different outcomes, but all voting models describe a mechanical process in which voter preferences are given, and then an algorithm is used to aggregate those preferences. These voting models examine potentially interesting issues in vote aggregation, but when looking at the relationship between political preferences and public policy, those issues are minor. Two more significant issues have been emphasized in the previous two chapters. First, as Chapter 2 demonstrated, the preferences that voters express through voting are expressive rather than instrumental, and second, as Chapter 3 demonstrated, the preferences voters express are subject to manipulation. Both of those issues enlarge the gap between preferences Pi and collective choices C. 9

Anthony Downs, An Economic Theory of Democracy (New York: Harper & Row, 1957). James M. Enelow and Melvin J. Hinich, The Spatial Theory of Voting: An Introduction (Cambridge: Cambridge University Press, 1984), provide an in-depth discussion of this type of voting model.

https://doi.org/10.1017/9781009323178.005 Published online by Cambridge University Press

64

Preference Aggregation through Voting

4.5  Expressive versus Instrumental Preferences Voting models that attempt to relate collective choices back to the preferences of individual voters implicitly assume that voters express instrumental preferences when they vote. They will vote for the alternative they would choose if the choice were theirs alone, or they may vote strategically (for example, not voting for a hopeless candidate so they can vote for their favorite among likely winners) under the thought that their vote can affect the election outcome. But as Chapter 2 noted, voters must realize that the likelihood that their individual vote will affect an election outcome is vanishingly small, so voters have no incentive to vote instrumentally. They cast their votes based on their expressive preferences, not their instrumental preferences.10 If instrumental voters weigh the cost of going to the trouble to vote against the possible benefit they could get if they cast the decisive vote, their utility-maximizing option is to abstain. Brennan and Lomasky argue that expressive voting undermines the logic behind formal voting models, saying these models are “fundamentally misconceived.”11 While there are good reasons to rethink economic models of voting in light of the incentives facing voters, the logic of those models remains intact. Voters express preferences that are then aggregated to produce collective decisions, as represented by equation 4.1, and the models merely explain, given voter preferences, how those preferences are aggregated. The issue raised by Brennan and Lomasky is how voter preferences are formed, not how they are aggregated, so their analysis applies to a completely different issue (how voters determine their expressive preferences) than the subject of formal voting models (which analyze, given voter preferences, how preferences that voters express are aggregated). As Brennan and Lomasky point out, there is good reason to think that the way people form political preferences is different from the way they form their preferences in markets, because they face different incentives in politics when compared to markets. The link between voter preferences 10

11

Joseph MacMurray, “Ideology as Opinion: A Spatial Model of Common-Value ­Elections,” American Economic Journal: Microeconomics 9, no. 4 (November 2017), pp. 108–140, develops a model in which if a voter’s “vote does not turn out to be pivotal, then his vote choice does not influence his utility.” This assumes instrumental voting, and as noted, the utility-maximizing instrumental voter will abstain. The voter’s utility from voting comes from expressing a preference, not from potentially affecting the election outcome. Geoffrey Brennan and Loren Lomasky, Democracy and Decision: The Pure Theory of Electoral Preference (Cambridge: Cambridge University Press, 1993), p. 77.

https://doi.org/10.1017/9781009323178.005 Published online by Cambridge University Press

4.6  Aggregating Expressive Preferences

65

and public policy outcomes hinges crucially on the way that political preferences are formed. As Chapter 2 explained, voters vote based on their expressive preferences, which differ from their instrumental preferences. Voters express preferences at the ballot box that differ from what they would choose if the choice were theirs alone.

4.6  Aggregating Expressive Preferences The expressive preferences of voters, which determine how they vote, are likely to be different from those they would act on if the choice were theirs alone, as Chapter 2 explained. Voters are not necessarily expressing a preference for their most preferred outcomes, because they are not choosing outcomes. Rather than voting preferences Di, voters are voting their expressive preference, which can be represented as Ei. If voters vote instrumentally, then in equation 4.1, Pi = Di, whereas if they vote expressively, Pi = Ei. The problem with interpreting collective choices as being determined by citizen preferences is that Di ≠ Ei. If voters vote expressively rather than instrumentally, the outcome of an election is determined by voters’ expressive preferences Ei, so

CI ≠ fE(E1, E2, E3, … En). (4.3)

and the result of the election is CE, the aggregation of expressive ­preferences. But because Di ≠ Ei,

CI ≠ CE. (4.4)

The election result cannot be interpreted as showing the relationship between the outcomes voters prefer and the actual collective choice. The problem with formal voting models is not the logic by which they depict the aggregation of votes, but rather with their implied assumption that they depict the outcome as the aggregation of instrumental preferences, when actual votes are cast expressively. The outcome is analyzed as if it shows the relationship between voters’ preferences for outcomes and the actual collective choice, but

CI ≠ fE(E1, E2, E3, … En). (4.5)

For example, the median voter model depicts the aggregation of preferences by majority rule voting as the preference of the median voter. In that model, the election outcome is what the median voter most prefers. But the collective choice is determined by the expressive preferences of the voters, not the voters’ instrumental preferences. The model is interpreted

https://doi.org/10.1017/9781009323178.005 Published online by Cambridge University Press

66

Preference Aggregation through Voting

as if equation 4.5 is an equality, but it is not. The actual collective choice is CE, not CI, as voting motels are usually interpreted. Because ­expressive preferences are different from instrumental preferences, even if the model accurately depicts how preferences are aggregated, the outcome is not the one the median voter most prefers.12

4.7  Social Choice and Social Welfare Social choice and public choice are closely related areas of inquiry in the social sciences, yet they are fundamentally at odds with each other. Public choice uses economic methods to analyze the political decisionmaking process, to explain how government, and collective action more generally, operates. Social choice examines methods of aggregating individual preferences to identify aggregation procedures that can come closest to maximizing social welfare by producing collective decisions that reflect the individual preferences of those who participate in the collective decision-making process. One well-recognized problem with the line of reasoning in social choice theory is that without the ability to make interpersonal utility comparisons, one can never weigh the utilities of some against the utilities of others to get a measure of social welfare. This chapter points out another issue. The analysis of social choice through voting is undermined if voters vote expressively rather than instrumentally. Nobel laureate and leading social choice theorist Kenneth Arrow says, “In the context of social choice, each individual may be assumed to have a preference ordering over all possible social states.”13 Arrow assumes that this preference ordering is expressed when voting, which is not the case when people vote expressively. If people do not reveal their instrumental preferences when they vote, there is no way to relate aggregate outcomes of elections back to the underlying preferences of voters for actual outcomes. One cannot identify the degree to which social choices reflect the instrumental preferences of voters, because the social choice mechanisms being analyzed do not aggregate instrumental preferences; they aggregate expressive preferences. Kenneth Arrow depicts the challenge in making social choices as finding a method of aggregating preferences to arrive at a collective choice 12 13

It may be worth a mention that the individual with the median instrumental preference is not necessarily the individual with the median expressive preference. Kenneth J. Arrow, “General Economic Equilibrium: Purpose, Analytic Techniques, Collective Choice,” American Economic Review 64, no. 3 (June 1974), p. 269.

https://doi.org/10.1017/9781009323178.005 Published online by Cambridge University Press

4.8  The Link between Preferences and Outcomes

67

that maximizes welfare,14 but social choice theory attempts to do this based on the assumption that the preferences voters reveal at the ballot box are instrumental preferences, when in fact they are expressive preferences. The title of Arrow’s book, Social Choice and Individual Values, indicates that he is analyzing the relationship between collective choices, represented as C here, and the individual values of voters, but the implied assumption is that their individual values are instrumental values whereas when people vote, they do so based on their expressive preferences, not instrumental preferences.

4.8  The Link between Preferences and Outcomes Voting models are not wrong in the way they describe preference aggregation, but it would be an error to believe that the models describe a link between voters’ instrumental preferences – preferences for actual outcomes – and collectively chosen outcomes. Voting models provide a link between expressive preferences and collective choices, not instrumental preferences and collective choices. Any conclusions regarding collective choices and instrumental preferences would require an understanding of how instrumental preferences are linked to expressive preferences, assuming that there is a functional link g between them. Assuming that instrumental preferences and expressive preferences are somehow related,

Ei = gi (Di), (4.6)

and the challenge is to discover that functional relationship, g. Equation  4.6 indicates that people’s instrumental preferences in some way determine people’s expressive preferences, but that is not necessarily the case. Before one can accept the relationship expressed in e­ quation 4.6, one must understand how expressive preferences are formed. The following chapter s­ uggests that any link between Ei and Di is likely to be very indirect. Voting models describe the way that the expressive preferences of voters are related to collective choices. They do not show any relationship between voter preferences for actual outcomes and collective choices. Voting models are interpreted with the implied assumption that C = CI when in fact C = CE. Before any conclusions can be drawn about a link 14

Arrow, Social Choice and Individual Values.

https://doi.org/10.1017/9781009323178.005 Published online by Cambridge University Press

68

Preference Aggregation through Voting

between instrumental preferences and election outcomes, some link between instrumental preferences and expressive preferences must be established. The aggregation procedures are modeled with mathematical precision, but the interpretations given to the results are precisely wrong.

4.9  Elites and Masses Voting models depict the aggregation of the preferences of voters whose only impact on public policy is through their effect on election outcomes. All voters are treated equally in those models,15 and voters have no impact beyond the impact of the votes they cast. For reasons already discussed, their political preferences are expressive. This applies to most citizens and voters. However, an elite few individuals make public policy and can affect outcomes, so those individuals – members of the political elite – will have instrumental preferences. Some voting models have depicted certain individuals whose preferences do affect election outcomes. These individuals, to use the terminology in the models, are agenda setters. A common example in many organizations is a nominating committee that chooses candidates that will be voted on for election. The nominating committee proposes candidates, and the organization members vote on which of the nominated candidates will be elected. The nominating committee clearly has an instrumental effect on the outcome, because they choose the slate of candidates who can be elected. In some organizations – the American Economic Association and Southern Economic Association are two examples – only one candidate is nominated for president of the organization. While write-in campaigns could be mounted, the nominating committee’s candidate always wins, for obvious reasons. Even when multiple candidates are on the ballot, the nominating committee has the power to restrict the choices available to voters, and so to have an instrumental effect on the election outcome. Under some circumstances, someone who has control of the agenda can manipulate the electoral process so that the outcome of majority 15

In some cases, voters may be weighted so the votes of some count more than the votes of others. For example, when stockholders vote, each share gets one vote, so voters who own more shares get more votes. In homeowner associations, different classes of homes may have their votes weighted differently. In a condominium association, for example, a threebedroom condo may have more voting power than a two-bedroom condo. Still, each voter affects the election outcome only through the preferences expressed through voting.

https://doi.org/10.1017/9781009323178.005 Published online by Cambridge University Press

4.9  Elites and Masses

69

rule voting, rather than being the preference of the median voter, is the ­preference of the agenda controller.16 More generally, public policy is made by a few well-connected elites who are able to bargain with each other to produce public policy outcomes. Government ministers, members of legislatures and parliaments, and high-level bureaucrats (and maybe low-level bureaucrats!) are examples. These people make public policy, so their policy preferences are instrumental, and the way they form preferences will be different from the way voters form preferences. While government elections typically have more democratic foundations than those organizations that use small nominating committees to select candidates, voters in government elections also have limited choices, and the influences of elites, and the campaign contributions and public support of candidates, can influence and limit the choices available to voters. California voters, for example, initiated recall elections for Governors Gray Davis in 2003 (successful) and Gavin Newsom in 2021 (unsuccessful). What type of collective decision-making procedure allows someone to gain enough support to be the collective choice of the voters, and a few years later to be unpopular enough that voters express a preference for that individual to be removed from office before serving a full term? This surely raises a question about the degree to which the collective choices people make through voting represent their instrumental preferences. This chapter deals with models of voting procedures in which voters express preferences among clearly defined options. The political elite, who determine the options available to voters, are not incorporated into those models. The elite heavily influence the choices presented to voters, but voting models only depict the aggregation of preferences, given those preferences and given the options among which they are able to choose. Given the choice between A and B, will voter preferences be aggregated to choose A or B? Those models do not describe how the alternatives of A or B come to be placed before voters – something to be considered in later

16

One model of agenda control is presented by Thomas Romer and Howard Rosenthal, “Political Resource Allocation, Controlled Agendas, and the Status Quo,” Public Choice 33, no. 4 (1978), pp. 27–43. R. D McKelvey, “Intransitivities in Multidimensional Voting Models and Some Implications for Agenda Control,” Journal of Economic Theory 12, no. 3 (June 1976), pp. 472–482, offers a model in which the agenda controller can manipulate voting outcomes to get the outcome the agenda controller most prefers. Along the same lines, see Norman Schofield, “Instability of Simple Dynamic Games,” Review of Economic Studies 45, no. 3 (October 1978), pp. 575–594.

https://doi.org/10.1017/9781009323178.005 Published online by Cambridge University Press

70

Preference Aggregation through Voting

chapters. This chapter deals only with the way that social scientists depict the aggregation of preferences, given those preferences and the alternatives presented to voters. The main point is that there is at best a tenuous relationship between the preferences voters express when they vote and their preferences for actual political outcomes. They vote expressively rather than instrumentally. In contrast, a small political elite choose the alternatives available to voters and have a substantial amount of control over which outcomes they choose, and when holding political power, determine public policy. Their political decisions are instrumental. To understand the complete process, beginning with the options available to voters, requires an understanding of the political actions of both the elite and the masses. Voting models depict the masses, who choose among options given to them, without depicting the process by which those options are determined. Chapter 7 discusses this in more detail.

4.10 Conclusion There is some tendency, after an election, to think that the people have spoken and their preferences have been taken into account. Voting models developed by social scientists depict this process with mathematical precision, and scholars often conclude that if the math is correct, the case is proven. There always is the issue of relating mathematical symbols to activities in the real world, which is why it is important to see how voting models have the potential to mislead scholars into believing things about the real world that are not true. This chapter demonstrates why voting models do not show any connection between voters’ instrumental preferences and election outcomes. The presumed link between the two – preferences and outcomes – goes deeper than just academic models of voting. Those models reinforce what is already widespread opinion: that election outcomes somehow reveal the general will or the public interest. Voting models show why this is not always the case, but even there, the models make too strong a link between preferences and collective choices. Voters do not necessarily express a preference for the election outcomes they would prefer to see when they vote. To understand what link, if any, exists requires an understanding of how voters form their expressive preferences. The formal analysis of voting has focused on the relationship between underlying voter preferences, represented above in equation 4.1 as Pi, and the collective choices that different methods of vote aggregation produce,

https://doi.org/10.1017/9781009323178.005 Published online by Cambridge University Press

4.10 Conclusion

71

represented by C. The assumption, almost always implied rather than explicitly stated, underlying those models is that the preferences people express are instrumental preferences, denoted above as Di, so election outcomes are CI, the aggregation of instrumental preferences. However, because voter preferences are expressive, Ei, rather than instrumental, there is no clear relationship between voting outcomes, CE, and instrumental preferences, Di. The use of this notation clarifies the issue. The method of preference aggregation matters little to establishing a connection between voters’ underlying preferences and the collective choices they make in elections, because when they vote, their preferences are expressive rather than instrumental. The purpose of this chapter is to question the link between the outcomes that voters actually prefer and the outcomes they vote for, but questions about voting (rather than voting models) go deeper than this. Some models, like the median voter model, conclude that there is a mechanism that directs election outcomes to conform to the preferences of the voters (the median voter), but if those preferences are derived from the preferences of the elite, as later chapters argue, this further breaks the link between voters’ instrumental preferences and election outcomes. The choices voters are presented with in elections would then be those offered to them by the elite, and preferred by the elite. The academic literature on voting has discussed expressive voting in detail, and while some issues are unsettled, the idea that the preferences voters reveal when they vote are expressive rather than instrumental is not new. The academic literature has not done much to take the next step and examine how voters form their expressive preferences, and how they might differ from instrumental preferences. This issue is pursued in the next chapter.

https://doi.org/10.1017/9781009323178.005 Published online by Cambridge University Press

5 The Formation of Political Preferences

The utility people get from having and expressing political preferences comes solely from their having and expressing them, because one individual’s political preferences will have no effect on political outcomes. People may experience real consequences as a result of the political preferences they express. For example, if someone expresses support for a candidate the person’s peer group dislikes, the person could suffer the disapproval of those peers. Similarly, expressing a preference for a candidate the person’s peer group supports can enhance that person’s sense of belonging to the group. Factors like these can influence the political preferences people express, but these consequences arise only because the individual has expressed those preferences, not because the individual’s preferences have any effect on public policy. For those outside of the political elite who actually make public policy, an individual’s expressed political preferences will have no effect on actual political outcomes. When people make choices in the marketplace, they get what they choose. The choices people make at the ballot box have no effect on election results, so casting a vote for an alternative the voter would not choose if the choice were instrumental imposes no material cost on the voter. Voters are unlikely to reason “I would prefer A to B, but because my vote is only expressive, I will vote for B.” But they are very likely to reason that they feel an affinity for others with whom they associate who favor B, so they will vote for B to validate a connection they have with those others, while giving little consideration to whether they would be better off if A won the election, or even whether others would be better off. 72

https://doi.org/10.1017/9781009323178.006 Published online by Cambridge University Press

The Formation of Political Preferences 

73

Voting is expressive, so voters choose the options that make them feel best, rather than the options that would be best for them if those options were chosen. The utility-maximizing question for the individual voter is “Would I feel better voting for A or B?” not “Would I be better off if A won the election or if B won?” Political decisions, unlike market decisions, do not even give individuals a good opportunity to evaluate alternative outcomes. When people make market choices, they typically have the opportunity to evaluate the alternatives they did not choose. They might experience regret when they observe that someone else got a better meal at a restaurant than they did. They might experience regret when their neighbor’s car looks like it would have been a better value for them than the car they actually bought. The same is not true in politics because when one candidate or policy is chosen, voters never see what the actual results of choosing a different candidate or policy would have been. This leaves citizens and voters with little tangible evidence to suggest that they would have been better off had a different collective choice been made. Voters who voted for a winning candidate may see some problems and issues after the election, but still can think that the winning candidate is better than the one who lost, with no tangible evidence to suggest otherwise. Meanwhile, those who voted for the losing candidate can see those same problems and issues and believe that had the candidate they favored been elected, outcomes would have been better. This lack of a direct feedback mechanism to identify beneficial public policies allows citizens and voters to retain their political beliefs, even when they are inconsistent with the outcomes they would prefer if they could choose those outcomes themselves. Such preferences might be called irrational, although it is difficult to see the irrationality of holding an opinion that has no instrumental consequences.1 Voters will tend to minimize cognitive dissonance by believing that the choices they made when voting were the best options. Chapter 2 concluded that the preferences people reveal when voting are expressive, not instrumental, and that people may vote for an outcome even if they would prefer that a different outcome win the election. They may vote for outcomes they would not choose if the choice were theirs alone. Chapter 3 concluded that people’s preferences are subject to manipulation. There is good evidence that even their instrumental 1

Bryan Caplan, The Myth of the Rational Voter: Why Democracies Choose Bad Policies (Princeton: Princeton University Press, 2007).

https://doi.org/10.1017/9781009323178.006 Published online by Cambridge University Press

74

The Formation of Political Preferences

preferences can be manipulated through advertising, peer pressure, and other means. The instrumental choices they make affect the outcomes they get. So it is even more likely that people’s expressive preferences are subject to manipulation, because when voting, the choices they make have no effect on the outcomes that are collectively determined. Chapter 4 showed that because political preferences are expressive, there is at best a tenuous link between people’s instrumental preferences and the collective choices that are made through democratic political institutions. That leads to the question of how people form their political preferences. The political preferences people express will have a strong influence over the public policies produced by government. There may be many factors that influence people’s public policy preferences. Building on the preceding chapters, this chapter discusses those factors. To repeat the first equation from the previous chapter,

C = f(P1, P2, P3, … Pn). (5.1)

The collective choice C is the result not only of the preferences of individuals but also the way they are aggregated. To understand what is chosen collectively requires understanding both the aggregation procedure, represented by f, and how the preferences that are aggregated are formed. This chapter turns to the issue of preference formation. People may, indeed, form their political preferences on instrumental grounds. If asked, most people surely will explain how their political preferences are not only instrumentally beneficial to themselves, but also instrumentally beneficial to a larger group. Few people will claim they hold their political preferences because those are the preferences of all of their friends. However, people are more likely to believe that the political preferences of others are subject to influence from outside sources. Other people adopt the political preferences of friends. Other people form their political preferences based on information they get from biased news sources. Many factors affect people’s political preferences, and an examination of those factors demonstrates why expressive preferences differ from instrumental preferences and how expressive political preferences are formed. The idea that the preferences voters express at the ballot box can differ from their instrumental preferences is not new. But the academic literature on expressive preferences has done little to describe how people form their expressive preferences. This chapter reviews an extensive body of literature on political preference formation to identify the factors that lie behind the formation of the political preferences of citizens and voters.

https://doi.org/10.1017/9781009323178.006 Published online by Cambridge University Press

5.1  Preferences and Choices

75

5.1  Preferences and Choices Economists tend to describe individuals as utility maximizers. They have utility functions that constitute their preferences, and they refer to those utility functions to make choices that maximize their utility. In fact, the process works in the other direction, as James Buchanan explains. People make choices and the choices they make define their preferences. “Individuals do not act so as to maximize utilities, described in independently-existing functions. They confront genuine choices, and the sequence of decisions taken may be conceptualized, ex-post, (after the choices), in terms of ‘as if’ functions that are maximized. But those ‘as if’ functions are, themselves, generated in the choosing process, not separately from such process.”2 This is completely consistent with the way utility functions are derived from revealed preference experiments by economists. Paul Samuelson offers an example where, in theory, people can be offered choices along different budget constraints with different relative prices to reveal a utility function.3 Using the same methodology, John Kagel and coauthors were able to offer different choices to rats and pigeons by varying their budget constraints to derive animal demand curves.4 These experiments did not start with utility functions to predict choice, but rather observed choices to conjecture an underlying utility function. Preferences were revealed as a consequence of the choices people (and animals!) made and did not exist prior to the choices. Considering the process by which individual choices are made, Buchanan continues, “The potential participants do not know until they enter the process what their own choices will be.” Individuals do not have preferences and make choices based on them. Rather, individuals make choices, and their preferences are defined by the choices they make. People often do not know what they would choose until actually faced with a choice, in markets and in politics5 From a methodological 2 3 4

5

James M. Buchanan, The Logical Foundations of Constitutional Liberty (Indianapolis, IN: Liberty Fund, 1999), p. 245, italics in original. Paul A. Samuelson, “A Note on the Pure Theory of Consumer’s Behavior,” Economica ns. 5, no. 17 (February 1938), pp. 61–71. John H. Kagel, Raymond C. Battalio, Howard Rachlin, Leonard Green, Robert L. Basmann, and W. R. Klem, “Experimental Studies of Consumer Demand Behavior Using Laboratory Animals,” Economic Inquiry 13, no. 1 (March 1975), pp. 22–38. I would not pretend to defend this idea with an anecdote, but many years ago I developed a fondness for a certain brand of mixed nuts. I would eat some daily and when I ran low would stop by the grocery store to buy more. Doing so was not a budget breaker to be

https://doi.org/10.1017/9781009323178.006 Published online by Cambridge University Press

76

The Formation of Political Preferences

standpoint, the challenge is not to identify the preferences people have, but rather to identify why they make the choices they do. People do not have utility functions, as economic models, including voting models, assume. They are faced with alternatives and they make the choices they believe will give them the most utility. When examining the link between political preferences and public policy, the question is what entices them to choose one option over another. Those choices then define their preferences.

5.2  Instrumental Preferences As the previous chapter observed, even though the concepts of expressive preferences and expressive voting are well established, models of voting tend to assume, sometimes implicitly, that the preferences voters express at the ballot box are instrumental. Indeed, there is some incentive, however small, for people to favor parties, candidates, and policies that would make them better off. Even here, however, people may be reluctant to express support for policies that make them better off if they do so at the expense of others. People might prefer to express otherregarding preferences even when they actually would prefer outcomes that benefit themselves. People could express preferences that they believe would benefit a larger group – for example, a nation. People may feel a patriotic duty to vote for candidates and policies that appear to be beneficial to their nation. People might feel compelled to vote for policies that appear to instrumentally benefit a subset of the population, such as the poor, or ethnic minorities. But this view of instrumental preferences shades very close to being expressive. Those preferences would be instrumental only if individuals would choose them if the choice were theirs alone. Otherwise, what a voter claims is an instrumental preference actually is expressive. Beyond the intentions of voters, they may not have a good understanding of whether public policies actually are instrumentally beneficial. For example, as this is being written, there is a national debate in the United States about establishing a $15 nationwide minimum wage. The argument is that this would be beneficial to low-wage people because sure, and I recall thinking to myself that I liked the nuts so much that I would continue buying them even if the price were much higher. But then the price actually did rise substantially, and much to my surprise I quit buying them. I could certainly afford them, but refused to pay the higher price. I was very surprised that I did not know even my own preferences prior to being presented with the choice of whether to buy at the higher prices.

https://doi.org/10.1017/9781009323178.006 Published online by Cambridge University Press

5.2  Instrumental Preferences

77

they would be paid more. However, it would also price low-productivity people out of the labor market. Employers will hire employees only if those employees can produce more revenue for the employer than it costs to hire them. Even if employers could “afford” to pay $15 an hour, they will not hire employees who cost them more than they bring in to the firm. Phrased differently, a $15 an hour minimum wage would make it illegal for low-productivity people to have jobs. The policy mandates that if someone wants to work but cannot negotiate a wage of $15 an hour or more, that person is not allowed to work. Most people do not look at the minimum wage this way, but it is descriptive of the effect of the law. A law that appears on the surface to benefit one group of people – low productivity workers – might actually work to their detriment. There is no instrumental consequence to supporting this policy, or any other, however, because political preferences are expressive rather than instrumental. People might feel they are being kind to low-income workers by voting to have their employers pay them more. Without passing judgment on the merits of a $15 minimum wage, the way people view the policy depends on how it is framed. When people are asked whether they think it should be illegal for people to accept jobs that pay less than $15 an hour, support is considerably lower than if they are asked whether they support a mandated $15 minimum wage. While both questions are the same instrumentally, the first seems to penalize those who want to work but cannot find higher-paying jobs, while the second seems to express support for low-income workers. Most voters will not be in a good position to evaluate the actual consequences of even a relatively straightforward issue such as this. Would a $15 minimum wage help low-income workers or harm them? Many political issues are far more complex. But because political preferences are expressive, voters have no good reason to devote their time to an intense study of the issues. They vote for the option that makes them feel good. While people will argue that their own political preferences are based on instrumental values, there may be a tenuous connection. People surely want what is good for themselves and others, and even more likely, they will want to express views to others that are other-regarding, so they will offer instrumental justifications for their political preferences. Instrumental values may be a source of political preferences, but there are reasons to think that expressive preferences will be based on more than just instrumental values.

https://doi.org/10.1017/9781009323178.006 Published online by Cambridge University Press

78

The Formation of Political Preferences

5.3  The Status Quo Because political preferences are expressive, people do not have instrumental reasons to change them. People are exposed to political points of view early in life, from family, friends, and educational institutions. Children pick up some political preferences from their parents, so it is not surprising that political party affiliation tends to run in families. Children whose parents identify as Democrats are likely to identify themselves as Democrats, at least initially. The idea of family values extends to political preferences. Because political preferences are expressive, there are no negative instrumental consequences from holding a particular set of preferences, and children are likely to gain more utility from adopting their parents’ preferences than some alternative. Indeed, where would children even be exposed to alternative points of view? Achen and Bartels note that “partisan loyalties often carry across generations. … Children tend to adopt the partisanship of their parents, and those attachments tend to persist into adulthood.”6 One place outside the home that might influence political preferences is school. One of the merits claimed for government-run schools (as opposed to private schools) is that they do a better job of socializing students. Socialization includes educating students about political preferences. Civics classes are largely about how government institutions enable governments to act in the best interest of their citizens, and history classes tilt heavily toward political history, sometimes to emphasize the merits of a nation’s political predecessors. By instilling patriotic preferences in students, and increasingly in the twenty-first century pointing out shortcomings of earlier generations, schools can push preferences onto students that undermine certain features of the political landscape.7 Students are an easy mark for instilling political preferences, because outside of family influences, they likely have only vague ideas about 6

7

Christopher H. Achen and Larry M. Bartels, Democracy for Realists: Why Elections Do Not Produce Responsive Government (Princeton: Princeton University Press, 2016), p. 233. The social environment offered by schools may also affect political preferences. Stephen B. Billings, Eric Chyn, and Kareem Haggag, “The Long-Run Effects of School Racial Diversity on Political Identity,” American Economic Review Insights 3, no. 3 (September 2021), pp. 267–284, find that a 10 percentage point increase in the share of minority enrollment in K–12 schools decreases the likelihood of voters registering Republican by 12 percent fifteen years later.

https://doi.org/10.1017/9781009323178.006 Published online by Cambridge University Press

5.3  The Status Quo

79

different political ideologies and policies. Educational institutions do not have to change students’ minds, because many students do not have clearly formed preferences. Educational institutions provide students with facts to support conclusions, and educators, like everyone else, tend to look for facts that support the conclusions they want to reach. A selective presentation of facts can influence young minds that are unaware of facts that support opposing views. A second advantage that educators have in creating the political preferences of their students is that students are examined on what they have learned, so one motivation for learning is to be able to do well on examinations. People are much more convinced of arguments and viewpoints when they make those arguments themselves than when they just hear them from others. A student who hears a particular viewpoint in class is less likely to embrace that viewpoint than a student who repeats that viewpoint to answer an examination question. There is no doubt that education, especially about public policy issues, is powerful propaganda. As this is being written, there is a lively debate about teaching critical race theory in schools. The arguments both for and against it rest on the assumption that the political preferences of students will be influenced by what they hear in the classroom. There would be no reason to debate whether it should be part of the curriculum if those debating did not believe that educators can influence the political views of their students. The power of propaganda in a classroom setting was illustrated to me as a college student in what was probably the single classroom hour that influenced me more than any other. It was a biology class in which the instructor showed two movies about the cross-Florida barge canal that was then under construction. The instructor did no speaking at all. He just showed the two movies. One was produced by the Army Corps of Engineers and explained all the benefits of the canal. It would facilitate shipping and would provide recreational benefits like fishing and boating. It would help the environment by facilitating drainage of flooded areas. The second movie was produced by an environmental group and explained that the canal would cause much damage to the natural environment and was too narrow for most shipping that goes around the tip of Florida. While the first movie showed happy citizens fishing in the sections that had been constructed, the second showed acres of dead trees partially submerged where construction was taking place.

https://doi.org/10.1017/9781009323178.006 Published online by Cambridge University Press

80

The Formation of Political Preferences

Both movies, on their own, were very convincing. I am not sure what my biology teacher hoped I would learn from those two movies, but what I did learn was that if you hear only one side of an argument from a knowledgeable party, that argument will be very convincing. Had I seen only one of those movies, I would have been convinced by that side. Early in people’s lives they pick up political preferences, from family members, from school, and perhaps from other sources that are discussed below. Because political actions of citizens and voters are expressive, there is no feedback mechanism to prompt them to change those preferences. It costs them nothing to cling to their preferences even if they are irrational, and even if the policies they advocate would harm them (and others).8 People need persuasive reasons to change their expressive preferences because they have no instrumental consequences. A wellestablished bias noted by behavioral economists is status quo bias.9 That applies to political preferences as well as for goods and other life situations. People feel comfortable with the status quo, including the political preferences they currently hold.10

5.4  The Endowment Effect One of the significant discoveries in behavioral economics is the endowment effect, which finds that people place a higher value on things they own, just because they own them. In one experiment revealing the endowment effect, people were given a mug and later asked how much they would have to be paid to sell it, while others were asked how much they were willing to pay to buy the same mug. Sellers wanted much higher prices to sell their mugs than buyers were willing to pay to purchase a mug.11 The endowment effect concludes that the higher value sellers placed on the mug compared to buyers was the utility they received from ownership.12 8

This is, again, a major point made by Caplan, The Myth of the Rational Voter. David Gal, “A Psychological Law of Inertia and the Illusion of Loss Aversion,” Judgment and Decision Making 1, no. 1 (July 2006), pp. 23–32. This status quo bias and other behavioral anomalies are discussed by Daniel Kahneman, Jack L. Knetsch, and Richard H. Thaler. “Anomalies: The Endowment Effect, Loss Aversion, and Status Quo Bias,” Journal of Economic Perspectives 5, no. 1 (Winter 1991), pp. 193–206, who are leading authorities in behavioral economics. The endowment effect is explained by Daniel L. Kahneman, Jack L. Knetsch, and Richard H. Thaler, “Experimental Tests of the Endowment Effect and the Coase Theorem,” Journal of Political Economy 98, no. 6 (December 1990), pp. 1325–1348. The endowment effect is one of the behavioral anomalies discussed by Kahneman et al., “Anomalies.”

9 10

11

12

https://doi.org/10.1017/9781009323178.006 Published online by Cambridge University Press

5.5  Peer Pressure

81

One study found that people who had tickets to the NCAA Final Four basketball tournament valued their tickets fourteen times higher than the value potential buyers placed on the tickets.13 In many settings people reveal that they would require a much higher payment to sell something they own than they would be willing to pay for that same thing if they do not own it. The fact that their willingness to pay to acquire objects is less than their willingness to accept payment to sell offers evidence that part of the value of things they own comes solely from the fact that they own it. This is the endowment effect. While there is good evidence of an endowment for goods, the same principle applies to political preferences. People view their political preferences as a part of their identities, and value them more because those preferences belong to them. Once people have adopted preferences as their own, the higher value they place on them makes people reluctant to change them. Bernheim et al. conclude that “people are naturally drawn to inflexible outlooks because inflexibility helps them to resolve the core time-inconsistency problem.”14 People are especially reluctant to reconsider their political preferences, because those preferences have no effect on the political outcomes they experience. Those preferences belong to them, and just like the mug in the experiment described above, they are reluctant to give them up.

5.5  Peer Pressure People are social creatures and have a desire to conform with the views of the groups with which they identify. Bernheim et al. note that “Accumulating evidence shows that people often adjust their behavior to conform with the choices of others, either partially or fully. … One common response to ‘social dissonance’ is to conform, i.e. mimic observable manifestations of the predominant attitude. That response creates unpleasant dissonance between the individual’s actions and their true attitudes. … the individual resolves this internal dissonance by gradually bringing their outlook in line with their observable behavior.

13

Ziv Camon and Dan Ariely, “Focusing on the Forgone: How Value Can Appear So Different to Buyers and Sellers,” Journal of Consumer Research 23, no. 3 (December 2000), pp. 360–370.

14

Douglas Bernheim, Luca Braghieri, Alejandro Martinez-Marquina, and David Zuckerman, “A Theory of Chosen Preferences,” American Economic Review 111, no. 2 (February 2021), p. 722.

https://doi.org/10.1017/9781009323178.006 Published online by Cambridge University Press

82

The Formation of Political Preferences

Thus,  a  predominant attitude is eventually internalized and become a general motivation for behavior.”15 People’s desire to fit in with their social group shifts their preferences to conform with those of the group. This extends well beyond politics. Sports fans of the same teams bond together, and people find solidarity with their peers who share their musical tastes, fashion sense, and food preferences, as well as with those who attended the same schools they did.16 That desire for solidarity and group membership may be stronger with regard to political views for two reasons. First, because political preferences are expressive rather than instrumental, there are no instrumental consequences for adopting the political views of those with whom they associate. Political outcomes will be the same regardless of one’s political preferences. Second, actual political outcomes do have instrumental effects. In that sense, they are more important than people’s tastes for music or food. Thus, it is more important for group solidarity to have the same political preferences than, for example, the same preferences for food.17 People have an incentive to agree with their peers to get along with them. On choices as unimportant as what food to eat or what music to listen to, they run the risk of eating food they do not like or listening to music they do not enjoy if they go along with the group’s preferences when they are at odds with their own. If they modify their political preferences to conform with the group, political outcomes will be unaffected, so they can gain the expressive value of conforming with the views of their group without bearing any instrumental cost. Christopher Achen and Larry Bartels conclude that “group loyalties matter a great deal and the details of policy positions not very much”18 in the political preferences people express. While one sees this sort of tribalism in sports, all that is at stake is the outcome of a game. Still, sports rivalries can become heated, to the extent that fans of opposing teams can engage in physical fights with one another. Politics is competitive like sports, but it is not just a game.

