The Oxford Handbook of Law and Economics: Volume 1: Methodology and Concepts [online version ed.] 019968426X, 9780199684267

Citation preview

The Oxford Handbook of Law and Economics: Volume 1: Methodology and Concepts

The Oxford Handbook of Law and Economics: Volume 1: Methodology and Concepts   Francesco Parisi The Oxford Handbook of Law and Economics: Volume 1: Methodology and Concepts Edited by Francesco Parisi Print Publication Date: Apr 2017 Subject: Economics and Finance Online Publication Date: May 2017

(p. iv)

Great Clarendon Street, Oxford, OX2 6DP, United Kingdom Oxford University Press is a department of the University of Oxford. It furthers the University’s objective of excellence in research, scholarship, and education by publishing worldwide. Oxford is a registered trade mark of Oxford University Press in the UK and in certain other countries © Oxford University Press 2017 The moral rights of the authors have been asserted Impression: 1 All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, without the prior permission in writing of Oxford University Press, or as expressly permitted by law, by licence or under terms agreed with the appropriate reprographics rights organization. Enquiries concerning reproduction outside the scope of the above should be sent to the Rights Department, Oxford University Press, at the address above You must not circulate this work in any other form and you must impose this same condition on any acquirer Published in the United States of America by Oxford University Press 198 Madison Avenue, New York, NY 10016, United States of America British Library Cataloguing in Publication Data Data available Library of Congress Control Number: 2017935284 ISBN 978–0–19–968426–7 (Volume 1) Page 1 of 2

The Oxford Handbook of Law and Economics: Volume 1: Methodology and Concepts ISBN 978–0–19–968420–5 (Volume 2) ISBN 978–0–19–968425–0 (Volume 3) ISBN 978–0–19–880373–7 (Set) Printed and bound by CPI Group (UK) Ltd, Croydon, CR0 4YY Links to third party websites are provided by Oxford in good faith and for information only. Oxford disclaims any responsibility for the materials contained in any third party website referenced in this work.

Page 2 of 2

List of Figures

List of Figures   Francesco Parisi The Oxford Handbook of Law and Economics: Volume 1: Methodology and Concepts Edited by Francesco Parisi Print Publication Date: Apr 2017 Subject: Economics and Finance Online Publication Date: May 2017

(p. ix)

List of Figures

2.1 The General Transaction Structure 22 11.1 Horizontal Policy Competition 229 11.2 Horizontal Policy Competition with Law Instruments 230 11.3 Two-Dimensional Horizontal Policy Competition with Law 231 11.4 Vertical Policy Competition with Legal Doctrines (Rules and Standards) 233 11.5 Vertical Policy Competition with Law Decision Instruments 236 11.6 Internal Policy Competition with Doctrines 240 (p. x)

Page 1 of 1

List of Tables

List of Tables   Francesco Parisi The Oxford Handbook of Law and Economics: Volume 1: Methodology and Concepts Edited by Francesco Parisi Print Publication Date: Apr 2017 Subject: Economics and Finance Online Publication Date: May 2017

(p. xi)

List of Tables

5.1 Sample Decision Structure 91 17.1 Mention of “Benefit–Cost” or “Cost Benefit” in Legal Cases 356 17.2 Law Review and Related Journals: Number of Articles with the Term “Cost Ben­ efit” or “Benefit Cost” 356 17.3 The Measurement of Benefits and Costs in Terms of Gains and Losses 359 17.4 Present Values of One Million 360 17.5 Basic Steps in Benefit–Cost Analysis 361 17.6 Coleman’s Preference Reversal Example 369 17.7 Certainty Equivalent Rates Within the Burgess–Zerbe Range 374 24.1 An Example of the Doctrinal Paradox 508 (p. xii)

Page 1 of 1

List of Contributors

List of Contributors   Francesco Parisi The Oxford Handbook of Law and Economics: Volume 1: Methodology and Concepts Edited by Francesco Parisi Print Publication Date: Apr 2017 Subject: Economics and Finance Online Publication Date: May 2017

(p. xiii)

List of Contributors

The late Gary Becker

Brian H. Bix University of Minnesota Law School

John Bronsteen Loyola University Chicago School of Law

Christopher Buccafusco Cardozo School of Law, Yeshiva University

Emanuela Carbonara University of Bologna

Giuseppe Dari-Mattiacci University of Amsterdam

Gerrit De Geest Washington University, St. Louis

Page 1 of 4

List of Contributors

David M. Driesen Syracuse University College of Law

Daniel A. Farber University of California, Berkeley

Jonah B. Gelbach Professor of Law, University of Pennsylvania

Werner Güth Max Planck Institute of Economics

Charles A. Holt University of Virginia

Christine Jolls Yale Law School

Jonathan Klick University of Pennsylvania Law School

Robin Paul Malloy Syracuse University College of Law

Jonathan S. Masur University of Chicago Law School

Thomas J. Miceli University of Connecticut

Pam A. Mueller Princeton University

Page 2 of 4

List of Contributors

Janice Nadler Northwestern Pritzker School of Law

Shmuel Nitzan Bar Ilan University

Jacob Paroush Bar Ilan University

Richard Posner United States Seventh Circuit Court of Appeals and University of Chicago Law School (p. xiv)

Shruti Rajagopalan State University of New York

Mario J. Rizzo New York University

Chris William Sanchirico University of Pennsylvania Law School

Sean P. Sullivan Federal Trade Commission

Emerson H. Tiller Northwestern University

Tom R. Tyler Yale Law School

Page 3 of 4

List of Contributors Georg Vanberg Duke University

Viktor Vanberg Albert Ludwigs Universitaet, Freiburg

Stefan Voigt University of Hamburg

Georg von Wangenheim University of Kassel

Tess Wilkinson-Ryan University of Pennsylvania Law School

Donald Wittman University of California, Santa Cruz

Richard O. Zerbe University of Washington

Page 4 of 4

The Future of Law and Economics

The Future of Law and Economics   Gary S. Becker and Richard A. Posner The Oxford Handbook of Law and Economics: Volume 1: Methodology and Concepts Edited by Francesco Parisi Print Publication Date: Apr 2017 Subject: Economics and Finance, Law and Economics, Econometrics, Experimental and Quanti­ tative Methods Online Publication Date: May 2017 DOI: 10.1093/oxfordhb/9780199684267.013.002

Abstract and Keywords This exchange between Judge Posner and Professor Becker — two founding fathers of our discipline — was composed almost three years ago, but the perspective and wisdom of this exchange remains most relevant to this day. Since then, Professor Becker has died; Judge Posner has added a brief remembrance of Becker at the end of this exchange. Keywords: law and economics movement, Chicago school, Gary Becker, Richard Posner

THIS exchange between Judge Posner and Professor Becker—two founding fathers of our discipline—was composed almost three years ago, but the perspective and wisdom of this exchange remains most relevant to this day. Since then, Professor Becker has died; Judge Posner has added a brief remembrance of Becker at the end of this exchange. POSNER: The future of an evolving academic field belongs to the young. They know what their elders know, and their careers depend on their being able to build on existing knowledge in creative ways. The old are likely to be in a defen­ sive crouch, fearing that the young will build their careers in part on rejecting, or at best superseding, the work of their elders. So, in reading what follows: caveat emptor. The modern field of “law and economics” (that is, of economic analysis of law) dates from the 1960s. Until then, Jeremy Bentham’s economic analysis of criminal law having been forgotten, economics was thought relevant to only a few fields of law, all commercial—antitrust law, public utility and common carrier regu­ lation, and tax law. By the end of the 1960s, as a result of articles (and the occa­ sional book) by William Baxter, Gary Becker, Guido Calabresi, Ronald Coase, Harold Demsetz, William Landes, Henry Manne, and others, economics was un­ derstood to be relevant to the entire domain of the law—relevant both to under­ standing the law (positive analysis) and to reforming it (normative analysis). That was half a century ago. In the intervening period the evolution of law and econom­ ics has been shaped by a number of forces: the increased mathematization of eco­ Page 1 of 6

The Future of Law and Economics nomics (including advances in techniques of statistical analysis); the increased availability of statistical data usable in empirical analyses utilizing the latest sta­ tistical techniques, as a result of the computer revolution; the broadening of the scope of economics both conceptually (as in the rise of game theory and the ad­ vent of behavioral economics—the invasion of economics by psychology) and in the areas of human activity that are studied by economists (marriage and divorce, for example); the increased size and “academification” of the legal professoriat; and, related to a number of these developments, increased specialization of acade­ mic law. The early contributors to the field of law and economics were economists and lawyers—not lawyer-economists—and they tended to write across legal do­ mains. So Becker, for example, studied both racial discrimination and criminal law enforcement, and Coase both tort law and communications regulation, and Baxter both patent law and environmental law. Very little of the work of this early period was either mathematical or statistical (or empirical at all). But beginning in the 1970s economists such as Steven Shavell built increasingly sophisticated mathe­ matical models of legal phenomena. It began to be felt that legal training alone would not enable a lawyer to do sophisticated economic analysis of law, and so economic analysis of law increasingly became the province of law professors who had a Ph.D. in economics, as well as of economists specializing in the application of economics to law who did not have a law degree. During the 1970s, economic analysis of law began to permeate legal teaching as well as scholarship, and eco­ nomic consultants and expert witnesses became fixtures of commercial litigation in a variety of fields—in part because lawyers were learning in law school how economics could be used in legal analysis. Most of these consultants and witness­ es were not and are not economic analysts of law, but rather analysts of business practices challenged in litigation. The trend toward increased economic sophisti­ cation in the 1970s, which has continued ever since, has had a side effect of in­ creasing the separation between academic economic analysis of law and the prac­ tice of law. The two-degree scholars generally don’t have time to engage in law practice to any significant extent (often no more than a one-year clerkship with a judge) before beginning their academic careers. The increased formalization of economics makes it difficult for lawyers who do not have training in economics to collaborate with economists or lawyer-economists. Increasingly, economic analysts of law write for each other, in specialized journals, rather than for the larger pro­ fession. Increasingly, indeed, they write not for economic analysts of law as a whole but for economic analysts of the writer’s subspecialty. The expansion of a field leads to the multiplication of its subspecialties. These developments have in­ creased, and will continue to increase, the rigor of economic analysis of law. The search for new worlds to conquer that is a hallmark of a progressive research pro­ gram has already paid off. One example is increased attention to the economics of foreign and international law and, concomitantly, increased exploitation of the op­ portunities that cross-country comparison provides for empirical study. Another example is the empirical study of judicial behavior, where insights from labor eco­ nomics and the economics of organizations are being used to interpret the large Page 2 of 6