15

16

17 18

Douglas Bernheim, Luca Braghieri, Alejandro Martinez-Marquina, and David Zuckerman, “A Theory of Chosen Preferences,” American Economic Review 111, no. 2 (February 2021), p. 736. See Robert Ackerlof, “‘We Thinking’ and Its Consequences.” American Economic Review 106, no. 5 (May 2016), pp. 415–419, for a discussion of the effects of group identity. Pamela Johnston Conover, “The Influence of Group Identifications on Political Perception and Evaluation,” Journal of Politics 46, no. 3 (August 1984), pp. 760–785. Achen and Bartels, Democracy for Realists, p. 233.

https://doi.org/10.1017/9781009323178.006 Published online by Cambridge University Press

5.5  Peer Pressure

83

The public policies that emerge from political competition have real effects on people’s lives, so it is easy to see why disagreements can become heated. Those disagreements can be avoided if people adopt the views held by those close to them. Academic studies document the effect of peer pressure on the formation of political views. People want to fit in with their friends and their peer groups, so will be inclined to adopt the political preferences of their peer groups.19 They feel good when they do so, and it costs them nothing in terms of actual political outcomes. Political views form a part of people’s group identity, so they conform to the political views of others in their group.20 Because the adoption of and acting upon political views is expressive rather than instrumental, this is easy to do, costs people nothing in an instrumental sense, and brings them utility by reinforcing their group identity. People do adopt the political views of their group, and often are completely unaware that they are doing so.21 Peer pressure may also affect whether individuals are sufficiently motivated to vote at all.22 Because their one vote will have no effect on an election outcome, something else must motivate voters to cast a ballot, and peer pressure can be that motivation.23 If one is a member of a peer group that supports a particular candidate, the concern about disapproval of one’s peers might lead one to vote rather than abstain. This has an obvious effect on political outcomes. For example, if two candidates were each favored by an equal number of voters, the candidate who motivates more supporters to actually show up and vote will win. Peer pressure from supporters could make that difference. Enrico Cantoni and Vincent Pons offer interesting empirical evidence on the peer effect in both preferences and the propensity to turn out to vote by looking at the voting behavior of individuals who move from one political jurisdiction to another. They find that those who relocate are more likely to adopt the public policy preferences of the majority in 19 20 21

22 23

This is the conclusion of Stephanie Chen and Oleg Urminsky, “The Role of Causal Beliefs in Political Identity and Voting,” Cognition 188 (July 2019), pp. 27–38. Jay J. Van Bavel and Andrea Pereira, “The Partisan Brain: An Identity-Based Model of Political Belief,” Trends in Cognitive Psychology 22, no. 3 (March 2018), pp. 213–224. Geoffrey L. Cohen, “Party over Policy: The Dominating Impact of Group Influence over Political Beliefs,” Journal of Personality and Social Psychology 85, no. 5 (2003), pp. 808–822. David K. Levine and Andrea Mattozzi, “Voter Turnout with Peer Punishment,” American Economic Review 110, no. 10 (October 2020), pp. 3298–3314. This is discussed in detail by Meredith Rolfe, Voter Turnout: A Social Theory of Political Participation (Cambridge: Cambridge University Press, 2012).

https://doi.org/10.1017/9781009323178.006 Published online by Cambridge University Press

84

The Formation of Political Preferences

their new locations, and are more likely to adopt their turnout behavior as well.24 A substantial academic literature supports the conclusion that people’s public policy preferences are influenced by their peers. People may identify with particular interest groups, such as teachers, unions, or law enforcement, and support candidates who have the backing of those interests. They may focus on narrow issues such as abortion or gun control, to the exclusion of other issues that have larger effects on their lives. If that makes them feel good, they can express those political preferences with no instrumental consequences, because their political preferences have no effect on political outcomes.

5.6  Cognitive Dissonance People act in ways to reduce cognitive dissonance. They do not want to hold conflicting views, and they readily accept, and even seek out, evidence that supports their current views, while being skeptical of any evidence that conflicts with their views.25 While this is true in general, it is even more true when the views people hold have no instrumental consequences. When people engage in instrumental acts, they tend to judge facts more accurately than when they engage in expressive acts.26 People tend to adjust their political beliefs to minimize inconsistency and tend to accept evidence supporting what they already believe.27 Joseph McMurray concludes that “a weakly informed voter – is therefore more likely to do harm than good by voting, and so prefers to abstain.”28 McMurray’s conclusion that a weakly informed voter may do more harm than good seems reasonable, but his conclusion that therefore the voter prefers to abstain is questionable, for two reasons. It equates the voter’s level of information with the voter’s intensity of preferences,

24

25

26 27

28

Enrico Cantoni and Vincent Pons, “Does Context Outweigh Individual Characteristics in Driving Voting Behavior? Evidence from Relocations within the United States,” American Economic Review 112, no. 4 (April 2022), pp. 1226–1272. Russel Golman, George Loewenstein, Karl Ove Moene, and Luca Zarri, “The Preference for Belief Consonance,” Journal of Economic Perspectives 30, no. 3 (Summer 2016), pp. 165–188. Brendan Nyhan, “Facts and Myths about Misperceptions,” Journal of Economic ­Perspectives 34, no. 3 (Summer 2020), pp. 220–236. Mikael Elender, “Correcting Mistakes: Cognitive Dissonance and Political Attitudes in Sweden and the United States,” Public Choice 153, nos. 1/2 (October 2012), pp. 235–249. Joseph McMurray, “The Paradox of Information and Voter Turnout,” Public Choice 165, nos. 1/2 (2015), p. 123.

https://doi.org/10.1017/9781009323178.006 Published online by Cambridge University Press

5.6  Cognitive Dissonance

85

and it suggests that voters vote because they are other-regarding. A potential third problem is determining whether the information voters have is correct. There is some evidence that voters make better choices when they are informed by experts, but just about any public policy position is supported by experts, presenting voters with the problem of determining whether experts on one side or the other of a public policy position are correct.29 People tend to choose their experts to minimize cognitive dissonance. People can have intense political preferences without having detailed knowledge about the alternatives. Often, increased information can weaken political preferences. One reason is that it is unlikely that a voter will favor every policy position a candidate espouses, so more information about a preferred candidate may weaken a more informed voter’s support. Another reason is that on most issues, there are persuasive arguments on both sides. Issues would not be controversial if this were not the case. So, gathering more information can weaken a voter’s attachment to a particular policy position. For these reasons, more information can increase cognitive dissonance. By remaining relatively uninformed, the voter can remain enthusiastic by minimizing cognitive dissonance. There is a tendency to link lack of voter information with indifference among alternatives,30 but this is not necessarily the case. Voters may have strong preferences even with little information, as Caplan indicates,31 and a lack of information does not appear to dissuade people with strong preferences from expressing them.32 The lack of consequences for expressing political preferences by uninformed voters leads to voter overconfidence, because the personal cost of making a poor choice in the voting booth is zero.33 Part of the utility a voter gets from voting comes from the very fact that the voter knows that one vote will not be decisive. This allows voters to hold expressive preferences that differ from their instrumental preferences. They can vote for outcomes they would not choose if the choice were theirs alone. 29 30 31 32 33

Archman Chakraborty, Parikshit Ghosh, and Jaideep Roy, “Expert-Captured Democracies,” American Economic Review 110, no. 6 (June 2020), pp. 1713–1751. Timothy J. Feddersen and Wolfgang Pesendorfer, “The Swing Voter’s Curse,” American Economic Review 86, no. 3 (June 1996), pp. 408–424. Bryan Caplan, The Myth of the Rational Voter: Why Democracies Choose Bad Policies (Princeton: Princeton University Press, 2007). Martin Gilens, “Political Ignorance and Collective Policy Preferences,” American Political Science Review 95, no. 2 (June 2001), pp. 379–396. Pietro Ortoleva and Erik Snowberg, “Overconfidence in Political Behavior,” American Economic Review 105, no. 2 (February 2015), pp. 504–535.

https://doi.org/10.1017/9781009323178.006 Published online by Cambridge University Press

86

The Formation of Political Preferences

Sendhil Mullainathan and Ebonya Washington undertake a study of political preferences and conclude that “the results of two estimation strategies together provide evidence that the act of voting strengthens future opinions of the chosen candidate.”34 Having voted for a candidate, people reduce cognitive dissonance by thinking more highly of that candidate in the future. There is no instrumental cost to doing so, because an individual vote has no effect on the outcome. There are many reasons to think that even if instrumental preferences remain the same, expressive preferences can change as a result of voting, and as a result of election outcomes. Mullainathan and Washington argue that “dissonance effects on voting would suggest that election efficiency is not necessarily increasing in turnout as high turnout today implies that a large body of the electorate will be biased in their evaluations of the incumbent in future contests.”35 One might question whether any election procedure can be judged to be efficient, but one interpretation of Mullainathan and Washington’s conclusion is that cognitive dissonance leads expressive preferences to differ from instrumental preferences. There is a tenuous relationship between instrumental preferences and expressive preferences. Voters adopt political preferences that conform with those of their peer group, and modify those preferences to reduce cognitive dissonance. The reduction of cognitive dissonance is easier with political preferences than with market preferences, because voters see little evidence that they have made poor choices. Voters can see the results of the collective choice actually made, but cannot observe what would have occurred had the losing alternative won. It is easy for those who supported the winner to think that things would be worse had the other side won, and easy for those who supported the loser to think that things would be better had the other side won. Whether that actually is the case is always a matter of speculation. There is no way to know what would have happened had an election outcome been different.

5.7  The Bandwagon Effect People like to side with winners, so there may be a bandwagon effect in politics. Thorsten Veblen noted that people choose to consume goods 34

35

Sendhil Mullainathan and Ebonya Washington. “Sticking with Your Vote: Cognitive Dissonance and Political Attitudes,” American Economic Journal: Applied Economics 1, no. 1 (2009), p. 108. Mullainathan and Washington, “Sticking with Your Vote,” p. 109.

https://doi.org/10.1017/9781009323178.006 Published online by Cambridge University Press

5.8  Political Advertising

87

that are popular with others, jumping on the bandwagon, and the same tends to be true in politics.36 Candidates and policies that appear popular may gain additional support simply because of that appearance. The same may be true in markets. People may choose to eat at a particular restaurant because it is popular, but a negative experience at that restaurant is likely to make them choose differently next time. Political preferences do not bring with them any negative consequences, so the bandwagon effect may be a larger determinant of political preferences than for preferences for goods and services. One academic study found that voters who support losing candidates modify their views after an election has taken place to reduce cognitive dissonance.37 They jump on the bandwagon and after the election express views closer to those of the winners than they did before. The bandwagon effect offers a direct explanation for why candidates want to claim that they are leading in pre-election polls. A message that tells potential supporters “We are winning, so let’s keep that momentum going” will be more effective than saying, “We are behind in the polls, so help me catch up.” If analogies between sports teams and political candidates hold up, one can see that winning teams can attract more fans to their games, and politicians have the same motivation for trying to get supporters on the bandwagon.

5.8  Political Advertising Chapter 3 discussed the influence of advertising on preferences in general, and specifically on the formation of political preferences. Advertising appears very effective in influencing people’s choices in markets, where their preferences have a direct and instrumental impact on outcomes, and it is not unreasonable to believe that advertising is even more effective in forming political preferences, which, because they are expressive rather than instrumental, tend to be based more on emotion than on reason. The goals of advertising in markets versus politics differ in a notso-obvious way. The goal of market advertising is to entice consumers

36 37

Thorstein Veblen, The Theory of the Leisure Class (New York: Macmillan, 1899). Austin D. Eubanks, Scott Eidelman, Derrick F. Till, David Sparkman, Patrick Stewart, and Robert H. Wicks, “Outcome-Based Dissonance and Morton’s Fork: Evaluative Consequences of Unfavorable Alternatives in the 2016 U.S. Presidential Election,” Journal of Theoretical Social Psychology 4, no. 1 (January 2020), pp. 21–31.

https://doi.org/10.1017/9781009323178.006 Published online by Cambridge University Press

88

The Formation of Political Preferences

to buy a product. The goal of political advertising is to win elections. This may mean trying to entice voters to support a candidate, but elections can also be won by discouraging people from supporting opposing candidates. As Chapter 3 discussed, this explains the persistence of negative political advertising despite frequent claims that people dislike negative ads. One problem political advertisers face is that because there are no instrumental consequences from casting a vote that, instrumentally, is against a voter’s interest, changing voter opinions is difficult. People want to minimize the discomfort of cognitive dissonance, which causes voters to reinforce their current political views rather than change them.38 Voters who have political preferences are inclined to stick with them. Political advertising will not, for most voters, change their minds about the candidates and policies they support. It is more likely to affect whether they vote or abstain. To minimize cognitive dissonance, voters are less likely to change their support than they are to decide they are not enthusiastic enough about their political preferences to express them at the polls. Political advertising is less about changing people’s minds on how to vote and more about changing their minds on whether to vote. As Christopher Achen and Larry Bartels note, “Political campaigns consist in large part of reminding voters of their partisan identities – ‘mobilizing’ them to support their group at the polls.”39 Positive political advertising is designed to make a candidate’s supporters feel good about supporting the candidate so voters will want to turn out to vote for that candidate, and to make those who support a candidate’s opponent feel bad about supporting the opponent, pushing those voters to abstain. Turnout is often the key to winning an election, and political advertising is geared toward getting supporters to turn out and nonsupporters to stay home on election day. Most voters will stick with their political preferences to minimize cognitive dissonance. For this reason, negative ads, unpopular as they might be, are effective. Voters will not expend the effort to cast an expressive vote for candidates who do not make them feel good about casting that vote. Voters do not want to reconsider their political views because that introduces the discomfort of cognitive dissonance. But they will readily 38

39

Ryan K. Beasley and Mark R. Joslyn, “Cognitive Dissonance and Post-Decision Attitude Change in Six Presidential Elections,” Political Psychology 22, no. 3 September (2001), pp. 521–540. Achen and Bartels, Democracy for Realists, p. 311.

https://doi.org/10.1017/9781009323178.006 Published online by Cambridge University Press

5.9  Candidate Charisma

89

reconsider whether to vote at all, because the act of voting has no instrumental value. If advertising is very effective, it could push supporters of a candidate beyond abstaining toward voting for another candidate. But even if this is the case, political advertising carries very little actual policy content and is designed to appeal to voter emotions. Reinforcing the conclusions of the previous section, voters will form, and change, their political preferences based on emotions and expressive factors, not on logical and instrumental factors.

5.9  Candidate Charisma People’s political preferences are more inclined to be based on emotional factors than instrumental factors when making political choices than when making market choices because in markets they get what they choose whereas political outcomes are unaffected by what they choose. Therefore, charismatic politicians should have an easier time selling their ideas to voters than charismatic salespeople have selling products to buyers. The expansion of visual mass media has increased the importance of charisma. Prior to the twentieth century, candidate views would be relayed in print media. The twentieth century first brought radio and then television as major methods of political exposure. Formal voting models depict voters choosing among candidate platforms and casting instrumental votes, but the incentives created by voting divert voter preferences away from the platforms of candidates and parties and toward the personal characteristics of the candidates. As the personal characteristics of candidates have become a more important component of electability relative to platforms because of the increased importance of visual media, voters are more inclined to support those individuals rather than their policies. Voters are unlikely to support a personable candidate who holds views contrary to their own, but this conclusion depends on individuals having predetermined political preferences, and on candidates having well-defined policies in their platforms. Both conditions are questionable. This chapter offers many reasons why voters’ preferences can be reshaped, from advertising to peer pressure. Voters who have specific policy preferences are unlikely to change them, but for voters whose preferences come from peer pressure, the desire to reduce cognitive dissonance and other factors discussed here can lead them to be attracted to charismatic candidates. Meanwhile, candidate platforms tend to be vague, with uplifting messages rather than actual policy proposals, to the

https://doi.org/10.1017/9781009323178.006 Published online by Cambridge University Press

90

The Formation of Political Preferences

point that voters can be attracted to the uplifting messages of charismatic candidates more than to specific policy proposals. Platforms have many different planks, and voters who find an emotional connection with a candidate can likely also find parts of the candidate’s platform agreeable. At that point, there is no instrumental cost to voters from adopting preferences that correspond more closely to the preferred candidate’s entire platform. Indeed, voters do not even need to know a candidate’s entire platform. Just knowing a few elements with which they can agree can be enough to gain their support. This will make voters feel good by reducing cognitive dissonance. There are psychological reasons why voters may prefer not to know about areas in which they disagree with their preferred candidate. Voter preferences can be determined by which candidate a voter favors, rather than voters favoring a candidate based on underlying voter preferences. Once individuals identify a candidate whose ideas are congenial, voter loyalty can shift from supporting the policies the individual advocates to supporting the individual him/herself. Vladimir Putin in Russia seems to be a good example, as does Donald Trump in the United States. People support the person more than the person’s policies. The result is that public policy debates become increasingly partisan and personal, rather than policy oriented. Citizens support policies because their preferred candidate supports them, and oppose policies supported by their opponents, regardless of the content of the policies. Their support shifts to an individual rather than a set of policies because it reduces cognitive dissonance. They can feel conflicted if they support a candidate and like many of that candidate’s policy proposals but dislike some others, so they avoid questioning policies of the candidate they favor. And again, there are no personal costs to be borne by doing so because an individual’s political preferences will never be decisive. The link between what people choose and what people get is severed in political decisions, so the utility people get from their political preferences comes solely from the satisfaction they get from expressing them. They modify their own preferences to correspond to those of the candidates they favor to prevent them from feeling uneasy about preferring a candidate who does not always agree with their policy views. Party activists tend to shift their political views to correspond with current party thinking. “Due to their commitment to their parties, many continuing activists have brought their attitudes on at least some policy dimensions into line with the positions emerging among party candidates,

https://doi.org/10.1017/9781009323178.006 Published online by Cambridge University Press

5.10  Current Conditions

91

leaders, and other activists.”40 People adopt the policy preferences of the political elite. They follow their leaders. Unlike the assumptions in formal voting models, voters do not start with a set of preferences and vote for candidates and parties who are closest to what they prefer. When they identify their preferred candidates and parties, they then modify their preferences to correspond with those candidate and party platforms. Because political preferences are expressive and not instrumental, there is no instrumental cost to doing so. It reduces cognitive dissonance and makes voters feel better about their preferences for candidates and parties. The incentives involved in political choice lead voters to favor the policy positions of candidates, rather than choosing candidates who share their policy positions. Voters do prefer candidates who hold policy positions similar to their own, but once voters identify a preferred candidate or party, they reduce cognitive dissonance by shifting their own views to be closer to those of the candidate or party they support. This can lead to candidates having more extreme political platforms than is suggested by spatial models in which candidate positions converge on the median voter’s preference. More informed voters can lead to greater differentiation of candidate platforms.41 Voters are often unaware of the specific policy positions of candidates partly because candidates deliberately do not express them. Candidates run based on their competence, and their commitment to the public interest, rather than on specific policy proposals. Parties and candidates do have policy orientations, but campaign based on their ability to do the job rather than on what, specifically, they intend to do.

5.10  Current Conditions Many of the factors that influence voter preferences pull them toward maintaining the preferences they have adopted in the past, but some voters do change their positions on parties and candidates. One factor that influences aggregate voter outcomes is current conditions. If things, generally, 40

41

Geoffrey C. Layman, Thomas M. Carsey, John C. Green, Richard Herrera, and Rosalyn Cooperman, “Activists and Conflict Extension in American Party Politics,” American Political Science Review 104, no. 2 (May 2010), pp. 342–343. This is the conclusion of Raphael Boleslavsky and Christopher Cotton, “Information and Extremism in Elections,” American Economic Journal: Microeconomics 7, no. 1 (February 2015), pp. 165–207.

https://doi.org/10.1017/9781009323178.006 Published online by Cambridge University Press

92

The Formation of Political Preferences

look good, voters tend to favor incumbents. If conditions are bad, they will vote against incumbents. At least a part of the aggregate result – that incumbents are more likely to win when conditions are good – is a result of turnout. Supporters of incumbents are more likely to show up and vote when conditions are good, and opponents of incumbents are less likely to turn out and vote against incumbents when conditions are good. But some voters will vote differently depending on current conditions. Political scientists call this type of voter preference formation retrospective voting. Rather than focusing on the platforms and policies proposed by parties and candidates, voters vote based on their past performance, assuming that conditions they experience are (at least in part) due to the performance of those in office. They may look at overall conditions, but also focus on how they, as individuals, are faring. If their lot in life has improved, they will tend to favor incumbents; if it has deteriorated, they will vote against the incumbent. Morris Fiorina argues that there is some sense in this type of retrospective voting.42 Voters cannot hope to be knowledgeable in many public policy areas. There is just too much to know. Even public policy specialists tend to focus on one or a few areas. So it makes some sense to judge the performance of elected officials based on the outcomes their policies produce rather than on looking ahead to try to forecast the likely outcomes of their proposals. Achen and Bartels devote four chapters of their book to retrospective voting, presenting much evidence that it has a significant influence on election outcomes.43 To the extent that it does occur, politicians have an incentive to exploit the tendency toward retrospective voting by manipulating current conditions so that things are improving before elections. With regard to economic policy, there is evidence that governments tend to increase government spending and implement easier monetary policy prior to elections, improving economic conditions to gain political support.44 Even more directly, incumbents are in a position to offer cash subsidies, tax breaks, and regulatory advantages to groups of voters to gain political support. Voters are rationally ignorant regarding public policy in general, but know first-hand about their own particular situations. 42 43 44

Morris P. Fiorina, Retrospective Voting in American National Elections (New Haven, CT: Yale University Press, 1981). Christopher H. Achen and Larry M. Bartels, Democracy for Realists. There is a substantial economic literature on political business cycles that result from this type of policy, initiated by William D. Nordhaus, “The Political Business Cycle,” Review of Economic Studies 42, no 2 (April 1975), pp. 169–190.

https://doi.org/10.1017/9781009323178.006 Published online by Cambridge University Press

5.11  Mass Media

93

5.11  Mass Media Advertising is a component of mass media, but mass media influence over political preferences extends well beyond advertising that is done by parties and candidates. Evidence suggests that exposure to political information from the mass media increases turnout,45 and turnout is an important factor in determining election outcomes. Favorable coverage helps candidates by mobilizing their base, while unfavorable coverage harms candidates by discouraging their base from voting. Information does not have to change people’s political preferences to affect elections because it affects who turns out. Tianyi Wang looks at the effect of a 1930s Populist radio show by Father Coughlin and finds that listeners exposed to Father Coughlin’s show were less likely to vote for President Roosevelt in the 1936 election.46 The effect of Father Coughlin’s radio broadcasts persisted beyond the airing of his show. Those in his listening area were also less likely to buy war bonds, and more likely to join pro-German groups. It does not seem unreasonable that the slant put on interpretation of current events by mass media would affect people’s political preferences. Mass media has more of a polarizing effect in the twenty-first ­century than it did in the twentieth, because there are more media outlets with more diverse political orientations. In the second half of the t­wentieth century, television had an increasingly dominant effect in conveying news about politics, but there were only a few television networks that had relatively homogeneous viewpoints on current events. All voters tended to get the same points of view from mass media. In the twenty-first century, a larger number of television networks with more diverse viewpoints offer viewers political coverage targeted toward their ­political preferences. Arye Hillman observes, “The media can profit by catering to ­identity-conforming interpretations and perspectives of expressive populations.”47 The diversity of viewpoints offered by twenty-first-century mass media helps reinforce the political biases that individuals have by offering viewers supporting evidence for their views while downplaying 45 46 47

Valentino Larcinese, “Does Political Knowledge Increase Turnout? Evidence from the 1997 British General Election,” Public Choice 131, nos. 3/4 (June 2007), pp. 387–411. Tianyi Wang, “Media, Pulpit, and Populist Persuasion: Evidence from Father Coughlin,” American Economic Review 111, no. 9 (September 2021), pp. 3064–3092. Arye L. Hillman, “Expressive Behavior in Economics and Politics,” European Journal of Political Economy 26, no. 4 (December 2010), p. 414.

https://doi.org/10.1017/9781009323178.006 Published online by Cambridge University Press

94

The Formation of Political Preferences

contrary evidence. While one might think that a wide diversity of viewpoints in mass media would lead to voters who are better informed, that does not happen when individuals consume only the subset of mass media sources that support their views and reject other sources (that they likely do not see anyway) as fake news and lies. Because voting is an expressive act, voters do not receive any instrumental benefit from becoming better informed. The result is that they have an incentive to collect information that supports their views, which reduces cognitive dissonance, and ignore information that conflicts with their views. Expressive preferences are based more on emotion than on facts. The mass media reinforces the views voters already have, and benefits charismatic candidates who appear attractive as personalities. This became more true in the twentieth century when television supplanted newspapers as a primary source of news, and even more true in the twenty-first century when media outlets target their political coverage to appeal to a subset of the population.

5.12 Conclusion This chapter has reviewed a large number of studies on how voters form their political preferences, and for the most part, those studies suggest that emotional reasons such as group identity, peer pressure, reduction of cognitive dissonance, and irrationally held beliefs play a large role. As Christopher Achen and Larry Bartels note, “Most of the time, voting behavior merely reaffirms voters’ partisan and group identities.”48 Expressive preferences are not determined primarily from instrumental preferences. Individuals adapt their preferences to be more like those of their peers, and can shift preferences to conform with those of successful parties and politicians because people get more utility from supporting winners than losers – the bandwagon effect. A substantial amount of research in the social sciences concludes that people’s preferences are subject to outside influence, and factors such as peer pressure, the desire to conform, and attempts to minimize cognitive dissonance all work to shape people’s preferences, including their political preferences. These findings, backed by academic studies and conforming with causal observation, go part-way toward understanding how people form their public policy preferences. If people’s preferences tend to conform with those of those with whom they associate, this pushes 48

Achen and Bartels, Democracy for Realists, p. 294.

https://doi.org/10.1017/9781009323178.006 Published online by Cambridge University Press

5.12 Conclusion

95

the question back to understanding what influences groups to adopt the policy preferences they do. Achen and Bartels observe that people are concerned with other more pressing issues than being well-informed citizens. They say, “Human beings are busy with their lives. … For most, leisure time is at a premium. Sorting out which presidential candidate has the right foreign policy toward Asia is not a high priority for them. Without shirking more immediate and more important obligations, people cannot engage in much well-informed, thoughtful political deliberation, nor should they.”49 But then they go on to say, “Well-informed citizens, too, have come in for their share of criticism, since their well-organized ideological thinking often turns out to be just a rather mechanical reflection of what their favorite group and party leaders have instructed them to think.”50 People get their policy preferences from the political elite. They follow their leaders. The idea that voters have preferences and that candidates and parties develop their platforms to conform with voter preferences is, at a minimum, on shaky ground, because voter preferences for public policies are subject to influence from so many forces unconnected to how those policies will actually affect the voter. The direction of causation goes (mostly) in the other direction. Candidates and parties sell their policy preferences to voters, who adopt the policy preferences of the candidates and parties with whom they identify.

49 50

Achen and Bartels, Democracy for Realists, p. 9. Achen and Bartels, Democracy for Realists, p. 12.

https://doi.org/10.1017/9781009323178.006 Published online by Cambridge University Press

6 Anchor Preferences and Derivative Preferences

Chapter 5 reviewed an extensive literature covering many academic disciplines which showed that people’s political and public policy preferences are influenced by many factors not directly related to public policy. People’s preferences are subject to manipulation, as Chapter 3 explained, and political preferences are more susceptible to manipulation because they are expressive preferences. For the masses, the utility people gain from their political preferences comes solely from having and expressing them. Therefore, the preferences they express are not necessarily an indicator of the public policy outcomes they would most prefer. This chapter builds on those that went before to explain that citizens and voters adopt a political identity that defines their perception of their own political views. They anchor on this identity, which could be a political party, an ideology, an issue, or a political candidate. Having adopted this political identity, most of their public policy preferences derive from those of their anchors. There are many complex public policy issues. Few voters will have the competence to analyze many of them, and there is little incentive to do so anyway, because their policy preferences have no effect on policy outcomes. They get more satisfaction from holding policy preferences that minimize cognitive dissonance by adopting policy views consistent with those of their anchors. Having identified with an anchor, most of their public policy preferences are derivative of the anchor’s. People follow their leaders.

6.1  Public Policy Preferences Shopping for political outcomes is different from shopping in the ­market, for several reasons. The previous five chapters have emphasized the 96

https://doi.org/10.1017/9781009323178.007 Published online by Cambridge University Press

6.1  Public Policy Preferences

97

difference between instrumental preferences and expressive preferences, and why that difference is important for understanding how voters form their preferences and how they express them at the ballot box. This difference means that voters may vote for outcomes they would not choose if the choice were theirs alone. Another difference between voting and shopping is that voters are offered choices between bundles of policies rather than being able to express preferences for policies individually. By voting for a party or candidate, they are supporting all of the policies in the platform of that party or candidate. They cannot pick and choose individual policies to support, as they can choose individual goods in the market.1 To feel good about their political choices – their anchors – they will want to agree with them on most things. To minimize cognitive dissonance, they will want to like all of the items in the market basket, and there is no instrumental reason standing in their way. Shoppers who shop at a supermarket take their carts from isle to isle, placing goods in their carts that they want to purchase. Every item in the cart is chosen by the shopper because the shopper wants the item, and items the store stocks that the shopper does not want do not go into the shopper’s cart. Shoppers get exactly the bundle of goods they want. If shopping were done in supermarkets as it is done in elections, competing candidates would fill shopping carts with items they wanted to offer the voters, and voters would then be offered the choice of a cart filled by one candidate or another. Rather than shoppers personally deciding what would go in their carts, candidates would decide, and shoppers would be offered only the choice of carts filled by one of the candidates. To extend the analogy, supporting a party or candidate means expressing a preference for everything in that candidate’s cart. If supermarket shopping were done this way, it is apparent that shoppers would not get exactly what they wanted in their shopping carts. Even in the cart they liked the best, there would be some items shoppers would never buy, and their favorite cart would likely be missing other items the shopper would have chosen if allowed the individual choice. In politics, supporting a party or candidate surely means supporting some policies the voter would not favor, and perhaps favoring a candidate who does not support other policies the voter supports. The voters are offered one total bundle of public policies or another and cannot customize their political shopping carts the way they can their market shopping carts. 1

This point is made by James M. Buchanan, “Individual Choice in Voting and the Market,” Journal of Political Economy 62, no. 4 (August 1954), pp. 334–343.

https://doi.org/10.1017/9781009323178.007 Published online by Cambridge University Press

98

Anchor Preferences and Derivative Preferences

To minimize cognitive dissonance, citizens can adjust their preferences to conform with the contents of their anchors’ carts. There is no reason not to do so, because the cart they actually get will be the same regardless of the preference they express. Even this analogy paints too rosy a picture because it is never entirely clear what is in the political market basket. As Chapter 3 noted, politicians deliberately make their platforms vague. To keep with the supermarket analogy, imagine that shoppers can only see what is on the top of the basket. Some items below are not visible so the shopper is not entirely sure about what is in the basket. Voters must choose among baskets they know have some things they do not want, and have some things they cannot see, so they do not know exactly what they are favoring when they choose one basket over another. This uncertainty makes it easier for citizens to support the policies of their anchors. They support general goals, such as higher standards of living, environmental preservation, and world peace, rather than specific policies to achieve those goals. This compounds the issues raised by the fact that they are making an expressive rather than instrumental choice. When offered the opportunity to express a preference for one shopping cart over another, voters know that the basket they get may not be the one they say they prefer, and know that they will get the same basket regardless of their expressed preference. This is why they could express a preference for a basket they would not choose if the choice were theirs alone. They could vote for the basket with all the fruits and vegetables and other healthy food, being virtuous in their expressed preference, when if the choice were theirs alone, they would have taken the basket with the beer and junk food. Political platforms are multidimensional, encompassing a large number of discrete issues. Just as people have preferences over beverages, snacks, and main courses, people have views on tax policy, the size of government, the appropriate role of government in health care, national security issues, what should be protected under freedom of speech, foreign policy, the degree to which women should be able to choose to have an abortion, and much more. Political preferences encompass all of these dimensions. This chapter builds on the material that precedes it to describe how people form their political preferences for public policies. The political preferences citizens and voters express can be divided into anchor preferences and derivative preferences. Anchor preferences are those that embody a person’s political identity. People may identify as members of a political party, a political movement, an ideology, an

https://doi.org/10.1017/9781009323178.007 Published online by Cambridge University Press

6.2  Anchor Preferences

99

issue, an individual candidate, or a religion.2 Their political preferences anchor on this identity. Most policy preferences are derivative preferences, derived from the preferences associated with the person’s anchor. People’s political identity forms an anchor, and most of their policy preferences are derived from that anchor. They follow their leaders.

6.2  Anchor Preferences The previous chapter reports results from a large number of studies on the formation of political preferences, and this chapter applies the findings of those studies to provide a framework that explains how people form their political preferences. Anchor preferences are those that define people’s political identities. They define how people see themselves, and how they want others to see them.3 The previous chapter offered some insight into how anchor preferences are formed. Typically, people identify with a political group, and their anchor preferences are the preferences of that group. People want to fit in with their peers, so adopt similar views. Alan Hamlin and Colin Jennings refer to this as social identification.4 There is no instrumental reason to debate with others about expressive preferences, as there is over instrumental preferences. People can disagree with members of a group about where that group should lunch together, because it makes an instrumental difference in the quality of their lunch, but there is no instrumental reason to debate among friends about who should be elected to a political office; the same person will be elected regardless of whether they agree, so they might as well agree. That will make their personal interactions more pleasant, and will solidify their sense of group identity.5 Christopher Achen and Larry Bartels, discussing the ideas of sociologists, say that “groups are fundamental to social life. People took their views from the groups to which they belonged, often because the people

2 3

4 5

Norman Podhoretz, Why Are Jews Liberals? (New York: Doubleday, 2009). See Stephen Ansolabehere and M. Socorro Puy, “Identity Voting,” Public Choice 169, nos. 1/2 (October 2016), pp. 77–95, and Moses Shayo, “A Model of Social Identity with an Application to Political Economy: Nation, Class, and Redistribution,” American Political Science Review 103, no. 2 (May 2009), pp. 147–174, for a discussion. Alan Hamlin and Colin Jennings, “Expressive Political Behaviour: Foundations, Scope, and Implications,” British Journal of Political Science 41, no. 3 (July 2011), pp. 645–670. Further discussion can be found in Dennis J. Snower and Steven J. Bosworth, “IdentityDriven Cooperation versus Competition,” American Economic Review 106, no. 5 (May 2016), pp. 420–424.

https://doi.org/10.1017/9781009323178.007 Published online by Cambridge University Press

100

Anchor Preferences and Derivative Preferences

around them made it difficult not to do so. Ideas that individuals had picked up elsewhere were disapproved of, and most people were eventually convinced to discard them.”6 People adopt political preferences that conform with those they associate with, because they want to get along with others in their group. Socialization influences political preferences more than careful reasoning does. Anchor preferences often come from one’s family. A person’s parents supported a particular political party, so they do. Peer pressure from a group of friends or associates is another source of anchor preferences. Once those preferences are formed, the endowment effect suggests that people will place a higher value on those preferences just because the preferences are theirs.7 To reduce cognitive dissonance, people will readily accept information supporting their anchors and will tend to dismiss information that questions their anchors.8 The bandwagon effect further encourages them to adopt and retain the preferences of those around them as their anchor.9 While some people do analyze and evaluate issues to adopt their anchor, there is a path dependency to anchor preferences for the reasons stated in the previous paragraph, and those anchors tend to form for reasons other than the instrumental value of the public policies behind them. This does not mean that people cannot change anchors when compelling reasons arise to do so, but that because anchor preferences are expressive, their origins often lie with factors outside of the particular policies espoused by their anchors. People view their beliefs as assets, and are reluctant to depreciate their value by questioning them.10 Anchor preferences often are rooted in ideologies. People might think of themselves as progressives, conservatives, libertarians, or socialists. Even more specifically, people may anchor to political parties, thinking of themselves as Democrats or Republicans, for example. Sometimes people 6

Christopher H. Achen and Larry M. Bartels, Democracy for Realists: Why Elections Do Not Produce Responsive Government (Princeton: Princeton University Press, 2016), p. 219. 7 Daniel Kahneman, Jack L. Knetsch, and Richard H. Thaler, “Experimental Tests of the Endowment Effect and the Coase Theorem,” Journal of Political Economy 98, no. 6 (December 1990), pp. 1325–1348. 8 Sendhil Mullainathan and Ebonya Washington, “Sticking with Your Vote: Cognitive ­Dissonance and Political Attitudes,” American Economic Journal: Applied Economics 1, no. 1 (2009), pp. 86–111. 9 Thorstein Veblen, The Theory of the Leisure Class (New York: Macmillan, 1899). 10 Roland Benabou and Jean Tirole, “Identity, Morals and Taboos: Beliefs as Assets,” Quarterly Journal of Economics 126, no. 2 (2011), pp. 805–855.

https://doi.org/10.1017/9781009323178.007 Published online by Cambridge University Press

6.2  Anchor Preferences

101

will anchor on specific individuals. Donald Trump is a good example of someone who developed a group of supporters that went beyond anchoring to an ideology or party and anchored specifically on him. After he lost the 2020 presidential election, his supporters stormed the Capitol to try to prevent the Senate from ratifying his loss. They were anchoring on Donald Trump, the individual, more than the Republican party or conservative ideology. President Trump’s supporters could not have realistically thought that their demonstrations in the nation’s capital could have altered the course of the election and kept him in the presidency. But they did not have to consider the possibility seriously, because their activities were expressive, showing support for Trump himself even as most members of Trump’s own party acknowledged that Trump did not win the election. Thousands of Trump supporters participated in the demonstration that breached the Capitol building, so even if an individual demonstrator did think there was a chance the election result could be overturned, one additional demonstrator would have no effect on the outcome. People acting instrumentally would see that, just as with voting, the presence of one additional person would have no effect, so purely instrumental actors would stay home, and perhaps follow the events on television just as sports fans watch their favorite teams on television. Regardless of one’s views on the efficacy of the demonstration, each individual participant was acting expressively, following the same logic that identifies voting as an expressive act. More generally, participating in demonstrations and political rallies of any type is expressive action. One more participant will make no actual difference, so the low-cost course of action – and perhaps the safest course of action – is to stay home. People participate not because they think one more participant will make a difference but because they want to express solidarity with the group and have a sense of belonging. They want to be able to say, “I was there.” Anchor preferences sometimes do derive from instrumental preferences, in particular when individuals are directly affected. In the 1960s, for example, Southern Democrats supported segregation, which provided an anchor for individuals with those views, even though Southern Democrats tended to be fiscally conservative and so had a lot in common with Republicans on most issues. The civil rights movement that ended forced segregation destroyed that anchor for Southern Democrats, and the once solidly Democratic South became increasingly Republican. This example is an exception in that people do not often change their anchors, but it also shows that when they do, they exchange one anchor

https://doi.org/10.1017/9781009323178.007 Published online by Cambridge University Press

102

Anchor Preferences and Derivative Preferences

for another. While anchors can be chosen for instrumental reasons, they most often are rooted in non-policy-related reasons, like peer group identification. Regardless of how they are formed, once people identify with an anchor, the endowment effect, the bandwagon effect, and attempts to minimize cognitive dissonance work to reinforce that anchor.

6.3  Principled Preferences Beyond a doubt, some people have principled reasons for their anchor preferences. Consider the contentious issue of abortion. Some people may hold strong views that women have the right to determine whether to continue a pregnancy. As the slogan goes, “My body, my choice.” Others may hold the strong view that abortion is murder. They will anchor on candidates and parties that reflect their strong views. Factors discussed in the previous chapter may still be relevant. Where did those voters get their strong preferences? They may have come from friends, family, religious organizations, or political messages sent by parties and candidates. Nonetheless, some voters anchor on specific policy issues based on strongly held values. Other voters may anchor on policy preferences that are broader rather than single issues. American voters might feel that public policy should be more oriented toward helping the less fortunate, and view the Democratic party as sharing those values, so they anchor on Democrats. Other voters may feel that a strong economy with a favorable business climate increases the well-being of everyone, and view the Republican party as sharing those values, so they anchor on Republicans. When asked, almost everyone will claim to have public policy preferences based on principles, as opposed to adopting the preferences of their family and friends, or being manipulated by political advertising. But they are still subject to these influences, and still have no instrumental reason for their choice of anchor, because their one vote will not affect an election outcome. Michael Barber and Jeremy Pope present evidence that people tend to anchor on a party and derive their policy preferences from their anchors.11 When their anchors change their policy positions on issues, citizens tend to change policy preferences to conform with their anchors. Voters may have more confidence in the wisdom of their 11

Michael Barber and Jeremy C. Pope, “Does Party Trump Ideology? Disentangling Party and Ideology in America,” American Political Science Review 113, no. 1 (2019), pp. 38–54.

https://doi.org/10.1017/9781009323178.007 Published online by Cambridge University Press

6.4 Abstention

103

policy preferences than is warranted, because they bear no cost from being overconfident. Regardless, the conclusions of this volume remain the same even if citizens and voters anchor on well-researched and strongly held convictions.

6.4 Abstention In keeping with the idea that that one vote will not affect an election outcome, instrumental voters will conclude that making the effort to vote fails a cost–benefit test – that the cost of voting exceeds the expected benefit, so the “rational” voter will abstain. Rational individuals should not engage in behavior that costs more than they receive in return, and there is no instrumental return to voting because election outcomes are not determined by a single vote. These rational voters may have strong preferences, but choose not to vote because voting fails the cost–benefit test. Voters may also choose to abstain because they are apathetic. This is different from the rational abstainers in that apathetic voters just do not care enough to be motivated to cast a ballot. Stepping away from voting for the moment, imagine someone living in an apartment who has a friend living in a nicer apartment but paying 50 percent more in rent. The person living in the lesser apartment would rather live in the nicer one, but to do so might require giving up some leisure to work more and earn more income, and the person living in the lesser apartment just does not care enough about moving to a nicer apartment to be motivated to do anything about it. Similarly, the apathetic voter may prefer one candidate or party over another but is not sufficiently motivated to actually cast a vote. Upon closer examination, the voter apathy motivation for not voting does not hold up well, because there is nothing an individual voter can actually do to affect the outcome of an election. If the voter was told, “It is your choice to pick the party or candidate who will hold power,” and the voter says, “I do not care enough to make a choice,” that would be a sign of voter apathy. But the voter does not actually have that choice, so what appears to be apathy is more likely to be the voter who wants to cast an instrumental vote but realizes that one vote will have no instrumental effect. The idea that low voter turnout is the result of voter apathy is suspect. Abstention may be a rational expressive preference objecting to the political process rather than to particular candidates. To vote may be viewed as signaling that the voter approves of the process that assigns

https://doi.org/10.1017/9781009323178.007 Published online by Cambridge University Press

104

Anchor Preferences and Derivative Preferences

some individuals the ability to exercise political power over others, and some individuals may choose to abstain to express their views that they do not approve of the process. They do not view government institutions as a legitimate way to give some people power over others. Abstention is an expression of the anchor preferences of these voters. If no candidates are acceptable to a voter, or for those who view the whole electoral process as illegitimate, the voter may express a preference to abstain rather than to vote for what the individual views as the lesser evil, or support of an illegitimate process. A libertarian anarchist might view all government as illegitimate, and refuse to participate in the process for expressive reasons.12 Some will argue that citizens have a moral responsibility to vote. The anarchist, or anyone who believes government has overstepped its moral bounds, may conclude that it is a moral responsibility to abstain – to avoid participating in a system they view as illegitimate and immoral. Those who hold political power like high voter turnout, and may argue that citizens have a moral duty to vote. They want to encourage high turnout because large numbers of voters may symbolize widespread support for the system, and legitimize their claims on power.13 Without passing judgment on the argument, not everyone agrees, and some express their disagreement by choosing not to vote. Just as voters make the choice to vote, abstainers make the choice not to vote. A voter who chooses to abstain is expressing the view that casting an empty ballot is preferable to voting for any other alternatives offered to voters.14 For these voters, offering the alternative to vote for “None of the Above” might draw them to the polls. Voter apathy is not a good explanation for abstention, because voters must have a reason they choose to do something other than vote. Choosing to abstain because one vote has no instrumental consequence is a reason. Choosing not to vote because the individual views the process as illegitimate is a reason. 12

13 14

See, for example, Murray N. Rothbard, The Ethics of Liberty (Atlantic Highlands, NJ: Humanities Press, 1982). Many individuals have told me personally that they do not vote because they do not approve of government’s exercise of power in various spheres. See Murray Edelman, The Symbolic Uses of Politics (Urbana: University of Illinois Press, 1964). Dominik Klein, “Expressive Voting, Graded Interests, and Participation,” Public Choice 188 nos. 1/2 (July 2021), pp. 221–239, shows that different vote aggregation procedures offer different payoffs to expressive voters. For example, if more alternatives are offered to voters, they are more likely to find that one of them more closely corresponds to their preferences. This assumes that voters have preferences and cast votes reflecting those preferences, a premise that is questioned in this volume.

https://doi.org/10.1017/9781009323178.007 Published online by Cambridge University Press

6.5  Derivative Preferences

105

For those who view voting as an action that violates their personal values, abstention is an expression of their anchor preferences. One of the challenges of those who are running for elective office is to convince their supporters they should vote rather than abstain. Abstention is the rational choice for those who want to cast an instrumental vote, so the challenge is to turn those who favor the candidate or party into expressive voters. Candidates must convince voters that they will gain utility simply from going to the polls to express their support, because voters must realize that their one vote will not affect the election outcome.

6.5  Derivative Preferences Public policy has many different dimensions. Within a broad area such as foreign policy there are many subsidiary issues. How should support for Israel versus Palestinians be balanced? What military and trade policies should be applied to China? How much support should the United States offer to NATO? Within domestic policy, how aggressively should government fight a war on drugs? When should abortion be allowed? What is the appropriate role of government in health care? What regulatory environment is appropriate for corporations? There are persuasive arguments to support different views on all these issues, the evidence being that many knowledgeable people assert strong but very different views on public policy issues. And there are many more policy issues than just those listed in this paragraph. There are expert opinions supporting all sides of these issues. Which experts should voters follow? The political issue space is multidimensional – there are many issues. When this is the case, in theory there is no stable winning policy platform under majority rule. No matter what the status quo is, there is always a different set of policies that would be favored by a majority.15 The cyclical majority illustrated in Chapter 1 offers a simple example, which has been shown to hold generally when there is more than one policy dimension. With many policy dimensions, there is no one political platform that is a majority rule equilibrium, in the sense that it is preferred by a majority of voters to any alternative.16 15

16

R. D. McKelvey, “Intransitivities in Multi-Dimensional Voting Models and Some Implications for Agenda Control,” Journal of Economic Theory 12, no. 3 (June 1976), pp. 472–482. William H. Riker, “Implications from the Disequilibrium of Majority Rule for the Study of Institutions,” American Political Science Review 74, no. 2 (June 1980), pp. 432–446, discusses some implications of this conclusion.

https://doi.org/10.1017/9781009323178.007 Published online by Cambridge University Press

106

Anchor Preferences and Derivative Preferences

Despite theoretical models showing that there is no stable equilibrium set of policies supported by a majority over all others, Gordon Tullock observes that in the real world, as opposed to the abstract world of theory, political outcomes are very stable.17 Theoretical models of political instability suggest that incumbents should never get reelected, yet they do most of the time, and government policies and programs seem to be even more durable than private sector firms, despite celebrated proofs of the uniqueness and stability of competitive market equilibrium.18 The median voter model, described by Anthony Downs, depicts voter preferences lying on a left–right continuum, and concludes that majority rule voting produces the outcome that is most preferred by the median voter. That model, which depicts political preferences as existing on a single dimensioned left-to-right continuum, seems more descriptive of actual political outcomes than McKelvey’s multidimensional model. In the median voter model, there is a unique stable equilibrium outcome. But the model relies on the assumption that political preferences lie on a single-dimensioned continuum. One explanation, building on the material from the previous chapters, is that people do not independently develop their policy preferences on individual issues. Rather, they have anchor preferences and they derive their preferences on most issues from their anchors. With expressive political preferences, people’s preferences are determined by the choices they make rather than their choices being determined by their preferences. Gabriel Lenz offers evidence that voters’ preferences are rooted in an anchor that they are reluctant to abandon, and that they adopt the public policy positions of their anchors. Their preferences follow those of the leaders they choose as anchors.19 Models of political instability assume voters have a given set of preferences and that political candidates try to design their platforms to conform to those preferences. However, if people’s political preferences on most issues are derived from their anchor preferences, preferences for most public policy issues are absorbed into voters’ anchor preferences. People anchor on a candidate or party and then derive their policy preferences on most issues from those held by their anchor.

17 18 19

Gordon Tullock, “Why So Much Stability?” Public Choice 37, no. 2 (1982), pp. 189–202. See, for example, Kenneth J. Arrow and Gerard Debreu, “Existence of an Equilibrium for a Competitive Economy,” Econometrica 27, no. 3 (1954), pp. 265–290. Gabriel L. Lenz, Follow the Leader? How Voters Respond to Politicians’ Policies and Performance (Chicago: University of Chicago Press, 2012).

https://doi.org/10.1017/9781009323178.007 Published online by Cambridge University Press

6.5  Derivative Preferences

107

Individuals who anchor as Democrats will tend to support more government gun control, more government involvement in health care, and a woman’s right to have an abortion. People do not start with those preferences and then decide, “I am a Democrat.” Rather, they start with their political identity as Democrats and conclude, “I am a Democrat, so I favor gun control, more government involvement in health care, and a woman’s right to have an abortion.” These preferences are derivative preferences, derived from the policy positions advocated by the ­individual’s anchor.20 The endowment effect indicates that when individuals adopt anchor preferences as their own, they will value those preferences more highly. When their derivative preferences conform with the policies held by their anchors, that reduces cognitive dissonance. To derive public policy preferences from a voter’s anchor imposes no instrumental cost on individuals, because political choices are expressive actions and the political preferences individuals express have no effect on the political outcomes. These are well-established ideas in social science. It would seem unlikely that there would be a strong correlation between people’s views on gun control and their views on government’s role in health care, or on military spending and legalization of recreational drugs, yet there is. One explanation is that most public policy preferences are derivative. People identify with their anchor preferences and then adopt most of their public policy preferences based on those advocated by their anchors. As Bernheim et al. observe, “[B]ecause worldviews are complicated objects, it is difficult to invent them … society offers a menu of prefabricated worldviews to its members, most of whom will limit themselves to combinations of these options.”21 People are not inclined to research many public policy issues to draw independent conclusions about which are best. This is even more true because their policy preferences have no effect on policy outcomes. Rather, they adopt the policy preferences of their anchors on most issues, which reduces cognitive dissonance and makes them feel better about their political choices. Otherwise, it would be difficult to support any party or candidate with much enthusiasm, 20

21

One source of political identity is conformance with members of organizations individuals choose to join. See Jean-Paul Carvalho, “Identity-Based Organizations,” American Economic Review 106, no. 5 (May 2016), pp. 410–414, for a discussion. Douglas Bernheim, Luca Braghieri, Alejandro Martinez-Marquina, and David Zuckerman, “A Theory of Chosen Preferences,” American Economic Review 111, no. 2 (February 2021), p. 724.

https://doi.org/10.1017/9781009323178.007 Published online by Cambridge University Press

108

Anchor Preferences and Derivative Preferences

because it is unlikely that one candidate or party would be a close match to an individual’s preferences on all issues. Consider the variety among the market baskets of grocery shoppers. How satisfied would they be if instead of filling their own baskets, they were offered the choice of only one of a few baskets filled by others? But this is what happens in politics, and many voters do exhibit intense preferences for individual candidates and parties. It is implausible to think that this could happen without voters adopting derivative policy preferences from their anchors. Because political preferences are expressive, people can search for information to support their political beliefs and limit cognitive dissonance, and “when it comes to rationalizing away contradictory evidence, compartmentalizing knowledge, and deluding oneself, more educated, attentive, and analytically able people often display greater propensities toward such behaviors.”22 As Christopher Achen and Larry Bartels observe, “In fact, political rationalization is often most powerful among people who are well-informed and politically engaged.”23 People protect their beliefs, and adopting derivative preferences that support their anchor preferences helps people do this.

6.6  Most Political Preferences Are Derivative In the aggregate, almost all political preferences are likely to be derivative. People choose their anchors based on expressive rather than instrumental reasons and derive their policy preferences from their anchors. How many voters will view any one issue as an anchor issue? To offer a simple (and unrealistic) example, assume there are ten voters and ten dimensions in political issue space, and each voter has a different anchor issue. Voter 1 anchors on issue 1, voter 2 anchors on issue 2, and so forth. The result is that on any one issue, political preferences are derivative for nine of the ten voters. Even if anchor preferences are instrumental – and often they are not – there will be almost no correspondence between voters’ instrumental preferences and the preferences they express at the ballot box. A voter who anchors on a party adopts political preferences that correspond with that party’s platform. The result is that political preferences 22 23

Roland Benabou and Jean Tirole, “Identity, Morals and Taboos: Beliefs as Assets,” Quarterly Journal of Economics 126, no. 2 (2011), p. 145. Christopher H. Achen and Larry M. Bartels, Democracy for Realists: Why Elections Do Not Produce Responsive Government (Princeton: Princeton University Press, 2016), p. 294.

https://doi.org/10.1017/9781009323178.007 Published online by Cambridge University Press

6.6  Most Political Preferences Are Derivative

109

of party supporters will be determined almost entirely by candidates and parties. Their preferences on most issues will be determined not by the utility they would get from policy outcomes, but rather by the positions of those to whom they are anchored. Consider an earlier example of principled voters who anchor on an issue like abortion. American voters who favor a woman’s right to make the choice are likely to favor the Democratic party, and policy preferences on other issues like gun control, the tax structure, government involvement in health care, and redistribution programs are likely to be derivative of those of their anchors. Those who oppose abortion, likewise, are likely to have derivative preferences that follow the Republican party. It is not a coincidence that people who tend to be pro-choice on the abortion issue also tend to favor stronger gun control. Having chosen an anchor, most policy preferences are derivative. One element emphasized in Donald Trump’s 2016 presidential campaign was that China was taking advantage of the United States by engaging in predatory export policies and stealing intellectual property from US companies, among other things. After Donald Trump was elected president in 2016, individuals were much more willing to express xenophobic views that echoed the president’s, and those who disagreed with those views were less likely to express their disagreement.24 Expressed political preferences tend to shift to reflect the preferences of political leaders, which is consistent with the endowment effect, the bandwagon effect, and attempts to minimize cognitive dissonance. More generally, this example illustrates that people are willing to modify their derivative preferences to correspond with those of their anchors. The Republican party, at least since Ronald Reagan’s presidency, supported free trade, but after President Trump won on a protectionist platform aimed at China, Mexico, and other countries, most Republicans did not push back and argue that Trump’s protectionist policies were out of step with the party’s values. Rather, they supported Trump’s trade policies. The avoidance of cognitive dissonance will prevent many voters from changing their party affiliations and anchor preferences, because to do so would be to admit that they were wrong in the past. The endowment effect suggests that people place a higher value on the party they identify 24

Leonardo A Bursztyn, Georgy Egorov, and Stefano Fiorin, “From Extreme to Mainstream: The Erosion of Social Norms,” American Economic Review 110, no. 11 (November 2020), pp. 3522–3548.

https://doi.org/10.1017/9781009323178.007 Published online by Cambridge University Press

110

Anchor Preferences and Derivative Preferences

with simply because they view it as their party. Because there are many public policy dimensions, voters who are committed to their anchor preferences then adopt policy preferences in other dimensions to conform with their anchor, which again minimizes cognitive dissonance. Group identity has a strong influence on individual views and behavior.25 People who anchor on a particular party, say, Democrats in the United States, tend to accept the party’s positions on all (or at least, most) policy issues. That is what it means to identify as a Democrat. So, their policy preferences are derivative of their anchor. But it goes the other way too. Someone might anchor on an issue, such as legalization of recreational drugs. Those in favor would support Democrats and those opposed would support Republicans, and preferences on other issues will tend to be derivative of those parties and candidates whose policy views correspond with their anchor. Almost all policy preferences are derivative. Supporting candidates on one particular issue may mean supporting other areas of candidates’ platforms the voter may not favor.26 There is no reason to think that people who are opposed to the legalization of recreational drugs would also favor restricting women’s ability to have abortions, yet both were components of Republican platforms in 2020. People who feel strongly about one issue but have weak or poorly defined preferences on the other will anchor as Republicans and, minimizing cognitive dissonance, will adopt the party’s views on the other. Such support is costless in the instrumental sense because nobody casts a decisive vote. The particular example given here may be dated by the time the reader reads it. The earlier example of the Republican party once strongly supporting free trade but turning to support tariffs and protectionism when Donald Trump was elected president in 2016 illustrates how derivative preferences can shift when an anchor’s policies shift. Policy views on other derivative issues are just as subject to change. They change because people’s policy preferences on most issues are derivative and conform with the preferences of their anchors, so when anchors change their policy views on some issues, voters who have those anchors also make that change. Those preferences are derivative and expressive, so there is no instrumental cost from doing so. 25 26

George A. Akerlof and Rachel E. Kranton, “Economics and Identity,” Quarterly Journal of Economics 115, no. 3 (August 2000), pp. 715–753. Randall G. Holcombe and Robert J. Gmeiner, “Interest Group Support for Non-Group Issues,” Constitutional Political Economy 29, no. 3 (September 2018), pp. 303–316.

https://doi.org/10.1017/9781009323178.007 Published online by Cambridge University Press

6.7  Defining the Issue Space

111

6.7  Defining the Issue Space Public policy has many dimensions. How much should be spent on national defense? How much should government spend on health care? How much regulation of business is appropriate? These are examples of the many different political issues that define the issue space, and each issue represents a separate dimension. There are so many issues that it would be difficult to come up with a count of them all, and in this sense the issue space is multidimensional. This is important, in theory, because if the issue space is multidimensional, there is no one set of public policies that will be favored by a majority of voters over all others. Whatever the status quo, there is some alternative set of policies that a majority would favor.27 There are many possible explanations for why political outcomes and public policies appear more stable than in theory they might be. One is that the issue space is not as multidimensional as it seems when just counting up issues. Preferences on particular issues tend to be correlated with preferences expressed by the anchors voters have, which collapses the issue space down, perhaps to a single dimension. Politicians and parties commonly are identified by their positions on a political spectrum from left to right, because preferences on a large number of issues fit that left-to-right continuum. If preferences fall on this left-to-right continuum, then there is a stable outcome. The platform preferred by the median voter is preferred by a majority of voters over all others.28 Even this framework may be overstating the dispersion of voter preferences. Consider a two-party system as in the United States. Whether voters anchor on a party, a candidate, or an issue, most preferences are derivative and voter preferences will shift to one of the platforms of the two major party candidates. Voters choose their anchors, and their derivative preferences fall in line with that anchor. The left-to-right distribution of preferences collapses down to two points. Voters prefer the policies of one candidate or the other. The winning candidate is the one who is able to get the most votes, which is a two-step process.29 First, candidates must convince voters to favor them, and second, candidates must convince those supporters to 27 28 29

This conclusion is explained in the theoretical work of R. D. McKelvey, “Intrasitivities in Multi-Dimensional Voting Models and Some Implications for Agenda Control.” This is explained by Anthony Downs, An Economic Theory of Democracy (New York: Harper & Row, 1957). In US presidential elections, it is the candidate who gets the most electoral votes, but the process that follows is the same.

https://doi.org/10.1017/9781009323178.007 Published online by Cambridge University Press

112

Anchor Preferences and Derivative Preferences

turn out and vote. This is done by becoming the expressive choice of voters, which is not necessarily their instrumental choice. Candidates win over voters with charisma and emotion. They bolster their chances of winning by convincing voters they will win. The bandwagon effect attracts supporters who want to side with a winner, so candidates who appear to have a lot of support gain even more for that reason alone. Instrumental concerns are secondary, and emotional and expressive factors pull voters to favor the policy preferences of one candidate or another. Regardless of the number of issues in play, political preferences of voters center around the two points that are the points that are the platforms of the two candidates. To go back to the shopping cart analogy, they express a preference for everything in one of the two shopping carts, which reduces cognitive dissonance, gives them a feeling of ownership as described by the endowment effect, and has no negative instrumental consequences because one vote is never decisive. In the supermarket, shoppers express preferences over every available item, but in the voting booth, voters express a preference for one fully loaded basket of policies over another. When most voter preferences center around one of two points in a multidimensional issue space, there is no third alternative that can gain a majority of the votes and lead to cycles. Depending upon how persuasive the candidates are at turning out their bases, there can be a cycle back and forth between the two candidate positions from one election to the next, but not a third alternative that could beat one of the two. Voters do have a choice of anchors, but to return to the shopping cart analogy, that is not the same as being able to choose everything in the cart. They choose a cart on which they anchor, and express a preference for the contents of that cart. Compared to the median voter model in which candidate platforms converge on the preference of the median voter, the existence of anchor preferences that in turn define voters’ derivative preferences opens the opportunity for candidates to differentiate their platforms. Unlike that median voter model where candidates move their platforms to correspond with voter preferences, voter preferences move to correspond with candidate platforms. This conclusion rests on the premise that candidates and parties are able to shape the political preferences of voters, and the preceding chapters explained why this is the case. It does not have to be the case, if parties and candidates are unable to persuade voters to express preferences for their policies, but it tends to be the case because political preferences are expressive, and because people form those preferences for other than instrumental

https://doi.org/10.1017/9781009323178.007 Published online by Cambridge University Press

6.7  Defining the Issue Space

113

reasons. This view of political preferences corresponds more closely with observed political outcomes than the view that voters have preferences and candidates try to tailor their platforms to those preferences. The result is that the issue space collapses down into two points, regardless of how many issues are in play. Voter preferences converge on one of those points. There have been some notable exceptions in US presidential elections. Ross Perot claimed 19 percent of the popular vote in 1992, George Wallace received 14 percent of the popular vote in 1968, and Robert La Follette gained 17 percent of the popular vote in 1924. These elections show that sometimes a third-party alternative will arise that will cause voters to shift their anchors, but in all cases, those candidates did not win and their appeal faded to insignificance by the time of the next election. These exceptions show that political preferences are not always formed by voters who hang onto an anchor and derive most of their policy preferences from their anchor. Voters (and abstainers) are not machines, and are not programmed to make their political choices. They are individuals who can be persuaded to change their minds if offered persuasive reasons. So, there will be exceptions to the general observation that voters tend to stick with their anchors. Even the exceptions in the previous paragraph, however, are examples in which some member of the political elite – a La Follette, a Wallace, a Perot – was able to shift some voters’ anchors. The expressed preferences of voters who supported those candidates still came from the political elite. Proportional voting, in which members of parliament are elected based on the percentage of the vote their parties receive (rather than a winner-take-all election), allows for a larger number of anchor positions. In proportional systems, a party that receives 20 percent of the votes gets 20 percent of the seats in parliament, whereas in plurality systems (like the United States), parties that get 20 percent of the votes get nothing. Proportional voting allows for a greater variety of anchors, but the general principle remains the same. Voters select an anchor, whether a party, a candidate, or an issue, and most of their policy preferences are derivative of that anchor. Unlike theoretical models of voting in which voters can have policy preferences spread throughout a multidimensional issue space, the way that voters determine their political preferences means that the preferences of most voters locate on a few points, and those points are defined by parties and candidates. Models that assume voter preferences as given exogenously do not recognize the process by which voters determine their expressive political preferences. Voter preferences are not exogenously

https://doi.org/10.1017/9781009323178.007 Published online by Cambridge University Press

114

Anchor Preferences and Derivative Preferences

given. They are determined as a part of the political process and are heavily influenced by the policy positions of the political elite. The tendency for voters to anchor on parties or candidates collapses many issues into a single-dimensioned issue space. Layman et al. say, “Due to their commitment to their parties, many continuing activists have brought their attitudes on at least some policy dimensions into line with the positions emerging among party candidates, leaders, and other activists.”30 The Republican shift on trade policy is but one example. The issue space runs in a single dimension between parties, from left to right, despite there being many discrete policy issues. Bernheim et al. echo this idea, saying that “there is a robust tendency for people to adopt pure worldviews that credit only one point of view, rather than mixed worldviews that credit many. Polarization follows directly from this ‘purist’ tendency.”31 Unlike voting models in which voters have preferences and candidates adopt their platforms to correspond with voter preferences, candidates and parties offer platforms and voters adopt those platforms as their anchors, with preferences on most policies being derivative of their anchors. Platforms often do not converge on some median preference, but remain separated (polarized), and the single dimension of citizen political preferences runs from one platform to the other. Voters adopt the policy positions of candidates rather than choosing candidates who share their policy positions. While voters do tend to anchor on parties and candidates with views similar to those in their peer group, once they identify an anchor, they reduce cognitive dissonance by adopting the policy preferences of their anchors.

6.8 Conclusion The idea that voter preferences are expressive rather than instrumental, and that expressive preferences can differ from instrumental preferences, is well established. This means that attempts to connect the preferences voters express at the ballot box with their instrumental interests are, at best, misleading. To do so would require that some connection be drawn between instrumental and expressive preferences, and this has not (yet) been done. The notion that democratic political processes somehow reveal the general will, as Rousseau suggested, does not hold up. 30

31

Geoffrey C. Layman, Thomas M. Carsey, John C. Green, Richard Herrera, and Rosalyn Cooperman, “Activists and Conflict Extension in American Party Politics,” American Political Science Review 104, no. 2 (May 2010), pp. 342–343. Bernheim et al., “A Theory of Chosen Preferences,” p. 723.

https://doi.org/10.1017/9781009323178.007 Published online by Cambridge University Press

6.8 Conclusion

115

While that idea of expressive preferences is well established, an explanation of how expressive preferences are formed is not. Academic discussion on expressive preferences has tended to note that they differ from instrumental preferences without explaining how they differ, or how they are formed. The analysis in this and preceding chapters casts doubt on finding a robust connection between instrumental and expressive preferences, because the motivations for forming those two types of preferences are different. There is, at best, a tenuous connection between the two. The utility of expressive preferences comes solely from having them and expressing them, because they have no effect on instrumental outcomes. People adopt the preferences of their families and friends because they value the feeling of group solidarity. They tend to side with parties and candidates who look like winners, as explained by the bandwagon effect. Once they adopt political preferences, the endowment effect reinforces those preferences. People’s political preferences tend to anchor on a party, a candidate, or a policy, not necessarily because it would provide them with any instrumental benefit but for emotional and expressive reasons. Having chosen an anchor, preferences on most policy issues are derivative. People adopt preferences that conform with their anchors, to reduce cognitive dissonance, and to conserve on information gathering. As Christopher Achen and Larry Bartels say, “We conclude that group and partisan loyalties, not policy preferences or ideologies, are fundamental in democratic politics.”32 Many policy issues are complex, and the general population is not in a good position to do the research to understand the details, so they adopt their preferences by yielding to authority. Their anchors have views on those policies, so voters adopt preferences derived from the views of their anchors. They follow their leaders. In an early study on political preferences, Angus Campbell and coauthors found that most voters did not have stable public policy preferences; rather, they anchored on parties or ideologies and their policy preferences shifted along with those of their anchors.33 Achen and Bartels present more recent evidence supporting this conclusion.34 Public policy issues are complex, there are many issues to consider, and the preferences people hold do not affect instrumental outcomes. Most people will claim that their policy preferences are well reasoned, but will also claim that this is 32 33 34

Achen and Bartels, Democracy for Realists, p. 18. Angus Campbell, Philip E. Converse, Warren E. Miller, and Donald F. Stokes, The American Voter (New York: John Wiley, 1960). Christopher H. Achen and Larry M. Bartels, Democracy for Realists.

https://doi.org/10.1017/9781009323178.007 Published online by Cambridge University Press

116

Anchor Preferences and Derivative Preferences

not true of those who have preferences different from theirs. They should consider that those with different policy preferences think the same way. People adopt political preferences for social and emotional reasons not closely connected to their instrumental preferences, and sometimes clearly at odds with them. Gordon Tullock’s “Charity of the Uncharitable” is a good explanation of why people can feel good about voting for policies they would not choose if the choice were theirs alone.35 They develop an attachment to an anchor, and most political preferences are derived from that anchor. Achen and Bartels conclude that “[p]eople tend to adopt beliefs, attitudes, and values that reinforce and rationalize their partisan loyalties.”36 Their policy preferences are derivative of their anchor preferences. Ultimately, citizens and voters adopt preferences that are sold to them by the political elite. Voting models that begin with the assumption that voters have preferences and that parties and candidates design their platforms to conform with voter preferences are not logically flawed. In theory, voting could work this way. In practice, however, voter preferences are subject to influence for reasons this and previous chapters have described. It is possible, as voting models depict, that voters have preferences and politicians tailor their platforms to conform to them. It is also possible that voters anchor on parties, candidates, and issues, and that most political preferences are derived from those of their anchors. This second possibility more closely corresponds with observed political reality. Everyone is different, and different people can form their political preferences in different ways. But substantial evidence supports the idea that for the most part, people get their political preferences from the political elites on which they anchor rather than the other way around. Rather than candidates and parties adopting platforms that correspond with the preferences of the voters, voters adopt the policy preferences of the candidates and parties on whom they anchor. Political preferences of the masses are derived from those of the political elite. The political elite tell voters what to think, and voters follow their leaders. That raises the question of how the policy preferences of political elites are formed.

35 36

Gordon Tullock, “The Charity of the Uncharitable,” Western Economic Journal 9, no. 4 (December 1971), pp. 379–392. Achen and Bartels, Democracy for Realists, p. 296.

https://doi.org/10.1017/9781009323178.007 Published online by Cambridge University Press

7 Preferences of Elites and Masses

The political preferences of citizens and voters are expressive, not ­instrumental, but the preferences of the political elite – those who actually determine public policy – are instrumental. The political elite determine public policy outcomes, so when members of the political elite express a preference for a particular policy outcome, that increases the likelihood of that outcome occurring. One way to define the political elite is that they are the individuals who design public policy. The relationship between their preferences and political outcomes is not quite as direct as with market choices. The diner who chooses the salad over the pizza gets the salad, but public policy decisions are rarely the product of one individual’s preferences alone. Members of the political elite negotiate with one another to design public policies. The result is often a compromise in which nobody gets everything they want. But individual members of the elite have the power to affect policy outcomes, which distinguishes them from members of the masses whose individual policy preferences have no effect on policy outcomes. The division between elites and masses has been long recognized in the social sciences. Karl Marx saw a division between the bourgeoisie and the proletariat, predating the twentieth-century terminology of elites and masses. In the twenty-first century the Occupy Wall Street movement that began in 2010 divided the population into the 1 percent and the 99 percent. This chapter examines that division to look at its impact on political preferences. If the masses get their public policy preferences from the elite, the next issue is to establish what public policies the elite promote. In his book The Power Elite, C. Wright Mills observes, “The powers of ordinary men are circumscribed by the everyday world in which 117

https://doi.org/10.1017/9781009323178.008 Published online by Cambridge University Press

118

Preferences of Elites and Masses

they live. … But not all men are in this sense ordinary. As the means and ­information and power are centralized, some men come to occupy positions in American society from which they can look down upon., so to speak, and by their decisions mightily affect, the everyday world of ordinary men and women.”1 Mills calls those people the power elite. They write the rules. Ordinary men and women – the masses – follow the rules. The people who actually determine public policy face different incentives when forming their policy preferences than those who have no say in its design. Earlier chapters have discussed in detail how voters’ expressive preferences can be affected by various factors ranging from peer pressure to cognitive dissonance. The policy preferences of the political elite – those who actually influence the design of public policy – are also subject to various influences, but the incentives are different for them because their expressed preferences have instrumental effects. This chapter focuses on the factors that determine the policy preferences of those who actually make public policy. Economists often assume that people make choices to maximize their incomes or their wealth. While people surely are motivated to improve their material well-being, they have other goals as well. According to Maslow’s well-known hierarchy of needs, once people’s basic needs for food, clothing, and shelter are satisfied, material needs become less important.2 They seek safety, and after that, they seek goals that involve their relationships with others. Among those goals, people seek power over others. The preferences elites have over public policy outcomes are motivated by many things, but one factor that always underlies those preferences is the quest for power. The more power they have, the more influence they will have over public policy. If they lose their power, they will lose their elite status and descend to become members of the masses.

7.1  The Quest for Power The desire to exercise power over others is present in most people to varying degrees, but is more present in the political elite because positions in government give those who hold them the ability to use the force of government to coerce others. There is a self-selection mechanism that leads those who have the highest desires to exercise power over others to 1 2

C. Wright Mills, The Power Elite (New York: Oxford University Press, 1956), p. 3. A. H. Maslow, “A Theory of Human Motivation,” Psychological Review 50, no. 4 (July 1943), pp. 370–396.

https://doi.org/10.1017/9781009323178.008 Published online by Cambridge University Press

7.1  The Quest for Power

119

seek positions in government. They have to seek power to get power. The selection mechanism that favors the power-hungry ascending to become the political elite works in two ways. Those who most want to exercise power over others will seek positions that bring with them political power. In addition, those who are most willing to exercise political power – to command people to act in ways they would otherwise choose not to – will be most successful. This is why in politics and government, Friedrich Hayek says that the worst get on top.3 The only reason to have a government is to force people to act in ways they would choose not to if not for the government’s threat of force. If people would voluntarily pay for public goods, there would be no need for government to force people to pay taxes. If people would choose to act in ways specified by government regulations, government would not be necessary to enforce those regulations. The purpose of regulation is either to force people to do things they otherwise would choose not to do, or to prohibit them from doing things they otherwise would choose to do. As all government action is backed by the threat of force against those who do not comply, those who will be most successful are people who are best able to disregard the preferences of individuals they govern, to enforce mandates that constrain people’s actions. The more ruthless the individual, the more successful the individual will be at exercising the powers of government. This is not to say that the exercise of political power is bad for the masses. For example, most people would agree that people should not assault or kill other people, and people should not take the property of others. Some opportunistic people, however, might do those things, and one of the activities of government is to enforce laws against those activities. While not completely successful, it is remarkable that one can walk among strangers in a big city and remain (mostly) safe from predation. Anthropologist Jared Diamond, who studies and has lived within hunter–gatherer societies, notes that in those societies, encounters with strangers are likely to be violent. Strangers are viewed as potential predators, and as potential prey. Modern societies with strong governments, in contrast, enable people to encounter people they do not know on a routine basis, without resorting to violence.4 3 4

The title of Chapter 5 in Friedrich A. Hayek, The Road to Serfdom (Chicago: University of Chicago Press, 1944) is “Why the Worst Get on Top.” Jared Diamond, The World until Yesterday: What Can We Learn from Traditional Societies? (New York: Viking, 2012). See also Steven Pinker, The Better Angels of Our Nature: Why Violence Has Declined (New York: Viking, 2011), for an argument that

https://doi.org/10.1017/9781009323178.008 Published online by Cambridge University Press

120

Preferences of Elites and Masses

People may like everything that their government does, but no matter how much people like their government, everything government does is backed by the threat of force, and those who will be most successful at using the tools of government are those who are least reluctant to use the threat of force against those who do not comply with government’s mandates. People, in general, like to have power over others, and there are selection mechanisms that push the most power-hungry people into government. Bertrand Russell, in his Nobel lecture “What Desires Are Politically Important?” said, “Man differs from other animals in one very important respect, and that is that he has some desires which are, so to speak, infinite, which can never be fully gratified.”5 One of those infinite desires is the desire to exercise power. No matter how much power someone has, the power-seeker always wants more, and indeed, the more power someone has, the more intensely they are inclined to desire more. Russell goes on to say that “a great many men will cheerfully face impoverishment if they can thereby secure complete ruin for their rivals.” Max Weber says that “the career of politics grants a feeling of power. The knowledge of influencing men, of participating in power over them, and above all, the feeling of holding on one’s hands a nerve fiber of historically important events can elevate the professional politician above everyday routine even when he is placed in formally modest positions.”6 John Kenneth Galbraith says, “In all societies, from the most primitive to the ostensibly most civilized, the exercise of power is profoundly enjoyed. … power is pursued not only for the service it renders to personal interests, values, or social perceptions, but also for its own sake, for the emotional and material rewards inherent in its possession and exercise.”7 Power over others comes from many sources. Some people have economic power that derives from resources under their control. Some people have political power that enables them to create binding rules enforced on others. People have religious power based on authority from religious institutions. People have social power that can influence others through persuasion, without coercion or payment. Political power differs from

5 6 7

the development of strong government over the past several hundred years has reduced interpersonal violence in all dimensions, from wars to personal assaults. Bertrand Russell, “What Desires Are Politically Important?” Nobel Lecture, December 11, 1950, found at www.nobelprize.org/prizes/literature/1950/russell/lecture. Max Weber, “Politics as a Vocation,” in H. H. Gerth and C. Wright Mills, eds., From Max Weber: Essays in Sociology (New York: Oxford University Press, 1946), p. 115. John Kenneth Galbraith, The Anatomy of Power (Boston: Houghton Mifflin, 1983), p. 10.

https://doi.org/10.1017/9781009323178.008 Published online by Cambridge University Press

7.2  Economic Power and Political Power

121

other types of power because the institutionalized use of force threatens those who do not comply with it.8 People enjoy having power for its own sake, and there are good reasons to think that those who find careers in government enjoy exercising power more than most people. There is a process of self-selection. Those who enjoy the exercise of power gravitate toward positions where they can exercise it.