The Future of Law and Economics quantities of statistical data that are available (or readily obtainable) concerning the activities of courts and judges. But the gap between academic law and eco­ nomics and the law as it is practiced and administered and created and applied is troublesome. Economic analysis of law has intrinsic intellectual interest (like ju­ risprudence) and is an invaluable component of a modern legal education, but one would also like to see it contribute to the solution of legal problems and the re­ form of our costly and cumbersome legal institutions. And for that the economic analyst needs to understand law from the inside, which no one, however bright, can do without legal experience (though it might be acquired, on the side as it were, after one had begun an academic career) as well as legal training, for law is like a foreign language. And to avoid the amateurishness of underspecialization, there is a pressing need for greater collaboration between law professors with re­ al legal experience and economists or lawyer-economists who have the analytic tools but not the insider’s understanding of the law in action and in its manifold in­ stitutional forms. There is also need for economic analysts of law, whether they are lawyers or economists or lawyer-economists, to interact with, and sometimes collaborate with, economists in economic departments and business schools who may be interested in law and may have special economic skills to bring to its study, an example being Andrei Shleifer of the Harvard economics department. The limited amount of such interaction and collaboration is reflected in the slow reaction of economic analysts of law to the financial crash of September 2008 and the ensuing downward spiral of the economy. The legal profession was deeply in­ volved in the creation of the complex financial instruments that crashed and, of course, in the creation, un-creation, and administration of the regulatory laws and institutions governing finance. Yet about these instruments and practices and reg­ ulations—the Federal Reserve Act, for example, or the debt ceiling, or the euro­ zone—economic analysts of law have been largely silent (though not entirely— think of Lucian Bebchuk at Harvard Law School and the finance group at Colum­ bia Law School), even though the economic crisis is about to enter its fourth year (fifth, if we count from December 2007, when GDP first began to dip). Predictions of the future are almost always just extrapolations. I will conclude in that vein. I expect economic analysis of law to grow in rigor, expand in scope, and becoming increasingly empirical as statistical databases become ever easier to create and analyze. Up to now the ratio of theoretical to empirical economic analysis of law has been very high, in part because theoretical papers can be produced much more quickly than empirical studies, and (unfortunately) number of papers pub­ lished is given undue weight in hiring and tenure decisions. That is a concern and another (which is related however to reluctance to undertake empirical studies) is that economic analysis of law may lose influence by becoming too esoteric, too narrow, too hermetic, too out of touch with the practices and institutions that it studies. BECKER: Posner gives an excellent discussion of the evolution of the field of law and economics, with the glaring omission of his own monumental contributions Page 3 of 6

The Future of Law and Economics that fundamentally helped define the approaches, techniques, and scope of this field. I will go over some of the same ground as he does on its evolution and likely future, and I will also add brief comments on the emerging and exciting subfield of macro law and economics. The first stage of research on law and economics was mainly theoretical. Economists and lawyers used and adapted concepts and analy­ sis from economics to show that legal rules and doctrines often had a clear eco­ nomic rationale, and to show how these concepts illuminated how laws and legal systems affect behavior and the efficiency of economic outcomes. These early studies had an enormous influence on how some lawyers and economists began to think about property, contracts, negotiations, trials and settlements, torts, an­ titrust, corporations and securities, crime, racial and gender discrimination, and other areas of the law. Yet it took a while for these ideas to spread into law schools since academic and other lawyers initially had little exposure to the economic way of thinking. Partly for this reason, considerable opposition developed to many of the ideas espoused by the economists and other pioneers in law and economics. Gradually, however, opposition weakened (although it has not disappeared) as newer generations of lawyers became better qualified to appreciate and evaluate the contributions of this new field. At the same time, the gain from mainly arbi­ traging theoretical ideas from economics into the field of law began to lose steam. This was in good part because the early contributions were mainly theoretical, with only occasional support from legal cases, and with still less frequent support from quantitative analysis. Theory alone cannot keep a field vibrant, although it can substantially shift the approaches to different issues. No field that deals with human behavior has ever remained exciting and innovative by relying on theoreti­ cal ideas alone, no matter how valuable these ideas are. A vibrant and creative field requires a continual dialogue between theory and evidence from the real world that not only helps test existing theories, but also suggests new theories that can then also be tested and extended, or rejected. Fortunately, as the first theoretical stage was slowing down, perhaps because opportunities to arbitrage economic ideas into law were shrinking, a second stage began that collected and analyzed quantitative data. This quantitative approach uses statistical techniques also drawn mainly from economics to analyze antitrust cases, contracts, litigation, intellectual property, divorce, crime, discrimination, and many other areas of the law. Quantitative analysis has become one of the most exciting frontiers in law and economics, since extensive legal data exists, often in rudimentary forms. These da­ ta can test, discard, or help in the reformulation of the theories on the effects of laws and legal rulings on behavior. Most participants in any field, including law and economics, specialize. Some are mainly theorists, while others are mainly em­ piricists. For a field to remain relevant, however, many researchers must be both, relating theories to real-world data. Otherwise, theories become sterile as theo­ rists mainly discuss what other theorists said, and empiricists become mainly number crunchers, with little effort to interpret the data in other than ad hoc ways. This is why I believe an exciting further development in law and economics will involve extensive interactions between theory and empirical analysis. Some Page 4 of 6

The Future of Law and Economics lawyers will cooperate with economists, but even then it would be valuable for the lawyers to acquire not only the rudiments of economic theory, but also basic econometric and other techniques for analyzing data. Economists involved in this research should also acquire some knowledge of legal opinions and how legal sys­ tems operate. Indeed, a growing number of individuals with both a law degree and a Ph.D. in economics are beginning to bridge this gap. The great majority of re­ search in law and economics has been at the “micro” level in the sense of consid­ ering the behavior of parties to contracts, torts, crime, and other individual and business behavior. This research has been fundamental, but a newer and also im­ portant research focus considers the interactions between legal systems and the macro economy. This research, pioneered by economists Daron Acemoglou and Andrei Shliefer, among others, analyzes the connections between legal systems and long-term rates of growth, the degree of economic inequality, aggregate in­ vestments, and other macroeconomic variables. To a lesser extent, this burgeon­ ing literature also analyzes how macroeconomic developments affect the evolution of legal systems. Scarcity of data often limits how much can be achieved empiri­ cally in understanding the macro interactions between laws and economics, al­ though the creation of long time series for many countries on relevant legal and economic variables is widening the database. I expect the macro interaction be­ tween law and economics to become another major frontier as the discipline of law and economics pushes its boundaries and insights into uncharted territories. POSNER: Gary Becker—In Memoriam. The death of Gary Becker, on May 3, 2014, at the age of 83, was a great loss to economics, including to economic analysis of law, a field of economics in which he was a pioneer (one of a number of important economic fields in which he was a pioneer—others being the economics of discrim­ ination, of crime, of education, of human capital, of labor, of addiction, of altruism, of organ donation, and of marriage and the family). He was unquestionably one of the greatest economists of the twentieth century (and the beginning of the twentyfirst). We had been friends for almost half a century and I learned a great deal from him. His particular importance to economic analysis of law derived from the emphasis he placed in his economic work on non-market behavior, such as crime and discrimination and the family, because so much of law is concerned with regu­ lating that behavior. The regulation of economic activity in the narrow sense is an important concern of law, but an economics of law that focused only on the regula­ tion of such activity, and therefore on fields such as antitrust, securities, tax, labor, bankruptcy, and commercial law, would be seriously incomplete. He died rather suddenly, his mental powers unimpaired by age. He will be missed.

Acknowledgments Source: (2011). This exchange between Judge Posner and Professor Becker was composed in the

Page 5 of 6

The Future of Law and Economics fall of 2011. Becker had thanked William Landes and William Hubbard for helpful com­ ments; Posner had thanked Landes for helpful comments. (p. 6)

Gary S. Becker

Gary S. Becker, University Professor, University of Chicago; Professor, University of Chicago Graduate School of Business. Richard A. Posner

Richard A. Posner, Judge, United States Seventh Circuit Court of Appeals; Senior Lecturer, University of Chicago Law School.

Page 6 of 6

Economic Models of Law

Economic Models of Law   Thomas J. Miceli The Oxford Handbook of Law and Economics: Volume 1: Methodology and Concepts Edited by Francesco Parisi Print Publication Date: Apr 2017 Subject: Economics and Finance, Law and Economics, Econometrics, Experimental and Quanti­ tative Methods Online Publication Date: May 2017 DOI: 10.1093/oxfordhb/9780199684267.013.003

Abstract and Keywords This article discusses the use of economic models for understanding law. It begins by de­ scribing the nature of economic models in general, and then turns to the specific applica­ tion of economic models to law. It distinguishes between ‘economic analysis of law’, which concerns the use of economic theory for describing the incentive effects of legal rules (positive analysis) and for prescribing better rules (normative analysis); and ‘law and eco­ nomics’, which concerns the relationship between law and markets as alternative institu­ tions for organizing economic activity. The article concludes with some comments on the actual process of building economic models of law. Keywords: economic models, economic analysis of law, law and economics

2.1 Introduction PHYSICISTS have long pondered the epistemological question of why mathematics ap­ pears to be so effective in describing the natural world. Einstein, for example, once asked, “How is it possible that mathematics, a product of human thought that is indepen­ dent of experience, fits so excellently the objects of physical reality?” (Livio, 2009, p. 1). The remarkable power of mathematics has even prompted some philosophers to wonder whether mathematical truths are invented or discovered by humans. Social scientists were somewhat later in recognizing the usefulness of mathematics, but they have also successfully employed mathematical models to describe human behavior. The application of economic models to law is an even more recent development, reflecting the general ex­ pansion of economic reasoning to behaviors outside of the traditional market setting. This trend represents a recognition that economics is defined not by the subject matter that it studies (markets), but by the methodology that it brings to bear in describing rational de­ cision-making in whatever context it occurs.1

Page 1 of 21

Economic Models of Law Law is an ideal subject matter in this respect because, like economics, it is largely con­ cerned with incentives. According to the axiom of rationality, which is the foundation of economic decision-making, individuals act to maximize their well-being (however that is measured), subject to the various constraints that they face. In a market setting, these constraints consist of income and prices; in law they consist of legal sanctions. The appli­ cation of economic analysis to law specifically presumes that individuals view the threat of sanctions—whether in the form of fines, damages, or imprisonment—as implicit prices that can be set to guide behavior in certain socially desirable directions. In this view, no­ tions of “legality” and “illegality” are stripped of their moral or ethical connotations and instead are interpreted according to a functionalist view of law. The great American legal scholar and jurist Oliver Wendell Holmes, Jr. anticipated this approach when, in his classic article “The Path of the Law,” he described his so-called “prediction theory” of law (Holmes, 1897). The key figure in this theory is the “bad man,” who represents the classic rational maximizer from economic theory. The bad man is not an immoral agent, but rather is amoral in the sense that he will break the law when the gains from doing so exceed the costs. Thus, “[i]f you want to know the law and nothing (p. 10)

else, you must look at it as a bad man, who cares only for material consequences which such knowledge enables him to predict” (Holmes, 1897, p. 459). This view of law may strike many as crass, especially those who obey the law out of a sense of moral obligation rather than a fear of punishment, but to examine the incentive effects of law, it is neces­ sary to focus on those individuals for whom it is a binding constraint. That is what eco­ nomic models of law seek to do. The remainder of this chapter examines the use of economic models for understanding law as a social institution. It begins by describing the nature of economic models in gen­ eral, and then turns to the specific application of economic models to legal relationships. It concludes with some practical comments on building economic models of law.

2.2 The Nature of Economic Models As a social science, economics relies heavily on models, mostly mathematical, as a way of simplifying the world. Without the use of models, it would be impossible to disentangle the myriad causal relationships that characterize the operation of a complex social sys­ tem like the marketplace or the legal system. To be sure, there is a necessary sacrifice in reality when using a model to isolate a particular phenomenon, but the hope is that the factors that are excluded from the model are extraneous to the particular question of in­ terest and so can be safely ignored, at least as a first approximation. The extent to which a model succeeds in isolating the essential features of an issue is an indication of its “goodness.” Economists use two basic approaches to evaluate the goodness of their mod­ els:2 the first is to evaluate the quality of the assumptions, and the second is to use empir­ ical evidence to test whether the model accurately predicts behavior.

Page 2 of 21

Economic Models of Law The assumptions of an economic model define the specific relationships that will be stud­ ied and the factors that will be excluded from analysis—in other words, what variables are endogenous and what variables are exogenous. In this sense, they are the counter­ part to controls in a laboratory experiment. The validity, or quality, of the assumptions de­ termines how believable the conclusions of the model are, because once the assumptions are in place, the results follow logically from the structure of the model. Most debates over economic models therefore focus on the assumptions: in particular, has the essence of the issue been captured, and are only inessential factors excluded? (p. 11) In many cas­ es these are subjective questions, and so the skillful choice of assumptions is often char­ acterized as an “art.”3 The second test of a model is whether it can predict or explain the real world.4 This is where theory meets empiricism; that is, where data or other forms of empirical analysis, like case studies, are used to test the predictions of a model. As in physics, economists tend to specialize in either theoretical or empirical analysis, but even those scholars who are exclusively interested in theory need to develop their models with an eye toward mak­ ing testable predictions, for according to the scientific method, a model that fails the em­ pirical test, no matter how elegant, should either be discarded or revised. Although economic models can vary widely in their specific techniques and methodolo­ gies, they usually share certain characteristics. First, they nearly always start by positing rationality on the part of the relevant decision-makers, at least some of the time.5 Second, the setting in which agents act is limited by the assumptions of the model, which, as dis­ cussed above, allows the researcher to focus on the particular question(s) of interest. Fi­ nally, economic models distinguish between positive and normative analysis; that is, be­ tween analysis that is meant to describe or predict behavior in a particular institutional setting, and analysis that is meant to prescribe a better outcome or policy based on some articulated social norm such as efficiency. Economic models of law come in both varieties, depending on whether the goal is to understand the origin or describe the impact of a particular legal rule, or to propose a better one. I will provide examples of both types of analysis below.

2.3 Economic Models of Law The application of economic analysis to law has a long history, dating back at least to the criminal law theories of Beccaria (1986 [1764]) and Bentham (1969 [1789]). Surprisingly, however, the idea that laws could be interpreted as creating incentives for behavior did not seem to resurface again until the 1960s with the work of Coase (1960) on externali­ ties, Calabresi (1961) on accident law, and Becker (1968) on criminal law. Since then, of course, the application of law to economics has blossomed into a major field in both law and economics.6 (p. 12)

Most critics of the economic approach to law argue that it is inappropriate to eval­

uate the law based on the norm of efficiency. The goal of the law should instead be to pur­ sue justice or fairness, however those objectives are defined. Richard Posner, one of the Page 3 of 21

Economic Models of Law founding fathers of law and economics, has responded to this criticism in two ways. First, he argues that justice in the sense of distributional equity is a value that most economists also recognize, along with efficiency, as relevant in judging the performance of the mar­ ket or the legal system. And although economists have no special insight into what de­ gree of distributional equity is desirable, they have a lot so say about the feasibility of at­ taining different outcomes and the amount of sacrifice of overall wealth that would be necessary to achieve a particular distributional goal. Second, he argues that one meaning of “justice” may in fact be efficiency because “in a world of scarce resources waste should be regarded as immoral” (Posner, 2003, p. 27). Kaplow and Shavell (2002) have argued that social welfare, which they define as the ag­ gregation of some index of the well-being of all members of society, should be the sole ba­ sis for evaluating legal policy. According to this view, fairness matters for legal rule-mak­ ing, but only insofar as it affects people’s well-being (that is, to the extent that they care about fairness). At the same time, narrow concepts of efficiency like pure wealth maxi­ mization (or Kaldor–Hicks efficiency) are inappropriate because they exclude factors (like fairness) that people value. It is nevertheless the case that most law and economic analy­ sis focuses on wealth maximization as the objective. Although changes in legal rules will often affect the distribution of income, it will usually be quite difficult to ascertain these effects, and in any case, tinkering with legal rules will not generally be the best means of achieving a more equitable distribution of wealth, or will entail a trade-off between fair­ ness and efficiency. For example, the assignment of liability for product-related damages will likely affect both the distribution of wealth between businesses and consumers, and their incentives to avoid accidents. The fact that there exist alternative mechanisms for redistributing wealth that have no relation to products liability (such as the income tax and welfare systems) suggests that the liability system should be used solely for creating incentives to minimize accident costs.7 Thus, the mainstream approach to law and eco­ nomics, most often associated with the “Chicago school,” has for the most part employed the wealth-maximization approach to the development of economic models of law. The following two sections provide specific examples of the use of economic models for understanding law. I divide the discussion into two sections based on what I perceive to be two distinct approaches to modeling law. The first, which I call “economic analysis of law,” uses economic models to describe the structure and function of law; and the second, which I call “law and economics,” examines the relationship between law and markets as alternative social institutions for organizing economic activity.

(p. 13)

2.4 Economic Analysis of Law

By economic analysis of law I mean the use of economic models to evaluate legal rules. The positive version of this approach uses economic methods to examine the law as it is. It thus addresses questions about how the law affects behavior, but often goes further to argue that the various doctrines of the common law reflect an underlying economic logic. This view, first put forward by Richard Posner, maintains that the goal of efficiency has Page 4 of 21

Economic Models of Law somehow become embedded in the law. Normative analysis, in contrast, asks how the law can be improved to better achieve the goal of efficiency, though as noted, other norms can be incorporated into the analysis. Mainstream economic analysis of law encompasses both positive and normative analysis. I first illustrate positive analysis as it has been ap­ plied to the analysis of tort law, but then extend the logic derived from study of that area to other areas of law. I then turn to the economic model of criminal law, as first formal­ ized by Becker (1968), to illustrate normative analysis.

2.4.1 The Economic Model of Tort Law The use of tort liability to internalize harmful externalities is a straightforward extension of the theory of Pigovian taxation. The main difference is that tort law is administered pri­ vately through the judicial system rather than publicly by a regulatory authority. In the context of a simple externality model, however, there is no difference (at least in terms of deterrence) between a tax and an equivalent imposition of liability. The real insights, I contend, emerge from the application of the model to negligence law. Ironically, the first use of an explicitly mathematical model of negligence was by a judge in the famous case of U.S. v. Carroll Towing Co. 8 In that case, Judge Learned Hand pro­ posed his eponymous test for determining negligence, which says that an injurer who fails to take care should be found negligent if B ru.

Page 18 of 25

The “Law” and Economics of Judicial Decision-Making: A Positive Political Theory Perspective

Figure 11.6 Internal Policy Competition with Doc­ trines.

For this illustration, assume the counter-judge is more closely preference-aligned with the higher court (HC) than with the other members of the panel. Let and represent the best achievable policies (from the high court’s perspective) under rule and standard regimes, respectively. If the counter-judge dissents, and policy returns through higher court reversal to the best possible location for the counter-judge (and higher court) under the given doctrine form (that is, to point under a rule regime or under a standards regime), the counter-judge’s utility will improve, although suffering the related cost ru or st, thus reducing the counter-judge’s overall utility at the new (reversal) policy location. Recall that under a rule, there is relatively little discretion as to the policy location when applied ( ) compared to a standard, which could allow the lower court to set the policy within a broader range of policy options. We see in the configuration of preferences, doctrines, and decision costs of Figure 11.6, that the sincere application of the governing rule, if that were the doctrine in place at the time of the case, would produce a relatively high utility for the counter-judge (point 1), and a low utility for the preference-majority (point 2). Ideally, the preference-majority would like to disobey the rule and move policy near its ideal point PM. That would not pose a problem for a preference-uniform panel—that is, where higher court scrutiny is (p. 241)

weak because there are no likely dissenters on the panel. When the panel is a preferencemixed panel, however, a counter-judge is a whistleblowing threat and a reversal—moving the policy to —is possible if the dissent is exercised. Such a reversal is a worst-case situation for the preference-majority as the outcome will be moved dramatically back to a location that greatly diminishes the preference-majority’s utility (point 2). The solution is for the pref­ erence-majority to move policy out to the point nearest its ideal point that will not trigger a dissent and reversal—that is, . While not ideal for the preference-majority (utility at point 3), it is better than accepting Page 19 of 25

The “Law” and Economics of Judicial Decision-Making: A Positive Political Theory Perspective initially (utility at point 2) or having such imposed later on reversal by the higher court. Because the utility the counter-judge receives at (point 4) is as good as it would receive by dissenting and getting a reversal under a rule (back to ), when dissent costs are calculated into the effort (utility at point 5), outcome is stable. That is, . The whistleblowing threat has caused the preference-majority to moderate its outcome choice from PM to . Now consider the results under a doctrinal standard regime. As the standard itself allows for a broad range of outcomes (in this scenario, there is no natural limit), the only limit on the outcome is the cost (st) to the counter-judge to dissent (that cost being one sufficient to trigger a reversal under the standard, a high cost indeed). The cost for dissent under a standard (with any hope of higher court reversal) is substantially greater than that under a rule, thus widening the acceptable case outcome zone for the preference-majority. If a dissent were to ensue, the reversion point would be the counter-judge’s ideal point, CJ (with utility for the counter-judge at point 6). The indifference point for the counter-judge then becomes (with utility at point 7). At that point, the counter-judge will not dissent because the counter-judge’s overall utility (less cost) is just as great at as (points 6 and 7, respectively)—that is, . This is a good result for the preference-majority as is near the preference-majority’s ideal point (with utility at point 8 for the panel prefer­ ence-majority). What we learn from this illustration is that whistleblowing—through the dissent threat of a counter-judge—is a substantially stronger threat under a rule than a standard, when the counter-judge is more closely aligned with the higher court than with the other panel judges. Under this scenario, it is the costs of dissent under various doctrinal forms that expand the discretionary range of the preference majority. Indeed, it is through doctrinal form that the power of a potential dissenter to constrain the policy choice of the prefer­ ence-majority on the panel is enhanced or diminished. Thus, we see once more that the law (and the form of law) conditions the competition of institutional actors, in this case in­ ternal actors (judges) of an institution (court).

(p. 242)

11.4 Conclusion

This chapter brings focus to the active use of law by judges in their competition over poli­ cy. Whether in competition with other branches of government (horizontal), other courts in a judicial hierarchy (vertical), or among themselves on a panel (internal), the structure and form of legal reasoning—for example, decision instruments and legal doctrines—not only conditions the behavior of judges and their competitors, but provides strategic tools Page 20 of 25

The “Law” and Economics of Judicial Decision-Making: A Positive Political Theory Perspective to drive the outcomes to the desired position of judges. Without matching “law” with the more general rules of the game, along with the resources and the preferences of the vari­ ous institutions, much of the real action and strategic decision-making of judges is missed; indeed, without an understanding of how law conditions strategy, and itself be­ comes the strategy, law merely appears as an inert artifact of black box institutions in competition over policy space. When law is analyzed for its impact on decision resources and decision transparency, and for its impact on institutional rules of the game, in addi­ tion to its independent normative role on which preferences are often formed, we gain a better understanding of the power and use of law in judicial decision-making. To be sure, there are many other features of law not discussed here that bear on the vari­ ous theaters of policy competition involving courts. Recently, scholars have identified the use of citations to legislative history by judges in their published opinions (Abramowicz and Tiller, 2009) and word choice by judges in describing facts or case reasoning in their published opinions (Hinkle et al., 2012) as possible legal features of judicial decision-mak­ ing used to impact the dynamics of the internal and vertical games between judges. There are certainly even more judicial decision-making features to consider. Until we explore the broader uses of  “law” as strategy, a complete understanding of political-institutional competition involving the judiciary will not be fully understood. Fortunately, the frame­ work of Law and PPT invites an opportunity to broaden this approach, and the work con­ tinues to progress.