7.2  Economic Power and Political Power Some differences between economic power and political power help illuminate what differentiates the political elite from the masses. Economic power comes from control over resources. That control gives people with economic power the ability to have higher standards of living, quite obviously, and also gives them the ability to use their economic power to entice others to exchange with them. Someone who owns a company, or even manages a division within one, has the power to offer people jobs, and if there are many applicants, has the power to decide which among the many will get the jobs. People selling goods and services, if there are few alternatives available, can use their market power to give themselves a bargaining advantage. Along these lines, note the difference between economic power as it is used here – as control over resources – and market power, which is some degree of monopoly or monopsony power. Someone in a rural area might own the only grocery store within fifty miles, giving that grocer substantial market power, but if the store is in a sparsely populated area, the grocer may not have much economic power, and little control over resources. In contrast, companies like General Motors and Toyota have substantial economic power – control over resources – but limited market power, because there are a sufficient number of competitors who exercise countervailing power.9 People in that rural area may have more choices when buying an automobile than when buying their groceries. Regardless of the degree of market power that people have, the only way people can use economic power to gain more is through voluntary exchange with others. The only way the grocer, or the automobile dealer, 8

9

Galbraith, The Anatomy of Power, discusses various types of power and the way they are exercised. The concept of power is discussed in more detail in Randall G. Holcombe, Coordination, Cooperation, and Control: The Evolution of Economic and Political Power (London: Palgrave-Macmillan, 2021), ch. 1. The concept of countervailing power is discussed by John Kenneth Galbraith, American Capitalism: The Concept of Countervailing Power (Boston: Houghton Mifflin, 1952).

https://doi.org/10.1017/9781009323178.008 Published online by Cambridge University Press

122

Preferences of Elites and Masses

can get someone’s money is if they voluntarily agree to tender it. As a result, to use economic power, people must offer others something that looks attractive in exchange. This applies to employees who must convince employers that it would be mutually beneficial to give an employee a job. This applies to businesses who must convince their customers that they would be better off with the company’s products in exchange for payment. People are not forced to enter into economic transactions. Economic power is exercised through voluntary agreement, and those agreements signify that all parties to the exchanges believe they are better off by making their trades. Political power is exercised through the threat of force against those who do not comply. Those in the political elite make the rules, and the masses are forced to comply with them. The stark difference is voluntary agreement versus the threat of force. Strong government may be a good thing – this will be discussed further in Chapter 9 – but regardless, government power comes from the ability to make a credible threat to use force against those who do not comply. In contrast, the exercise of economic power comes from enticing others into exchanges that all parties view as beneficial. The power of the economically powerful is, in this sense, not comparable to the power of the politically powerful. Absent political power, the economically powerful cannot force anyone to do anything. Nobody is forced to buy a product, or to take a job. Those who have political power have the threat of force standing behind their power.10 Another difference between economic and political power is that the exercise of economic power is decentralized, whereas political power is centralized. There is no top-down order placed on a market economy. Rather, individuals decide what exchanges to make in a decentralized manner. When buying a car, some people may decide to transact with General Motors and others may decide to transact with Toyota. Large economic entities might engage in transactions with each other, for example, when General Motors buys steel from U.S. Steel. But small economic units can also transact with large ones, for example, when an individual buys a car from General Motors, or when an individual buys cookware at Walmart. Political power is centralized with a top-down organization. Those who have political power make the rules, and the masses must abide by them. Most individuals are not in a position to be able to bargain with 10

The differences between economic and political power are discussed in more detail in Holcombe, Coordination, Cooperation, and Control.

https://doi.org/10.1017/9781009323178.008 Published online by Cambridge University Press

7.3  Economic Power Can Buy Political Power

123

their governments, as C. Wright Mills noted. The rules are made by those at the top, and those below are forced to obey them. Individuals lower in the hierarchy may be able to exercise substantial discretion in their actions. For example, a building inspector does not write the building code, but has substantial latitude in its enforcement. And that building inspector may even engage in transactions, such as taking bribes, that influence the use of that discretion. Still, those subject to the rules must ultimately abide by them, to the degree that they are enforced.11 When comparing the desire for economic versus political power, Adam Smith said, “The pride of man makes him love to domineer, and nothing mortifies him so much as to be obliged to condescend to persuade his inferiors.”12 Economic power can enhance one’s material well-being, but that quest for power entices people to use political power when it is available to them. Why try to persuade people when the option exists to force them?

7.3  Economic Power Can Buy Political Power People cannot use economic power alone to force others to transact with them, but those with economic power can use it to coerce others when they have the cooperation of those who have political power. The contractor who bribes the building inspector is one example, but other less blatant examples abound. A pharmaceutical company that has a patent on a life-saving drug that someone needs gets power from its patent. That patent is a grant of monopoly power to the pharmaceutical company. Without the patent, competing companies could offer the drug, bringing prices down. The patent enables the drug company to use the force of government to keep competitors out of its market. The mandate that motor fuels in the United States contain ethanol is another example. Ethanol comes mostly from corn, and corn farmers and corn processors were able to lobby Congress to mandate that people who want to buy gasoline for their cars also must buy ethanol mixed in with it. Surely, most motorists would choose not to buy gasoline that contains 11

12

Richard E. Wagner, Politics as a Peculiar Business: Public Choice in a System of Entangled Political Economy (Cheltenham, UK: Edward Elgar, 2016), suggests that there is more spontaneous order in the top-down order of government than is commonly ­recognized. This is likely true, but does not negate the fact that ultimately those who hold government power can back their mandates by force. Adam Smith, The Wealth of Nations (New York: Modern Library, 1937 [orig. 1776]), p. 365.

https://doi.org/10.1017/9781009323178.008 Published online by Cambridge University Press

124

Preferences of Elites and Masses

ethanol if given the choice. Otherwise, there would be no reason to force them. The corn lobby was able to use its economic power to buy political power to further its agenda. People with economic power are able to force others into transactions only when political power stands behind those forced transactions. Absent political power, nobody is forced to buy anything from anyone, sell anything to anyone, employ anyone, or be employed by anyone. This relationship between economic power and political power is well known, and a substantial academic literature has analyzed the effects of cronyism and corruption. Those with economic power often lobby government to provide them with tax breaks, subsidies, trade barriers, and regulatory barriers to hinder competitors from entering their markets. Firms with economic power can offer those with political power campaign contributions, jobs for friends and family, and other benefits to legislators and regulators in exchange for favorable treatment.13 Regulated firms are able to use their economic power to shape regulations for their benefit, often imposing costs on the general public.14 The general public typically knows nothing about the regulatory process, while regulated firms know the process well and have substantial financial incentives to lobby regulators for favorable treatment. A revolving door of employment, where employees of regulated firms move to positions in regulatory agencies and employees of regulatory agencies get jobs at the firms they once regulated, establishes personal connections that help them obtain favorable treatment. More generally, concentrated special interests have advantages in organizing when compared to larger and more diffuse groups, so industry lobbying groups, large firms, and organized special interest groups tend to be more successful in using their economic power to buy political power. Meanwhile, the general public is not well organized and not knowledgeable about most legislative and regulatory activity, giving concentrated interests the ability to influence public policy to benefit themselves.15

13

14

15

A substantial literature on this rent-seeking activity was initiated by Gordon Tullock, “The Welfare Cost of Tariffs, Monopolies, and Theft,” Western Economic Journal 5 (June 1967), pp. 224–232, and Anne O. Krueger, “The Political Economy of the RentSeeking Society,” American Economic Review 64 (June 1974), pp. 291–303. This idea of regulatory capture was explained by George J. Stigler, “The Theory of Economic Regulation,” Bell Journal of Economics and Management Science 2, no. 1 (Spring 1971), pp. 3–21. This is explained by Mancur Olson, Jr., The Logic of Collective Action” Public Goods and the Theory of Groups (Cambridge, MA: Harvard University Press, 1965.).

https://doi.org/10.1017/9781009323178.008 Published online by Cambridge University Press

7.4  Political Power Can Extort Economic Power

125

Seeing those with economic power using it to buy political favoritism, it is easy to be critical of this abuse of power. However, one must recognize that it is not the economic power that is being abused, but the political power that has been bought with it. Economic power can be used only if people on both sides of transactions agree that it is in their mutual interest. Economic power gives those who have it no authority over the actions of others. Political power does convey that authority. When those with economic power use it to buy political power, it is the political power that is being used to force people to make transactions they otherwise would not choose to make. Nobel laureate Joseph Stiglitz stated the issue well, saying, “It’s one thing to win a fair game. It’s quite another to be able to write the rules of the game – and write them in ways that enhance one’s prospects of winning. And it’s even worse when you can choose your own referees.”16 Stiglitz’s complaint is about the cronyism that enables those with economic power to buy the favors of those who have political power. The issue Stiglitz raises is only indirectly about economic power. The real problem is the ability to use it to buy favorable rules, and to buy the referees who enforce them. The same creative destruction of the market mechanism that allows innovative entrepreneurs to amass economic power also threatens those who have succeeded in acquiring it.17 To protect themselves from the creative destruction of capitalism, those with economic power negotiate with those who have political power to create barriers to entry against competitors, and to provide them with other advantages. The evolutionary nature of a market economy can displace those who have acquired economic power, providing them with the incentive to use some of their economic power to buy political power to solidify their positions in the economic hierarchy.

7.4  Political Power Can Extort Economic Power The political elite do not volunteer to legislate and regulate for the benefit of the economic elite. They expect to receive something in return. Because the political elite have the force of government behind them, they are able 16 17

Joseph E. Stiglitz, The Price of Inequality: How Today’s Divided Society Endangers the Future (New York: W.W. Norton, 2012), p. 59. Creative destruction is the phrase Joseph A. Schumpeter, Capitalism, Socialism, and Democracy (New York: Harper & Brothers, 1942), used to describe the evolutionary nature of capitalism.

https://doi.org/10.1017/9781009323178.008 Published online by Cambridge University Press

126

Preferences of Elites and Masses

to extract payment from the economic elite in exchange for favorable treatment. Peter Schweizer observes that people often look at the lobbying process as bribery – lobbyists paying off legislators and regulators in exchange for favorable treatment. Schweizer says the process is more like extortion.18 If lobbyists ask for legislation favorable to their clients, legislators will hold up that legislation until payment is made. But political power often is exercised in other ways. Legislators often threaten to pass legislation that will harm specific businesses, solely for the purpose of extracting payments from them in exchange for killing the legislation. If lobbyists do not pay up, bad things happen to them. Schweizer offers many examples, but a typical scenario plays out like this. A lobbyist makes a request to a legislator, either to support legislation that would benefit their clients or to kill proposed legislation that is unfavorable to the lobbyist’s clients. The legislator agrees to consider the lobbyist’s request, and also mentions to the lobbyist that the legislator’s political action committee is having a reception in a few days and he hopes the lobbyist can attend. The lobbyist knows that to get favorable legislative treatment, attendance is mandatory, and at the reception there is a bowl in the center of the room in which the lobbyist is expected to deposit a check as a “contribution.” The procedure is not written down anywhere or formally stated, but everyone knows how the system works. Depositing the check in the bowl facilitates favorable treatment. No check, no favorable treatment. The possibilities for benefiting those who hold political power are obvious. Legislators make the rules and regulators enforce them. Unlike economic power, which is exercised through voluntary exchange, there is no escaping the exercise of political power. Those who are reluctant to buy political favors will not be favored, while those who are willing to do so get favorable treatment. The greater state oversight is over economic activity, the more the profitability of firms depends on political connections rather than producing value for consumers. Firms have little alternative to participating, because refusing to play that game puts them at a disadvantage to their rivals. Schweizer describes some legislative proposals that are referred to as milker bills. They are proposed not because legislators want to pass them, but because interest groups will be willing to pay up to kill them. The intention of milker bills is not to pass legislation, but to coerce 18

Peter Schweizer, Extortion: How Politicians Extract Your Money, Buy Votes, and Line Their Own Pockets (Boston: Houghton Mifflin, 2013).

https://doi.org/10.1017/9781009323178.008 Published online by Cambridge University Press

7.5  The Discontinuity in Political Power

127

payments from lobbyists who do not want their clients to be harmed by the contents of the bills. Those who have political power have the force of government standing behind them, and the quest for power means that when possible, they will use their power to solidify their hold on it, and to gain more of it. Those with political power can enhance their power by using it to extract payment from those who have economic power.19 The distinction between economic power and political power often is not made by those who criticize the abuse of power. Economic power, by itself, can only be used to persuade others to engage in voluntary transactions. It does not give people the authority to coerce others. Political power forces others to comply with it. People who have economic power often use it to buy access to political power to create barriers to entry against potential competitors, solidifying their positions in the economic hierarchy. When those with economic power bargain with those who have political power, they provide more power to the holders of political power. Those with political power can demand continued payment in exchange for continued protection. The ever-closer relationships that develop over time between the economic and political elite tend to undermine market institutions and lead to economic stagnation, but the elite pursue those relationships because doing so solidifies and enhances their hold on power.20

7.5  The Discontinuity in Political Power The decentralized nature of economic power means that there is a continuous distribution of economic power between those who have little and those who have a lot. Some people have more economic power than others, but when you go to Starbucks, your $10 has just as much buying power as when Jeff Bezos brings his $10 into Starbucks. Sure, Mr. Bezos

19

20

Fred S. McChesney has described the process in “Rent Extraction and Rent Creation in the Economic Theory of Regulation,” Journal of Legal Studies 16, no. 1 (January 1987), pp. 101–118 and Money for Nothing: Politicians, Rent Extraction, and Political Extortion (Cambridge, MA: Harvard University Press, 1997). I have also discussed these institutional arrangements in Randall G. Holcombe, Political Capitalism: How Economic and Political Power Is Made and Maintained (Cambridge: Cambridge University Press, 2018). This process of increasing cooperation between the economic and political elite is discussed by Mancur Olson, Jr., The Rise and Decline of Nations (New Haven, CT: Yale University Press, 1982), and Randall G. Holcombe, Political Capitalism.

https://doi.org/10.1017/9781009323178.008 Published online by Cambridge University Press

128

Preferences of Elites and Masses

has many more $10 bills than you do, but that $10 has the same buying power whether it is in your hands or his. People who have little economic power have ways to get more. Someone with a low-paying job can work overtime, or take a second job, and get command over more resources. People can build their skills and with more human capital get a better job and gain more economic power. The distribution of economic power is continuous. Someone with $20 has twice the economic power as someone with $10. Someone with $10 million has ten times the economic power as someone with $1 million. Because of the decentralized nature of economic exchange, people with little economic power are able to transact with those who have a lot of economic power. Ordinary citizens are able to strike mutually agreeable bargains with economic giants like General Motors and Walmart. The same is not true of political power. There is a discontinuity in political power that prevents those with little political power from gaining more. Someone with little economic power can work overtime or take a second job and get more economic power. There is nothing ordinary people can do to gain more political power. Someone with little political power can donate money to a political party or candidate, or can volunteer to work for a party or candidate, but doing so only gives that party or candidate a bit more power. The individual still has no bargaining power. One can contribute to an interest group like the Nature Conservancy or the National Rifle Association, but the individual contributor has no say in what those organizations do and gains no political power from donating. The leaders of the organizations gain political power from the sum total of their donors and supporters, but the individuals have no more power than if they had not donated or volunteered.21 Some people have a lot of political power; most people have none. There is a discontinuity between those two groups. The difference that creates a continuous distribution of economic power but a discontinuity in political power is the bottom-up versus top-down nature of economic versus political institutions. Economic institutions are designed so that it is easy for individuals to engage in mutually advantageous exchange with each other. Walmart is a big company, but markets are designed so that they can deal with millions 21

For arguments that slant somewhat in the other direction, see Donald A. Wittman, “Why Democracies Profuse Efficient Results,” Journal of Political Economy 97 (1989), pp. 1395–1424, and Wittman, The Myth of Democratic Failure (Chicago: University of Chicago Press, 1995). Wittman does not argue that individuals have political power, but that institutional mechanisms like interest groups give them political power as a group.

https://doi.org/10.1017/9781009323178.008 Published online by Cambridge University Press

7.6  Transaction Costs and Public Policy

129

of economic entities every day, most of whom have much less economic power than Walmart. Political institutions are designed so that a few people make the rules that most people are obligated to follow. This is the nature of public policy formation, in which rules are made that apply to (almost) everyone. When one person goes to Walmart to buy a microwave oven, that transaction is between the buyer and Walmart. Others are not affected. Others are not involved, but they do not need to be involved, because the transaction applies only to the parties who are engaging in it. When those with political power pass laws, they apply to many others who have no part in approving those laws and no say in determining their contents. Institutions are designed to facilitate many bilateral transactions in the decentralized marketplace. The same is not true in the centralized topdown organization of government, where a few at the top make public policy while the many at the bottom must comply with the policies created by the political elite. While one can talk about an economic elite, there is not a clear dividing line between those who are in it and those who are not. People who would not qualify as economic elite transact with the elite on a regular basis, creating a continuum of economic power. The same is not true of the political elite. Some people are in it, and have the power to affect the direction of public policy; most people are not. There are very few who stand in some ambiguous middle ground. Those who are not members of the political elite, who have no political power, are the masses whose public policy preferences are derived from the elites on whom they anchor.

7.6  Transaction Costs and Public Policy The factor that separates the political elite from the masses is transaction costs. A transaction cost is anything that stands in the way of a mutually beneficial exchange. When transaction costs are low, people are able to bargain with each other for their mutual advantage. When transaction costs are high, they prevent people from bargaining with each other even when there are exchanges that would benefit them in the absence of transaction costs. The importance of low transaction costs to facilitating mutually advantageous exchange, and of high transaction costs standing in the way of mutually advantageous exchange, was pointed out by Ronald Coase in an article he wrote about externalities: costs that are imposed

https://doi.org/10.1017/9781009323178.008 Published online by Cambridge University Press

130

Preferences of Elites and Masses

on some people by the activities of others.22 One lesson from Coase’s article is commonly called the Coase theorem, and in one of its many variants can be stated as “In the absence of transaction costs, resources are allocated to their highest-valued uses.”23 The common sense of the Coase theorem is that if there are no impediments to mutually advantageous exchange, the person to whom any resource is worth the most will buy it. Anything that prevents such an exchange from occurring is a transaction cost. People who face low transaction costs can engage in exchange to maximize the value of resources to those in that low transaction cost group. Those who face high transaction costs are prevented from bargaining, so for them, resources may not be allocated to their highest-valued uses. To illustrate the importance of transaction costs in differentiating the political elite from the masses, consider the parallel between political transactions and market transactions. A standard example of an externality in market activity is a smokepolluting industry that generates air pollution that those around the polluters must breathe. Consider a steel town where a group of steel mills are producing steel, and as a by-product – the externality – they are also producing air pollution that imposes costs on the surrounding residents who breathe the polluted air. The steel mills are selling steel to automobile manufacturers (and others), who are able to strike bargains with each other because they face low transaction costs. Markets make it easy for those who want to buy steel to do so from those who want to sell steel. They face low transaction costs and can maximize the value of resources to themselves. The tens of thousands of residents in the vicinity of the mills might be willing to bargain with steel mills to reduce their pollution if it were possible, but fashioning a bargain that would require tens of thousands of people to participate would be difficult. Transaction costs are high enough to prevent a bargain from being struck. As a result, the surrounding residents bear the external cost of breathing polluted air. The value of resources is maximized for those who face low transaction costs – the steel mills and automobile manufacturers – but not for those who face high transaction costs – the people who breathe the polluted air. 22 23

Ronald H. Coase, “The Problem of Social Cost,” Journal of Law & Economics 3 (1960), pp. 1–44. The Coase theorem has many different and sometimes conflicting interpretations. See Steven G. Medema, “The Coase Theorem at Sixty,” Journal of Economic Literature 58, no. 4 (December 2020), pp. 1045–1128, for a discussion. The theorem as stated here will be the basis for the discussion that follows.

https://doi.org/10.1017/9781009323178.008 Published online by Cambridge University Press

7.6  Transaction Costs and Public Policy

131

That same idea applies to the production of public policies. The e­ conomic and political elite face low transaction costs that enable them to bargain with one another to produce public policies. Those people in the low transaction cost group are legislators, well-connected lobbyists, and others who have connections that enable them to negotiate to produce public policies. Most people face high transaction costs, and so like those who breathe the polluted air around steel mills, bear the costs of the bargains made by those who face low transaction costs: the political elite. The political elite are those who face low transaction costs. The masses face high transaction costs, so are unable to enter the bargaining process that creates public policy. The political elite are not trying to impose costs on the masses, just as steel mills are not trying to impose costs on the surrounding residents. But when those in the low transaction cost group design public policies that maximize the benefits to themselves, sometimes as a by-product, costs are imposed on others who are unable to enter the bargaining process because they face high transaction costs. The mandate that motor fuels contain ethanol is a good example. The corn lobby was able to negotiate with legislators to impose the mandate because lobbyists and legislators face low transaction costs. Those buying motor fuels bear a cost from the mandate, not because anyone wanted to impose costs on them, but as a by-product of a bargain that was designed to produce benefits for lobbyists and legislators. The decentralized nature of markets makes transaction costs low for most people. Anyone can enter a Walmart, or a Toyota dealership, and make mutually advantageous exchanges that benefit the buyer and seller. The centralized nature of political decision-making creates high transaction costs for most people. Thousands of people cannot simultaneously participate in designing public policy, so by necessity public policies will be designed by a few individuals who face low transaction costs and can bargain with each other. Those legislators, lobbyists, and others who are well connected can bargain to create public policies that maximize the value of those policies to themselves, while most people are unable to enter those bargains. People in the low transaction cost group are the political elite who design public policy. People in the high transaction cost group are unable to enter the negotiating process. Transaction costs separate the political elite from the masses. The discontinuity in political power is the result of transaction costs. The masses are in the high transaction cost group and have no political power. The political elite are those who are in the low transaction cost

https://doi.org/10.1017/9781009323178.008 Published online by Cambridge University Press

132

Preferences of Elites and Masses

group and participate in the bargaining to create public policy. Those policies created by the political elite extend their reach of power because they can continue to extort payment from beneficiaries, under threat of reversing the policies. For example, the ethanol mandate that provides economic benefits to corn farmers and processors was enacted by the political elite and can be repealed by them. As a result, the political elite are in a position to demand that those beneficiaries continue to pay up, or the mandate could be reversed. This furthers the goal of the political elite to maintain and extend their hold on power.

7.7  Controlling the Agenda Political scientists and economists have long understood that those who have control over the political agenda are able to have considerable control – and perhaps even completely determine – political outcomes. In a democratic setting, control over the agenda entails controlling the alternatives voters are offered, and the order in which those alternatives are brought up for a vote. To see the power of agenda control, consider some simple illustrative examples. Some organizations have nominating committees that determine candidates for positions as officers of the organizations. By choosing the candidates that members can vote for, the nominating committee has substantial power over determining who will hold those positions. In some organizations – the American Economic Association and the Southern Economic Association are two examples – the nominating committee nominates only one candidate for the president of the organization. Members are allowed to write in the names of other choices, but unsurprisingly, the sole candidate chosen by the nominating committee is always elected president of the organization. It appears that members elect their officers, but in fact they are chosen by a small nominating committee. Another example of agenda control is a tactic Arthur Burns used when he was chair of the Board of Governors of the Federal Reserve System from 1970 to 1978. The Board is composed of seven members who vote on the Bank’s policies. When items came up for a vote, Burns would explain his position on them, and then go around the table asking the other members, “If we were to vote on this item now, how would you vote?” If a majority said they would vote against him, he would continue the discussion, explaining his positions again, and ask the same question. This would continue until a majority said they would vote with Burns, at which time Burns, as chair, would say, “Okay, we are now going to

https://doi.org/10.1017/9781009323178.008 Published online by Cambridge University Press

7.7  Controlling the Agenda

133

vote.” By postponing any vote until Burns knew the votes would go his way, he effectively controlled the outcomes of all votes because as chair, he controlled the agenda.24 The same thing happens in legislatures and in legislative committees. Committee chairs will check with members and count votes before an actual vote takes place, which allows political leaders the opportunity to negotiate for votes to win an issue, prior to any vote being taken. Negotiations to gain votes can occur because members in the legislature can bargain with each other with low transaction costs. Control over the agenda is one factor that gives the speaker of the US House of Representatives and the president of the Senate their power. Academic study of agenda control shows that its power is much deeper than these simple examples would suggest. When voters are offered candidates for office, or ballot initiatives to determine public policy, the ability to determine the alternatives offered to voters has a substantial influence on what alternative eventually is selected.25 Under certain assumptions, someone who controls the agenda can offer a sequence of alternatives to voters such that no matter what the status quo when voting begins, the agenda controller can get the exact outcome he or she prefers.26 The primary system gives voters the illusion that they have ultimate control over who runs for, and who wins, political office. But voters are only able to vote for candidates who are on the ballot, and party leaders often negotiate with candidates to put them on, or off, the ballot to affect the election. In the 2020 Democratic primary, for example, Bernie Sanders had substantial support, but party leaders believed he would fare poorly as a candidate against incumbent Donald Trump. They preferred the more moderate Joe Biden, and convinced Biden’s closest competitors to drop out of the primary elections to give Biden more support, and draw support away from Sanders. Biden ultimately won and paid off those who helped him. Kamala Harris was chosen as his vice presidential candidate and Pete Buttigieg was appointed secretary of transportation. 24 25

26

I do not know if this story is completely accurate because the Board’s minutes are not complete transcripts, but it is a good example. This idea was developed by Thomas Romer and Howard Rosenthal, “Political Resource Allocation, Controlled Agendas, and the Status Quo,” Public Choice 33, no. 4 (1978), pp. 27–43. This idea is explained in mathematical terms by R. D. McKelvey, “Intransitivities in Multi-Dimensional Voting Models and Some Implications for Agenda Control,” Journal of Economic Theory 12, no. 3 (June 1976), pp. 472–482. A more accessible explanation is found in Randall G. Holcombe, An Economic Analysis of Democracy (Carbondale: Southern Illinois University Press, 1985), ch. 4.

https://doi.org/10.1017/9781009323178.008 Published online by Cambridge University Press

134

Preferences of Elites and Masses

In ways that are often the result of behind-the-scenes negotiations, the choices offered to voters are determined by the political elite.27 Voters see an electoral process in which they, the voters, determine the outcome. They do not see the behind-the-scenes negotiations in which the political elite control the agenda to limit the options offered to the masses. The elite have more influence over the choices offered to the masses than is commonly recognized. As noted in Chapter 4, academic analyses of political processes take the political preferences of the masses as given and undertake their analyses from there. The power to control the agenda is much greater when the political elite are able to shape the political preferences of the masses to conform with their own, and there are good reasons to think that this is the case. Voters express their preferences at the ballot box, unaware of the degree to which they have derived those preferences from those of their anchors, who control the agenda. How do citizens decide what policy positions they should favor when there are many complex policy issues? They get their information from the political elite. When an organization’s nominating committee offers its members only one candidate for the office of president, the influence of the agenda setter is obvious. In more complex settings, the power of the agenda setter remains even though it is less obvious to those who are subject to that power. The masses have no incentive to look for that exercise of power, because they have no ability to control its use. The elite have every incentive to downplay it, to create the illusion that democratic institutions give control of government to the people.

7.8  Elite Influence on the Preferences of the Masses The previous chapters have explained why the political preferences of most people are purely expressive. The political preferences of individual citizens have no impact on political outcomes, so the utility from forming 27

This is not always the case. In 2016 Florida governor Charlie Crist decided to run for a seat in the US Senate, and was the choice of the Republican party elite. Marco Rubio, outgoing speaker of the Florida House of Representatives, also decided to run for that Senate seat. The Republican elite tried to convince Rubio not to run, to leave a clear path for Crist to win the seat, but Rubio stayed in the race, won the Republican primary, and won the general election to become a senator. The political elite have much control, but not complete control. Also, note that Rubio, as a former speaker of the Florida House, was also a member of the political elite.

https://doi.org/10.1017/9781009323178.008 Published online by Cambridge University Press

7.8  Elite Influence on the Preferences of the Masses

135

and acting on their political preferences comes solely from their having them and expressing them. If people undertook political action only to have an effect on political outcomes, their rational response would be to do nothing. Whether voting, volunteering on campaigns, or making political contributions to candidates and parties, their individual actions have no effect on political outcomes. While it is true that all the votes taken together determine election outcomes and that all the contributions taken together can have an effect on outcomes, each individual’s contribution to that effort, whether through voting or contributing financially or volunteering, will not have a perceptible effect. That means, to put it plainly, that people adopt and act on political preferences because doing so makes them feel good. Because their preferences have no instrumental effect, citizens tend to be rationally ignorant regarding the facts about candidates, platforms, and policies, and the political preferences that make them feel good are those that minimize cognitive dissonance and disagreements among those with whom they interact the most. Recognizing that the demand for accurate and detailed information on the part of citizens is low, parties and candidates provide very little information of this type. Platforms are deliberately vague to broaden their appeal. Citizens will find little to disagree with in a vague platform. Meanwhile, because citizens adopt political preferences that make them feel good, parties and candidates try to deliver feel-good messages. Citizens anchor mostly on parties and candidates, so parties and candidates try to project an image of competence, of promoting the interests of the masses, and of being good people who can be trusted to exercise the power of government responsibly. Public policy issues fall to secondary importance, because everybody likes competent, responsible leaders who promote their interests, but not everybody agrees on what public policy measures would be appropriate. Citizens buy into the image of their anchors, and their public policy preferences are derivative of those anchors. While the elite want to be cautious about expressing specific policy preferences, they have good reason to be informed about policy alternatives and the expected consequences of implementing public policies. The elite determine public policy, so their preferences are instrumental. The preferences they express are vague and oriented toward appealing to citizen feelings and emotions, but the preferences they act on are specific and well informed. Citizens buy into the policy preferences of the elite not necessarily because they agree with them or even know what they are, but because they feel good about supporting their anchors, so they

https://doi.org/10.1017/9781009323178.008 Published online by Cambridge University Press

136

Preferences of Elites and Masses

support their anchors’ policy preferences. Often, they have only the vaguest idea about what they are supporting. On complex issues, it is easy for voters to buy into the preferences of someone who has gained their trust. That trust is reinforced through the electoral process in which citizens buy into the idea that political power is legitimately held by those who have obtained it through a democratic process. Leaders have the legitimate right to make public policy, because the voters chose them to do so. Citizens who support the process have some reason to support the outcome, even if their most preferred candidate did not win. This is even more true for citizens who have voted for those political leaders. They anchor on those leaders, and minimizing cognitive dissonance, their political preferences mirror those of the leaders. The views of individual citizens have no effect on policy outcomes anyway, and it is rare for people to admit they made a mistake in voting for some candidate or policy. Why should they? As Bryan Caplan says, the design of political institutions gives voters an opportunity to be rationally irrational, at no cost to themselves.28 Chapter 3 described how people’s preferences are subject to outside influences such as advertising and peer pressure. This is true for their instrumental preferences as well as their expressive preferences. Firms spend money to advertise their products because advertising is effective at altering people’s preferences. Chapter 5 explained why public policy preferences, as expressive preferences, are even more susceptible to outside influences. Chapter 6 explained that people’s political preferences tend to anchor on a party or candidate, or sometimes an ideology or even a public policy, and that citizen preferences on most issues derive from the preferences their anchors express on those issues. The masses get their political preferences by adopting those held by the political elite. This stands in contrast to the way that social scientists have analyzed the operation of democratic political institutions, as Chapter 4 explained. Social scientists tend to assume that people have political preferences and that candidates and parties shape their platforms to conform with the political preferences of the masses. The process actually works the other way around. The masses adopt the policy preferences of the elites to whom they anchor. Elites have instrumental policy preferences, and the masses express support for the bundle of preferences promoted by their elite anchors. 28

Bryan Caplan, The Myth of the Rational Voter: Why Democracies Choose Bad Policies (Princeton, NJ: Princeton University Press, 2007).

https://doi.org/10.1017/9781009323178.008 Published online by Cambridge University Press

7.9  Elite Objectives

137

7.9  Elite Objectives This goes partway toward understanding the formation of public policy. If the masses adopt the public policy preferences of the elite, the next step is understanding what determines the policy preferences of the elite. To begin to understand elite policy preferences, note that most public policies do not affect the elite directly. At the federal level, the political elite design a nation’s military policy and foreign policy, which has little direct effect on the elite. Social Security and health care programs make up a large share of national budgets, but the elite have their own pension and health care services and do not rely on the programs they design for the masses. The elite do not collect welfare, and are not bound by the regulations they impose on everyone else. The same is true at the state and local levels of government. Courts, law enforcement, health care programs, and most of what those governments provide do not affect the elite directly. They want personal security, but they tend to live and work in safe areas, they (mostly) do not spend time in prison, and are not recipients of the welfare programs their governments oversee. The political elite have policy preferences, but for the most part, they are not instrumental in the sense that the policies they design will have a direct effect on them. Perhaps the largest direct effect comes through the political marketplace, in which demanders of public policy outcomes lobby to get favorable treatment. As Peter Schweizer notes, this gives them an opportunity for personal gain through obtaining the financial and political support of lobbyists and their clients.29 For the most part, the personal instrumental gains that the political elite receive come from payoffs and support they get from furthering the political objectives of others. To maximize their direct gain, they should auction off policy outcomes to the highest bidders. The specifics of those policies are of secondary importance. Nobel laureate Gary Becker has depicted the legislative process in just this way.30 He describes a legislature as a political marketplace in which competing interests bid against each other to get the public policy outcomes they want. Becker takes a positive view of this activity, suggesting that lobbyists are willing to bid more for policies they value more highly, leading high-valued policies to be produced. This would be more likely to be true if those policies did not impose costs on the masses, who are 29 30

Peter Schweizer, Extortion, cited earlier. Gary S. Becker, “A Theory of Competition among Pressure Groups for Political Influence,” Quarterly Journal of Economics 98, no. 3 (August 1983), pp. 371–400.

https://doi.org/10.1017/9781009323178.008 Published online by Cambridge University Press

138

Preferences of Elites and Masses

excluded from being able to bargain. Nevertheless, this depiction of the public policy process suggests that the policy preferences of the political elite are those that can gain them the highest personal payoffs. Ultimately, the political elite want public policies that give them power. At the most basic level, this is what gives them their elite status. The political elite can continue in their positions only if they retain them. For elected officials, the reelection motive must rank high, because unless they can retain their offices, or move up to higher ones, they will lose their ability to create public policy. Even the most public-spirited elected officials can act in the public interest only if they retain their offices. The same is true of appointed officials. Their first goal is to retain and increase their power, because everything else comes because of the power they hold. This chapter began by discussing the quest for power and made the argument that people want power for its own sake in addition to what they can accomplish because of it. Government positions, by their nature, attract people who have the greatest desire to exercise power over others. Power is valued for its own sake, but also has instrumental value. Human motivations are complex, but if the public policy preferences of the political elite are distilled into a simple framework, the policies they prefer are the policies that give them power.

https://doi.org/10.1017/9781009323178.008 Published online by Cambridge University Press

8 Policies That Maximize Political Power

Surely the political elite have some policy preferences themselves – some ideological foundation upon which they build their policy proposals, some concept of the public interest. The ideological preferences of Margaret Thatcher and Ronald Reagan were apparent, for example. But the previous chapter offered good reasons why the political elite will prefer public policies that give them power, regardless of the specific policies. Policies that give the political elite more power have instrumental value to them. Policies that dilute their power work against their interests. The political elite, like everyone, look out for their personal interests, and their personal interests lie with increasing their power. To understand their motivations, one will not be led far astray by assuming that the political elite are power maximizers. The specific policies that give them power are of secondary importance to them. A more public-spirited way to look at this is that the social role of the political elite is to carry out the public interest – to implement the policies that further the interests of the citizens. They can do that only if they retain the power they have, and they can do even more good if they have even more power. Thus, the first goal of even the most public-spirited of politicians must be to retain and increase their power, in order to retain and increase their ability to further the public interest. If competing politicians are more self-centered and self-confident in their policy stances, it would be even more important for those who have power to retain it, to prevent their being displaced by people who do not have the same vision of the public interest. Whether one believes that the political elite are self-interested or public-spirited, they have strong incentives to seek power. 139

https://doi.org/10.1017/9781009323178.009 Published online by Cambridge University Press

140

Policies That Maximize Political Power

Most public policies have little instrumental effect on the political elite anyway, except to the degree that those policies give them power. They are making decisions that affect others for the most part, and that have only a small direct effect on themselves. This provides a public interest justification for the political elite to cater to the policies their constituents demand, as long as those policies do not limit the power of the political elite. The biggest direct effect on the elite their policy decisions have is that they provide the opportunity to solicit campaign contributions and other political favors from lobbyists and rent-seekers, and may give the political elite inside information that enables them to profit from investments.1 For the most part, the political elite create public policies that constrain the masses, but not themselves. The elite live under a different set of rules. Fortunately for the political elite, constituents are likely to demand policies that increase the power of the political elite. People want the elite to enact policies that give them benefits at the expense of others, which requires that others be forced to comply with those policies. Public policy demanders are, to use the terminology of economists, rent-seekers, and policies that provide rents, which might better be referred to as coerced transfers, require the force of government to stand behind them – to enforce them The political elite do not interact with the masses when they negotiate to produce public policy. They interact with the economic elite and their lobbyists, who are well connected because they face low transaction costs so are able to bargain for specific public policies. In practice, this means selling public policies to the highest bidder.2 Doing so provides resources – money and votes – to the political elite, enabling them to maintain and gain power. Rent-seekers are, in the short run, grateful to the political elite for the rents they receive, and provide political support in exchange. In the long run, the creation of an ongoing flow of rents creates an ongoing dependency of rent recipients on the political elite that keeps those rents flowing. This is a major source of elite power that comes out of the public policy process. If the rent recipients do not continue to buy the favors of the political elite, 1

2

See Peter Schweizer, Extortion: How Politicians Extract Your Money, Buy Votes, and Line Their Own Pockets (Boston: Houghton Mifflin, 2013), for some evidence and Randall G. Holcombe, “Political Incentives for Rent Creation,” Constitutional Political Economy 28, no. 1 (March 2017), 62–78, for a further discussion. This may be a way to discover which policies best promote the public interest, according to Gary S. Becker, “A Theory of Competition among Pressure Groups for Political Influence,” Quarterly Journal of Economics 98, no. 3 (August 1983), pp. 371–400.

https://doi.org/10.1017/9781009323178.009 Published online by Cambridge University Press

8.1  Principles and Politics

141

they risk losing their flow of rents. The greater the dependence of people on government policies, the more power that gives the political elite. This chapter discusses how the political elite can act to maximize their power, and what public policies will be produced as a result.

8.1  Principles and Politics A lack of principles is an asset in the game of politics, because it enables politicians to negotiate to get outcomes favorable to them; that is, outcomes that give them more power. Principled politicians have little room for negotiation. If a policy is proposed that conforms with the principled politician’s ideals, the politician will support it. If a policy is proposed that is at odds with the principled politician’s ideals, the politician will oppose it. There is no need to try to buy the vote of a principled politician on an issue that the politician supports on principle. The principled politician will support it regardless. And there is no point in trying to buy the principled politician’s vote on an issue that is at odds with the politician’s principles. The principled politician will oppose it on principle and cannot be bought. Unprincipled politicians have to be paid for their votes regardless of their underlying policy preferences. They can be paid to support policies they would otherwise oppose. Even for policies an unprincipled politician would prefer, the politician can be paid off to vote the other way. So, unprincipled politicians must be paid to vote for even those policies they would favor if they acted on principle. Otherwise, the other side can make an offer to get the unprincipled politician to vote the other way. Unprincipled politicians sell their support to the highest bidder, not necessarily for money, but often to accumulate favors to build their political capital. This is one way the politically powerful amass their power. Principles and politics are like oil and water.3 They do not mix. A successful politician must engage in political negotiations to accomplish anything. Democratic institutions require majority support, so politicians cannot accomplish anything alone. They must negotiate to gain the support of others to further their public policy initiatives. The requirement that they be willing to bargain implies a willingness to trade their votes on issues that are less important to them to gain votes for issues they value more. Voting on principle stands in the way of engaging in political exchange. 3

Randall G. Holcombe, “Principles and Politics: Like Oil and Water,” Review of Austrian Economics 22, no. 2 (June 2009), pp. 151–157.

https://doi.org/10.1017/9781009323178.009 Published online by Cambridge University Press

142

Policies That Maximize Political Power

The previous chapter noted that government positions attract people who have a greater desire to exercise power over others. Added to that, those who are less principled have an advantage in politics, because their principles do not stand in the way of negotiating to get an outcome favorable to themselves. Those personal characteristics – a lust for power and the absence of principles – enable members of the political class to respond more effectively to the incentives inherent in political institutions.

8.2  The Exercise of Power in Markets and in Politics Most people cannot rise to the status of the political elite. The political elite must be a small subset of society because transaction costs prevent large numbers of people from negotiating with each other to design public policy. The idea that democratic political institutions allow everybody to have a say in the policies enacted by their governments is a myth. Citizens and voters recognize this, which is why they tend to be rationally ignorant about public policy, and why their expressive votes do not necessarily represent their instrumental interests. Successful individuals, in politics and elsewhere, tend to be articulate, quick-thinking, and persuasive. Effective leaders project a sense that they care about those they lead, so followers want to follow. To succeed in politics, people must also be ruthless enough to be willing to sacrifice the interests of some to accomplish their own goals. Political power ultimately is based on the threat of force. Those with economic power must persuade others to cooperate with them. Those with political power can force others to obey their commands. However, the most successful political leaders are able to persuade people to prefer the policies that coerce them. Actual force is a last resort. People tend to follow their leaders. This willingness to use political power despite the objections of those who are subject to it stands in contrast with the requirements to obtain and use economic power. Some who hold economic power can be quite ruthless. In the late 1800s many who amassed substantial economic power were commonly referred to as robber barons. But those ruthless capitalists could use their economic power obtain the economic resources of others only if those others agreed. In a market economy, people become wealthy by enticing others to trade with them. Capitalists can buy out their rivals only if those rivals agree to sell. The use of economic power requires the consent of others for it to be exercised. Not so with political

https://doi.org/10.1017/9781009323178.009 Published online by Cambridge University Press

8.2  The Exercise of Power in Markets and in Politics

143

power. Those who hold political power use the threat of force to coerce those subject to it to comply. For those who want to exercise power, the use of force is indeed enticing. Why ask others to cooperate with you when you can force them to comply? One implication is that those with economic power often use it to buy political power – to get government to force others to do what the economically powerful want, and just as often, to prevent others from doing what the economically powerful do not want. Economic competitors might be able to win over a market with better products and lower prices, but they also might be able to do so by cooperating with government to create barriers to entry that prevent others from being allowed to compete. Those nineteenth-century industrialists and financers – the robber barons – became rich by raising the standards of living for everyone.4 The price of petroleum products fell substantially thanks to John D. Rockefeller’s Standard Oil, and Andrew Carnegie’s U.S. Steel offered more affordable steel to consumers. Railroads increased the scope of markets and lowered transportation costs. That is how markets work. At the same time, many of those progressive-era capitalists used their economic power to negotiate with the political elite to buy political power, in particular to establish a regulatory environment that prevented potential competitors from entering their markets.5 The use of economic power requires those who hold it to persuade others to cooperate. The use of political power allows those who hold it to force others to comply. In markets, people can use their resources to buy what they need to undertake their business projects. Steve Jobs, co-founder of Apple Computer, was reputed to have an unpleasant personality. Whether this is accurate or not, he could enter the market to buy the inputs he needed to make the computers and phones that his company sold. With his money, he did not need the support of those he dealt with. He just had to offer them a good deal. In politics, nobody can get anything done without the cooperation of others. As occasional coups show, even those who appear 4

5

Burton W. Folsom, Jr., The Myth of the Robber Barons (Reston, VA: Young America’s Foundation, 1991), describes the benefits that those so-called robber barons provided for the masses. For arguments along these lines, see Gabriel Kolko, The Triumph of Conservatism: A Reinterpretation of American History, 1900–1916 (New York: The Free Press, 1963), and Murray N. Rothbard, The Progressive Era (Auburn, AL: Ludwig von Mises Institute, 2017). With regard to railroads in particular, see Gabriel Kolko, Railroads and Regulation, 1877–1916 (Princeton: Princeton University Press, 1965), and Richard White, Railroaded: The Transcontinentals and the Making of Modern America (New York: W.W. Norton, 2011).

https://doi.org/10.1017/9781009323178.009 Published online by Cambridge University Press

144

Policies That Maximize Political Power

to be all-powerful dictators can rapidly lose their power without a group standing behind to support them. This is where the willingness to bargain and the ability to persuade others comes into play. In the market, people just need to offer value to others to enter into transactions. In politics, a different set of skills comes into play to build the support of a coalition. The process of becoming a member of the political elite is the process of obtaining and using power over others. There is good reason for those in the political elite to want a more secure hold on their power, because that is how they maintain their elite status, and there is good reason for those who want power to aspire to membership in the political elite, because that is where power resides. But politics is competitive. In elections, some people win while others lose. In markets, entrepreneurs have an incentive to entice customers to transact with them. In politics, people have an incentive to defeat rivals. In the market for soft drinks, there is room for both Coke and Pepsi to succeed. In the market for automobiles, there is room for both General Motors and Toyota to succeed. In electoral politics, one side wins and the other loses. Some people have a comparative advantage in those qualities that enable them to gain economic power. Some people have a comparative advantage in those qualities that enable them to gain political power. Economic power can be used only through the voluntary cooperation of others, so people who are persuasive in enticing others to deal with them have a comparative advantage in exercising market power. Indeed, as Chapter 3 noted, an entire profession – advertising – is devoted to exercising this type of persuasion. Political power is exercised through the threat of force against those who do not comply. Persuasion is helpful, but ultimately those who are willing to be ruthless in their enforcement of mandates have a comparative advantage in politics. McDonalds has been criticized for pushing fatty, sugary, unhealthy foods on customers, but the only way McDonalds can get people’s money is if they voluntarily tender it. Government demands that you turn over your money to them whether you like what they are buying with it or not. Ultimately, the difference between the exercise of economic and political power comes down to the difference between persuasion and coercion – between cooperation and control.

8.3  Politics and Power All people, including members of the political elite, have multiple goals in their lives. For the political elite, however, the quest for power is

https://doi.org/10.1017/9781009323178.009 Published online by Cambridge University Press

8.3  Politics and Power

145

particularly relevant, because power is necessary for them to remain in the political elite. In economic theory, firms are often assumed to have the goal of maximizing profit. On the one hand, this is overly simple, because those who run firms have other goals, but on the other hand, that singular goal makes sense in that firms that are profitable survive and grow, while those that are unprofitable wither and die.6 In the same sense that firms must pursue profits to survive, the political elite must pursue power to survive (as members of the political elite). But for the political elite, there is more to the pursuit of political power than just survival. As Chapter 7 explained, people pursue power for its own sake. Many people like to exercise power over others. And the quest for power is insatiable. Those who have power want more. Unlike the quest for food, there is no amount of power that will satiate the power-hungry. The competitive nature of politics makes power-seeking even more important. In markets, firms are trying to succeed for themselves, not destroy others, even if the creative destruction of capitalism produces that result. Firms are trying to increase their own profits, not decrease the profits of rival firms. In politics, competitors are actively trying to take power away from those who hold it. The political elite must be power seekers to retain their elite status. People who have power tend to acquire a taste for it. The more power people have, the more they want. One result is that the longer they are in power, the more authoritarian they tend to become. Examples of political leaders who became more authoritarian the longer they retained power are many. Names such as Hitler, Stalin, and Mao would stand out, but the same applies to US presidents such as Abraham Lincoln, Franklin Roosevelt, and – indeed – any US president. Examples in the 2020s are Presidents Putin in Russia and Xi in China. It is difficult to think of even one example of a political leader who became less authoritarian the longer they were in power.7 A selection bias is also at work. Some people may have a greater desire to exercise power over others. Those who have the greatest desire for power will gravitate toward government and politics; people who have little desire to exercise power over others will avoid working in politics and government. If the first goal of the power elite is to acquire and maintain power, the next step is to evaluate what public policies can accomplish that goal. 6 7

This idea is nicely developed by Armen A. Alchian, “Uncertainty, Evolution, and Economic Theory,” Journal of Political Economy 58, no. 3 (June 1950), pp. 211–221. If pushed to think of an example, I might consider George Washington.

https://doi.org/10.1017/9781009323178.009 Published online by Cambridge University Press

146

Policies That Maximize Political Power

8.4  The Adversarial Nature of Politics One stark difference between markets and politics is that market activity is cooperative, whereas political activity is adversarial. While it may be common to talk about competition in markets and cooperation in politics, the reality is that the terminology is backwards. Market activity takes place only when people see the mutual advantages of exchanges and cooperate with each other. Nobody is forced to buy from a seller (unless the government forces them), and nobody is forced to work for a particular employer. For market exchange to take place, everyone who is party to the exchange has to agree that the exchange is in their interest. In market activity, people cooperate with each other to make exchanges. One might talk about General Motors competing with Toyota, but those companies never deal with each other.8 The market works as General Motors and Toyota each try to entice potential auto buyers to buy from them rather than from another manufacturer. Market activity is cooperative. Political activity is adversarial. In elections, one party or candidate wins; another loses. When legislation comes up for a vote, it is rare that all legislators are in favor. When the vote occurs, some will win; others will lose. When interest groups find themselves on opposite sides of an issue, one wins; another loses. Because compromises often are made on public policy issues, typically it is the case that nobody ends up getting everything they want. Some people argue that abortion should be prohibited; others argue it should be the woman’s choice. The result is a policy that allows abortions in some cases and prohibits them in others. Nobody gets what they want, which means that everybody views those on the other side as adversaries. Some people argue that the right to bear arms gives people the right to own firearms without restriction. Others argue that nobody (or perhaps nobody but government employees) should be allowed to own firearms. The result is a policy that allows people to own firearms, but with restrictions so nobody gets what they want. People on the other side are adversaries. This outcome, that nobody gets everything they want, means that conflicts will be ongoing. Partial victories lead the victorious to want more and push the losers to try to regain what they lost. There is no alternative to the adversarial nature of politics, because government enforces one set of policies that apply to everyone, and everyone has different ideas about 8

This is not quite true. For example, in 1981 the two companies formed a joint venture to produce automobiles, which has since been dissolved. Even here, the two companies were cooperating with each other, not competing.

https://doi.org/10.1017/9781009323178.009 Published online by Cambridge University Press

8.5  Elites Want to Have Power, Not Use It

147

what they want. One result is that political victories are never permanent. Later negotiations can reverse a political victory. In politics, policy outcomes are not bought; they are rented. When people go to a store or a restaurant, customers and employees all get along with each other because they deal with each other only in cases that offer mutual advantages. Customers do not ask employees about their political views, their religious views, or their stances on various public policy issues. Customers and employees may disagree on almost everything, and yet they get along because in markets, people engage with each other only when it is to their mutual advantage. In politics, where public policies apply generally, small disagreements are difficult to solve cooperatively. One person favors one policy, another favors a different policy, and one of them will win while one loses. The adversarial nature of politics goes hand in hand with the adversarial nature of political power. The government threatens force to gain compliance with its mandates, which requires the power to coerce. People do not want to be subject to coercion themselves; they want government to implement policies that coerce others. In politics, some win while others lose. This applies to public policy in general. Some people are in favor; some are opposed. The adversarial nature of political activity produces policies that satisfy some but disappoint others, and even more likely, do not completely satisfy anyone. This points toward the type of policy that can solidify political power: Politicians must convince the masses that they are on the side of the masses. In market transactions, all parties to them are happy with the result. Otherwise, they would not transact. In public policy negotiations, some people win and are happy with the result; others lose and are unhappy. The political elite must accept the fact that their actions make some people unhappy. But not only do they accept it; they relish it. They celebrate when they win and others lose.

8.5  Elites Want to Have Power, Not Use It Political power enables those who have it to use the threat of force to make people comply with their mandates. But when power is used, it can be used up. Elites want to have power, but prefer not to use it. They want the threat of force to be sufficient to gain compliance, and prefer not to actually have to use force. They want the masses to be compliant, which is best accomplished when the masses weigh the costs and benefits of complying, and choose to comply to avoid sanctions that come with noncompliance.

https://doi.org/10.1017/9781009323178.009 Published online by Cambridge University Press

148

Policies That Maximize Political Power

If a few people violate the mandates of the political elite, those people can be singled out and punished as an example to others of the negative consequences people face when they do not comply. If there is mass resistance, using force against the resisters will cost the political elite support, and they may not be able to marshal sufficient force to get compliance from everyone. If a few people cheat on their income taxes, for example, the government can seek them out and penalize them, providing a lesson that might dissuade others who consider cheating on their taxes. If most people were to cheat on their taxes, there would be too many violators for the government to find and sanction them all. Power works best when the threat of force is sufficient to create compliance, so no actual force is needed. One implication is that those who have power always want more than the amount that would just be necessary to actually force people. They do not want people to be tempted to challenge their power. They want enough power that those who are subject to it perceive that noncompliance would make them worse off. They want people to choose to comply rather than face the inevitable sanctions that would come with noncompliance.

8.6  The Power of Populism Populist political movements are built on the adversarial nature of politics in several ways. To come to power, populist leaders argue that the status quo works against the masses. The common element in populist political leaders is that they claim to fight for the masses against those who oppress them. Beyond that, in the us against them adversarial nature of politics, they claim to represent a broad coalition of people – those in the us group – against a group of oppressors who have little political voice. Often, the oppressors – as populists depict them – are the elite, who are smaller in number than the masses. Foreigners are an even better target for populist rhetoric, because they have no say in a nation’s political leadership. Populist leaders vow to control the power of the elite for the benefit of the masses. The populist message has had success in a variety of political settings – even though populist leaders must be considered a part of the political elite. Populism is not a movement that is attached to a particular set of public policies. Populist leaders have spanned from political left to right. Hannah Arendt described both Josef Stalin and Adolf Hitler as

https://doi.org/10.1017/9781009323178.009 Published online by Cambridge University Press

8.6  The Power of Populism

149

populists,9 and in the twenty-first century Donald Trump in the United States, Boris Johnson in Britain, and Jait Bolsonaro in Brazil have been described as populists. The term has also been applied to leaders as varied as Daniel Ortega in Nicaragua, Hugo Chavez in Venezuela, and Benjamin Netanyahu in Israel. It is not associated with a particular set of public policies, but rather with the political message that populist leaders will stand against the elite to promote the interests of the masses. The term populism was first used in the United States in the late 1800s to describe an agrarian movement which perceived that as the nation industrialized, those with concentrated economic power were able to use it to take advantage of the masses, and in particular, take advantage of those who made their living in agriculture.10 It began as a reaction against the abuse of concentrated economic power, and that is often a theme in populist movements from one end of the political spectrum to the other. Donald Trump depicted foreign trading partners such as China and Mexico as adversaries, and also promised to “drain the swamp” in Washington, DC, painting the political elite as adversaries. Hitler argued that the allies, victorious in World War I, were taking advantage of their victory to oppress the German nation.11 Left-leaning populists from Lenin to Castro to Chavez depicted the economic elite – capitalists – as oppressors of the masses. The common element in populist leaders is their promotion of the us against them message of adversarial politics.12 Politics is by its nature adversarial. The power of populism is that populist leaders claim to support the masses in the us against the elite in the political arena. This maximizes their support. While the advantage of a political strategy that attempts to maximize the base of support seems almost too obvious to state, in the US presidential election of 2016, Hillary Clinton characterized the supporters of her opponent Donald Trump as a basket of deplorables. Meanwhile, Donald Trump was vilifying foreigners, who do not vote, as the adversaries. In the us

9 10 11 12

Hannah Arendt, The Origins of Totalitarianism (Cleveland, OH: World Publishing, 1958). This movement is discussed in Randall G. Holcombe, Liberty in Peril: Democracy and Power in American History (Oakland, CA: Independent Institute, 2019), ch. 8. John Maynard Keynes, The Economic Consequences of the Peace (New York: Harcourt Brace and Howe, 1920), agreed. I discuss populist movements in more detail in Randall G. Holcombe, “Populism: Promises and Problems,” The Independent Review 26, no. 1 (Summer 2021), pp. 27–37.

https://doi.org/10.1017/9781009323178.009 Published online by Cambridge University Press

150

Policies That Maximize Political Power

against them adversarial nature of politics, Clinton was depicting some citizens (and voters) as us but other citizens (and voters) as them. Trump was depicting Americans as the us group and foreigners (who do not get a vote) as them. Clinton was promoting a message of elitism; Trump was promoting a message of populism. The populist message starts with an adversarial positioning of elites against masses, but elites are the ones who hold the power in society, so to counter powerful elites requires a powerful government to control them. Thus, the populist message on the one hand is built on its opposition to the power elite, and on the other claims the need to increase its own power to control the power elite. The populist message, as is common with effective political messages, is designed to appeal to the emotions of citizens and voters. The populist message says others are holding the masses down, and the populists will control those others if they are given the power to do so. The populist message builds on the zero-sum thoughts of citizens by saying that giving government the power to control elites will shift benefits to the masses. The populist message is designed not to promote any particular set of policies, but to give those who control the government more power so they can control elites. The populist message ultimately is one about power rather than policy.

8.7  Elites Design Public Policy One antipopulist message is that government is a complex institutional framework that should be designed and run by experts: by the elite. Universities have departments that are devoted to the science of public administration. Woodrow Wilson, president of the United States from 1913 to 1921, and the only US president to have earned a PhD degree, was a leader in the development of the ideas of public administration.13 From the late 1800s on, the idea that government’s day-to-day operations should be run by a civil service largely independent of political control has taken hold, leading to a government bureaucracy that has substantial independence and a permanent workforce that can outlast those who are elected to oversee it. This antipopulist viewpoint is that government should be run by experts. Public policy should be designed by the elite. 13

See Woodrow Wilson, The State (Boston: D.C. Heath & Company, 1889), for some of Wilson’s ideas.

https://doi.org/10.1017/9781009323178.009 Published online by Cambridge University Press

8.8  Popular Policies

151

The populist message argues against elite control of government, but ultimately elites must design public policy because, if for no other reason, large numbers of people cannot all have a meaningful voice in policy design. Transaction costs are too high. So, while populist policy does not say so, it argues for replacing one set of elites with another. When populists control government, they become the political elite, almost by definition. They are the ones who design public policy. The elitist argument against populism is that experts know more about the design of public policy than the masses. This has a persuasive economic basis in that everyone specializes in certain activities, which gives them a comparative advantage over others in the areas in which they specialize. Public policy design should be left to the elite – the policy experts. The elite deserve the power they have, and it is in the public interest. The populist argues that government should control the power of the elite for the benefit of the masses, but to do so requires a powerful government. Independent of any particular policy preferences, both sides of this argument push for a more powerful government. From a public policy standpoint, then, the issue populism raises is not what policies should be promoted, but how much power government should have to design and implement them. Arguments on all sides point toward increasing the power of government, so it can implement desirable policies designed by experts, or so it is able to counter the power of elites who oppress the masses. The arguments on both sides, which have emotional appeal, ultimately look to shift more power to the already powerful.

8.8  Popular Policies Policies favored by the political elite are policies that give them more power. Specific policies that do so will vary over time and from place to place. Think about the way that successful political platforms are designed. They are vague about specific policies, but have the underlying theme that things are not as good as they could be, and by electing this candidate or party, things will get better. The problem is with the status quo, and political campaigns promise to improve it in some unspecified way. Once governments are in power, they gain more power by designing policies to address perceived problems with the status quo. Different policies address different problems. When in the late 1800s the masses viewed the abuse of economic power by the economic elite as a problem, government stepped up with

https://doi.org/10.1017/9781009323178.009 Published online by Cambridge University Press

152

Policies That Maximize Political Power

regulations and antitrust laws to counter that problem. In reaction to Upton Sinclair’s The Jungle, an exposé of the meat-packing industry, the Pure Food and Drug Act was passed in 1906, the very year Sinclair’s book was published.14 Periodic bank crises toward the end of the nineteenth century and into the twentieth led to the creation of the Federal Reserve Bank in 1913. More banking regulation followed in the 1930s in reaction to the Great Depression, and a financial crisis in 2008 led to additional banking regulation – the Dodd–Frank Act – in 2010. Periodic military threats justify building the government’s military power. Threats from spies and terrorism justify increasing the power of the surveillance state. Popular policies are those that give government more power to react to perceived problems. In all these cases, the policy response has been to give the government more power. The process works as follows. People perceive a problem and say that government should do something to address it. In response, government enlarges its scope of power to address that problem. The result is that popular policies are, in general, policies that give government more power. Political elites respond to the demands of citizens by expanding government power, because that is what citizens say government should do. Popular policies do not fit neatly on the left–right political spectrum. They are policies that convey power to the power elite. Military conflict deserves special emphasis. A more militarized state has the potential to convey substantial power to the power elite. Military conflict is clearly an us against them issue that is based on nationalism, drives patriotism, and tends to unify citizen opinion that enemies should be countered by an increase in government power. Christopher Coyne and Abigail Hall document the way that military intervention abroad increases government power at home.15 Weapons and tactics developed for military activities overseas find their way into domestic policing – both in terms of hardware and surveillance technology. When military conflicts erupt, patriotic citizens increase their support for the exercise of government power. The political elite have limited control over the types of events listed above, but they have extensive control over the narratives that surround them. Are demonstrations in the streets peaceful protests or civil unrest? 14 15

Upton Sinclair, The Jungle (New York: Doubleday, Page, & Co., 1906). Christopher J. Coyne and Abigail R. Hall, Tyranny Comes Home: The Domestic Fate of U.S. Militarism (Stanford, CA: Stanford University Press, 2018). See also Manufacturing Militarism: U.S. Government Propaganda in the War on Terror (Stanford, CA: Stanford University Press, 2021) by the same authors.

https://doi.org/10.1017/9781009323178.009 Published online by Cambridge University Press

8.8  Popular Policies

153

Are businesspeople really robber barons, or are they prospering by providing consumers jobs and consumption opportunities? Are foreign nations exercising their own sovereignty, or engaging in hostile acts? In all these cases, not only do the political elite have the ability to spin the news their way (fake news?), they also have some ability to control events because of their control over public policy. Political leaders can provoke foreign countries into confrontational actions, and can encourage domestic groups to act in ways that justify policy responses. Citizens can see problems and are quick to demand that the government do something to solve them. They are not very good at understanding what the best responses might be to those problems, and rely on political elites to tell them. When they see government action, they perceive that the government is doing something, which is what citizens wanted. When historians are asked who the greatest US presidents are, it is interesting to note that consistently at the top of the list are presidents who have presided over crises such as wars and depression. Abraham Lincoln, Franklin Roosevelt, and George Washington are always among the top. Presidents who presided over peace and prosperity are rarely ranked highly. Problems arise, the political elite tells the masses that the way to address them is to give the government more power, and the masses adopt the policy preferences of the elite. They follow their leaders. Franklin Roosevelt was elected president in the depths of the Great Depression, and took bold action in response, enlarging the powers of government, but never ending the Depression. His popularity seems to be based on the fact that he took action, even though that action failed to solve the problem. Some have argued that the policies he enacted actually made matters worse,16 but he was able to control the narrative and convince a substantial majority of the masses to adopt his policy preferences. He was first elected in 1932 and the Depression persisted for the rest of the decade, but Roosevelt was reelected by a substantial margin in 1936 and again in 1940. Apparently, his popularity was built on the fact that he pursued bold policies, not on whether those policies worked. After the terrorist attacks on September 11, 2001, President George Bush threatened Afghanistan, saying that if the Taliban government of the country did not turn over Osama bin Laden, the US military would 16

For arguments that Roosevelt’s policies were both counterproductive and designed to increase the power of government, see Robert Higgs, Depression, War, and Cold War: Challenging the Myths of Conflict and Prosperity (Oakland, CA: Independent Institute, 2006).

https://doi.org/10.1017/9781009323178.009 Published online by Cambridge University Press

154

Policies That Maximize Political Power

invade the country. It was a deliberately provocative threat that the Bush administration correctly anticipated would be rejected by the Taliban and justify a US invasion. That invasion was followed by the invasion of Iraq, which was also pushed by the Bush administration. While there was debate at the time as to the wisdom of these invasions, they were popular with the masses and produced patriotic support, strengthening the power of the Bush administration and enhancing the power of the political elite. The masses want government to respond to problems – to look out for their interests – but the nature of the problems they face and the appropriate policy responses are defined by the political elite. The Interstate Commerce Act was passed in 1887 and the Sherman Antitrust Act in 1890. With those pieces in place and identifying a popularly perceived problem, Teddy Roosevelt, who rose to the presidency in 1901, used them to develop his reputation as a trust-buster. When inflation picked up in the late 1960s and early 1970s, Richard Nixon imposed economywide wage and price controls in 1971. Most economists agree that wage and price controls are ineffective in controlling inflation (opinions of economists are more mixed about the efficacy of antitrust), but the effectiveness of those policies is of secondary importance to the masses. The masses want the elite to respond to problems. The elite want to enhance their hold on power, and they do so by enacting public policies that enhance their power, in response to publicly perceived problems. Often, the way that the masses perceive those problems is defined by the elite. From the vantage point of the political elite, the fact that the Great Depression did not end for another eight years as Roosevelt created policy after policy in response to it, or that inflation continued upward for a decade after Nixon’s attempts to stop it, is of secondary importance. The masses saw that there was a problem and that the elite were acting to deal with it. Those elite responses were popular, even though ineffective, because the masses adopted the policy preferences fed to them by the political elite. The masses favor policies that increase the power of government because the elite tell them they should favor those policies. The masses follow their leaders.

8.9  Power Leads to More Power The elite quest for power provides the mechanism for acquiring more power. The problems government responds to are often temporary. The economy goes into a recession, but eventually recovers. The nation fights a foreign war, but eventually that war ends. The policies enacted to

https://doi.org/10.1017/9781009323178.009 Published online by Cambridge University Press

8.10  Perpetual Crises

155

address those problems tend to stay after the problem fades. Government programs, once started, tend to remain. Government grows, one crisis after another. Crisis may be too strong a term here. People see problems and want the government to respond, even if these problems do not grow to the magnitude of crises. When problems arise, government expands to address them, but does not shrink back to its former level after the issues pass. This idea has been applied mainly to major crises such as wars and depressions,17 but it applies more generally to any increase in the scope and power of government. Milton Friedman once quipped, “Nothing is more permanent than a temporary government program.” The political elite use any opportunity of public discontent on some issue to increase their power in response to that discontent, and once they have that power, they do not let it go. Often, government programs that respond to problems bring with them unintended negative consequences that prompt calls for additional government intervention. People perceive a problem and demand that the government do something to address it. Government does something, and unintended negative consequences create new problems and a new demand that government do something to address those new problems. Those new policies have more unintended consequences, which leads to demands for more government action. Government control increases and the political elite gain increasingly more power.18 One intervention leads to the demand for another, and the scope and power of government grows. This process works well for the political elite who want to expand their power.

8.10  Perpetual Crises After World War I, Warren Harding campaigned for the presidency with the slogan “Return to normalcy.” Normalcy, in the sense Harding used the term, is not normal. Government moves from one crisis to another, responding by increasing its power, crisis by crisis. Harding’s presidency followed the crisis of World War I, and after less than a decade of “normalcy” the world faced the crisis of the Great Depression in the 1930s. 17

18

This idea is put forward by Alan T. Peacock and Jack Wiseman, The Growth of Public Expenditures in the United Kingdom (Princeton: Princeton University Press, 1961), and is the foundation for Robert Higgs, Crisis and Leviathan: Critical Episodes in the Growth of American Government (Oxford: Oxford University Press, 1987). This idea is developed by Ludwig von Mises, Interventionism: An Economic Analysis (Indianapolis: Liberty Fund, 2011). See also Sanford Ikeda, Dynamics of the Mixed Economy (London: Routledge, 1997), for a development of this idea.

https://doi.org/10.1017/9781009323178.009 Published online by Cambridge University Press

156

Policies That Maximize Political Power

Before the Depression ended, World War II followed in the 1940s. No sooner had that war ended than the Cold War began, and lasted four decades. The 1970s ushered in the energy crisis. The 1990s was the most “normal” decade since the 1920s, although it was disturbed by a savings and loan crisis that ran from 1986 to 1995 (when the last savings and loans were closed), and a global financial crisis in 1997. The terrorist attacks on September 11, 2001, brought with them another crisis and a substantial increase in government authority, along with a twenty-year war in Afghanistan. The financial crisis after the housing market collapse in 2008 brought with it more government intervention, and the global COVID pandemic that began in 2020 again prompted a crisis atmosphere and another expansion of government power. Amidst all these crises, climate change offers another impending crisis that cries out for more government intervention – more power to the political elite. A look at history suggests that crises are the normal state of affairs. The next crisis begins before the previous one has ended. In keeping with the theme of this chapter, Rahm Emanuel, then chief of staff for President Obama, said in 2008, “You never want a serious crisis to go to waste. And what I mean by that [is] it’s an opportunity to do things that you think you could not before.”19 Emanuel directly refers to crises as opportunities to expand the scope and power of government, which is the primary public policy goal of the political elite. The ratchets hypothesis of government growth says that when a crisis appears, government expands and after the crisis does not shrink back to its old level. But crises are not unusual phenomena. They are the normal state of affairs. History also shows the permanent expansion of government power that accompanies each passing crisis.

8.11  Rule of Law and Political Power Rule of law limits political power. Rule of law means that a society is governed by an objective set of set of laws that constrain people’s actions, and that those laws are uniformly enforced on everyone. A society governed by rule of law limits the discretionary use of power by the political elite. They must follow the law, just like everyone else. Rule of law implies that the political elite do not have the discretionary power to design public policy as they please. The exercise of political power under rule of law is limited to enforcing well-established rules. 19

From an interview with the Wall Street Journal, November 19, 2008.

https://doi.org/10.1017/9781009323178.009 Published online by Cambridge University Press

8.11  Rule of Law and Political Power

157

Even where it is most consistently practiced, rule of law is under constant threat because new situations arise that were not anticipated when the old laws were written. The masses push for government responses to unanticipated problems, and those government responses inevitably increase the scope and power of government. They also open the door to discretionary policy-making that tends to be designed to benefit those who make the policies. The previously noted tendency for the political elite to sell policies to the highest bidder undermines rule of law, and undermines market institutions. High bidders are often those who want to enlist government to impose costs on their rivals and provide targeted benefits to themselves. With rule of law, entrepreneurial individuals have the incentive to engage in market activity to create value for others, to enhance their economic power. When rule of law is undermined, entrepreneurial individuals can get ahead by seeking public policies that benefit themselves at the expense of others. They have the incentive to develop political connections, to join the low transaction cost group of cronies who make public policies for their own benefit. When rule of law is absent, institutional incentives encourage entrepreneurial individuals to engage in unproductive or destructive activities.20 If they can gain entry into a favored class, they can use political power to create public policies that benefit themselves but impose costs on others. Rule of law constrains the political elite. The absence of rule of law gives those with political power the discretion to use it as they see fit, and they can use it to their best advantage by selling public policy outcomes to the highest bidder. Without rule of law, the profitability of business increasingly becomes related to political connections, requiring firms to participate in lobbying and political negotiations to remain profitable. That is the goal of the power-maximizing political elite. If they can make the economic elite dependent on the political elite for their continued profitability, the political elite can continue to extract payment from those who benefit from those public policies. The elite viewpoint that government is best run by policy experts who have the knowledge to create effective public policy gives discretion to 20

This idea is explained by William J. Baumol, “Entrepreneurship, Productive, Unproductive, and Destructive,” Journal of Political Economy 98, no. 5, Part I (October 1990), pp. 893–921, and Entrepreneurship, Management, and the Structure of Payoffs (Cambridge, MA: MIT Press, 1993). An empirical test of this idea is undertaken by Russell S. Sobel, “Testing Baumol: Institutional Quality and the Productivity of Entrepreneurship,” Journal of Business Venturing 23, no. 6 (November 2008), pp. 641–655.

https://doi.org/10.1017/9781009323178.009 Published online by Cambridge University Press

158

Policies That Maximize Political Power

those who have political power. It is an argument that those who know best should design public policy. This gives the political elite more discretion and more power. It undermines rule of law with the idea that if the elite are given more discretion, they will use it to further the public interest. This idea is based on wishful thinking rather than an analytical look at the incentives of those who have power over others. If government were omniscient and benevolent, allowing those experts to design public policy could be a good idea. When government power lies in the hands of a power-maximizing political elite, the incentives the elite face point toward less desirable outcomes.