References Abramowicz, Michael and Emerson H. Tiller (2009). “Citation to Legislative History: Em­ pirical Evidence on Positive Political and Contextual Theories of Judicial Decision Mak­ ing” Journal of Legal Studies 38: 419–443. Boyd, Christine L., Lee Epstein, and Andrew D. Martin (2009). “Untangling the Causal Ef­ fects of Sex on Judging.” American Journal of Political Science 54: 389–411. Christiansen, Matthew R. and William N. Eskridge (2014). “Congressional Overrides of Supreme Court Statutory Interpretation Decisions, 1967–2011.” Texas Law Review 92: 1317–1541. Cohen, Linda R. and Matthew L. Spitzer (1994). “Solving the Chevron Puzzle.” Law & Contemporary Problems 57: 66–110. Cox, Adam B. and Thomas J. Miles (2008). “Judging the Voting Rights Act.” Colum­ bia Law Review 108: 1–54. (p. 243)

Cross, Frank B., Tonja Jacobi, and Emerson H. Tiller (2012). “A Positive Political Theory of Rules and Standards.” Illinois Law Review 2012: 1–41. Cross, Frank B. and Emerson H. Tiller (1998). “Essay, Judicial Partisan and Obedience to Legal Doctrine: Whistleblowing on the Federal Courts of Appeals.” Yale Law Journal 107: 2155–2176. Page 21 of 25

The “Law” and Economics of Judicial Decision-Making: A Positive Political Theory Perspective Epstein, Lee, William M. Landes, and Richard Posner (2011). “Why (And When) Judges Dissent: A Theoretical and Empirical Analysis.” Journal of Legal Analysis 1: 101–137. Epstein, Lee, William M. Landes, and Richard Posner (2013). The Behavior of Federal Judges: A Theoretical and Empirical Study of Rational Choice. Cambridge, MA: Harvard University Press. Eskridge, William N. (1991). “Overriding Supreme Court Statutory Interpretation Deci­ sions.” Yale Law Journal 101: 3321–3455. Eskridge, William N. and John Ferejohn (1992). “Making the Deal Stick: Enforcing the Original Constitutional Structure of Lawmaking in the Modern Regulatory State.” Journal of Law, Economics, & Organization 8: 165–189. Ferejohn, John and Charles Shipan (1990). “Congressional Influence on Bureaucracy.” Journal of Law, Economics, & Organization 6: 1–20. Fischman, Joshua B. (2013). “Interpreting Circuit Court Voting Patterns: A Social Interac­ tions Framework.” Journal of Law, Economics, & Organization 31(4): 808–842. Gely, Rafael and Pablo T. Spiller (1990). “A Rational Choice Theory of Supreme Court Statutory Decisions with Applications to the ‘State Farm’ and ‘Grove City’ Cases.” Journal of Law, Economics & Organization 6: 263–300. Hanks, Eva H., Michael E. Herz, and Steven S. Nemerson (1994). Elements of Law. Cincinnati, OH: Anderson Publishing Company. Hazelton, Morgan, Kristin Hickman, and Emerson H. Tiller (2013). “Panel Effects in Ad­ ministrative Law: A Study of Rules, Standards, and Judicial Whistleblowing.” Unpublished manuscript. Northwestern University. Hinkle, Rachael K., Andrew D. Martin, Jonathan D. Shaub, and Emerson H. Tiller (2012). “A Positive Theory and Empirical Analysis of Strategic Word Choice in District Court Opinions.” Journal of Legal Analysis 10: 407–444. Jacobi, Tonja and Emerson H. Tiller (2007). “Legal Doctrine and Political Control.” Journal of Law, Economics, & Organization 23: 326–345. Kastellec, Jonathan (2007). “Panel Composition and Judicial Compliance on the US Courts of Appeals.” Journal of Law, Economics, & Organization 23: 421–441. Kastellec, Jonathan (2011). “Hierarchical and Collegial Politics on the U.S. Court of Ap­ peals.” Journal of Politics 73: 345–361. Kim, Pauline T. (2009). “Deliberation and Strategy on the United States Court of Appeals: An Empirical Exploration of Panel Effects.” University of Pennsylvania Law Review 157: 1319–1381.

Page 22 of 25

The “Law” and Economics of Judicial Decision-Making: A Positive Political Theory Perspective Korobkin, Russell B. (2000). “Behavioral Analysis and Legal Form: Rules vs. Standards Revisited.” Oregon Law Review 79: 23–59. Lax, Jeffrey (2007). “Constructing Legal Rules on Appellate Courts.” American Political Science Review 101: 591–604. Lax, Jeffrey (2012). “Political Constraints on Legal Doctrine: How Hierarchy Shapes the Law.” Journal of Politics 74: 765–781. McCubbins, Matthew D., Roger G. Noll, and Barry R. Weingast (1994). “Legisla­ tive Intent: The Use of Positive Political Theory in Statutory Interpretation.” Law and Con­ temporary Problems 57: 3–37. (p. 244)

Miles, Thomas J. and Cass R. Sunstein (2006). “Do Judges Make Regulatory Policy? An Empirical Investigation of Chevron.” University of Chicago Law Review 73: 823–881. Miles, Thomas J. and Cass R. Sunstein (2008). “The Real World of Arbitrariness Review.” University of Chicago Law Review 75: 761–814. Revesz, Richard L. (2004). “Environmental Regulation, Ideology, and the D.C. Circuit.” Virginia Law Review 83: 1717–1772. Rodriguez, Daniel B. and Barry R. Weingast (2003). “The Positive Political Theory of Leg­ islative History: New Perspectives on the 1964 Civil Rights Act and Its Interpretation.” University of Pennsylvania Law Review 151: 1417–1542. Schanzenbach, Max M. and Emerson H. Tiller (2007). “Strategic Judging Under the Unit­ ed States Sentencing Guidelines: Positive Political Theory and Evidence.” Journal of Law, Economics, & Organization 23: 24–56. Schanzenbach, Max M. and Emerson H. Tiller (2008). “Reviewing the Sentencing Guide­ lines: Judicial Politics, Empirical Evidence, and Reform.” University of Chicago Law Re­ view 75: 715–760. Schlag, Pierre (1985). “Rules and Standards.” UCLA Law Review 33: 379–430. Schwartz, Edward P., Pablo T. Spiller, and Santiago Urbiztondo (1994). “A Positive Theory of Legislative Intent.” Law and Contemporary Problems 57: 51–74. Smith, Joseph L. and Emerson H. Tiller (2002). “The Strategy of Judging: Evidence from Administrative Law.” Journal of Legal Studies 31: 61–82. Spiller, Pablo T. and Rafael Gely (1992). “Congressional Control or Judicial Independence: The Determinants of U.S. Supreme Court Labor-Relations Decisions, 1949–1988.” Rand Journal of Economics 23: 463–492. Spiller, Pablo T. and Matthew Sptizer (1992). “Judicial Choice of Legal Doctrines.” Journal of Law, Economics, & Organization 8: 8–46. Page 23 of 25

The “Law” and Economics of Judicial Decision-Making: A Positive Political Theory Perspective Spiller, Pablo T. and Emerson H. Tiller (1997). “Decision Costs and the Strategic Design of Administrative Process and Judicial Review.” Journal of Legal Studies 26: 347–370. Spitzer, Matthew and Eric Talley (2011). “Left, Right, and Center: Strategic Information Acquisition and Diversity in Judicial Panels.” Journal of Law, Economics, & Organization 13: 101–126. Sunstein, Cass R., David Schkade, Lisa M. Ellman, and Andres Sawicki (2006). Are Judges Political? An Empirical Analysis of the Federal Judiciary. Washington, D.C.: Brookings In­ stitution Press. Tiller, Emerson H. (1998). “Controlling Policy by Controlling Process: Judicial Influence on Regulatory Decision Making.” Journal of Law, Economics, & Organization 14: 114–135. Tiller, Emerson H. (2002). “Resource-Based Strategies in Law and Positive Political Theo­ ry: Cost-Benefit Analysis and the Like.” University of Pennsylvania Law Review 150: 1453–1472. Tiller, Emerson H. (2015). The Law and Positive Political Theory of Panel Effects, The Journal of Legal Studies 44(S1): S35–58. Tiller, Emerson H. and Frank B. Cross (1999). “A Modest Proposal for Improving Ameri­ can Justice.” Columbia Law Review 99: 215–234. Tiller, Emerson H. and Frank B. Cross (2006). “What is Legal Doctrine?” North­ western University Law Review 100: 517–534. (p. 245)

Tiller, Emerson H. and Pablo T. Spiller (1999). “Strategic Instruments: Legal Structure and Political Games in Administrative Law.” Journal of Law, Economics, & Organization 15: 349–377.

Notes: (1) The insights from the Law and PPT movement have also underpinned normative analy­ ses (Eskridge and Ferejohn, 1992; McNollgast, 1994; Spitzer and Talley, 2013) and innov­ ative policy prescriptions (Tiller and Cross, 1999; Schanzenbach and Tiller, 2007, 2008; Miles and Sunstein, 2008) relating to court structure and judicial decision-making. (2) While there may be “pure rules” and “pure standards,” most doctrines rest some­ where in between. (3) A classical standard would be a “balancing test,” under which a court considered the equities of both sides before deciding. Another common standard is the multi-factorial test, in which doctrine tells lower courts to consider a series of factors as relevant to the decision’s outcome but provides no explicit instructions about how those factors are to be weighed (Cross, Jacobi, and Tiller, 2012).

Page 24 of 25

The “Law” and Economics of Judicial Decision-Making: A Positive Political Theory Perspective (4) Legal logic and policy preference rarely map perfectly on to each other. A legal rule must conform to linguistic and continuity principles inherent in a complex system of legal precedents and norms of communication (“legal craft”) while policy preferences have the ability to stand independent of these contextual conditions. (5) One might consider that if the lower court and higher court are aligned, why would the lower court pay any attention to doctrines as the higher court would not reverse the lower court. If a level of credibility and legitimacy is at stake for the lower courts in fol­ lowing doctrine, and the higher courts for enforcing the boundaries of doctrine, then the doctrines do provide a constraint on lower courts even when aligned with the higher court. As will be discussed below with the internal theater of competition, this legitimacy constraint with doctrine is even more forceful when there is diversity within the lower court panel of judges hearing the case (and even more so when the doctrine in play is a rule), as one of the judges may dissent if doctrine is not followed, thereby bringing atten­ tion to failure of the lower court to perform its assigned function.

Emerson H. Tiller

Emerson H. Tiller, Associate Dean of Academic Initiatives, J. Landis Martin Professor of Law & Business, Northwestern Law.

Page 25 of 25

Contractarian Perspectives in Law and Economics

Contractarian Perspectives in Law and Economics   Georg Vanberg and Viktor Vanberg The Oxford Handbook of Law and Economics: Volume 1: Methodology and Concepts Edited by Francesco Parisi Print Publication Date: Apr 2017 Subject: Economics and Finance, Law and Economics, Econometrics, Experimental and Quanti­ tative Methods Online Publication Date: May 2017 DOI: 10.1093/oxfordhb/9780199684267.013.020

Abstract and Keywords This article sketches the distinct perspective that a contractarian approach can bring to law and economics. It focuses on a particularly important strand of the contractarian tra­ dition: the constitutional political economy (CPE) research program (also known as con­ stitutional economics), developed most fully in the work of Nobel laureate James Buchanan. Like law and economics, the CPE paradigm is primarily concerned with the comparative analysis of social, economic, and political institutions. But its foundational assumptions offer a distinct contrast to the mainstream neoclassical paradigm that has dominated law and economics as a field. The article first provides a brief overview of con­ tractarian approaches. It then describes the central features of the CPE paradigm. It con­ trasts the foundations of the CPE approach with those of neo-classical economics; ex­ plores the implications of these differences for the research foci at the heart of these two traditions; and discusses how mainstream and constitutional economics approaches may be reconciled. Keywords: contractarianism, law, economics, constitutional political economy, constitutional economics, neo-clas­ sical economics

12.1 Introduction CONTRACTARIANISM has a long and distinguished history in Western political and social thought, and continues to be an important element of contemporary political theory. In this chapter, we sketch the distinct perspective that a contractarian approach can bring to law and economics as it has emerged out of neoclassical economics. Naturally, given the breadth of the relevant literature, it is necessary to narrow the analysis. We do so by focusing on a particularly important strand of the contractarian tradition: the constitu­ tional political economy (CPE) research program, developed most fully in the work of No­ bel laureate James Buchanan. Like law and economics, the CPE paradigm (also known as Page 1 of 24

Contractarian Perspectives in Law and Economics constitutional economics) is primarily concerned with the comparative analysis of social, economic, and political institutions. But its foundational assumptions offer a distinct con­ trast to the mainstream neoclassical paradigm that has dominated law and economics as a field. As a result, the CPE research paradigm provides a unique perspective on subject areas traditionally treated within law and economics. We suggest an interpretation for how these two research traditions may be integrated in a manner that can fruitfully ex­ tend the domain of both. We proceed as follows. In the next section, we provide a brief overview of contractarian approaches. In section 12.3, we sketch the central features of the constitutional political economy paradigm. In section 12.4, we contrast the foundations of the CPE approach with those of neoclassical economics. In section 12.5, we explore the implications of these differences for the research foci at the heart of these two traditions. In section 12.6, we il­ lustrate this contrast through an application to the work of Richard Posner. Section 12.7 (p. 247) sketches how mainstream and constitutional economics approaches may be rec­ onciled, and the final section concludes.

12.2 Contractarianism—Preliminary Issues Contractarian theories have occupied a central place in Western political thought going back at least to Plato’s Republic. Modern contract theory traces its origins to the seven­ teenth and eighteenth centuries, and the publication of four seminal works in the contrac­ tarian tradition: Hobbes’ Leviathan (1651), Spinoza’s Tractatus Theologico-Politicus (1670), Locke’s Second Treatise (1688), and Rousseau’s Social Contract (1762). Notably, the contractarian tradition continues to be influential among contemporary scholars, as reflected, for example, in Buchanan and Tullock’s Calculus of Consent (1962), Rawls’ The­ ory of Justice (1971), Gauthier’s Morals by Agreement (1986), as well as the extensive lit­ erature derived from these foundational texts. A common feature of all contractarian approaches is that they are individualistic. Con­ tractarianism is concerned with the extent to which existing social or political arrange­ ments are consistent with voluntary agreements among the individuals who live under them. Put differently, the contractarian paradigm assumes that political and social arrangements can be viewed fruitfully through the lens of exchange among autonomous individuals for mutual benefit. This paradigm has been employed both in positive and in normative inquiry. As positive, explanatory accounts, contract theories investigate whether, and if so, how, social and political arrangements emerge from explicit or tacit consent among individuals. In this dimension, contractarianism rests on methodological individualism, that is, the principle that social phenomena must be explained with refer­ ence to the actions of individuals. As normative accounts, contractarian approaches at­ tempt to assess the legitimacy of existing social and political arrangements by asking whether, and to what extent, these arrangements are consistent with the (ongoing) tacit or explicit consent of the individuals involved.1 In focusing on the issue of legitimacy, con­ tractarian arguments are based on a normative individualism, that is, the principle that Page 2 of 24

Contractarian Perspectives in Law and Economics the relevant measuring rod for assessing the legitimacy or desirability of social arrange­ ments are the evaluations of the individuals who must live under them. One source of controversy surrounding contractarianism (in particular the social contract theory of the state) has been a failure clearly to distinguish the positive and normative us­ es of contractarian arguments. As descriptive accounts, contractarian (p. 248) approaches represent instances of “conjectural history” or “explanations of principle.” Such accounts do not make specific historical claims, but offer an explanation of how, in principle, social and political arrangements may have originated through reciprocal agreements among individuals. Such contractarian conjectural history has been subject to vigorous criticism. As is well known, David Hume offered an early critique of social contract theory, and ar­ gued that consent “has very seldom had place in any degree and never almost in its full extent” in the creation of particular governments (1777/1987a, p. 474). Instead, Hume of­ fered an alternative account in which chieftains, who acquire authority as “coordination devices” in times of tribal warfare, are gradually able to extend their power if they can provide effective dispute resolution in times of peace.2 More recently, in one of his final contributions, Mancur Olson (1993)—directly challenging contractarian arguments as ex­ planatory accounts—sketches a conjectural history in which roving bandits, who prey up­ on the population, realize that providing the population with protection against other roving bandits, and engaging in predictable theft, can lead to tremendous increases in productivity that provide greater opportunities for sustained “taxation.”3 Thus, self-inter­ est turns roving bandits into stationary bandits. As Olson concludes, “these violent entre­ preneurs naturally do not call themselves bandits but, on the contrary, give themselves and their descendants exalted titles. They sometimes even claim to rule by divine right” (1993, p. 568). In short, in Olson’s account, government does not emerge from agreement among individuals—it is imposed. These alternative accounts—which do not require collective action, as contractarian explanations do—offer plausible and even com­ pelling accounts of the origin of political institutions. Nevertheless, of course, the plausi­ bility of these accounts does not preclude the possibility that mutual agreement among individuals has historically been one manner in which cooperative social arrangements have come about.4 Regardless of its status as conjectural history, the primary aim of contractarianism has typically not been to provide a historical account of the origins of political arrangements. Rather, contractarian arguments are predominantly normative in orientation. They are in­ tended to assess the legitimacy of political arrangements, and to identify desirable re­ forms in social and political institutions. As Buchanan notes (in his solo-authored appen­ dix to The Calculus of Consent, the seminal work that laid the foundation for the CPE par­ adigm):

(p. 249)

The relevance of the contract theory must lie, however, not in its explanation of the origin of government, but in its potential aid in perfecting existing institutions of government. (Buchanan and Tullock, 1962, p. 318) Page 3 of 24

Contractarian Perspectives in Law and Economics At first blush, this normative endeavor may appear inseparable from the descriptive en­ quiry. Legitimizing current arrangements might seem to require demonstrating that these arrangements originated in agreement among autonomous individuals. This appears to have been part of Hume’s position, who concluded that because consent “has very seldom had place” in the foundation of governments, “some other foundation of government must also be admitted” (1777/1985a, p. 474). However, the historical (descriptive) question of origins can, and should, be separated from the normative question of legitimacy. To see this, it is useful to distinguish between original agreement and ongoing agreement, that is, the question whether social arrangements originated in consent, and whether they are currently based on consent. Even if it could be established that political arrangements emerged from an original agreement among founding members, the presumptive legit­ imizing force of such consent cannot extend to future generations.5 Conversely, if an arrangement enjoys the ongoing consent of its current members, the legitimacy provided by such consent to the constraints that the arrangement imposes on current members is independent of the fact that the arrangement may have non-consensual origins. An analogy between democratic polities and private organizations may be useful in this context. Whether a private organization is able to trace its origins to a voluntary contract signed by its founding members or not, such history is irrelevant in judging the legitima­ cy of its current operation. What is of relevance, instead, is the ongoing voluntary accep­ tance of the organization’s constitution by its members. If its current members have joined voluntarily, and keep up membership by voluntary choice, the organization can be considered legitimized as a cooperative enterprise to the mutual benefit of its members.6 Similarly, on the contractarian logic, it is the ongoing voluntary agreement of its membercitizens that provides legitimacy to the constitution of a democratic polity, not an original agreement that may or may not have existed at its founding. Of course, the notion of ongoing consent raises critical conceptual and practical issues. A key contribution of the CPE research paradigm is to confront these challenges by draw­ ing attention to the question of the extent to which existing institutional arrangements can be said to reflect the consent of member-citizens, as well as the (p. 250) question of which kinds of institutional reforms are likely to make such consent more meaningful. Put another way, the CPE research program creates a bridge between the positive and nor­ mative dimensions of contractarian theory: As we shall argue in more detail below, CPE rests on a bedrock commitment to normative individualism. But this commitment drives a positive research program concerned with the question of how institutional frameworks for political, economic, and social interactions can be designed to create conditions favor­ able to its normative vision. Before returning to these questions, we provide a sketch of the CPE paradigm, and contrast it with mainstream approaches in law and economics.

12.3 Constitutional Political Economy Constitutional political economy can best be described as the economics of rules.7 As a research program, it is most closely identified with the work of James M. Buchanan (e.g. Page 4 of 24

Contractarian Perspectives in Law and Economics Buchanan and Tullock, 1962; Brennan and Buchanan, 1985; Buchanan, 1991), but it draws on other sources as well, including F. A. Hayek’s thoughts on the limits of knowl­ edge and the reason of rules (Vanberg, 1994, pp. 109ff.). It is part of the broader spec­ trum of approaches in modern economics that can be summarily described as the new in­ stitutional economics (Van den Hauwe, 1999). To understand what is meant by “the economics of rules,” it is useful to begin with a criti­ cal distinction at the heart of the CPE enterprise. This is a distinction between two levels at which one can approach social and political interactions. The “sub-constitutional level” (sometimes also referred to as the “level of actions”) is concerned with how individ­ uals pursue their various aims within a given institutional setting. Because institutions (“rules”) shape the opportunities and constraints that individuals confront, they have sig­ nificant implications for how the interactions among individuals unfold, and what the character of the order that results from these interactions is. As F. A. Hayek put it, the or­ der of rules affects the resulting order of actions (Hayek, 1969). Put differently, sub-con­ stitutional analysis asks how individuals act subject to a given set of rules—just as one can think about how players in a game (an analogy often employed in the CPE tradition) choose strategies subject to the game’s rules. Much work in law and economics—along with much of contemporary social science in general—is engaged in such enterprises: In­ stitutional structures are taken as given, and used to explain the actions and interactions of individuals who are subject to them.8 There is, however, a second level at which one can approach social and political interactions. This is the “constitutional level,” in which the analytical focus is not on the choice of actions within a set of rules, but rather the choice of rules themselves. At this level, emphasis lies on the fact that even though some rules that prevail in historically evolved social groups or communities may be beyond deliberate control and willful change, individuals often can (and do) choose (to change) the rules under which they live, and such choices can be guided by the knowledge of how alternative systems of rules will structure the resulting order of actions. In this sense, constitutional economics engages in comparative institutional analysis, and focuses on choice among constraints. To return to the game analogy, rather than focus on how players can be more successful in playing a given game, the constitutional economist’s interest is in inquiring how people may be able to play better games by adopting better rules. In this sense, the CPE research per­ spective recalls Adam Smith’s concept of political economy as “the science of a legisla­ tor” (Smith, 1776/1981, p. 486), a science that can provide guidance to those who are to (p. 251)

choose the rules for a society. Modern law and economics is concerned with questions at the sub-constitutional as well as the constitutional level. Considerable scholarship focuses on the contractual relations that emerge within given market rules (e.g. the theory of the firm), while other scholarship is concerned directly with the choice among (or revision of) systems of rules (e.g. the literature on optimal negligence rules). Explicitly emphasizing, as contractarian approaches do, the distinction between the subconstitutional and constitutional levels is analytically significant for a number of reasons. First, the nature of the choice confronting individuals at the two levels is fundamentally Page 5 of 24

Contractarian Perspectives in Law and Economics different. In one instance, the question is how to act within a set of constraints. In the oth­ er, the question is which set of constraints is desirable. This implies that at the constitu­ tional level, individuals are concerned with how alternative rules will structure a series of interactions over time, rather than with their particular interests in any given interaction. Second, and perhaps more importantly, at the constitutional level, the prospects for a common interest among individuals are typically greater than at the sub-constitutional level. At the sub-constitutional level, individuals may often have (at least partially) con­ flicting interests. But they may nevertheless share a common interest in playing a better game, that is, in revision of the rules that is likely to result in an order of actions that pro­ duces gains for all individuals. As Buchanan observed, As it operates and as we observe it to operate, ordinary politics may remain con­ flictual … while participation in the inclusive political game that defines the rules for ordinary politics may embody positively valued prospects for all members of the polity. (1999/1990, p. 386) The final sentence of this quotation draws attention to a central issue that constitutes a second distinctive characteristic of the CPE paradigm. At the constitutional level, CPE is concerned with the evaluation of alternative systems of rules. Such comparative institu­ tional analysis raises the critical question of the standard by which sets of rules can be compared or evaluated. How can alternative rules be assessed in terms of their (p. 252) “positively valued prospects”? A distinctive feature of the CPE paradigm is its commit­ ment to the principle that the evaluation of social arrangements must proceed on the ba­ sis of the valuations and choices of the participants themselves. Its insistence on this in­ ternal normative standard sets the CPE paradigm apart from perspectives that rely on ex­ ogenous standards to directly assess the legitimacy of social arrangements. As we argue in more detail below, mainstream law and economics tends to rely on such exogenous standards, including, for example, the efficiency criterion (usually in the familiar Kaldor– Hicks formulation)9 or Richard Posner’s “wealth maximization” standard (1979a, 1979b). From the CPE perspective, external evaluations, even if based on values imputed to indi­ viduals, are insufficient for establishing the legitimacy of social arrangements. It is this commitment that gives rise to the characterization of CPE as “politics as exchange,” and marks it as a contractarian enterprise. This commitment also gives rise to the central role of the unanimity principle as a basis for establishing the legitimacy of social and political institutions in the CPE tradition. We now turn to a more detailed examination of this fea­ ture.