8.12  Creating Dependence on the Political Elite The political elite prefer public policies that increase the size and scope of government, because that increases their power. They prefer policies that differentially benefit well-defined special interests, because that makes those interests dependent on the people who possess political power. If those who benefit from targeted special interest policies fail to support the political elite, the political elite can eliminate those special interest policies. Different policies are directed toward the economic elite, who can provide the highest direct payments for favorable policies, and at the masses, who can deliver the highest number of votes to support the political elite. Welfare state policies make people dependent on government payments, whether the policies are intended to do so or not.21 But often, they are intended to do just that. Social Security provides income to the elder population, and Medicare funds their health care. These programs are not targeted to the needy, but to the elderly, to buy their political support. The greater the share of the population that is on the receiving end of government payments, the more the population relies on the political elite to maintain that support. Public policies target benefits to the masses to make them more dependent on government. Public policies also target the economic elite, who can make more direct payments for favorable policies.22 Subsidies, tax breaks, and

21 22

See Charles Murray, Losing Ground: American Social Policy, 1950–1980 (New York: Basic Books, 1984). The interaction between the political and economic elite is discussed in Randall G. Holcombe, Political Capitalism: How Economic and Political Power Is Made and Maintained (Cambridge: Cambridge University Press, 2018).

https://doi.org/10.1017/9781009323178.009 Published online by Cambridge University Press

8.12  Creating Dependence on the Political Elite

159

regulatory barriers to prevent competitors from accessing markets are common. Once in place, either the interest group continues to pay up to have the policy continued, or it will be terminated.23 The political elite design these policies to enhance their power. Their power increases in general as a result of the payments they receive, and they also gain power over specific individuals who benefit from those programs and must continue to pay to keep them in place. Groups that lobby for special interest benefits are an easy target for this type of political extortion, because the organized group is easy to identify. Consider the policy of mandating that ethanol be included in motor fuels. The corn lobby pushes for the policy, and those same lobbyists who push for the policy make themselves a clearly identified target to ask for payment to keep the policy in place. The policy imposes costs on the users of motor fuels, but they are not an organized political group, so it would be difficult to approach them to ask for payment in exchange for eliminating the mandate. Rather than approach millions of individual drivers to ask for payments to end the mandate, the political elite can approach one lobbying group to ask for payments to continue it. Simply by creating an organization, interest groups create a single target that the political elite can approach to extract payment in exchange for a continuation of policies that favor the group. Thus, public policies are designed to favor concentrated special interests rather than the general public interest.24 Public policies always have some public interest rationale behind them. Nobody justifies a policy by saying that it will convey benefits to one group by imposing costs on another. Bruce Yandle insightfully discusses this by looking at support for alcohol prohibition by bootleggers and Baptists.25 The Baptists favored prohibition because of the moral and social problems they perceived accompanied the consumption of alcohol. The bootleggers favored prohibition because they profited from selling

23

24 25

Several observers have referred to this as extortion, including Fred S. McChesney, “Rent Extraction and Rent Creation in the Economic Theory of Regulation,” Journal of Legal Studies 16, no. 1 (January 1987), pp. 101–118; McChesney, Money for Nothing: Politicians, Rent Extraction, and Political Extortion (Cambridge, MA: Harvard University Press, 1997); and Peter Schweizer, Extortion: How Politicians Extract Your Money, Buy Votes, and Line Their Own Pockets (Boston: Houghton Mifflin, 2013). This is explained by Mancur Olson, Jr., The Logic of Collective Action: Public Goods and the Theory of Groups (Cambridge, MA: Harvard University Press, 1965). Bruce Yandle, “Bootleggers and Baptists: The Education of a Regulatory Economist,” Regulation 7, no. 3 (May–June 1983), pp. 12–16.

https://doi.org/10.1017/9781009323178.009 Published online by Cambridge University Press

160

Policies That Maximize Political Power

illegal alcohol. Yandle notes that all demand for regulation tends to work the same way. Some argue it is in the public interest, but there is also a contingent that demands that same regulation because they personally benefit from it. Those who benefit have a strong incentive to push for it. The result is an increase in the scope and power of government, shifting more power to the political elite. The political elite support these public policies because the policies increase their power, and the masses support them because the political elite sell the masses on those policies. Citizens and voters follow their leaders.

8.13 Conclusion Chapter 3 noted that the most effective political messages are vague in their policy content, saying not much more than that things are not as good as they could be, and if the party or candidate offering the message is given political power, they will enact policies to make things better. The masses tend to anchor their policy preferences on parties and candidates, as Chapter 6 showed, so the policy preferences citizens and voters express are the policy preferences of the political elite. While some problems, such as plagues and natural disasters, are outside the control of the elite, in many cases the elite are in a good position to define the problems that should concern the masses in addition to offering policies in response. The policies promoted by the elite are consistently policies that give the political elite more power. Policies that favor well-organized special interests convey more power to the political elite because those policies make the interest groups dependent on the political elite for the continuation of the benefits they receive from the programs. But the same is true of transfer programs that support the masses. The more dependent people are on the political elite, the more power the political elite gains. The public policy preferences of the masses are derived from those of the elites on whom they anchor. The political elite favor policies that give them more power. Citizen demands that government address problems, and citizen demands that government provide transfers to them, shift power to the elite. The elite tell the masses this is what they should want, and the masses follow their leaders.

https://doi.org/10.1017/9781009323178.009 Published online by Cambridge University Press

9 Patriotism, Propaganda, and the Public Interest

Prior to the Enlightenment, people had no illusion that government served their interests. People were the subjects of their governments, and their obligation as citizens was to serve the interests of their ­governments  – effectively, the interests of the ruling class. Thomas Hobbes, writing in 1651, put forward the argument that it was in the best interest of the masses to abide by the rules of the political elite.1 A society without rules, and without someone to enforce those rules, would descend into anarchy, Hobbes argued, where life was solitary, poor, nasty, brutish, and short. The way out of that anarchy was for all citizens to agree to abide by the rules of the sovereign. That was the social contract. Even if people disagreed with some of the rules laid out by the political elite, they were still obligated by the social contract to abide by them. If people could choose what rules to abide by and what rules to disregard, society would digress into anarchy. As Hobbes saw it, an orderly and productive society required that the masses obey the rules created by the political elite. Their obedience was required not so much as an obligation to the elite as for the benefit of the masses, their fellow citizens, as Hobbes described it. The government’s subjects were obligated to follow their leaders, for the benefit of all. That view of government changed with the development and widespread acceptance of Enlightenment ideas, led by John Locke. In 1690, just a few decades after Hobbes wrote, Locke declared that people naturally have rights, and the role of government is to protect those rights.2 1 2

Thomas Hobbes. Leviathan (New York: E.P. Dutton, 1950 [orig. 1651]). John Locke, Two Treatises of Government (Cambridge: Cambridge University Press, 1960 [orig. 1690]).

161

https://doi.org/10.1017/9781009323178.010 Published online by Cambridge University Press

162

Patriotism, Propaganda, and the Public Interest

Locke also had a social contractarian view of government, but unlike Hobbes’s vision of the social contract, government was part of the contract as Locke described it. People naturally have rights, Locke asserted, and their obligation under the social contract is to respect the rights of others. But sometimes, opportunistic people will violate others’ rights, and government’s role in the social contact is to protect individual rights. If government does not uphold its obligation, Locke said, people have a right to replace their government with one that will protect their rights. The relationship of citizens with their governments depicted by Hobbes is backwards according to Locke. Hobbes depicted citizens as subjects of their governments, obligated to carry out the government’s mandates. Locke argued that government should serve its citizens rather than citizens serving their governments. This Enlightenment thinking made it more challenging for the political elite to maintain their positions of power. Indeed, Locke was writing shortly after the Glorious Revolution in Britain in 1688 that deposed King James, offering principles in defense of the king’s removal from power. In the next century, the ideas of Locke and other Enlightenment writers laid the foundation for both the American Revolution in 1776 and the French Revolution that began in 1789.3 The idea that government should serve the people rather than the people serving the government has become more firmly entrenched since those Enlightenment ideas were first advanced, challenging the power of the political elite. The concept of populism, discussed in the previous chapter, is both an acknowledgment of the Enlightenment idea that government should serve the people and a recognition that in fact the elite design public policy to benefit themselves. Populists demand a government that represents the interests of the masses rather than the elite, an idea that is flawed because populist governments merely replace one set of elites with another. Prior to the Enlightenment, the masses viewed that they had an obligation, as subjects of their governments, to further the interests of their leaders. There is no going back to pre-Enlightenment thinking, so the challenge facing the political elite is to discover mechanisms that will convince the masses that they should abide by the mandates of the elite. Brute force is a mechanism of last resort. Far better are mechanisms that lead citizens to agree to comply without overtly forcing them. Patriotism 3

Bernard Bailyn, The Ideological Origins of the American Revolution (Cambridge, MA: Harvard University Press, 1967), discusses the impact of Enlightenment ideas on the American Revolution.

https://doi.org/10.1017/9781009323178.010 Published online by Cambridge University Press

9.1  The Social Contract

163

and propaganda are tools the elite use to maintain control of the political views of the masses. Patriotic citizens want to follow their leaders.

9.1  The Social Contract Some concepts of the social contract theory of the state were introduced in Chapter 1. Hobbes’s claim that citizens have an obligation to abide by the rules of government is a persuasive one. The government imposes rules on everyone to create an orderly, productive, and safe society. Citizens have an obligation to their fellow citizens to abide by those rules. Everybody is better off when everyone follows the rules. Is there actually a social contract – a set of obligations that everyone has with respect to everyone else? One rule universally enforced by governments is that citizens should not kill each other. Most people would agree that this rule is in the public interest. Even murderers are well aware that they have violated a social norm when they kill. People agree to abide by this rule themselves in exchange for everyone else also abiding by it, but some people might opportunistically violate the rule, so citizens rely on government to enforce it, and to penalize those who violate it. Similarly, people generally agree that they should not steal each ­other’s property (even though they may not agree on what constitutes the legitimate ownership of property), and rely on government to enforce that rule. They are content to be bound by government enforcement of rules against theft, in exchange for everyone else being bound by those same rules. People do not have to be students of Hobbes to agree with the logic of Hobbes’s arguments. In this sense, there is a social contract. People generally recognize that they have certain rights, including their rights to their lives and property, and they agree that they have a social obligation not to violate the rights of others. And they agree that the proper role of government is to set and enforce those rules to further everyone’s interests. Steven Pinker and Jared Diamond present book-length arguments supporting the idea that strong governments have allowed large numbers of people to live peacefully with one another.4 Their books can be seen, at least in part, as documentation to support Hobbes’s argument that an orderly society requires that its members comply with the rules of a strong government. 4

Steven Pinker, The Better Angels of Our Nature: Why Violence Has Declined (New York: Viking, 2011); Jared Diamond, The World until Yesterday: What Can We Learn from Traditional Societies? (New York: Viking, 2012).

https://doi.org/10.1017/9781009323178.010 Published online by Cambridge University Press

164

Patriotism, Propaganda, and the Public Interest

The logic behind government enforcement of rules is that people have an incentive to be free riders. They might violate the rules themselves even though they want others to be bound by them. So, they are agreeable to having the rules enforced on themselves as long as they are enforced on everyone else. People buy into that argument for rules that not only protect their rights and property but also force them to pay taxes to finance government goods and services. This contractarian argument is a fiction, of course. Governments were not formed to provide goods and services for people and to protect their rights, even if governments do perform those functions. Governments were instituted by force, to enable some people to rule over and control others. One might cite the US government as an example of a government established to protect the rights of its citizens, as its Declaration of Independence states, but even here, the government of the United States was imposed by force, as a result of some people winning a war against others. The war was fought over which group would hold the powers of government – the power to use the threat of force to get citizens to ­comply with their rules. The US government, and every other government, was established by force. I ask my college students, “Why do we have government?” I write the question on the board. The answers they give are that the government provides good things for them. They mention schools, national defense, roads, courts, and police protection. Government does provide those things, but that is not why we have government. We have government because some people use force to rule over others. The answers my students give show that they buy into the social contract theory of the state, even if they are not familiar with it by that name. They have been propagandized into believing government does good things for them, so they should support the government and abide by its rules. In general, people accept the legitimacy of these obligations to others, but do not hold government to the same conditions. While most people would agree they should not kill others or take the property of others, governments routinely do these things, enforcing a death penalty for some crimes, and on a grander scale, engaging in wars with the intention of killing adversaries. They also claim the right to take people’s property through taxation. Murray Rothbard has argued that those in government should be held to the same ethical standards as everyone else.5 5

Murray N. Rothbard, The Ethics of Liberty (Atlantic Highlands, NJ: Humanities Press, 1982).

https://doi.org/10.1017/9781009323178.010 Published online by Cambridge University Press

9.1  The Social Contract

165

It is unethical for some people to coerce others, so government is behaving unethically when it forces people to pay taxes. Taxation is theft. Indeed, because all government activity is based on force, government itself is unethical, according to Rothbard. Most people dismiss Rothbard’s extreme libertarian views as unworkable in the real world, providing good evidence that while people believe they have an obligation to refrain from violent behavior against others, they do not hold government to that same standard. James Buchanan says, “The anarchist utopia must be acknowledged to hold a lingering if ultimately spurious attractiveness. Little more than casual reflection is required, however, to suggest that the whole idea is a conceptual mirage.”6 Whether one sides with Rothbard or Buchanan, the point is that most people agree with Buchanan’s conclusion and consider government a necessary component of an orderly society, not bound by the same rules it imposes on its subjects. A pro-government ideology encourages people to follow their leaders, and to accept the legitimacy of the power they exercise. Jean Jacques Rousseau is quoted in Chapter 1 as saying that democratic political institutions reveal the general will, so that when a collective decision is made through a democratic political process, it furthers the general will.7 Rousseau says that if he disagrees with a decision made by his democratic government, all that shows is that he is wrong about the general will. Rousseau’s concept of democracy essentially says that the policy preferences of the masses are the policy preferences of the elite. After all, it is the elite who make public policy, and Rousseau claims that when a democratic government does so, it is furthering the general will. Few citizens will be familiar with the writings of Rousseau or Hobbes, but the ideas in those writings resonate with the masses. The general argument extends beyond protecting people’s lives and property to apply to everything government does. People want roads, but might be free riders if asked to voluntarily contribute so they can be built. Therefore, government forces everyone to pay, and provides roads for everyone – and libraries, and fire departments, and more – and people are better off because of it. 6 7

James M. Buchanan, The Limits of Liberty: Between Anarchy and Leviathan (Chicago: University of Chicago Press, 1975), p. 3. Jean Jacques Rousseau, The Social Contract, or Principles of Political Right, translated by G. D. H. Cole, www.constitution.org/jr/socon.htm (1762).

https://doi.org/10.1017/9781009323178.010 Published online by Cambridge University Press

166

Patriotism, Propaganda, and the Public Interest

The argument even extends to government redistribution. The rich care about the poor, but charity might fall short because while people want to help those in need, they have a tendency to free ride off the contributions of others, resulting in a level of charitable giving that is too low according to everyone’s preferences. So, government collects taxes to engage in redistribution, which is better for everyone, rich or poor.8 Are these arguments correct? There is no way to tell. One could ask rich people if they want to pay taxes to redistribute income to the poor, and if they say no, the social contractarian response is that they are just trying to be free riders. The social contractarian argument legitimizes the government’s use of force by implying that it is a part of a social contract – something people have agreed to, even though they did not actually agree. This contractarian vision of government depicts government’s threat of force as a mechanism to overcome a prisoners’ dilemma situation. The prisoners’ dilemma, illustrated in Figure 9.1, is a two-player game which shows that while the two individuals would be better off cooperating with each other, each has the incentive to defect, regardless of what the other player does. The entries in the payoff matrix show the payoff to each strategy, depending on the other player’s action. Assume that player 1 cooperates. If player 2 then cooperates, each player gets a payoff of 5. If player 2 defects while player 1 cooperates, player 2 gets a payoff of 7 rather than 5 (and player I gets a payoff of only 1). Thus, if player 1 cooperates, the best strategy for player 2 is to defect. If player 1 defects, player 2 can cooperate, receiving a payoff of only 1, or also defect and get a payoff of 2. Player 2’s best strategy if player 1 defects is also to defect. The matrix is symmetrical, so the same analysis applies to player 1. No matter what player 2 does, player 1 gets a higher payoff from defecting rather than cooperating. Both players have an incentive to defect, in which case both get a payoff of 2. But if both cooperated, they would both get a payoff of 5. The challenge is to find a way to persuade the players to cooperate, and one answer is to create a government that forces them to cooperate. The social contract theory of the state is built on this idea that citizens find themselves in a prisoners’ dilemma situation where everyone wants other people to abide by the rules of government, but would choose to disregard the rules themselves if they could. People would not pay taxes 8

This argument is made by Harold M. Hochman and James D. Rodgers, “Pareto O ­ ptimal Redistribution,” American Economic Review 59, no. 4, Part 1 (September 1969), pp. 542–557.

https://doi.org/10.1017/9781009323178.010 Published online by Cambridge University Press

9.2  Patriotism and Propaganda

167

Player 1 Cooperate

Defect

7

5 Cooperate 1

5

Player 2

1

2

Defect 7

2

Figure 9.1  Prisoners’ dilemma game

without government forcing them to. People would not obey government regulations without the government’s threat that those violating regulations will have costs imposed on them by government. Third-party enforcement is necessary to force people to act cooperatively. The state is that third party. The theory is plausible enough that people think they should abide by government’s mandates. If the elite tried to rule as autocrats, the masses would resist. Employing actual force to make people comply is costly to the elite, and risky. Gaining compliance is both less costly and less risky when people believe they should comply. They agree to the rules that subject them to penalties when they act contrary to them because they believe they benefit from the rules imposed on them by the elite. They believe elite policy preferences are their preferences.

9.2  Patriotism and Propaganda The idea that people should abide by the government’s mandates is reinforced by the use of patriotism and propaganda. The use of patriotism

https://doi.org/10.1017/9781009323178.010 Published online by Cambridge University Press

168

Patriotism, Propaganda, and the Public Interest

and propaganda to support the power of the political elite is so much a part of the social order that it passes almost unnoticed. Students in government schools begin their school day reciting a pledge of allegiance to their government. The national anthem is played at the beginning of sporting events, sometimes with military personnel holding flags, and if the event is large enough, a fly-by of military aircraft. Governments are also heavy purchasers of advertising in all types of media.9 People who appear disrespectful of the symbols of government – the flags, the anthems – are labeled unpatriotic. The term is not just descriptive. It is intended to be pejorative. The institutions of government, in addition to providing mechanisms by which public policy is made, also have symbolic value. If public policies are viewed as mandates imposed on the masses by elites, they will be resisted. Democratic government provides symbolic value to elites by making it appear that government actions are legitimate activities sanctioned by the masses.10 Think again about Rousseau’s argument that collective decisions made through democratic processes reveal the general will. Government is the mechanism that allows people to work collectively to achieve their goals. Even if people do not view democratic government as an extension of the general will, democratic institutions provide the appearance of legitimacy to the actions of the political elite. People may disagree with actions the government takes, but democratic institutions legitimize those actions through the process by which the elites are given their powers. Legislators, presidents, prime ministers, and bureaucrats acquire their powers through a democratic process through which citizens choose who can legitimately exercise those powers. By supporting the process, the masses grant the legitimacy of the powers exercised by the political elite. To illustrate by way of example, when President George W. Bush made the decision to invade Iraq in 2003, that decision was very controversial. However, those who disagreed with the decision but supported the institutions of democracy found it difficult to disagree that President Bush had the authority to make that decision. After the invasion took place, both supporters and those who opposed the invasion fell in line to

9

10

A detailed development of these ideas is found in Christopher J. Coyne and Abigail R. Hall, Manufacturing Militarism: U.S. Government Propaganda in the War on Terror (Stanford, CA: Stanford University Press, 2020). This idea is developed in detail by Murray Edelman, The Symbolic Uses of Politics (Urbana: University of Illinois Press, 1964).

https://doi.org/10.1017/9781009323178.010 Published online by Cambridge University Press

9.3  Thank You for Your Service

169

support it. The messaging was not that people should support the decision to invade, but that people should support the troops. The elite were able to rely on patriotism and pro-military propaganda to win (sometimes grudging) support from the masses. Had the decision to invade been made by an autocrat, the symbolic support of democratic institutions would not have been in place, which would have made it more difficult to win the support of the masses. Autocrats must reorient the symbolic allegiance the masses have toward democratic institutions and turn it toward allegiance to the autocrats themselves. Stalin and Hitler were able to do this, making their actions appear to be an exercise of legitimate power. Donald Trump was unable to do so. Stalin and Hitler commanded power because they were Stalin and Hitler. Trump commanded power because he was the elected president, and he lost that power when he failed to win reelection. The political elite call on patriotism and propaganda to support the legitimacy of their actions, and political institutions are designed to convey this image of legitimacy. Elections make it appear that “the people have spoken” and the political elite reflect the preferences of the ­voters. But as the previous chapters have illustrated, the preferences they express at the ballot box are the preferences the elite have encouraged them to express.

9.3  Thank You for Your Service In the twenty-first century United States, military personnel are often given privileges, such as military discounts at businesses and priority boarding on airline flights, and are publicly told, “Thank you for your service.” With an all-volunteer military, they are being thanked and respected for the occupation they chose, raising the question of why they should be singled out to be treated with extra respect. Do they deserve more respect than a coal miner, an electric utility line worker, or an auto worker? Despite the fact that citizens benefit more from having automobiles to drive and electricity in their homes than from wars fought in poor countries half-way around the world from the United States, the answer is patriotism. Military personnel are giving service to their country, as opposed to doing jobs to earn a paycheck. The respect they are offered is respect given to the coercive power of the state. With a bit of reflection, it seems clear that the people who bring Americans food, electricity, and motorized transportation provide more benefit to citizens than those who fight in foreign wars. Yet critics will say

https://doi.org/10.1017/9781009323178.010 Published online by Cambridge University Press

170

Patriotism, Propaganda, and the Public Interest

that if not for the military, the nation’s very existence would be threatened. That retort is questionable. The United States fought a twentyyear war in Afghanistan, which it lost, and Americans seem no worse off because of the loss. But even accepting the retort as true, most military personnel have jobs similar to those in the private sector. Why would someone with a desk job in the military deserve more respect than someone with a desk job at a private company? Why would an air force mechanic deserve more respect than one who works for an airline? The answer is patriotism. Without thinking too much about it, people are propagandized to think that those military personnel are serving their country. They are, but so are farmers, factory workers, and medical personnel. Are military personnel really providing more service to their country than these other individuals? Are people who have desk jobs at the Pentagon providing more service to their country than people who have desk jobs at the Department of Health and Human Services? The patriotism that honors military personnel in this way is more than just pro-government propaganda. It is propaganda aimed at supporting the element of government that has a mission most oriented toward using force against others. It celebrates the coercive power of government.

9.4  Choosing Adversaries The very nature of political activity creates winners and losers. Everyone wants to be on the winning side. In politics, it is us against them. One strategy for maximizing power is to create divisions such that those who fall in the us group are citizens and voters, and those in the them group are not. Thus, nationalism is a politically potent sentiment. We, in our nation, are on one side. They, outside the border, are on the other. The previous chapter noted the contrasting strategies of Donald Trump and Hillary Clinton in the 2016 presidential election. Trump, the winner of the election, depicted foreigners as adversaries. Hillary Clinton campaigned using a different strategy, calling Trump supporters a basket of deplorables. Clinton’s calling Trump voters deplorables likely had the effect of motivating them to turn out to vote against her. Attacking an opposing candidate makes some sense. It may discourage supporters from turning out to vote for a candidate who appears to be flawed. Attacking that candidate’s supporters is questionable, because it is likely to motivate those who are being attacked to fight back: to turn out and vote against their attacker.

https://doi.org/10.1017/9781009323178.010 Published online by Cambridge University Press

9.5  Can’t We All Just Get Along?

171

Trump’s messaging on foreign trade was also noteworthy in the 2016 election. He said that the United States was being taken advantage of by bad trade deals, and that if he was elected, he would renegotiate them and get good deals. Trump never said what was wrong with the deals the United States had, or what changes in their provisions could make them better. This fits exactly with the campaign strategy discussed in Chapter 5. Claim that things are not as good as they could be, and if elected, this candidate will make them better. Everyone is in favor of that, but some people would disagree with specific policies, so the best campaign strategy is to not mention any. In 2020 Joe Biden, who defeated Trump that year to win the presidency, included tax increases in his campaign platform. But he pledged that the tax increases he had in mind would raise taxes only on the top 1 percent of income earners, and most people would pay the same or lower taxes. In dividing us against them, it is best to include all voters in the us group, but almost as good (and maybe better: see below) is including 99 percent of the voters in the us group and only 1 percent in the them group. That populist message told voters Biden would support the interests of the masses against the elite. That message entices citizens to follow their leaders. If the elite want to gain the support and cooperation of the masses, they have to offer the message that the public policies they favor will be in the public interest. Trump’s message on foreign trade and Biden’s message on taxes were intended to do this. People do see social situations as potential prisoners’ dilemmas in which people will act against the public interest unless they are sanctioned for their antisocial actions. The political elite have an incentive to promote policies that benefit everyone – that help the entire society escape from being trapped in a prisoners’ dilemma.

9.5  Can’t We All Just Get Along? Why does anyone have to be in the them group? Rather than campaign using an adversarial us against them strategy, why not campaign on a platform that envisions everyone cooperating for their mutual benefit? One reason this strategy is weak is that many people see politics – and more generally, see life – as a zero-sum game. If some people get more, that means others will get less. People want political leadership that will put them in the group that gets more. A second reason is that adversarial politics provides an avenue for people to blame others for holding them back. The idea that we would be better off if it were not for others taking

https://doi.org/10.1017/9781009323178.010 Published online by Cambridge University Press

172

Patriotism, Propaganda, and the Public Interest

advantage of us is an easy political sell. People are easily persuaded that they would be better off if others were not holding them back, rather than taking responsibility for their own lots in life. A campaign message of unity would seem to play well. People do want to get along with others, and in one sense a world at peace, and a nation at peace, would appeal to everyone. That message is challenged by the tribal instincts that have evolved in humans since the stone age. Hunter–gatherer societies had this us against them mentality that pushed them to cooperate with members of their own tribes and to view members of other tribes with hostility. Jared Diamond, an anthropologist who has lived among hunter–gatherers, notes that when strangers in those societies meet, the encounter is likely to be violent.11 People are likely to view strangers as potential predators that they should try to neutralize, or potential prey. Better to take what they have before they can take what you have. Having lived among hunter–gatherers, Diamond, who is a professor at the University of Southern California, marvels at being able to walk through the streets of downtown Los Angeles among throngs of strangers without feeling threatened by them. But tribal instincts remain strong, and modern societies have developed institutions to accommodate them. One example is organized sports competitions. People become fans of particular teams, bond with other fans of the same teams, take joy when their team triumphs, but feel a personal loss when their team loses. This is one way that people can channel their tribal instincts into activities that cause minimal harm to others. Politics has similar effects, though they are not always as benign. People choose political candidates and parties with enthusiasm similar to their choices of favorite sports teams, and also with similar enthusiasm want their side to defeat their opponents. In politics, one side wins and the other loses. Politics brings out people’s tribal instincts, but a difference between politics and sports is that there are instrumental consequences to the outcomes of political consequences. Some policies are chosen over others, and both the winners and losers in political contests must live under the political authority of those preferred by the winners. There are actual instrumental differences in the public policies that will be chosen by one side over another, and the tribal instincts people have brought with them from much earlier times make them want to be on the winning side of political contests. So, a political message that 11

Jared Diamond, The World until Yesterday: What Can We Learn from Traditional Societies? (New York: Viking, 2012).

https://doi.org/10.1017/9781009323178.010 Published online by Cambridge University Press

9.6  The Authority of the Elite

173

depicts an adversarial contest that will make us winners over them is a more powerful message than one saying we are all in this together so let’s work together to make things better. The adversarial message makes it appear more important to be on the winning side.

9.6  The Authority of the Elite Many factors give the elite substantial control over the public policy preferences of the masses. Previous chapters have focused heavily on the rational ignorance of most citizens and the incentives they have to express the policy preferences of their anchors – members of the political elite. As this chapter notes, people buy into the social contract theory of the state, although most will not recognize it by that name. They believe they have a social obligation to follow the government’s rules, which create an orderly society, and they believe that the general purpose of government is to act in the public interest – enough so that they express some degree of outrage when they think government has done otherwise. This gives those who hold political power the authority to make the rules, which the masses then are obligated to obey. The elite reinforce these beliefs with patriotic messages and propaganda supporting the right of elites to make public policy decisions. Propaganda may be aimed at gaining support for specific policies, but the message of propaganda is more general. It reinforces the idea that the elite have the authority to impose policies on the masses, and that when they do, they are acting for the public good. It is the message that Rousseau articulated more than two centuries ago. The message is that some people – the political elite – have the legitimate power to make the rules, and that the masses are obligated to follow the rules. The masses buy into that message, but as C. Wright Mills said, the elite see themselves as above the rules. The rules apply to the masses, but not to the elite who make them. The result is that government does not do as good a job at organizing society to escape a prisoners’ dilemma situation as that model seems to illustrate. Figure 9.2 suggests an ­alternative outcome. Figure 9.2 reproduces the payoff matrix from Figure 9.1. The only difference is the labels. Player 1 from Figure 9.1 represents the elite; player 2 represents the masses. The elite make the rules. The masses believe their obligation is to follow the rules. The rules do not apply to the elite. Because the masses follow the rules, the outcome of this game will be limited to the upper right or upper left cells in Figure 9.2. The elite can

https://doi.org/10.1017/9781009323178.010 Published online by Cambridge University Press

174

Patriotism, Propaganda, and the Public Interest Elites Follow Rules

Rules Don’t Apply to Me

7

5 Follow Rules 1

5

Masses

1

Rules Don’t Apply to Me 7

2 2

Figure 9.2  Masses and the elite

either follow the rules themselves and get a payoff of 5, or if the rules do not apply to them, get a payoff of 7. They make the rules for their benefit and the outcome is in the upper right quadrant, where the masses follow the rules and the rules do not apply to the elite. The elite get a payoff of 7 and the masses get a payoff of 1. The exact numbers, taken from the prisoners’ dilemma game, are for sake of illustration, but they do suggest two things. First, if both follow the rules, the combined payoff is a total of 10, whereas in the upper right quadrant where the rules do not apply to the elite, the combined payoff is 8. This suggests that the elite could take advantage of their elite status to design public policy in a way that gives them the highest payoff, but is not the best outcome for the masses, and does not produce the highest payoff for the entire group. This is not a new idea. Marx and Engels argued that the governments of capitalist societies act in the best interest of the elite (the bourgeoisie),12 C. Wright Mills said that the power elite 12

Karl Marx and Friedrich Engels, The Communist Manifesto (New York: International Publishers, 1948 [orig. 1848]).

https://doi.org/10.1017/9781009323178.010 Published online by Cambridge University Press

9.7  The Escape from Anarchy

175

write rules that constrain the masses but do not apply to themselves,13 and Joseph Stiglitz said the elite write the rules that favor themselves and then choose their own referees.14 Figure 9.2 just illustrates the point that has been made by many previous social scientists. A second thing Figure 9.2 illustrates is that by subjecting themselves to the authority of the political elite, the masses may be worse off than if they were not bound by the rules. The matrix in Figure 9.2 shows a payoff of 1 if the masses follow the rules while the rules do not apply to the elite, but a payoff of 2 if both the elite and masses act as if the rules do not apply to them. The propaganda that pushes the masses to buy into the idea that they are obligated to follow the government’s rules because of a social contract – an obligation to their fellow citizens – could lead to an absolute decline in their welfare.15 The masses may, or may not, believe they would be better off if they chose not to follow the rules, but whether that is the case can only be a matter of conjecture. The masses can see how they are faring under the status quo, but can only speculate about their welfare if circumstances were different. When they follow the rules, they can see (in Figure 9.2) that their payoff is 1, but can only guess how things might be different if they decided not to follow the rules.

9.7  The Escape from Anarchy Thomas Hobbes argued that citizens are obligated to abide by the rules of government to escape a situation of anarchy, in which life is nasty, brutish, and short. He thought that everyone would be better off following the rules of government, leading to an orderly and productive society, and in most cases, he surely is right. In that sense, Figure 9.2 overstates the case, and the numbers in Figure 9.3 likely better illustrate the outcome of government’s imposition of rules on the masses. The numbers are changed slightly to show this possibility. Hobbes said that people are obligated to follow the government’s rules, placing the outcome of the game shown in Figure 9.3 in the upper right or upper left cells of the matrix, and if the elite who make the rules 13 14 15

C. Wright Mills, The Power Elite (New York: Oxford University Press, 1956). Joseph E. Stiglitz, The Price of Inequality: How Today’s Divided Society Endangers the Future (New York: W.W. Norton, 2012). A prisoners’ dilemma model like this is used to analyze the ideology of the social contract in Randall G. Holcombe, “Contractarian Ideology and the Legitimacy of Government,” Journal of Institutional Economics 17, no. 3 (June 2021), pp. 379–391.

https://doi.org/10.1017/9781009323178.010 Published online by Cambridge University Press

176

Patriotism, Propaganda, and the Public Interest Elites Follow Rules

Don’t Follow

7

6 Follow Rules 6

3

Masses

3

Don’t Follow 7

2 2

Figure 9.3  Escape from Hobbesian anarchy

are not obligated to follow them, the outcome, as before, is in the upper right cell of the matrix. Both the elite and the masses are better off in the upper right cell than in the lower right, where nobody follows the rules, if Hobbes is right (and in most cases, he is). But the outcome is still slanted to favor the elite over the masses. As in Figure 9.2, the combined payoff if both follow the rules is higher than the combined payoff in which the masses follow the rules but the rules do not apply to the elite. The numbers in those figures were chosen only to illustrate some points, but they do help demonstrate the possibilities that exist for government-imposed order. Figure 9.2 shows that the masses are worse off abiding by the government’s rules than not. This possibility is considered by Peter Leeson, who provides empirical evidence that the citizens of Somalia were better off after their government collapsed in 1991, leaving them in a stateless society.16 Leeson looks at data – including life expectancy, infant mortality, access to clean water, literacy, and telephone 16

Peter T. Leeson, “Better Off Stateless: Somalia before and after Government Collapse,” Journal of Comparative Economics 35 (2007), pp. 689–710.

https://doi.org/10.1017/9781009323178.010 Published online by Cambridge University Press

9.8 Conclusion

177

service – to show that in a wide variety of dimensions, the masses were worse off with government than without. The outcome illustrated in Figure 9.2, in which the masses would be better off without the rules imposed on them by the elite, is a distinct possibility. Leeson refers to these types of governments as failed states, and Somalia is likely not the only case. Figure 9.3 surely shows the more common outcome, in which the masses are better off following the government’s rules than not. The figure makes it appear that this is a choice offered the masses, but of course it is not. The elite, in control of government, use the threat of force to compel the masses to abide by their rules. Still, as Hobbes suggests, they are almost always better off doing so.

9.8 Conclusion The public policy preferences of the masses are derived from those of the elite. Individuals anchor on a party, a candidate, or an ideology, and most of their preferences are derived from those of their anchors. To understand their preferences, then, requires understanding the policy preferences of their anchors. The political elite, on whom the masses anchor, do not systematically favor any particular ideologically based public policies. Yes, the ideological orientations of politicians are often apparent, but those policy preferences are a means toward the end of gaining, retaining, and increasing power. The primary goal of elites is not to achieve any particular policy outcome, but to maintain and improve their positions in the power hierarchy. This makes the attainment of any other goals they might have possible. The political elite favor the policies that appear to be the best means toward that end. Even if politicians do have strong ideological orientations, they can implement policies that further those ideological goals only if they maintain their power. Thus, in the same way that firms must be profitable to survive, so economists often assume firms maximize profit, politicians must maintain power to survive, so that is their first-order goal. To do this, they must convince the masses that their power is legitimate. Their appeal to patriotism and the use of propaganda helps them do so. People buy into the idea that they have some obligation to abide by the government’s rules – the rules designed by the political elite – and that doing so furthers the public interest. Institutions are designed to convey the appearance of legitimacy to the power exercised by the political elite, and the analysis in this chapter shows how the result is likely to differentially benefit the elite, while potentially imposing some cost on

https://doi.org/10.1017/9781009323178.010 Published online by Cambridge University Press

178

Patriotism, Propaganda, and the Public Interest

the masses. To look back at the figures in this chapter, the masses end up with a lower payoff in the upper right cell of the matrix than they would get if they ended up where social contract theory suggests they should, in the upper left. The mechanisms that leave the masses in that position have been analyzed by public choice theory – rent-seeking, regulatory capture, and interest group politics. The elite can maneuver to get this outcome because of the substantial control they have over the public policy preferences of the masses. Ultimately, the elite tell the masses what they should think, and for many reasons, the masses adopt the public policy preferences of the elite. They follow their leaders. Prior to the Enlightenment, governments clearly ruled by force and their subjects viewed it as their obligation to carry out their governments’ mandates. The Enlightenment idea that government should serve its citizens rather than citizens serving their governments created challenges for political elites who wanted to retain power. Democratic institutions have symbolic value by making it appear that democratic governments are accountable to their citizens, and act in their interests. Democratic institutions provide legitimacy to the political elite and bring with them the additional benefit that the patriotism and propaganda that accompanies those institutions pushes citizens to believe that they should follow their leaders. Patriotism and propaganda lower the cost of enforcing compliance because they lead citizens to believe that they should comply.

https://doi.org/10.1017/9781009323178.010 Published online by Cambridge University Press

10 Implications for Democracy

After the fall of the Berlin Wall and the dissolution of the Soviet Union, Francis Fukuyama declared that democracy represented the final stage in the evolution of human government.1 The influence of Enlightenment ideas which laid the foundation for the American and French Revolutions in the late eighteenth century established a framework for governance that has continued to spread throughout the world, and has proven its success as a method of governance. As Chapter 1 discussed, the broad term democracy can mean a number of different things. As a set of political institutions, democracy implies that political leaders are elected by citizens, and that government is accountable to those citizens. In a broader sense, democracy is often taken to mean a form of government that carries out the will of the people. The preceding analysis questions the degree to which democratic governments carry out the will of the people, and the degree to which democratic governments are accountable to their citizens. More than just being poorly informed, citizens and voters tend to adopt their policy preferences from the political elite, so it would be more accurate to say that democratic government carries out the will of the elite than the will of the people. Democratic accountability, in theory, comes from elections that give voters the power to determine who holds political power, and to replace those who abuse it. But the political elite are hardly accountable to the masses when the preferences the masses express at the ballot box are those they get from the elite. There are competing elites, to be sure, but citizens have an incentive to be rationally ignorant about public 1

Francis Fukuyama, The End of History and the Last Man (New York: Free Press, 1992).