12.3.1 Politics as Exchange and the Unanimity Principle In his Nobel lecture, James M. Buchanan characterized his research program as having three constitutive elements: “methodological individualism, homo economicus, and poli­ tics-as-exchange” (1999/1986, p. 456). Buchanan’s view of “politics as exchange” is root­ ed in his more general view that economics should be properly viewed as the science of the gains from trade, that is, as the science that examines how individuals can reap mutu­ al benefits from voluntary cooperation.10 The traditional focus of economics, of course, is Page 6 of 24

Contractarian Perspectives in Law and Economics on voluntary market exchanges as the paradigm case of mutually beneficial social trans­ actions. The CPE paradigm extends the “mutual gains from trade” notion to voluntary co­ operation more generally, most importantly, to the question of how individuals may real­ ize mutual gains by joint commitment to rules (Buchanan, 2001/1988a, pp. 60ff.).11 The most (p. 253) obvious instances where such joint commitment is mutually advantageous are social dilemma situations in which individually rational, unconstrained choices gener­ ate a pattern of outcomes inferior to what might result if all participants were appropri­ ately constrained in their choices. In extending the exchange paradigm to the analysis of institutions, the CPE approach adopts a consistently individualistic-subjectivist perspective: Whether an institutional (“rule”) change is socially beneficial, or represents mutual gain, can ultimately be judged only by the persons involved. It cannot be judged by an exogenously defined, independent standard. As Buchanan (1999/1986, p. 461) put it, “[i]mprovement in the workings of poli­ tics is measured in terms of the satisfaction of that which is desired by individuals, what­ ever this may be, rather than in terms of moving closer to some externally-defined, supraindividualistic ideal.” One consequence of this perspective—and a hallmark of the CPE approach—is that the relevant test for what qualifies as a “better game” is the voluntary agreement of the par­ ties involved to a change in the rules. Note the affinity between such consent to the rules and the notion of voluntary market exchange. In the case of market exchange, the claim that contracting parties are left better off by an exchange is based on no other evidence than their voluntary agreement to the transaction. Constitutional economics proceeds on the argument that, as a matter of consistency, the same reasoning should be applied to other forms of social cooperation. Ultimately, mutual advantage can be diagnosed only on the basis of voluntary agreement on the desirability of the respective arrangements. The unanimity principle thus “provides the test for the validity of specific propositions ad­ vanced by the political economist … concerning the existence of ‘mutual gains from trade’ through the political process” (Buchanan and Tullock, 1962, p. 94), and becomes, in this sense, the ultimate criterion of legitimacy for changes in social arrangements.12 This em­ phasis on voluntary agreement as the relevant criterion for the “goodness” of social transactions or arrangements is, of course, what gives the CPE paradigm its contractari­ an character. As Buchanan (2001/1988b, p. 150) observes, contractarianism … can be interpreted as little more than an extension of the par­ adigm of free exchange to the broader setting … By shifting “voluntary exchange” upward to the constitutional level of choices among rules, the consensual or gen­ eral agreement test may be applied. While unanimity is the ultimate test of mutual gain, and occupies a special place as the ultimate criterion for the legitimacy of social and political orders, it is critical to (p. 254)

distinguish between unanimity as a principle of legitimation and unanimity as an actually practiced decision rule in social choice. As Buchanan and Tullock already stressed in The Calculus of Consent, at the constitutional level, individuals will typically have an interest Page 7 of 24

Contractarian Perspectives in Law and Economics in consenting to decision-making rules for sub-constitutional matters that surrender their individual veto, even though doing so is predicted to result, at least on occasion, in out­ comes that impose losses on them. The reason for this is straightforward. As decision rules approach unanimity, they impose higher decision-making costs. In addition, they lead to larger opportunity costs for decisions that are not reached because smaller and smaller groups are able to block positive action. Anticipating these costs, individuals may therefore prefer decision rules well below unanimity if they anticipate that the loss from a departure from unanimity rule will be outweighed by gains that result from a reduction of decision-making costs, and by a reduction in the ability of others to block decisions that the individual favors.13 Specifically, individuals may well have prudential reasons unani­ mously to agree on “a majority rule for the operation of certain collectivized activities” (Buchanan and Tullock, 1962, p. 94), as well as on less-than-unanimity rules for the revision of decision-making procedures. Crucially, however, it is only from individuals’ joint agreement on these rules that deviations from unanimity can be concluded to be mu­ tually beneficial.14

12.4 Constitutional Economics vs. Welfare Eco­ nomics We are now in a position to return to our introductory claim that the CPE paradigm can provide a distinct perspective in law and economics. What differentiates this theoretical outlook from mainstream neoclassical economics, the paradigm that is predominant in law and economics? The two approaches share much in common. Both are individualistic in orientation. In their positive, explanatory endeavors, both are committed to (p. 255) methodological individualism, and employ the homo economicus model in investigating how individuals respond to incentives and constraints confronting them, including those embodied in legal rules and institutions. Moreover, in their normative orientation, the two paradigms also seem to share common ground, since both focus on individual evaluations as the relevant benchmark. As Kenneth Arrow (1987, p. 124) notes: It has been taken for granted in virtually all economic policy discussion since the time of Adam Smith, if not before, that alternative policies should be judged on the basis of their consequences for individuals. Critically, however, welfare economics—the normative branch of the neoclassical para­ digm—and CPE differ fundamentally in their approach to assessing these “consequences for individuals.” Welfare economics has developed largely within a “maximization para­ digm.” Within this paradigm—whether in the explicit formulation of a social welfare func­ tion in the Bergson–Samuelson tradition, or in the (hypothetical) compensation principle of the Kaldor–Hicks approach—social states, policies, and institutions are evaluated di­ rectly in search of a “social optimum.” The approach is “individualist” in the sense that the “social welfare function” operates on individual utilities.15 However, individual human agents count only as the “observation points,” so to speak, from which the welfare econo­

Page 8 of 24

Contractarian Perspectives in Law and Economics mist “reads” the utility inputs into his calculation. In this sense, this outlook can be char­ acterized as “utility individualism.”16 Buchanan argues that such utility individualism involves a tacit methodological shift. Economists analyze markets as the interplay of the actions of rational, self-interested in­ dividuals engaging in voluntary exchange. Welfare economics proceeds by generalizing the notion of “rational choice” from the level of individual human action in the market set­ ting to the level of collective organization, that is, it conceives of “society” as if it were a quasi-individual that evaluates policy alternatives in terms of its own collective “utility function.” Put differently, in transferring the concept of rationality as maximizing choice from the level of individual action to the level of collective political action, welfare eco­ nomics treats society as if it were a choosing entity with its own value scale. For Buchanan, this represents an implicit collectivist perspective that abandons the individu­ alism of the classic economic paradigm.17 In contrast—and consistent with the gains-from-trade perspective sketched in the previous section—the central claim of Buchanan’s contractarian constitutional economics is that the methodologically consistent way of extending the economic approach from the (p. 256)

study of markets to the study of collective arrangements lies in applying the procedural logic inherent in the study of market exchange to collective political arrangements, rather than in extending the outcome-oriented notion of utility maximization from individual to collective choice. Just as the “efficiency” of market outcomes can be inferred only from voluntary agreements among the parties involved, the outcomes of political processes can be judged only in terms of the nature of the choice processes from which they emerged. Here, too, the ultimate criterion of “efficiency” can be no other than the extent to which the choice processes from which outcomes result can reasonably be said to reflect the voluntary agreement of the parties concerned. Put differently, the choice individualism of the CPE paradigm places primary emphasis on the notion “that individuals are the ultimate sovereigns in matters of social organization” and that “the legitimacy of social-organizational structures is to be judged against the vol­ untary agreement of those who are to live or are living under the arrangements that are judged” (Buchanan, 1999/1991, p. 288).18 Social states or policies cannot—as in tradition­ al welfare economics—be assessed directly in terms of their societal value. Instead, pri­ mary emphasis must be placed on the procedures and institutions under which collective choices are made. The difference between these two outlooks—the utility individualism of welfare economics and the choice individualism of constitutional economics—implies that these two para­ digms lead to fundamentally different concepts of efficiency and legitimacy in social mat­ ters. Traditional welfare economics emphasizes the maximization of utility aggregates, in markets as well as in politics. Constitutional economics stresses voluntary agreement among sovereign individuals, in matters of collective political choice no less than in pri­ vate market choice.19 And this difference, in turn, leads to different emphases in research focus. Page 9 of 24

Contractarian Perspectives in Law and Economics

12.5 Difference in Research Focus: Out­ come Evaluation vs. Constitutional Reform (p. 257)

As we detailed in the previous section, a central feature that distinguishes contractarian constitutional economics from mainstream neoclassical economics is that the CPE para­ digm extends the procedural logic inherent in the economic approach from market ex­ change to collective political arrangements, rather than extending the outcome-oriented notion of utility maximization from individual to collective choice. In this section, we high­ light that these different conceptions lead to a divergence in the research foci of the two traditions. Because the focus of traditional welfare economics lies in the direct evaluation of “social states,” conceived as “a complete description of society, including every individual’s posi­ tion in it” (Sen, 1970, p. 152),20 two issues take center stage within this paradigm: (1) The problem of aggregating individual utilities into a properly defined measure of “social welfare,”21 and (2) the use of such constructions in the evaluation of social states. This approach leads naturally to an orientation towards the direct comparative evaluation of specific public policies and institutions in terms of the outcomes they are predicted to in­ duce, that is, individuals have “derived institutional preferences” that can be aggregated just like utilities over social states (see Riker, 1980; Bawn, 1993). For Buchanan, a major drawback of the outcome-oriented logic of this approach is that it shifts attention away from the institutional structure of an economy. That is, rather than focus on the nature of the processes that give rise to specific outcomes, and the institu­ tional framework within which policies are chosen, the approach evaluates outcomes di­ rectly. In contrast, the principal virtue of the gains-from-trade paradigm that undergirds constitutional economics is that its procedural logic highlights the role of the institutional framework within which individuals choose, whether in markets, politics, or organization­ al arrangements more generally. Buchanan (1999/1962, p. 229) stresses this difference, and the potential contribution of the constitutional perspective, this way: (T)he economist’s task is simply that of repeating in various ways the admonition, “there exist mutual gains from trade,” emphasizing the word mutual and forever keeping in mind that “trade” need not be confined to the exchange of goods and services in the marketplace. Welfare economics can make real progress through such (p. 258) a change in approach, which, quite literally, may be called the intro­ duction of “constructive institutionalism.” In contrast to a welfare economics that focuses on evaluating outcomes (or “social states”) directly in terms of “social welfare,” the CPE paradigm takes an indirect, proce­ dural approach to measuring improvement. It focuses on asking whether social, econom­ ic, and political frameworks can be expected to work to the mutual benefit of the persons involved, as judged by these persons themselves. In other words, it asks whether these frameworks enhance the ability of individuals to secure gains from trade. Moreover, the paradigm emphasizes the potential for changes in rules and institutions that enhance the Page 10 of 24

Contractarian Perspectives in Law and Economics prospects for free and equal individuals to successfully pursue their separate and com­ mon interests in mutually compatible ways. In short, this tradition places far greater em­ phasis on improving the “rules of the game” than on attempts at affecting outcomes di­ rectly. As Buchanan (2001/1989, p. 264) stresses, the appropriate domain for political economy, for politically directed reform as well as for discussion and analysis of that reform, is exclusively limited to struc­ ture. Efforts directed toward effectuating modifications of results that emerge on­ ly from complex interdependencies within structures are misguided.22 Connected back to the argument of the previous section, the central question for re­ search becomes: If voluntary agreement is the ultimate legitimizing principle in social and political affairs, how should constitutions be designed to best secure the prospects for mutually beneficial projects, and to limit the risk of projects that injure all citizens, or benefit part of the citizenry at the expense of the rest? As Buchanan (1999/1986, p. 462) observes, [t]he focus of evaluative attention becomes the process itself … The constitution of policy rather than policy itself becomes the relevant object of reform. In stressing the institutional frameworks of markets and collective choice processes, and in emphasizing that the prospects for mutual gain for individuals depend critically on these frameworks, the constitutional economics paradigm draws directly on the older tra­ dition of political economy associated with Adam Smith. A central insight of Smith’s Wealth of Nations—summarized in his famous metaphor of the invisible hand—is the proposition that under appropriate institutional rules, the pursuit of private interests by individuals can serve larger “societal” purposes, understood as serving the goals of other (often unknown) individuals.23 The Calculus of Consent, a foundational text in the (p. 259) CPE tradition, was self-consciously understood by Buchanan and Tullock as an extension of this insight to the realm of politics: “The question that we have posed in this work con­ cerns the possibility of extending a similar approach to political organization. Can the pursuit of individual self-interest be turned to good account in politics as well as in eco­ nomics?” (1962, p. 304). At bottom, an answer to this question requires attention to un­ derstanding the “rules for collective choice-making, the constitution, under which the ac­ tivities of political tradesmen can be similarly reconciled with the interests of all mem­ bers of the group” (1962, p. 304).

12.6 Wealth Maximization and Consent: the Case of Richard Posner Richard Posner is one of the most prominent and prolific representatives of the law and economics field. His approach to the field provides an instructive example to illustrate the differences between the two perspectives contrasted above—the choice individualism of constitutional economics and the utility individualism of mainstream economics. Central Page 11 of 24

Contractarian Perspectives in Law and Economics to Posner’s seminal contributions are two distinct, although related, claims. As a positive, descriptive matter, Posner has argued that—given certain conditions—legal systems will tend to evolve towards rules and institutions that are efficient in the sense of maximizing wealth, an argument he most famously elaborated with respect to the common law (Pos­ ner, 1972), but also with respect to primitive legal institutions (Posner, 1980).24 A second, normative claim is that wealth maximization constitutes an appropriate ethical goal or standard for the design and evaluation of legal rules and institutions (see, e.g., Posner, 1979a, 1979b), a position widely shared by other scholars in the law and economics field. A critical issue in connection with the claim that wealth maximization constitutes an ap­ propriate ethical standard is how this standard can be grounded as a normative principle. While this issue is usually not explicitly addressed, in a series of essays (1979a, 1979b) Posner confronts it head-on, arguing that wealth maximization can be derived from a Kantian commitment to individual autonomy and consent. In so doing, he explicitly re­ jects the notion that wealth maximization constitutes a utilitarian principle. As he con­ cludes, wealth maximization can be established as a normative principle “not (p. 260) by reference to Pareto superiority as such or its utilitarian premise, but by reference to the idea of consent” (Posner, 1979b, pp. 491f.). What is striking about Posner’s argument is that he explicitly rejects the utility individual­ ism of the neoclassical tradition, and instead, by invoking “the idea of consent,” appears to embrace a contractarian perspective. However, close analysis reveals that ultimately his argument remains rooted in the utilitarian tradition, rather than the contractarian al­ ternative offered by the CPE paradigm. Posner begins the analysis by focusing on transactions that occur in a “perfectly free mar­ ket.”25 In such a setting, legitimate transactions are those that the contracting parties consent to voluntarily, that is, those transactions that are free of “fraud and duress” (1979b, p. 492).26 What makes consent valuable in this setting, according to Pos­ ner, is that voluntary consent provides a reliable indication that in expectation, at the time of agreement, all parties to the transaction expect to benefit. Potential losses have been factored into an individual’s calculus, and are outweighed by potential gains.27 Accordingly, so Posner concludes, “involuntary, uncompensated losses experienced in the market … are fully compensated ex ante and hence are consented to” (1979b, p. 492). As Posner notes, when moving from market transactions to more complex “exchanges”— in particular to political and economic institutions—applying the same normative criteri­ on becomes more challenging: “A more difficult question is raised, however, by the simi­ lar attempt to ground nonmarket, but arguably wealth-maximizing institutions … in the principle of consent” (1979b, p. 492). This difficulty is rooted in the absence—at least in most circumstances—of express consent to a particular set of institutions that mirrors the consent given by participants to particular market transactions. As a result, the signal of adequate ex ante compensation for potential losses that voluntariness in market transac­ tion provides is missing. How then can institutions be legitimized in the absence of ex­ press consent? Page 12 of 24

Contractarian Perspectives in Law and Economics Posner answers this challenge through a subtle shift in the nature of the argument. Fo­ cusing on the example of a compensation scheme for automobile accidents, Posner em­ ploys textbook economic analysis to compare the expected wealth effects of a system of negligence with those of a strict liability system, concluding that “the sum of the liability and accident insurance premiums will be lower under negligence, and everyone will pre­ fer this” (1979b, p. 493). Posner argues that this imputed preference can be used to con­ clude that, although individuals have not de facto consented to the negligence standard, if confronted with a hypothetical choice among competing institutional arrangements, (p. 261) individuals would choose the negligence regime. That is, for Posner, fictitious consent, based on a (well-trained) outside observer’s expectations regarding the wealth effects of alternative institutional arrangements can be substituted for actual consent as expressed by individuals “to justify institutions” (1979b, p. 494). Related back to the distinction drawn in the previous section, Posner adopts an essential­ ly “choice individualist” position at the level of legitimizing market transactions: It is the express consent of autonomous individuals that is relevant for the legitimacy for their transactions. In shifting to the level of institutional justification, however, he abandons a choice-individualist conception in favor of a utility-individualist approach. At this level, evaluation of rules and institutions proceeds via an external analyst-observer who can judge which institutions and rules are preferable in terms of the evaluations imputed to individuals. What allows Posner to make it appear as if wealth maximization can bridge the utilitarian-contractarian divide is the fact that,28 when he claims to use “consent to justify institutions of wealth maximization” (1980, p. 495), he treats consent as an epis­ temic device rather than as the criterion that in and of itself confers legitimacy. For Pos­ ner, what ultimately “justifies” market transactions as well as those of institutions are the positive wealth effects that they produce (which in his particular conception means that transactions or institutions maximize wealth).29 At the level of market transactions, volun­ tary consent represents the most reliable indicator of wealth increase. Because practical difficulties make explicit consent a less convenient epistemic tool at the institutional lev­ el, Posner uses direct evaluation of the efficiency or wealth effects of institutions as a sub­ stitute indicator, assuming that institutions that are superior by this criterion would be consented to. The critical step in Posner’s shift from a process-oriented approach at the level of market transactions to an outcome-oriented approach at the institutional level is his invocation of “the hypothetical question whether, if transaction costs were zero, the affected parties would have agreed to the institution” (1979b, p. 494). On the surface, this question ap­ pears to continue to root institutional legitimacy in a contractarian perspective. However, it represents a paradigmatic shift because it implies that the actual choices of au­ tonomous individuals are replaced by an external analyst-observer’s assessment of whether a particular institution is preferable in terms of evaluations imputed to individu­ als. Although this approach is distinct from the Kaldor–Hicks compensation principle in the specific manner in which individual preferences are aggregated, the approach is (p. 262) methodologically identical: It focuses on direct evaluation of policies or outcomes Page 13 of 24

Contractarian Perspectives in Law and Economics in terms of imputed individual preferences, rather than focusing on the nature of the col­ lective decision process that gives rise to these policies and outcomes.

12.7 Reconciliation: Reasons for Consent vs. Legitimization by Consent Our central argument can be summed up succinctly. Both neoclassical economics—the dominant tradition within law and economics—and CPE share a contractarian foundation in the sense that both employ voluntary market exchange among autonomous individuals as the prototypical example of efficient exchange and gains from trade. The two tradi­ tions diverge in how they scale up the analysis of market exchange to more complex set­ tings. In the neoclassical paradigm, this is done by evaluating “social states” directly in terms of (appropriately) aggregated individual preferences as imputed by an analyst-ob­ server. In contrast, the CPE paradigm adopts the process orientation of market exchange by focusing on the institutional framework from which social and political outcomes emerge, and asking to what extent this framework allows individuals to pursue mutual gains from exchange in more complex social settings. In this paradigm, the critical test for the legitimacy of institutions, and of the outcomes these institutions shape, is volun­ tary consent by the individuals who must live under them. While the two traditions approach the problem of legitimizing institutions and social states in these divergent ways, there is, nevertheless, an important role to be played in the CPE research program by the kind of analysis typical of the neoclassical tradition, and the neoclassical paradigm can be given an interpretation that is compatible with contrac­ tarian foundations. We sketch this “reconciliation” in this section. To develop this argu­ ment, it is useful to return to Posner’s analysis of the negligence rule. Posner’s position, in simplified form, involves three steps: 1) A demonstration, based on economic reasoning, that a negligence standard will lower insurance premiums compared to strict liability; 2) The assumption that all individuals will prefer lower premiums; 3) The assumption that (2) implies that consent to the negligence regime can be pre­ sumed. From the CPE perspective, Posner’s argument is problematic as an attempt to legitimize a negligence regime because presumed consent based on imputed preferences cannot confer legitimacy. But Posner’s argument is not problematic, and may be useful, if inter­ preted as an attempt to provide reasons for individual consent, or to understand why (p. 263) individuals might choose a negligence standard if given the opportunity. Put dif­ ferently, the kind of analysis provided by Posner can be crucial for individuals in deciding what they, given their preferences, should consent to, or in helping an analyst-observer reconstruct what would lead individuals to consent to particular proposals.