179

https://doi.org/10.1017/9781009323178.011 Published online by Cambridge University Press

180

Implications for Democracy

policy issues, and often, rationally irrational. Ultimately, they follow their leaders rather than constraining them. This is what leadership means. Leaders are people who can organize those under them to work toward common goals. Individuals have their own individual preferences, and the challenge leaders face is to unify them behind common goals – goals that come from the leaders themselves. Corporate leaders motivate those under them to be productive employees who enhance the profitability of their corporations. Coaches motivate their players to work hard to win their competitions. Political leaders motivate citizens to support common public policy goals, but to do so requires that someone articulate those goals. Successful leaders do this. Leadership can be defined as the ability of an individual or group of individuals to influence and guide followers or other members of an organization. The overall message of this volume depicts political leaders as having the same qualities as leaders in other organizations. They influence and guide their followers. More bluntly, political leaders tell their followers what to think, and if they are effective leaders, their followers embrace the messages of the leaders on whom they anchor. One reason this idea faces resistance with regard to political leaders is that the ideology of democracy depicts those who hold political power as carrying out the will of the people. But one message the leaders themselves pass on to their followers is that the leaders are furthering the interests of those followers. They do so because followers are more likely to accept the public policy preferences of their anchors if they believe that their anchors are constrained by democratic institutions. Elections offer little in the way of accountability when incumbents are overwhelmingly likely to be reelected. Reelection rates for members of the US House of Representatives typically exceed 90 percent, and looking back at the period from 1900 to 2020, thirteen incumbent US presidents were reelected, compared with six who lost their reelection bids. Incumbent advantage also exists in state and local elections, one consequence being that fifteen states have enacted term limits for their legislators, and thirty-five states have term limits for their governors. The Twenty-Second Amendment to the Constitution of the United States establishes term limits for the president. Citizen desire for term limits shows that the masses recognize that the democratic political process does not make elected officials accountable to the electorate, at least to the degree that the electorate believes they should be accountable.

https://doi.org/10.1017/9781009323178.011 Published online by Cambridge University Press

10.1  Academic Analysis

181

10.1  Academic Analysis While this volume’s analysis is ultimately about the way that people form their public policy preferences, it also is a reaction against common assumptions that are made in academic analyses of democratic political institutions. Those analyses take citizen preferences as given and analyze how the political elite respond to those preferences. They assume that candidates and parties design their platforms to conform with citizen preferences, and that governments ultimately are accountable to their citizens. As Rousseau, quoted in Chapter 1, said more than 200 years ago, democratic governments enact policies that further the general will, and that ideology of democracy – that democratic governments act in the interest of the masses – carries over into the twenty-first century. The academic analysis of democratic political institutions saw a shift in the second half of the twentieth century, from describing the mechanisms through which democratic political institutions work toward more analytic models that relate citizen preferences to the public policies government enacts. One common element in these analyses is that they start with the assumption that voters have preferences and examine the way those preferences are aggregated to produce collective decisions and ­public policy. The median voter model has been a standard in that regard since the 1960s. Taking voter preferences as given and extending in a singledimensioned continuum from political left to political right, the model concludes that with majority rule voting, the collective choice of the group is the choice most preferred by the median voter.2 Things become more complicated if the set of issues is multidimensional.3 There may be no stable equilibrium outcome with democratic decision-making in that case.4 One common element in all of these models is that they take voter preferences as given and begin their analysis from that point. 2 3

4

Anthony Downs, An Economic Theory of Democracy (New York: Harper & Row, 1957), remains a standard exposition of this model. Spatial voting models extending to multiple dimensions are analyzed by James M. Enelow and Melvin J. Hinich, The Spatial Theory of Voting: An Introduction (Cambridge: Cambridge University Press, 1984). See R. D. McKelvey, “Intransitivities in Multi-Dimensional Voting Models and Some Implications for Agenda Control,” Journal of Economic Theory 12, no. 3 (June 1976), pp. 472–482 and William H. Riker, “Implications from the Disequilibrium of Majority Rule for the Study of Institutions,” American Political Science Review 74, no. 2 (June 1980), pp. 432–446, for discussions.

https://doi.org/10.1017/9781009323178.011 Published online by Cambridge University Press

182

Implications for Democracy

Models of social choice go beyond simply analyzing how voter preferences are aggregated and transformed into public policies to look for ways to aggregate preferences of citizens and voters so that the aggregate outcome best represents the underlying preferences that are being aggregated.5 But again, voter preferences are taken as given and are assumed to be instrumental. The analysis in Chapter 4 discusses the problem seen here. If voters express preferences that are different from the outcomes they would most prefer, voting mechanisms cannot reveal the relationship between voters’ expressed preferences and voter welfare. They cannot reveal the general will, to use Rousseau’s term. To interpret the outcomes of democratic decision-making processes requires an understanding of how citizens form the political preferences they express. The literature on expressive voting has taken the first step by noting that the preferences voters express at the ballot box do not necessarily reflect their preferences for public policy outcomes. Even here, however, the analysis takes preferences as given.6 The recognition that the preferences people express when voting may differ from the outcomes they would prefer entails a significant advance in the understanding of political preferences, but it leaves unanswered the question of how people form their expressive preferences. This volume explains why the expressed political preferences of citizens and voters are derived from the policy preferences of the political elite. This volume takes issue with standard academic models of democratic decision-making that depict the political elite as designing their platforms to correspond with the preferences of citizens and voters. The direction of causation mostly goes in the other direction. Citizens and voters adopt the political preferences of the elites on whom they anchor. The more important implications of this analysis do not lie with academic analyses of democracy, however, but with democratic institutions themselves. Still, academic models of the political process deserve scrutiny because they form the basis for the indoctrination students receive about their democratic governments. Democratic political institutions have a limited ability to act as a check on the exercise of political power by the elite because citizens adopt the political preferences of the elite. 5

6

Kenneth J. Arrow, Social Choice and Individual Values (New Haven, CT: Yale University Press, 1953), is a standard reference for this area of inquiry. See T. Nicolaus Tideman, Collective Decisions and Voting: The Potential for Public Choice (Aldershot, UK: Ashgate Publishing, 2006), for an extensive analysis of different voting mechanisms. This literature was developed from the work of Geoffrey Brennan and Loren Lomasky, Democracy and Decision: The Pure Theory of Electoral Preference (Cambridge: ­Cambridge University Press, 1993).

https://doi.org/10.1017/9781009323178.011 Published online by Cambridge University Press

10.2  Democracy Is Divisive

183

10.2  Democracy Is Divisive The political elite are not of a single mind. They offer competing ideas and competing messages to voters with the hope of winning their support. The nature of democracy dictates that citizens be offered these competing messages, because in electoral politics, some people win and others lose. To challenge one another for positions of political power, politicians must offer competing visions to citizens and voters with the hope that more voters will anchor on their ideas than on those of their competitors. The visions they offer are not clearly formulated policies, as Chapter 3 explained. They are aspirations designed to have emotional appeal, rather than specific public policies. But they must be designed to be divisive at their foundation, to convey the message to citizens and voters that it makes a difference who they vote for, that they would be better off if they anchored this party’s vision rather than the alternatives. Democracy is divisive because electoral politics creates winners and losers. Unlike economic activity, in which people buy and sell only when they expect to benefit, democratic politics, like sports contests and wars, are contests that people want to win, and want others to lose. Government action, by itself, is divisive because one set of rules applies to everyone, and not everyone agrees on which rules should apply. In its simplest form, democracy means that a majority can impose its preferences on a minority. It has been said that “democracy is two wolves and a sheep voting on what to have for dinner,”7 but the forgoing analysis suggests that the reality may be worse than this. Rather than a majority imposing its preferences on a minority, political institutions are designed to allow a minority to impose their preferences on the majority. The elite design public policy, and the masses are compelled to comply. In cases where everyone wants pretty much the same thing, democracy does a good job of finding those common interests, but so would most other methods of collective decision-making. Democracy breaks down when preferences differ. Go back to Chapter 1 and consider the example of the cyclical majority. Preferences in that example were divided over three alternatives, and majority rule voting was unable to identify a collective choice that would be the most preferred outcome for any majority of voters. When preferences differ, any system of government will be d ­ ivisive. Some people will be dissatisfied with the government’s actions. 7

This quip is often attributed to Benjamin Franklin, but there is no direct evidence that Franklin said or wrote it.

https://doi.org/10.1017/9781009323178.011 Published online by Cambridge University Press

184

Implications for Democracy

But democracy, seen as a system of government that carries out the will of the voters, pits some voters against others. This manifests itself in partisan politics. Citizens take their policy preferences from their anchors – the political elite – but democratic politics is a competition in which one group of elites competes against another. While competing elites offer different visions to citizens and voters, the platforms they offer have one thing in common. They envision a government in which they, the elite, gain power. Ultimately, the elite are after power. They want to win over other elites, and after they win, they want to enhance their power over others. As they enlist their followers for support, this pits some citizens against others. The more pervasive that ideology of democracy – that democratic governments carry out the will of the voters – the more intense will be those partisan divisions. Underlying this political competition is the quest for power. Democratic institutions encourage divisiveness, because people competing for political power must tell potential supporters that their ideas are better than those of their competitors, and to be better, their ideas must be different. To support one set of ideas is to oppose the others. It is remarkable that individuals can develop such animosities toward each other over differences in political preferences. Those preferences are expressive, and have no instrumental consequences. How one person votes has no impact on election outcomes, or on public policy more generally. If friends have different political preferences, people see this as a reason to look for a different set of friends. Why should one person care about another person’s political preferences when those preferences have no effect on any outcomes? But tribal instincts take hold, and in the twenty-first century it has become increasingly ­common for people to refuse to associate with others who have ­different political preferences.

10.3  The American Founders Did Not Create a Democracy The United States is often regarded as a prototypical democracy – the first democratic government established in the modern age. To express skepticism about democracy seems anti-American. But the American Founders did not design their government to be a democracy, if that term is to mean a government that carries out the will of the people. The government’s powers were designed to be constitutionally limited. The government was given limited and enumerated powers, and was prohibited from exercising powers not explicitly given to it by the Constitution. The Tenth

https://doi.org/10.1017/9781009323178.011 Published online by Cambridge University Press

10.3  The American Founders Did Not Create a Democracy

185

Amendment to the Constitution says in its entirety, “The powers not ­delegated to the United States by the Constitution, or prohibited by it to the states, are reserved to the states respectively, or to the people.” Neither the Declaration of Independence nor the Constitution of the United States describe the new American government as a democracy. The Declaration of Independence is largely a list of grievances against the king of England. It lists many ways in which the British government violated the rights of the colonists, and declares that it is the right of the colonists to establish their own government to protect its citizens’ rights – not to carry out the will of the people. Philosophically, the design of the new government followed the ideas of Locke, not Rousseau. Even the use of democratic elections was very limited as the Constitution was written. As originally designed, only members of the House of Representatives were elected by a vote of the citizens. Senators were chosen by their state’s legislature, and members of the judicial branch were appointed by the president and confirmed by the Senate. The president was to be chosen by an electoral college, and in the absence of an electoral majority, by the House of Representatives. The Constitution specifies that state legislatures choose the method by which the state’s presidential electors are selected. Article II, Section 1 of the Constitution says, “Each state shall appoint, in such a manner as the legislature thereof may direct, a number of electors, equal to the whole number of senators and representatives to which the state may be entitled in the Congress.” There is no provision for citizens to vote for president, although the Constitution does not prohibit states from choosing their electors by popular vote. The Founders thought that state legislators would choose electors who would be knowledgeable about presidential candidates, so they could cast informed votes – more informed than if the president were voted on by the general public. They also envisioned a system that insulated selection in presidential elections from popular opinion, but left it up to the state legislatures to specify the exact method by which electors were selected. They thought that most states would have a “favorite son” candidate from their own state, so specified that electors “vote by ballot for two persons, of whom one at least shall not be an inhabitant of the same state with themselves.” This was to prevent favorite son candidates from the largest states from receiving the most votes. The Founders mistakenly thought that in most cases, no candidate would receive votes from a majority of the electors, so the electoral college would serve as a search

https://doi.org/10.1017/9781009323178.011 Published online by Cambridge University Press

186

Implications for Democracy

committee, forwarding the names of the highest vote-getters to the House of Representatives, which would then choose the president. The Constitution specifies “The electors shall meet in their respective states …” and continues, “The Congress may determine the time of choosing the electors, and the day on which they shall give their votes, which shall be the same throughout the United States.” The electors were to meet in their states to prevent them from getting together to collude to select a president, and were required to vote on the same day for that same reason. In an era without a means of rapid communication – no telephones or even telegraphs – electors in different states would find it difficult to negotiate with each other to choose a president. The system was designed with the intention that, in most cases, no candidate would receive an electoral majority and the House of Representatives would ultimately select the president from the list of candidates provided by the electoral college. The system never worked as intended. In the election of 1800, the most common way in which states chose their electors was to have the state legislature make the choice. That election produced a tied electoral vote (in which the House chose Thomas Jefferson), and subsequent presidents won electoral majorities up until 1824. By that time, most states had moved to the present system in which state electors were chosen by popular vote. The last state to move to popular voting for president was South Carolina, in which citizens did not vote for presidential electors until after the Civil War.8 The election of 1824 saw Andrew Jackson get the highest number of electoral votes, but not a majority. The House chose John Quincy Adams as president, in what Jackson supporters called a corrupt bargain. Henry Clay had said he would support Adams if Adams would appoint him as secretary of state, which provided support for Adams’s selection as president. Jackson supporters were outraged, claiming that Jackson got the most electoral votes, so he should be chosen as president. Was this really a corrupt bargain? The Constitution specified that if nobody won an electoral majority, it was up to the House to choose the president, which is what they did. If the intention was to have the highest vote-getter be chosen if nobody received an electoral majority, there would be no need to have the House involved. Just select the candidate with the highest number of electoral votes. The House acted as the 8

More detail on the evolution of the electoral college is found in Randall G. Holcombe, Liberty in Peril: Democracy and Power in American History (Oakland, CA: Independent Institute, 2019), ch. 4.

https://doi.org/10.1017/9781009323178.011 Published online by Cambridge University Press

10.4  Constitutional Limits on Popular Opinion

187

Constitution specified, but after such a long period in which the presidency was won by electoral majorities, the ideology of democracy had made inroads, and Jackson’s supporters argued that he should have been elected because he had the most votes. Corrupt bargain or not, this is an example of how political elites negotiate among themselves to produce political outcomes. These events show that even early in the nation’s history, many Americans thought of American democracy as a method of implementing popular opinion. Immediately after the election of 1824, the modern Democratic party was formed for the purpose of electing Andrew Jackson to the presidency, which they did in 1828. The Whig party was formed to counter the Democrats, and the modern two-party system was established in the United States. With Lincoln’s election in 1860, the Republicans displaced the Whigs, and the partisan politics that the American Founders had hoped to avoid were well underway.

10.4  Constitutional Limits on Popular Opinion As the Founders designed it, the government specified in the Constitution was intended to be only one-sixth democratic, if democracy means that those who hold power are directly accountable to the voters. The government was designed with three branches of government to check and balance the powers of each other. To be able to do that, the branches must be of about equal power. Only one-half of one-third of the government was directly elected – the members of the House of Representatives. Senators were chosen by their state legislatures, the judicial branch was appointed by the president, and the president was intended to be selected by the House of Representatives from a slate of candidates forwarded by the electoral college. This system was purposefully designed to insulate those with political power from direct political pressure from citizens. The US government was not designed to be a democracy, either in the sense of carrying out the will of the people or having those with political power be directly accountable to citizen preferences. It has become more democratic in both senses since the Constitution was written. The original Constitution was designed so that only one-sixth of the federal government would be directly accountable to citizens. In the twenty-first ­century two-thirds are directly accountable through democratic elections. And the prevailing attitude of both the political elite and the masses is that government should be empowered to implement the ­policy p ­ references of the masses.

https://doi.org/10.1017/9781009323178.011 Published online by Cambridge University Press

188

Implications for Democracy

Even this analysis overstates the degree to which the new American government was designed to be accountable to the preferences of its citizens. Once in place, those who held the reins of power were intended to be limited in the ways they could use that power by constitutional constraints that allowed them only the limited and enumerated powers listed in the Constitution. The analysis in this volume emphasizes a major shortcoming in that vision of democratic government. If the masses acquire their public policy preferences from the elite to whom they anchor, it is the elite who design public policy, and the masses follow their leaders. Democratic government is not accountable to its citizens, and is not constrained to act in their interest, if the political preferences expressed by the masses are derived from those of the elite.

10.5  Eroding Constitutional Constraints The idea of limited and enumerated powers granted by the Constitution has fallen by the wayside, and in the twenty-first century the Tenth Amendment has no force. Perhaps the straw that broke the camel’s back on this was the Supreme Court’s 1937 decisions that the Social Security program was constitutional. Without passing judgment on the merits of the program itself, there is nothing in the Constitution to suggest that one of the enumerated powers of the federal government is to run a compulsory retirement program. It would seem that if the Social Security program is constitutional, there is no constitutional limit to the federal government’s powers.9 Reading the words of the Constitution, if Congress and the president wanted to establish a mandatory government retirement program, they should have amended the Constitution to allow it, as was done in 1913 to allow a federal income tax. This example shows how much attitudes toward constitutional constraints changed among both elites and masses between 1913 and 1937. This idea that democratic political institutions channel the public policy decisions of the political elite to implement policies to further the interests of the masses ultimately works against the masses, because the political elite are not an omniscient and benevolent group. They often have only a vague idea of what policies are in the best interest of the masses 9

I discuss the constitutionality of the Social Security program further in Liberty in Peril, ch. 10.

https://doi.org/10.1017/9781009323178.011 Published online by Cambridge University Press

10.6  Checks and Balances

189

because they tend to see things through their own eyes. And, following human nature, everyone is inclined to act to further their own interests. When Charles Wilson, then CEO of General Motors, was nominated by President Eisenhower to be secretary of defense in 1953, he was asked in his confirmation hearings whether his substantial holdings of General Motors stock would present a conflict of interest if he were appointed secretary. He famously replied that he thought that what was good for the nation was good for General Motors, and vice versa. Surely he was sincere in his statement, but this illustrates how the elite view the public interest from their own personal point of view. The elite are not unique in this regard. The idea of the American Founders to create a government with enumerated and constitutionally limited powers was designed to constrain both elites and masses from using the government to further their own interests. The policy menu, as originally designed, was not open-ended, but rather was limited by the enumerated powers given to the government in the Constitution. The twenty-first-century ideology of democracy gives legitimacy to anything the political elite does, and thus expands their latitude to undertake the public policy actions they prefer. This is what the Founding Fathers wanted to prevent by creating a government with enumerated and constitutionally limited powers.

10.6  Checks and Balances To control the power of political elites, the Founders built into the Constitution a system of checks and balances. The checks that are most celebrated are the three branches of government – legislative, executive, and judicial – that check and balance each other. For these checks to be effective, each of the three branches would have to be of about equal power, and the system has worked tolerably well for two and a half centuries. The balance of power has been threatened by an executive branch that has grown stronger over time,10 and by increasing judicial activism in which courts go beyond adjudicating cases to mandating policies.11 10 11

For a discussion of the growing power of the presidency, see Arthur M. Schlesinger, Jr., The Imperial Presidency (Boston: Houghton Mifflin, 1973). Schlesinger was also instrumental in calling attention to judicial activism. He originated the term in Arthur M. Schlesinger, Jr., “The Supreme Court, 1947,” Fortune (January 1947), pp. 202–208. A more complete discussion of judicial activism appears in Keenan D. Kmiec, “The Origin and Current Meanings of ‘Judicial Activism,’” California Law Review 92, no. 5 (October 2004), pp. 1441–1477.

https://doi.org/10.1017/9781009323178.011 Published online by Cambridge University Press

190

Implications for Democracy

If the executive and judicial branches have gained power, this suggests that the legislature has lost power, in a relative sense anyway. That is likely true, but it has not lost power in an absolute sense because the constitutional constraints imposed on the federal government have eroded over the centuries, giving the federal government more power overall. The checks and balances designed into government go deeper than this. As originally designed, senators were chosen by their state legislatures and represented the interests of their state governments. The passage of legislation requires the approval of both the House and Senate, which originally meant that it had to meet with the approval of the representatives of the people (in the House) and the representatives of state governments (in the Senate). That created a higher bar for the passage of legislation, and provided a mechanism by which the state governments could check the power of the federal government.12 This check was eliminated by the Seventeenth Amendment to the Constitution, ratified in 1913, that mandated the election of senators by popular vote. As originally designed, the interests of the people and the interests of the state governments could check and balance each other within the Congress, but after the passage of the Seventeenth Amendment, both the House and Senate represented the interests of the people, weakening the ability of state governments to check the power of the federal government, but increasing the concept of a democratic government accountable to the voters. The power of the state governments to check and balance the power of the federal government was reinforced by the Constitution’s Tenth Amendment that limited the federal government to exercise only those powers enumerated in the Constitution, reserving other powers to the states and to the people. This check has also eroded over time. Originally, the federal government was designed as a federation of states, which placed the states above both the federal and local governments in the hierarchy of power. That ceased to be true after the Civil War, which firmly established the supremacy of federal government power over that of the states. The analysis in this volume offers a compelling reason to structure a government so that the power of some who hold the reins of power is checked and balanced by the power of others. If the policy preferences of the masses are derived from those of the political elite, and if the masses 12

This mechanism is discussed and explained in James M. Buchanan and Gordon Tullock, The Calculus of Consent (Ann Arbor: University of Michigan Press, 1962), ch. 16.

https://doi.org/10.1017/9781009323178.011 Published online by Cambridge University Press

10.6  Checks and Balances

191

tend to be rationally ignorant and express policy preferences that differ from their instrumental interests, the masses will be in a poor position to control the power of the political elite. They will be following their leaders, not constraining them. Public policy is designed by the political elite, and the masses have at best a limited ability to influence it. A system of checks and balances creates a mechanism whereby subsets of the political elite check and balance the powers of each other. An effective system of checks and balances requires more than just a division of powers. It requires that the implementation of public policy take place only through cooperative effort among different groups of elites, each of which have an incentive to guard their political powers against encroachment by the others.13 Constraints on the exercise of political power come from what political scientists refer to as veto players: those who have the power to constrain the political actions of others who have political power.14 If political power is divided, and if those who hold it guard their own sphere of power, abuses in the use of power by some can be checked and balanced by the power of others. The powerless cannot control the powerful, even if the powerless far outnumber the powerful. To constrain abuses of power by the political elite, some members of the elite must be in a position to constrain the use of power by others. The Constitution of the United States requires legislation to be approved by both houses of Congress and the president, who can veto legislation.15 Congress can override a presidential veto with a two-thirds majority, but legislation can then be challenged in the courts, which have the power to determine if it is permitted under the Constitution. Legislation is passed by Congress, but it is executed by the executive branch. Not only do the branches of government have separate powers, they also have spheres

13

14

15

See Geoffrey Brennan and Alan Hamlin, “A Revisionist View of the Separation of ­Powers,” Journal of Theoretical Politics 6 (1994), pp. 345–368, for an elaboration of this idea. George Tsebelis, Veto Players: How Political Institutions Work (Princeton: Princeton University Press, 2002). Also see his “Veto Players and Institutional Analysis,” Governance: An International Journal of Policy and Administration 13, no. 4 (October 2000), pp. 441–474. Veto players may want to constrain their responses to enhance their ability to strike a favorable bargain, by delegating their veto power. See Navin Kartik, Andreas Kleiner, and Richard Van Weelden, “Delegation in Veto Bargaining,” American Economic Review 111, no. 12 (December 2021), pp. 4046–4087, for a discussion. See Charles M. Cameron, Veto Bargaining: Presidents and the Politics of Negative Power (Cambridge: Cambridge University Press, 2000), for a discussion of the president’s use of veto power as a check on the power of others.

https://doi.org/10.1017/9781009323178.011 Published online by Cambridge University Press

192

Implications for Democracy

of influence that provide the incentive to guard against encroachment from the other branches. The first goal of the powerful is to protect their power. The design of checks and balances in the Constitution give the political elite certain powers, and provides the incentive for those who have power to preserve it by preventing other elites from taking it. The basic logic of checks and balances rests on the recognition that it is the elite who make public policy, so an effective way to check their power is to design a system in which subsets of the elite check and balance the powers of other subsets.

10.7  Constraining Leviathan Three main mechanisms have been suggested as ways to constrain the power of government: democratic oversight that gives citizens the ultimate power over their governments, constitutional limits on government powers, and checks and balances. An examination of democratic institutions suggests that checks and balances are necessary, and the other two mechanisms will be ineffective in the absence of checks and balances. Democratic oversight is the mechanism that underlies the romantic notion of democratic government. In this romantic view of democracy, political elites are accountable to the citizens who elect them, and the democratic process is designed so that government responds to the will of its citizens. This volume is not unique in seeing the shortcomings of that romantic notion of democracy. Democratic oversight is ineffective because citizens realize that as individuals, they have no influence over the activities of government, or over who is elected to hold the reins of power in a democracy. Even though ultimately all the votes taken together determine an election’s outcome, no one vote will be decisive. Individuals realize that they have no political power, so voters tend to be rationally ignorant, and derive their political preferences from many factors other than what is in their best interest and the best interest of others. It bears emphasizing that the powerless cannot control the powerful, even if the powerless far outnumber the powerful. The masses cannot control the power of the elite. Democratic oversight is not an effective mechanism for constraining Leviathan government. Democratic elections offer a good mechanism for deciding who can exercise the powers of government, and of peacefully replacing those in power if there is general dissatisfaction with them. But even when this happens, one set of elites replaces another.

https://doi.org/10.1017/9781009323178.011 Published online by Cambridge University Press

10.8  Competition among Elites

193

Constitutional constraints limit the powers of government, if they can be enforced. The specification of limited and enumerated powers for government was one of the principles embodied in the Constitution of the United States.16 The problem lies in designing a mechanism to enforce those limits. If democratic oversight is ineffective, a constitution can list limited and enumerated powers, but there is no guarantee that those who hold power will remain within those limits. Indeed, looking at the twenty-first-century US government, its activities spill well beyond those enumerated in the Constitution. Neither democratic institutions nor constitutional constraints have the ability, by themselves, to constrain government power. That points toward the role of checks and balances. If political power is exercised by the elite, a division of power and oversight among different groups of elites offers the best mechanism for enforcing constitutional constraints. Democratic oversight and constitutional constraints are desirable mechanisms, to be sure, but they cannot work in the absence of checks and balances – a system in which the power of some members of the elite checks and balances the power of others. Adolf Hitler was democratically elected. Vladimir Putin was democratically elected. Without the presence of other elites who can exercise countervailing power, the desire for power on the part of those who have it overrides everything else.

10.8  Competition among Elites The logic of checks and balances is that they create a competition among elites. The masses have no effective power to control public policy. Realizing this, they are rationally ignorant and even rationally irrational. The preferences they express when they vote are often not representative of the political outcomes they would prefer. Their political preferences are expressive, not instrumental, and ultimately are derived from the preferences of the elite. They are following their leaders, not constraining them. The idea that the political elite are accountable to the masses is a fiction. The elite exercise power. The masses are forced to comply with it. While it is true that those who hold political power can lose it at the ballot box, the result is that one set of elites replaces another. Political elites are power seekers. Competing members of the elite do offer alternatives 16

An extended defense of constitutional constraints on the powers of government is given in Geoffrey Brennan and James M. Buchanan, The Reason of Rules: Constitutional Political Economy (Cambridge: Cambridge University Press, 1986).

https://doi.org/10.1017/9781009323178.011 Published online by Cambridge University Press

194

Implications for Democracy

to voters, but the alternatives are to choose one set of elites or another. The first goal of all members of the political elite is to maintain and increase their power. Competition among elites has its downsides. It can create divisiveness among the masses as they choose to anchor with one set of elites versus another. The upside is that elites have the incentive to jealously guard the power they have and always want to acquire more. In political competition, the way they do that is to constrain the power of their competitors. The benefits of democratic elections are obvious but limited, because after the election, the winners have the power. If government institutions are designed so that power is divided, the power of some members of the elite can constrain the actions of others. The necessity of checks and balances to constrain the abuse of power is clear once one recognizes the limited ability of the masses to control the elite. The romantic notion of democracy as a system in which the political elite are accountable to the masses works to the advantage of the elite, because it presents the appearance of a government constrained to act in the public interest. This is a vision of democracy the elite have an incentive to promote. They paint those who do not buy into that vision as unpatriotic. Thus, there is good reason to look back at the Founders’ vision of American government to recognize that they did not intend to create a democracy in this sense. They intended to design a government with limited and enumerated powers, and a system in which some elites check and balance the power of others.

10.9  Political Preferences and Public Policy The main conclusion of this volume is that the public policy preferences of the masses are derived from the political elite. This stands in contrast to much academic analysis of democratic processes, which assumes that voters have preferences and that candidates and parties design their platforms to correspond with the preferences of the voters. If citizens and voters adopt the policy preferences offered to them by the political elite, this implies that democratic institutions give citizens very little ability to constrain the actions of those who hold political power. The masses support the public policies the elites on which they anchor tell them they should support. The political elite are, for the most part, unconstrained by the citizens who elect them. The role of voters, in the aggregate, is to choose which members of the elite exercise the power of government.

https://doi.org/10.1017/9781009323178.011 Published online by Cambridge University Press

10.9  Political Preferences and Public Policy

195

The powerlessness of individual citizens is a consequence of democratic institutions in which a single individual’s political preferences have no impact on public policy outcomes. Individuals know their one vote will not affect an election outcome, so they tend to be rationally ignorant about public policy issues. Their political preferences are purely expressive, because the political choices they make have no effect on political outcomes. The same people will be elected regardless of who an individual votes for, or whether the individual even votes. The utility people get from their political preferences comes only from having them and expressing them. As a result, individuals may vote for outcomes they would not choose if the choice were theirs alone. They have no incentive to vote for the outcome they would prefer to see, because their vote has no effect on whether their preferred outcome will be chosen. They express preferences for parties, candidates, and policies that make them feel good, and those good feelings tend to be based more on emotion than on the outcomes they might anticipate from alternative political choices. They tend to adopt the political preferences of their family members, friends, and charismatic politicians. Public policy issues are often complex, and it is easier to identify parties and candidates who offer general messages that make them feel good than to identify policies that are good for themselves and their fellow citizens. Unlike shopping in the market, where individuals can choose the items they want to buy, politics makes them choose between one entire set of policy proposals or another. Citizens and voters tend to anchor on a candidate or party, and then support that candidate’s entire market basket of policy proposals. They follow their leaders because it reduces cognitive dissonance when they agree with their anchor’s policy proposals rather than sifting through them and agreeing with some but disagreeing with others. The political preferences they hold come from the elite on whom they anchor. Candidates and parties make it easy for citizens and voters to adopt their policy preferences by offering only vague policy proposals with feelgood messages. Rather than offer voters specific policies, they tell voters things are not as good as they could be, but if they are elected, they will make things better. They tell voters, for example, that climate change is a threat, and if elected, they will pursue policies that will limit greenhouse gas ­emissions – without explaining what policies they would enact to do that. They make vague statements about improving the performance of the economy without offering any specific policies to do that. Their goal is to

https://doi.org/10.1017/9781009323178.011 Published online by Cambridge University Press

196

Implications for Democracy

make voters feel good about supporting them, not to enumerate ­specific policy changes. The ultimate goal of the political elite is to gain and increase their power, and they propose public policies with that goal in mind. Citizens often perceive problems with the status quo, and demand that government do something to address those issues. The political elite happily extend the reach of their power to do so – in response to the popular demand that government do something. The elite suggest to those who anchor on them that government can improve their lives if only they are given the power to do so, and citizens then ask for those interventions suggested to them by the elite. The scope and power of government – of the political elite – thus increases in response to the policy preferences that the masses have adopted from the elites on whom they anchor. The political elite tell people their lives can be better if they give government more power, and the masses follow their leaders, asking for government to take more control over the society in which they live. The power-seeking political elite are happy to comply. Thus, the size and scope of government increases because the masses ask for it.17 The masses ask for it because the elite from whom they get their policy preferences have told them they should.

10.10  The Quest for Power The analysis in this volume paints a pessimistic picture regarding the ability of citizens to work within democratic political institutions to control the power of the political elite. Both academic analysis and conventional wisdom suggest that the political elite design public policies to conform with the policy preferences of the masses, but this method of citizen oversight of government is ineffective, because citizens adopt the policy preferences of the elite. The political elite do not design their policies to conform with the views of the masses; rather, the masses adopt their policy preferences from the elites on whom they anchor. The ability of democratic institutions to control government power is commonly overestimated and is supported by the patriotism and propaganda pushed on the masses by the elite. If there is reason for optimism, it is that the elite are not a homogeneous group. Members of the political elite compete among themselves 17

Citizen desire to have government take more control over their lives is discussed by James M. Buchanan, “Afraid to be Free: Dependency as Desideratum,” Public Choice 124, nos. 1/2 (July 2005), pp. 19–31.

https://doi.org/10.1017/9781009323178.011 Published online by Cambridge University Press

10.10  The Quest for Power

197

for power. A part of that competition is institutionalized in systems of checks and balances within the structure of government, but another part is the competition among different members of the elite to gain positions of power. In elections, some elites compete with others and offer different public policy platforms to voters. There is also competition for appointed positions within government. The ineffectiveness of the masses to control the power of the elite becomes apparent when competition among members of the elite is absent. Hitler, Stalin, Mao, Castro, Chavez, and Putin offer examples. In markets, where exchanges among individuals are voluntary, people have the incentive to look for opportunities for mutual benefit. In politics, some people win while others lose, so individuals have an incentive to look for ways to gain power for themselves by denying it to others. One can think of Toyota and General Motors as competitors, but they do not interact with one another. They interact with their customers for their mutual benefit. Competition is more descriptive of democratic politics than market activity. In elections, one candidate wins while another loses. In policy-making, views are rarely unanimous, so one side wins while another loses. Citizens and voters are offered alternatives because of competition among members of the political elite. However, their preferences in these alternatives are adopted from the elites on whom they anchor, and they choose their anchors based on expressive and emotional criteria rather than on the instrumental merits of various proposals. As a result, policy proposals are often vaguely formulated and designed for emotional appeal rather than to convey factual information. The common element in the policy proposals from all members of the elite is that they propose to increase the power held by the elite. In a more ideal world – the world depicted in many academic models of democratic decision-making – the competition among members of the elite would move public policy to more closely conform to the desires of the citizens. Even there, academic models depict reasons why democratic institutions fall short even with well-informed voters who have stable preferences, and with politicians who try to win their support. Special interests have an advantage in the political bargaining process, well-connected individuals and groups engage in rent-seeking and regulatory capture, and political institutions are prone to corruption. Those conclusions come from models in which citizens have instrumentally formulated stable preferences. When citizens adopt the policy preferences offered them by the political elite, the illusion that citizens

https://doi.org/10.1017/9781009323178.011 Published online by Cambridge University Press

198

Implications for Democracy

are able to constrain the power of the elite through democratic political institutions must fade. If voters adopt the policy preferences of their elite anchors, as this volume concludes, voters do at least have choices among anchors. One can hope that the more persuasive anchors propose policies that conform more closely with the instrumental preferences of citizens. Despite competition among members of the political elite, the one thing they all have in common is the quest for power. They would not be competing in the political arena otherwise. The aspiration to hold a government position is the aspiration to exercise power over others, even for people with the best of intentions. The thought that government could help the least fortunate in society by providing them with health care, education, adequate nutrition, or housing assistance is the thought that using the force of government to take from some to give to others would be a desirable policy. Without debating the merits of such policies, those who support them advocate giving the political elite the power to implement and enforce them. Those who exercise the power of government are able to do so because the masses demand it. People perceive problems and demand that the government do something to address those problems. To do something means expanding the scope and power of government. Where do the masses get their policy preferences for government intervention? They get them from the political elite. They are following their leaders.

https://doi.org/10.1017/9781009323178.011 Published online by Cambridge University Press

References

Acemoglu, Daron, and James A. Robinson. The Narrow Corridor: States, ­Societies, and the Fate of Liberty. New York: Penguin Press, 2019. Achen, Christopher H., and Larry M. Bartels. Democracy for Realists: Why ­Elections Do Not Produce Responsive Government. Princeton: Princeton Uni­ versity Press, 2016. Ackerlof, Robert. “‘We Thinking’ and Its Consequences.” American Economic Review 106, no. 5 (May 2016): 415–419. Akerlof, George A., and Rachel E. Kranton. “Economics and Identity.” Quarterly Journal of Economics 115, no. 3 (August 2000): 715–753. Alchian, Armen A. “Uncertainty, Evolution, and Economic Theory.” Journal of Political Economy 58, no. 3 (June 1950): 211–221. Anderson, Terry L., and P. J. Hill. The Birth of a Transfer Society. Stanford, CA: Hoover Institution Press, 1980. Ansolabehere, Stephen, and M. Socorro Puy. “Identity Voting.” Public Choice 169, nos. 1/2 (October 2016): 77–95. Arendt, Hannah. The Origins of Totalitarianism. Cleveland, OH: World ­Publishing, 1958. Arrow, Kenneth J. “General Economic Equilibrium: Purpose, Analytic Techniques, Collective Choice.” American Economic Review 64, no. 3 (June 1974): 253–272. Arrow, Kenneth J. Social Choice and Individual Values, 2nd ed. New Haven, CT: Yale University Press, 1963. Arrow, Kenneth J., and Gerard Debreu. “Existence of an Equilibrium for a Com­ petitive Economy.” Econometrica 27, no. 3 (1954): 265–290. Bailyn, Bernard. The Ideological Origins of the American Revolution. Cambridge, MA: Harvard University Press, 1967. Barber, Michael, and Jeremy C. Pope. “Does Party Trump Ideology? Disentan­ gling Party and Ideology in America.” American Political Science Review 113, no. 1 (2019): 38–54. Bator, Francis M. “The Simple Analytics of Welfare Maximization.” American Economic Review 47, no. 1 (March 1957): 22–59. 199

Published online by Cambridge University Press

200

References

Bator, Francis M. “The Anatomy of Market Failure.” Quarterly Journal of Economics 72, no. 3 (August 1958): 351–379. Baumol, William J. Entrepreneurship, Management, and the Structure of Payoffs. Cambridge, MA: MIT Press, 1993. Baumol, William J. “Entrepreneurship: Productive, Unproductive, and Destruc­ tive.” Journal of Political Economy 98, no. 5, Part 1 (October 1990): 893–921. Beasley, Ryan K., and Mark R. Joslyn. “Cognitive Dissonance and Post-Decision Attitude Change in Six Presidential Elections.” Political Psychology 22, no. 3 (September 2001): 521–540. Becker, Gary S., “A Theory of Competition among Pressure Groups for Political Influence.” Quarterly Journal of Economics 98, no. 3 (1983): 371–400. Benabou, Roland, and Jean Tirole. “Identity, Morals and Taboos: Beliefs as Assets.” Quarterly Journal of Economics 126, no. 2 (2011): 805–855. Benabou, Roland, and Jean Tirole. “Mindful Economics: The Production, Con­ sumption, and Value of Beliefs.” Journal of Economic Perspectives 30, no. 3 (Summer 2016): 141–164. Bernheim, Douglas, Luca Braghieri, Alejandro Martinez-Marquina, and David Zuckerman. “A Theory of Chosen Preferences.” American Economic Review 111, no. 2 (February 2021): 720–754. Billings, Stephen B., Eric Chyn, and Kareem Haggag. “The Long-Run Effects of School Racial Diversity on Political Identity.” American Economic Review Insights 3, no. 3 (September 2021): 267–284. Bernheim, Douglas, Luca Braghieri, Alejandro Martinez-Marquina, and David Zuckerman. “A Theory of Chosen Preferences.” American Economic Review 111, no. 2 (February 2021): 720–754. Black, Dunan. The Theory of Committees and Elections. Cambridge: Cambridge University Press, 1958. Boleslavsky, Raphael, and Christopher Cotton. “Information and Extremism in Elections.” American Economic Journal: Microeconomics 7, no. 1 (February 2015): 165–207. Bowen, Howard R. “The Interpretation of Voting in the Allocation of Economic Resources.” Quarterly Journal of Economics 58, no. 1 (November 1943): 27–48. Brennan, Geoffrey, and James M. Buchanan. The Reason of Rules: Constitutional Political Economy. Cambridge: Cambridge University Press, 1986. Brennan, Geoffrey, and Alan Hamlin. “Expressive Voting and Electoral Equilib­ rium.” Public Choice 95, nos. 1/2 (April 1998): 149–175. Brennan, Geoffrey, and Alan Hamlin. “A Revisionist View of the Separation of Powers.” Journal of Theoretical Politics 6, no. 3 (1994): 345–368. Brennan, Geoffrey, and Loren Lomasky. Democracy and Decision: The Pure Theory of Electoral Preference. Cambridge: Cambridge University Press, 1993. Buchanan, James M. “Afraid to Be Free: Dependency as Desideratum.” Public Choice 124, nos. 1/2 (July 2005): 19–31. Buchanan, James M. “Individual Choice in Voting and the Market.” Journal of Political Economy 62, no. 4 (August 1954): 334–343. Buchanan, James M. The Limits of Liberty: Between Anarchy and Leviathan. Chicago: University of Chicago Press, 1975.

Published online by Cambridge University Press

References

201

Buchanan, James M. The Logical Foundations of Constitutional Liberty. ­Indianapolis: Liberty Fund, 1999. Buchanan, James M. “Public Choice: Politics without Romance.” Policy 19, no. 3 (Spring 2003): 13–18. Buchanan, James M. “Public Finance and Public Choice.” National Tax Journal 28, no. 4 (December 1975): 383–394. Buchanan, James M. “The Relevance of Pareto Optimality.” Journal of Conflict Resolution 6, no. 4 (December 1962): 341–354. Buchanan, James M. “What Should Economists Do?” Southern Economic Journal 30, no. 3 (January 1964): 213–222. Buchanan, James M., and Gordon Tullock. The Calculus of Consent: Logical Foundations of Constitutional Democracy. Ann Arbor: University of Michigan Press, 1962. Bursztyn, Leonardo, Georgy Egorov, and Stefano Fiorin. “From Extreme to Mainstream: The Erosion of Social Norms.” American Economic Review 110, no. 11 (November 2020): 3522–3548. Campbell, Angus, Philip E. Converse, Warren E. Miller, and Donald F. Stokes, The American Voter. New York: John Wiley, 1960. Cameron, Charles M. Veto Bargaining: Presidents and the Politics of Negative Power. Cambridge: Cambridge University Press, 2000. Camon, Ziv, and Dan Ariely. “Focusing on the Forgone: How Value Can Appear So Different to Buyers and Sellers.” Journal of Consumer Research 23, no. 3 (December 2000): 360–370. Cantoni, Enrico, and Vincent Pons. “Does Context Outweigh Individual Char­ acteristics in Driving Voting Behavior? Evidence from Relocations within the United States.” American Economic Review 112, no. 4 (April 2022): 1226–1272. Caplan, Bryan. The Myth of the Rational Voter: Why Democracies Choose Bad Policies. Princeton: Princeton University Press, 2007. Carvalho, Jean-Paul. “Identity-Based Organizations.” American Economic Review 106, no. 5 (May 2016): 410–414. Chakraborty, Archman, Parikshit Ghosh, and Jaideep Roy. “Expert-Captured Democracies.” American Economic Review 110, no. 6 (June 2020): 1713–1751. Chen, Stephanie, and Oleg Urminsky. “The Role of Causal Beliefs in Political Identity and Voting.” Cognition 188, no. 1 (July 2019): 27–38. Coase, Ronald H., “The Problem of Social Cost.” Journal of Law & Economics, 3 (1960): 1–44. Cohen, Geoffrey L. “Party over Policy: The Dominating Impact of Group Influ­ ence over Political Beliefs.” Journal of Personality and Social Psychology 85, no. 5 (2003): 808–822. Conover, Pamela Johnston. “The Influence of Group Identifications on Politi­ cal Perception and Evaluation.” Journal of Politics 46, no. 3 (August 1984): 760–785. Coyne, Christopher J., and Abigail R. Hall. Manufacturing Militarism: U.S. Government Propaganda in the War on Terror. Stanford, CA: Stanford University Press, 2021. Coyne, Christopher J., and Abigail R. Hall. Tryanny Comes Home: The Domestic Fate of U.S. Militarism. Stanford, CA: Stanford University Press, 2018. Delgado, Richard, and Jean Stefancic. Critical Race Theory: An Introduction, 2nd ed. New York: New York University Press, 2012.

Published online by Cambridge University Press

202

References

Diamond, Jared. The World Until Yesterday: What Can We Learn from Traditional Societies? New York: Viking, 2012. Downs, Anthony. An Economic Theory of Democracy. New York: Harper & Row, 1957. Edelman, Murray, The Symbolic Uses of Politics. Urbana: University of Illinois Press, 1964. Elender, Mikael. “Correcting Mistakes: Cognitive Dissonance and Political Atti­ tudes in Sweden and the United States.” Public Choice 153, nos. 1/2 (October 2012): 235–249. Enelow, James M., and Melvin J. Hinich. The Spatial Theory of Voting: An Introduction. Cambridge: Cambridge University Press, 1984. Engelen, Bart. “Solving the Paradox: The Expressive Rationality of the Decision to Vote.” Rationality and Society 18, no. 4 (2006): 419–441. Eubanks, Austin D., Scott Eidelman, Derrick F. Till, David Sparkman, P ­ atrick Stewart, and Robert H. Wicks. “Outcome-Based Dissonance and Morton’s Fork: Evaluative Consequences of Unfavorable Alternatives in the 2016 U.S. Presidential Election.” Journal of Theoretical Social Psychology 4, no. 1 (­January 2020): 21–31. Feddersen, Timothy J., and Wolfgang Pesendorfer. “The Swing Voter’s Curse.” American Economic Review 86, no. 3 (June 1996): 408–424. Fiorina, Morris P. “Further Evidence on the Partisan Consequences of Legisla­ tive Professionalism.” American Journal of Political Science 43, no. 3 (1989): 974–977 Fiorina, Morris P. Retrospective Voting in American National Elections. New Haven, CT: Yale University Press, 1981. Folsom, Burton W., Jr. The Myth of the Robber Barons. Reston, VA: Young America’s Foundation, 1991. Frank, Robert H. Choosing the Right Pond: Human Behavior and the Quest for Status. New York: Oxford University Press, 1985. Frank, Robert H. Luxury Fever: Weighing the Cost of Excess. Princeton: Princ­ eton University Press, 2000. Fukuyama, Francis. The End of History and the Last Man. New York: Free Press, 1992. Gal, David. “A Psychological Law of Inertia and the Illusion of Loss Aversion.” Judgment and Decision Making 1, no. 1 (July 2006): 23–32. Galbraith, John Kenneth. American Capitalism: The Concept of Countervailing Power. Boston: Houghton Mifflin, 1952. Galbraith, John Kenneth. The Anatomy of Power. Boston: Houghton Mifflin, 1983. Gilens, Martin. “Political Ignorance and Collective Policy Preferences.” American Political Science Review 95, no. 2 (June 2001): 379–396. Gino, Francesca, Michael I. Norton, and Roberto A. Weber. “Motivated Bayes­ ians: Feeling Moral while Acting Egotistically.” Journal of Economic Perspectives 30, no. 3 (Summer 2016): 189–212. Golman, Russell, George Loewenstein, Karl Ove Moene, and Luca Zarri. “The Preference for Belief Consonance.” Journal of Economic Perspectives 30, no. 3 (Summer 2016): 165–188.

Published online by Cambridge University Press

References

203

Gordon, Thomas, and John Trenchard. Essays on Liberty, Civil and Religious, and Other Important Subjects, Ronald Hamowy, ed. Indianapolis: Liberty Fund, 1995. Graaf, J. de V. Theoretical Welfare Economics. London: Cambridge University Press, 1957. Hamlin, Alan, and Colin Jennings. “Expressive Political Behaviour: Foundations, Scope, and Implications.” British Journal of Political Science 41, no. 3 (July 2011): 645–670. Hayek, Friedrich A. “The Non-Sequitur of the ‘Dependence Effect.’” Southern Economic Journal 27, no. 4 (April 1961): 346–348. Hayek, Friedrich A. The Road to Serfdom. Chicago: University of Chicago Press, 1944. Hicks, John R. Value and Capital: An Inquiry into Some Fundamental Principles of Economic Theory. Oxford: Oxford University Press, 1939. Higgs, Robert. Depression, War, and Cold War: Challenging the Myths of Conflict and Prosperity. Oakland, CA: Independent Institute, 2006. Higgs, Robert. Crisis and Leviathan: Critical Episodes in the Growth of American Government. Oxford: Oxford University Press, 1987. Hillman, Arye L. “Expressive Behavior in Economics and Politics.” European Journal of Political Economy 26, no. 4 (December 2010): 403–418. Hobbes, Thomas. Leviathan. New York: E.P. Dutton, 1950 [orig. 1651]. Hochman, Harold M., and James D. Rodgers. “Pareto Optimal Redistribution.” American Economic Review 59, no. 4, Part 1 (September 1969): 542–557. Holcombe, Randall G. Advanced Introduction to Public Choice. Cheltenham, UK: Edward Elgar, 2016. Holcombe, Randall G. “Contractarian Ideology and the Legitimacy of Govern­ ment.” Journal of Institutional Economics 17, no. 3 (June 2021): 379–391. Holcombe, Randall G. Coordination, Cooperation, and Control: The Evolution of Economic and Political Power. London: Palgrave-Macmillan, 2021. Holcombe, Randall G. An Economic Analysis of Democracy. Carbondale: South­ ern Illinois University Press, 1985. Holcombe, Randall G. Liberty in Peril: Democracy and Power in American History. Oakland, CA: Independent Institute, 2019. Holcombe, Randall G. Political Capitalism: How Economic and Political Power Is Made and Maintained. Cambridge: Cambridge University Press, 2018. Holcombe, Randall G. “Political Incentives for Rent Creation.” Constitutional Political Economy 28, no. 1 (March 2017): 62–78. Holcombe, Randall G. “Populism: Promises and Problems.” The Independent Review 26, no. 1 (Summer 2021): 27–37. Holcombe, Randall G. “Principles and Politics: Like Oil and Water.” Review of Austrian Economics 22, no. 2 (June 2009): 151–157. Holcombe, Randall G., and Robert J. Gmeiner. “Interest Group Support for NonGroup Issues.” Constitutional Political Economy 29, no. 3 (September 2018): 303–316. Ikeda, Sanford, Dynamics of the Mixed Economy: Toward a Theory of Interventionism. London: Routledge, 1997.

Published online by Cambridge University Press

204

References

Kahneman, Daniel, Jack L. Knetsch, and Richard H. Thaler. “Experimental Tests of the Endowment Effect and the Coase Theorem.” Journal of Political Economy 98, no. 6 (December 1990): 1325–1348. Kahneman, Daniel, Jack L. Knetsch, and Richard H. Thaler. “Anomalies: The Endowment Effect, Loss Aversion, and Status Quo Bias.” Journal of Economic Perspectives 5, no. 1 (Winter 1991): 193–206. Kagel, John H., Raymond C. Battalio, Howared Rachlin, Leonard Green, Robert L. Basmann, and W. R. Klem. “Experimental Studies of Consumer Demand Behavior Using Laboratory Animals.” Economic Inquiry 13, no. 1 (March 1975): 22–38. Kartik, Navin, Andreas Kleiner, and Richard Van Weelden. “Delegation in Veto Bargaining.” American Economic Review 111, no. 12 (December 2021): 4046–4087. Keynes, John Maynard. The Economic Consequences of the Peace. New York: Harcourt Brace and Howe, 1920. Klein, Dominik. “Expressive Voting, Graded Interests, and Participation.” Public Choice 188, nos. 1/2 (July 2021): 221–239 Kmec, Keenan D. “The Origin and Current Meanings of ‘Judicial Activism.’” California Law Review 92, no. 5 (October 2004): 1441–1477. Kohn, Meir, “Value and Exchange.” Cato Journal 24, no. 3 (Fall 2004): 303–339. Kolko, Gabriel. Railroads and Regulation, 1877–1916. Princeton: Princeton Uni­ versity Press, 1965. Kolko, Gabriel. The Triumph of Conservatism: A Reinterpretation of American History, 1900–1916. New York: The Free Press, 1963. Kranton, Rachel E. “Identity Economics 2016: Where Do Social Norms Come From?” American Economic Review 106, no. 5 (May 2016): 405–409. Krueger, Anne O., “The Political Economy of the Rent-Seeking Society,” American Economic Review 64, no. 3 (June 1974): 291–303. Larcinese, Valentino. “Does Political Knowledge Increase Turnout? Evidence from the 1997 British General Election.” Public Choice 131, nos. 3/4 (June 2007): 387–411. Layman, Geoffrey C., Thomas M. Carsey, John C. Green, Richard Herrera, and Rosalyn Cooperman. “Activists and Conflict Extension in American Party Politics.” American Political Science Review 104, no. 2 (May 2010): 324–346. Leeson, Peter T, “Better Off Stateless: Somalia before and after Government Col­ lapse,” Journal of Comparative Economics 35, no. 4 (2007): 689–710. Lenz, Gabriel L. Follow the Leader? How Voters Respond to Politicians’ Policies and Performance. Chicago: University of Chicago Press, 2012. Levine, David K., and Andrea Mattozzi. “Voter Turnout with Peer Punishment.” American Economic Review 110, no. 10 (October 2020): 3298–3314. Lindsay, James, and Helen Pluckrose. Cynical Theories: How Activist Scholarship Made Everything about Race, Gender, and Identity. Durham, NC: Pluckstone Publishing, 2020. Locke, John. Two Treatises of Government. Cambridge: Cambridge University Press, 1960 [orig. 1690].

Published online by Cambridge University Press

References

205

Marx, Karl, and Friedrich Engels. The Communist Manifesto. New York: ­International Publishers, 1948. Maslow, Abraham H. “A Theory of Human Motivation.” Psychological Review 50, no. 4 (July 1943): 370–396. McChesney, Fred S. “Rent Extraction and Rent Creation in the Economic Theory of Regulation.” Journal of Legal Studies 16, no. 1 (January 1987): 101–118. McChesney, Fred S. Money for Nothing: Politicians, Rent Extraction, and ­Political Extortion. Cambridge: Harvard University Press, 1997. McKelvey, R.D. “Intransitivities in Multi-Dimensional Voting Models and Some Implications for Agenda Control.” Journal of Economic Theory 12, no. 3 (June 1976): 472–482. McMurray, Joseph. “Ideology as Opinion: A Spatial Model of Common-Value Elections.” American Economic Journal: Microeconomics 9, no. 4 (November 2017): 108–140. McMurray, Joseph. “The Paradox of Information and Voter Turnout.” Public Choice 165, nos. 1/2 (2015): 13–23. Medema, Steven G. “The Coase Theorem at Sixty.” Journal of Economic Literature 58, no. 4 (December 2020): 1045–1128. Mill, John Stuart. On Liberty, People’s Edition. London: Longmans, Green, and Co., 1913. Mills, C. Wright. The Power Elite. New York: Oxford University Press, 1959. Mises, Ludwig von. Interventionism: An Economic Analysis. Indianapolis: Lib­ erty Fund, 2011. Mullainathan, Sendhil, and Ebonya Washington. “Sticking with Your Vote: Cognitive Dissonance and Political Attitudes.” American Economic Journal: Applied Economics 1, no. 1 (2009): 86–111. Murray, Charles. Losing Ground: American Social Policy, 1950–1980. New York: Basic Books, 1984. Niskenen, William A. Bureaucracy and Representative Government. Chicago: Aldine-Atherton, 1971. Nordnaus, William D. “The Political Business Cycle.” Review of Economic Studies 42, no. 2 (April 1975): 169–190. Nyhan, Brendan. “Facts and Myths about Misperceptions.” Journal of Economic Perspectives 34, no. 3 (Summer 2020): 220–236. Olson, Mancur, Jr. The Logic of Collective Action: Public Goods and the Theory of Groups. Cambridge, MA: Harvard University Press, 1965. Olson, Mancur, Jr. The Rise and Decline of Nations. New Haven, CT: Yale Uni­ versity Press, 1982. Ortoleva, Pietro, and Erik Snowberg. “Overconfidence in Political Behavior.” American Economic Review 105, no. 2 (February 2015): 504–535. Packard, Vance. The Hidden Persuaders. London: Longmans, Green and Co., 1957. Pareto, Vilfredo. The Mind and Society: A Treatise on General Sociology. New York: Dover Publications, 1963. Peacock, Alan T., and Jack Wiseman. The Growth of Public Expenditures in the United Kingdom. Princeton: Princeton University Press, 1961.

Published online by Cambridge University Press

206

References

Persson, Torsten, and Guido Tabellini. Political Economics: Explaining ­Economic Policy. Cambridge, MA: MIT Press, 2000. Pinker, Steven. The Better Angels of Our Nature: Why Violence Has Declined. New York: Viking, 2011. Podhoretz, Norman. Why Are Jews Liberals? New York: Doubleday, 2009. Riker, William H. “Implications from the Disequilibrium of Majority Rule for the Study of Institutions.” American Political Science Review 74, no. 2 (June 1980): 432–446. Robbett, Andrea, and Peter Hans Matthews. “Partisan Bias and Expressive ­Voting.” Journal of Public Economics 157, no. 1 (January 2018): 107–120. Robbins, Lionel. The Nature and Significance of Economic Science. London: Macmillan, 1932. Rolf, Meredeth. Voter Turnout: A Social Theory of Political Participation. Cam­ bridge: Cambridge University Press, 2012. Romer, Thomas, and Howard Rosethal. “Political Resource Allocation, ­Controlled Agendas, and the Status Quo.” Public Choice 33, no. 4 (1978): 27–43. Rothbard, Murray N. The Ethics of Liberty. Atlantic Highlands, NJ: Humanities Press, 1982. Rothbard, Murray N. The Progressive Era. Auburn, AL: Ludwig von Mises Insti­ tute, 2017. Rothstein, Bo. “Epistemic Democracy and the Quality of Government.” European Politics and Society 20, no. 1 (2019): 16–31. Rousseau, Jean Jacques. The Social Contract, or Principles of Political Right. Translated by G. D. H. Cole. www.constitution.org/jr/socon.htm. 1762. Russell, Bertrand. “What Desires Are Politically Important?” Nobel Lecture, December 11, 1950, found at www.nobelprize.org/prizes/literature/1950/ russell/lecture. Samuelson, Paul A. Foundations of Economic Analysis. Cambridge, MA: Har­ vard University Press, 1947. Samuelson, Paul A. “A Note on the Pure Theory of Consumer’s Behavior.” Economica n.s. 5, no. 17 (February 1938): 61–71. Schofield, Norman. “Instability of Simple Dynamic Games.” Review of Economic Studies 45, no. 3 (October 1978): 575–594. Schumpeter, Joseph A., Capitalism, Socialism, and Democracy, 2nd ed. London: George Allen & Unwin, 1947. Schlesinger, Arthur M., Jr., “The Supreme Court, 1947.” Fortune (January 1947): 202–208. Schweizer, Peter. Extortion: How Politicians Extract Your Money, Buy Votes, and Line Their Own Pockets. Boston: Houghton Mifflin, 2013. Shayo, Moses. “A Model of Social Identity with an Application to Political Econ­ omy: Nation, Class, and Redistribution.” American Political Science Review 103, no. 2 (May 2009): 147–174. Sinclair, Upton. The Jungle. New York: Doubleday, Page, & Co., 1906. Smith, Adam, An Inquiry into the Nature and Causes of the Wealth of Nations. New York: Modern Library, 1937 [orig. 1776].

Published online by Cambridge University Press

References

207

Snower, Dennis J., and Steven J. Bosworth. “Identity-Driven Cooperation versus Competition.” American Economic Review 106, no. 5 (May 2016): 420–424. Sobel, Russell S. “Testing Baumol: Institutional Quality and the Productivity of Entrepreneurship.” Journal of Business Venturing 23, no. 6 (November 2008): 641–655. Stigler, George J. “The Theory of Economic Regulation.” Bell Journal of Economics and Management Science 2, no. 1 (Spring 1971): 3–21. Stigler, George J., and Gary S. Becker. “De Gustibus Non Est Disputandum.” American Economic Review 67, no. 2 (March 1977): 76–90. Stiglitz, Joseph E. The Price of Inequality: How Today’s Divided Society Endangers the Future. New York: W.W. Norton, 2012. Tideman, Nicolaus. Collective Decisions and Voting: The Potential for Public Choice. Aldershot, UK: Ashgate Publishing, 2006. Tideman, T. Nicolaus, and Gordon Tullock. “A New and Superior Process for Making Social Choices.” Journal of Political Economy 84, no. 6 (December 1976): 1145–1159. Tsebelis, George. Veto Players: How Political Institutions Work. Princeton: Princeton University Press, 2002. Tsebelis, George. “Veto Players and Institutional Analysis.” Governance: An International Journal of Policy and Administration 13, no. 4 (October 2000): 441–474. Tullock, Gordon. “The Charity of the Uncharitable.” Western Economic Journal 9, no. 4 (December 1971): 379–392. Tullock, Gordon, “The Welfare Cost of Tariffs, Monopolies, and Theft,” Western Economic Journal 5, no. 3 (June 1967): 224–232. Tullock, Gordon. “Why So Much Stability?” Public Choice 37, no. 2 (1982): 189–202. Tyran, Jean-Rovert, and Alexander K. Wagner. “Experimental Evidence on Expressive Voting,” Oxford Handbook of Public Choice, vol. 2. Roger D. Con­ gleton, Bernard N. Grofman, and Stefan Voigt, eds. Oxford: Oxford University Press, 2019, ch. 45, 928–940. van Bavel, Jay J., and Andrea Pereira. “The Partisan Brain: An Identity-Based Model of Political Belief.” Trends in Cognitive Psychology 22, no. 3 (March 2018): 213–224. Veblen, Thorstein. The Theory of the Leisure Class. New York: Macmillan, 1899. Wagner, Richard E. Politics as a Peculiar Business: Public Choice in a System of Entangled Political Economy. Cheltenham, UK: Edward Elgar, 2016. Wang, Tianyi. “Media, Pulpit, and Populist Persuasion: Evidence from Father Coughlin.” American Economic Review 111, no. 9 (September 2021): 3064–3092. White, Richard. Railroaded: The Transcontinentals and the Making of Modern America. New York: W.W. Norton, 2011. Williamson, Oliver E. “A Comparison of Alternative Approaches to Economic Methodology.” Journal of Institutional and Theoretical Economics 146, no. 1 (1990): 61–71. Wilson, Woodrow. The State. Boston: D.C. Heath & Company 1889.

Published online by Cambridge University Press

208

References

Wittman, Donald A. The Myth of Democratic Failure. Chicago: University of Chicago Press, 1995. Wittman, Donald A. “Why Democracies Produce Efficient Results.” Journal of Political Economy 97, no. 6 (1989): 1395–1424. Yandle, Bruce. “Bootleggers and Baptists: The Education of a Regulatory Econo­ mist.” Regulation 7, no. 3 (May–June 1983): 12–16. Yeager, Leland B. Ethics as a Social Science: The Moral Philosophy of Social Cooperation. Cheltenham, UK: Edward Elgar, 2001. Yeager, Leland B. “Rights, Contract, and Utility in Policy Espousal.” Cato Journal 5, no. 1 (Summer 1985): 259–294.

Published online by Cambridge University Press

Index

abstention, 25, 103 Acemoglu, Daron, 7 Achen, Christopher H., 82, 92, 94, 99, 115 Adams, John Quincy, 186 advertising in markets, 45 negative, 52 political, 46 agenda control, 68, 132 American Economic Association, 132 American Founders, 185 anarchy, Hobbesian, 3 anchor preferences, 96, 98, 99 confidence in, 102 and elite power, 160 and group identity, 99 and ideology, 100 as instrumental preferences, 101 principled, 102 anchors choice among, 198 choice of, 112 Anderson, Terry, 13 Arendt, Hannah, 148 Army Corps of Engineers, 79 Arrow, Kenneth J., 56, 58 and social choice assumptions, 66 bandwagon effect, 45, 86, 100, 112 Bartels, Larry M., 82, 92, 94, 99, 115 Becker, Gary S., 39, 137 Berlin Wall, 16, 179 Biden, Joe, 133, 171 Bolsonaro, Jait, 149

bootleggers and Baptists, 159 Brennan, Geoffrey, 21, 32, 64 Buchanan, James M., 2, 6, 7, 41, 42, 59, 165 and utility maximization, 75 Burns, Arthur, 132 Bush, George W., 23, 153, 168 California recall elections, 69 Caplan, Bryan, 27, 136 Carnegie, Andrew, 143 Castro, Fidel, 149 Cato’s Letters, 5 Chavez, Hugo, 149 checks and balances, 193 in the Constitution, 189 to control elite power, 191 effect of the Civil War on, 190 necessity of, 194 role of the states, 190 circular reference; See reference, circular civics education, 78 as propaganda, 79 Clinton, Hillary, 149, 170 Coase theorem, 130 applied to politics, 131 Coase, Ronald H., 129 cognitive dissonance, 31, 73, 84, 97, 107, 112, 114 and anchor preferences, 100, 109 and preference formation, 91 reduced by lack of information, 90 and voter information, 85 Cold War, 156 competing elites, 179

209

https://doi.org/10.1017/9781009323178.012 Published online by Cambridge University Press

210

Index

conspicuous consumption, 43, 44 Constitution of the United States, 11 constitutional constraints, 192, 193 corrupt bargain, 186 corruption, 124 COVID pandemic, 156 Coyne, Christopher J., 152 creative destruction, 125, 145 critical race theory, 40, 79 cross-Florida barge canal, 79 cyclical majority, 9, 105 De Beers company, 49 Declaration of Independence, 12, 15 democracy and freedom, 16 ideology of, 1, 12, 15, 17 instability in, 10 legitimacy of, 12 and progressivism, 14 and the public interest, 11 romantic notion of, 194 symbolic value of, 178 in the United States, 184 democratic oversight, 192 Democratic party founding of, 187 dependence effect, 48, 49 derivative preferences, 96, 99, 105 derived from anchors, 107 Diamond, Jared, 119, 163, 172 Dodd–Frank Act, 152 Downs, Anthony, 2, 26, 63, 106 economic elite membership in, 129 electoral college, 41, 62, 185 elites competition among, 193 in sports, 35 Emanuel, Rahm, 156 endowment effect, 80, 107, 112 Enlightenment, 1, 4–6, 19, 161, 178, 179 ethanol mandate, 18, 123, 131, 132 expressive preferences compared with instrumental preferences, 62 formation of, 74 utility from, 72 expressive voters and sports fans, 34 expressive voting, 182 instrumental component of, 33

Father Coughlin, 93 Federal Reserve Bank, 152 Frank, Robert, 44 free riders, 165 Friedman, Milton, 155 Fukuyama, Francis, 179 Galbraith, John Kenneth, 48, 55, 120 General Motors, 189 Glorious Revolution, 162 Gore, Al, 23 government function of, 119 government failure, 59 Hall, Abigail R., 152 Harding, Warren, 155 Hayek, Friedrich A., 50, 119 Hicks, John R., 42 hierarchy of needs, 118 Hill, P.J., 13 Hillman, Arye, 33, 93 Hitler, Adolf, 148, 169 Hobbes, Thomas, 3–5, 161, 175 Hochman, Harold M., 30 Hong Kong, 17 Humphrey, Hubert, 52 incumbency advantages of, 180 instrumental preferences, 64 and altruism, 76 and the political elite, 68, 117 instrumental voting, 23 interest groups, 18, 197 and political power, 128 Interstate Commerce Act, 154 Interstate Commerce Commission, 13 Iran, 17 Jackson, Andrew, 186 Jefferson, Thomas, 186 Jobs, Steve, 143 Johnson, Boris, 149 judicial activism, 189 Kagel, John H., 75 King, Martin Luther, 16 La Follette, Robert, 113 leadership meaning of, 180

https://doi.org/10.1017/9781009323178.012 Published online by Cambridge University Press

Index Leeson, Peter T., 176 Lenin, Vladimir, 149 Lincoln, Abraham, 153, 187 Locke, John, 4–6, 15, 161 compared with Rousseau, 13 Lomasky, Loren, 21, 32, 64 Machiavelli, Niccolo, 47 market failure, 59 market preferences and political preferences, 97 Marx, Karl, 117, 174 Maslow, A. H., 118 mass media and charismatic politicians, 89 median voter model, 8, 63, 106, 112, 181 milker bills, 126 Mill, John Stuart tyranny of the majority, 7 Mills, C. Wright, 117, 123, 173, 174 minimum wage, 76 multidimensional issue space, 105 Munn v. Illinois, 13 Nader, Ralph, 23 Napoleon Bonaparte, 47 National Rifle Association, 128 nationalism, 152 Nature Conservancy, 128 negative advertising, 52 Netanyahu, Benjamin, 149 Nixon, Richard M., 154 normalcy, 155 Obama, Barack, 156 Occupy Wall Street, 16, 117 Ortega, Daniel, 149 Packard, Vance, 45 Pareto, Vilfredo, 21, 47 patriotism, 167 and support for military personnel, 169 peer pressure and voter turnout, 83 Perot, Ross, 113 Pinker, Steven, 163 plurality voting, 8 policy experts and progressivism, 150 and rule of law, 157 political advertising and cognitive dissonance, 88

negative, 88 and voter turnout, 88 political choices compared with market choices, 73 political competition, 144, 146 political elite, 129 and agenda control, 134 barriers to entering, 142 choice among, 193 dependence on, 158 as experts in governance, 151 ideology of, 139 and the ideology of democracy, 17 legitimacy of, 168 membership in, 69 policy preferences of, 137 and power maximization, 151, 177, 184 principles of, 141 and the prisoner’s dilemma, 174 political issue space, 111, 113 political marketplace, 137 political polarization, 114 political power backed by force, 122 centralized, 131 discontinuity in, 127 legitimacy of, 136 purchase of, 123, 143 and rule of law, 156 political preferences and advertising, 87 and candidate charisma, 89 and civics education, 78 compared to preferences for sports teams, 82 and elite power, 140 expressive nature of, 135 and framing effects, 77 instrumental importance of, 82 instrumental justifications for, 77 manipulation of, 47 and mass media, 93 multidimensional, 98 and peer pressure, 82 are predominantly derivative, 108 and retrospective voting, 92 and status quo bias, 78 utility from, 115 political rallies as expressive action, 101 political slogans, 51

https://doi.org/10.1017/9781009323178.012 Published online by Cambridge University Press

211

212

Index

politics adversarial nature of, 146, 147, 149, 171 and compromise, 146 divisiveness of, 183 populism, 148, 162 power countervailing, 121 decentralized, 122 expansion of, 154 and extortion, 126 political compared with economic, 121, 125 types of, 120 power elite, 118 power maximization, 118, 145 as the objective of the political elite, 138 and political elites, 151, 177 and the political elite, 184, 196 and the public interest, 139 and public policy, 198 and ruthless individuals, 119 selection bias for, 145 self-selection for, 118 preference aggregation and the median voter model, 9 preference formation, 74 preference manipulation, 74 preferences expressive. See expressive preferences subjective, 20 principal-agent relationship, 36 principled politicians, 141 principled preferences, 102 prisoners’ dilemma, 166 progressivism, 13 and democracy, 14 propaganda, 168 and legitimacy of government, 173 proportional voting, 8, 113 public choice, 1, 2 compared to social choice, 66 public choice theory, 178 public interest and individual interests, 18 Pure Food and Drug Act, 152 Putin, Vladimir, 90, 145 rational ignorance, 26, 28 rational irrationality, 27 Reagan, Ronald, 139

redistribution, 31 reference, circular; See circular reference regulation capture theory of, 124 purpose of, 119 regulatory capture, 197 rent-seekers and elite power, 140 and political extortion, 140 rent-seeking, 197 retrospective voting, 92 Revolution, American, 5 rights natural, 4 Source of, 4 robber barons, 142 Robbins, Lionel, 41 Robinson, James, 7 Rockefeller, John D., 143 Rodgers, James D., 30 Roosevelt, Franklin D., 153 Roosevelt, Teddy, 154 Rothbard, Murray N., 164 Rousseau, Jean Jacques, 12–14, 17, 58, 165, 181 rule of law, 156 Russell, Bertrand, 120 Samuelson, Paul A., 42 and revealed preference, 75 Sanders, Bernie, 133 Schumpeter, Joseph A., 125 Schweizer, Peter, 126, 137 September 11, 2001, 156 Seventeenth Amendment, 190 Sherman Antitrust Act, 13, 154 Sinclair, Upton, 152 slavery, 4 Smith, Adam, 48, 123 snob effect, 45 social choice, 182 and expressive voting, 66 social choice compared to public choice, 66 social contract, 3–5, 162, 163, 173 and prisoners’ dilemma, 166 as fiction, 164 Somalia, 177 Southern Economic Association, 132 Soviet Union, 179 Stalin, Josef, 148, 169

https://doi.org/10.1017/9781009323178.012 Published online by Cambridge University Press

Index Stigler, George J., 39 Stiglitz, Joseph E., 125, 175 strategic voting, 23, 64 Tenth Amendment, 185, 188, 190 term limits, 180 Thatcher, Margaret, 139 third-party candidates, 24 transaction costs and the political elite, 129 tribalism, 82, 172, 184 in sports, 172 Trump, Donald, 90, 109, 110, 133, 149, 169, 170 as a political anchor, 101 Tullock, Gordon, 6, 29, 106, 116 tyranny of the majority, 7 unintended consequences, 155 Veblen, Thorstein, 43 Veblen, Thorsten, 86 Ventura, Jesse, 32 veto players, 191 Vietnam war, 16 voter apathy, 103 voter preferences and the bandwagon effect, 86 and current conditions, 91 voter turnout, 112 elite preference for, 36

elite support for, 104 and negative advertising, 53 and peer pressure, 83 and political advertising, 88 in presidential election years, 53 voters rational, 37 votes wasted, 24, 25 voting charitable, 30 ethical, 35 ethical obligation for, 36 feedback from, 28 instrumental consequences of, 25 as preference aggregation, 56, 57 utility from, 22, 25, 37, 72 voting models, 61 Voting models and expressive preferences, 67 Wallace, George, 113 Washington, George, 153 Weber, Max, 120 welfare economics, 59 welfare maximization, 58 welfare state, 158 Wilson, Charles, 189 Wilson, Woodrow, 150 Yandle, Bruce, 159

https://doi.org/10.1017/9781009323178.012 Published online by Cambridge University Press

213