Page 14 of 24

Contractarian Perspectives in Law and Economics More generally, this discussion highlights the distinction between the legitimizing role of consent, which is central to constitutional economics, and the problem of providing rea­ sons for consent, that is, developing arguments to explain why a specific policy or institu­ tional change represents an opportunity for individuals to improve their position. In “searching” for changes in the institutional framework that represent mutual gains, and explaining why they do so, rigorous analysis rooted in the mainstream tradition can play a critical role. Institutional changes are usually complex, and involve second- (and higher-) order effects that may make their ultimate repercussions difficult to evaluate for non-spe­ cialists. In this context, “expert” analysis is often critical. Such analysis can provide infor­ mation on the working properties of alternative institutional frameworks and their likely consequences, and it can attempt to correct erroneous beliefs that individuals may hold in these regards. However, from a contractarian perspective, such an inquiry is always conjectural, and can only be offered as prudential advice: It represents an analyst’s hy­ potheses about what it is in the interests of individuals to consent to, given her assump­ tions about their preferences, and her analysis of alternative institutional arrangements. The “test” of such hypotheses—and the step that confers legitimacy on an institutional change—is whether individuals, informed by the analyst’s arguments, do in fact consent if given the opportunity. Buchanan (1999/1959, pp. 207f.) expresses this point as follows: In a sense, the political economist is concerned with discovering “what people want.” The content of his efforts may be … summed up in the familiar statement: There exist mutual gains from trade. His task is that of locating possible flaws in the existing social structure and in presenting possible “improvements.” His spe­ cific hypothesis is that mutual gains do, in fact, exist as a result of possible changes (trades). This hypothesis is tested by the behavior of private people in re­ sponse to the suggested alternatives. Since “social” values do not exist apart from individual values in a free society, consensus or unanimity (mutuality of gain) is the only test which can insure that a change is beneficial.30 In other words, the type of work that dominates mainstream (law and) economics is fully compatible with the constitutional economics approach, and can serve an important func­ tion in informing and shaping public debate regarding the effects of current policies, pos­ sible improvements, as well as identifying potential reforms in the general (p. 264) eco­ nomic and political framework. The only point that must be stressed from the contractari­ an perspective is that such analysis is conjectural and can suggest reasons for individuals to consent, but it cannot directly legitimatize institutions or institutional reforms.31 Understood in this circumscribed manner, the neoclassical approach is fully compatible with the CPE paradigm. Highlighting the conjectural nature of arguments that provide reasons for consent also draws attention to an element of the CPE paradigm that can fruitfully extend the re­ search focus of law and economics. The test of conjectures regarding reforms that repre­ sent mutual gain lies in their ability to command consent, which serves to underscore the importance of the constitutional framework within which collective decisions are taken, and the process by which the “rules for making rules” are adopted. Frameworks for col­ Page 15 of 24

Contractarian Perspectives in Law and Economics lective decision-making that enhance the ability of individual citizens meaningfully to ex­ press or withhold consent, and that make it more difficult to adopt collective choices that do not represent mutual gains, represent more rigorous tests of the ability of proposed changes to garner consent. From a contractarian perspective, such frameworks are both more legitimate, and enhance the legitimacy of policy choices that emerge from them. Understanding how constitutional frameworks can be improved to enhance both of these features has been a hallmark of research in constitutional economics. For example, sys­ tems of jurisdictional competition that lower exit costs for citizens generally make resi­ dency a more meaningful expression of consent (Vanberg, 2000; Vanberg and Kerber, 1994). Similarly, returning to themes already evident in The Calculus of Consent, Buchanan’s late work focused on the possibility of imposing a “generality” constraint on democratic politics, thus making discriminatory policies more difficult to adopt (see Buchanan and Congleton, 1998). In closing, it is useful to stress that such a research focus on improvements in constitu­ tional frameworks highlights a critical point often overlooked in critiques of contractarian arguments. For obvious reasons, “idealized consent,” in which individuals are fully free to choose the set of rules under which they wish to live, is unattainable in a constrained world of overlapping generations, and physical and political boundaries. Nevertheless, political and economic systems differ significantly in the extent to which citizens can meaningfully register or withhold consent.32 From a contractarian perspective, this im­ plies that the relevant task—as much as reforms at the sub-constitutional level, and per­ haps more so—is to identify opportunities for improving constitutional frameworks in this respect.

(p. 265)

12.8 Conclusion

Perhaps the most significant contribution of law and economics as a discipline has been to stress the importance of institutions for social and economic outcomes—an insight that had largely been lost in the “institutions-free” mainstream economics of the early twenti­ eth century. Law and economics shares this institutional focus with the constitutional eco­ nomics paradigm that emerged at roughly the same time. However, while both paradigms emphasize the central role of institutions, they diverge in their approaches to institutional evaluation. Law and economics remains predominantly rooted in welfare economics in the sense that it focuses on the direct evaluation of social states, policies, and institutions in terms of the aggregated, imputed preferences of individuals. Constitutional economics, in contrast, eschews the direct evaluation of outcomes in favor of a contractarian founda­ tion. In this tradition, consent by the individuals who must live under a set of institutions becomes the relevant criterion for assessing their desirability and legitimacy, as well as for evaluating proposed reforms. Nevertheless, the two paradigms do “speak to each oth­ er.” Interpreted as prudential advice that can identify reasons for consenting to particular institutions, work in the mainstream tradition can play an important role in identifying potential improvements at the sub-constitutional and constitutional level. Constitutional economics, in turn, can enrich law and economics by more explicitly expanding its re­ Page 16 of 24

Contractarian Perspectives in Law and Economics search focus to incorporate work that is concerned with constitutional-level reforms that enhance the ability of citizens to secure mutual gains from trade, and to express mean­ ingful consent to the frameworks that structure social, economic, and political interac­ tions.

References Arrow, Kenneth (1951). Social Choice and Individual Values. New Haven, CT: Yale Univer­ sity Press. Arrow, Kenneth (1987). “Economic Theory and the Hypothesis of Rationality,” in J. Eatwell, M. Milgate, and P. Newman, eds., The New Palgrave Dictionary of Economics, 69–75. London: Palgrave Macmillan. Barry, Brian (1965). Political Argument. London: Routledge and Kegan Paul. Bawn, Kathleen (1993). “The Logic of Institutional Preferences: German Electoral Law as a Social Choice Outcome.” American Journal of Political Science 37: 965–989. Brennan, Geoffrey and James M. Buchanan (1985). The Reason of Rules—Constitutional Political Economy. Cambridge: Cambridge University Press. Brennan, Geoffrey and Alan Hamlin (1998). “Constitutional economics,” in Thomas Ulen and Peter Newman, eds., The New Palgrave Dictionary of Economics and the Law, 401– 410. London: Palgrave Macmillan. Buchanan, James M. (1991). The Economics and Ethics of Constitutional Order. Ann Ar­ bor, MI: University of Michigan Press. Buchanan, James M. (1999/1954). “Social Choice, Democracy, and Free Markets,” in The Logical Foundations of Constitutional Liberty, Vol. I, The Collected Works of James M. Buchanan, 89–102. Indianapolis, IN: Liberty Fund. (p. 266)

Buchanan, James M. (1999/1959). “Positive Economics, Welfare Economics, and Political Economy,” in The Logical Foundations of Constitutional Liberty, Vol. I, The Collected Works of James M. Buchanan, 191–209. Indianapolis, IN: Liberty Fund. Buchanan, James M. (1999/1962). “The Relevance of Pareto Optimality,” in The Logical Foundations of Constitutional Liberty, Vol. I, The Collected Works of James M. Buchanan, 210–229. Indianapolis, IN: Liberty Fund. Buchanan, James M. (1999/1964). “What Should Economists Do?” in: The Logical Founda­ tions of Constitutional Liberty, Vol. I, The Collected Works of James M. Buchanan, 28–42. Indianapolis, IN: Liberty Fund. Buchanan, James M. (1999/1986). “The Constitution of Economic Policy,” in The Logical Foundations of Constitutional Liberty, Vol. I, The Collected Works of James M. Buchanan, 455–468. Indianapolis, IN: Liberty Fund. Page 17 of 24

Contractarian Perspectives in Law and Economics Buchanan, James M. (1999/1990). “The Domain of Constitutional Economics,” in The Log­ ical Foundations of Constitutional Liberty, Vol. I, The Collected Works of James M. Buchanan, 377–395. Indianapolis, IN: Liberty Fund. Buchanan, James M. (1999/1991). “The Foundations for Normative Individualism,” in The Logical Foundations of Constitutional Liberty, Vol. I, The Collected Works of James M. Buchanan, 281–291. Indianapolis, IN: Liberty Fund. Buchanan, James M. (2001/1977). “The Use and Abuse of Contract,” in Choice, Contract, and Constitutions, Vol. XVI, The Collected Works of James M. Buchanan, 111–123. Indi­ anapolis, IN: Liberty Fund. Buchanan, James M. (2001/1988a). “Contractarian Political Economy and Constitutional Interpretation,” in Choice, Contract, and Constitutions, Vol. XVI, The Collected Works of James M. Buchanan, 60–67. Indianapolis, IN: Liberty Fund. Buchanan, James M. (2001/1988b). “Economists and the Gains-from-Trade,” in Ideas, Per­ sons, and Events, Vol. XIX, The Collected Works of James M. Buchanan, 135–152. Indi­ anapolis, IN: Liberty Fund. Buchanan, James M. (2001/1989). “On the Structure of an Economy. A Re-emphasis of Some Classical Foundations,” in Federalism, Liberty, and the Law, Vol. XVIII, The Collect­ ed Works of James M. Buchanan, 263–275. Indianapolis, IN: Liberty Fund. Buchanan, James M. and Roger D. Congleton (1998). Politics by principle, not interest— Towards nondiscriminatory democracy. Cambridge: Cambridge University Press (reprint­ ed as Vol. XI, The Collected Works of James M. Buchanan, Indianapolis, IN: Liberty Fund, 2003). Buchanan, James M. and Gordon Tullock (1962). The Calculus of Consent: Logical Foun­ dations of Constitutional Democracy. Ann Arbor, MI: University of Michigan Press (reprinted as Vol. III, The Collected Works of James M. Buchanan, Indianapolis, IN: Liber­ ty Fund, 1999). Downie, R. S. (1987). “Moral Philosophy,” in J. Eatwell, M. Milgate, and P. Newman, eds., The New Palgrave Dictionary of Economics, 551–556. London: Palgrave Macmillan. Ellickson, Robert (1989). “A Hypothesis of Wealth-Maximizing Norms: Evidence from the Whaling Industry.” Journal of Law, Economics, & Organization 5(1): 83–97. Friedman, David (2000). Law’s Order. Princeton, NJ: Princeton University Press. Gauthier, David (1986). Morals by Agreement. Oxford: Oxford University Press. Hayek, F. A. (1960). The Constitution of Liberty. Chicago: University of Chicago Press. Hayek, F. A. (1969). “Rechtsordnung und Handelsordnung,” in F. A. Hayek, Freiburger Studien. Tuebingen: J.C.B. Mohr. (p. 267)

Page 18 of 24

Contractarian Perspectives in Law and Economics Hume, David (1777/1987a). “Of the Original Contract,” in Eugene F. Miller, ed., Essays, Moral, Political, and Literary. Indianapolis, IN: Liberty Fund Classics, pp. 465–487. Hume, David (1777/1987b). “Of the Origin of Government,” in Eugene F. Miller, ed., Es­ says, Moral, Political, and Literary. Indianapolis, IN: Liberty Fund Classics, pp. 37–41. Olson, Mancur (1993). “Democracy, Dictatorship, and Development.” American Political Science Review 87(3): 567–576. Patty, John and Elizabeth Maggie Penn (2014). Social Choice and Legitimacy: The Possibil­ ities of Impossibility. Cambridge: Cambridge University Press. Posner, Richard (1972). Economic Analysis of Law. Boston, MA: Little, Brown. Posner, Richard (1979a). “Utilitarianism, Economics, and Legal Theory.” Journal of Legal Studies 8: 103–140. Posner, Richard (1979b). “The ethical and political basis of the efficiency norm in com­ mon law adjudication.” Hofstra Law Review 8: 487–507. Posner, Richard (1980). “A Theory of Primitive Society, with Special Reference to Law.” Journal of Law and Economics 23(1): 1–53. Priest, George (1977). “The common law process and the selection of efficient rules.” Journal of Legal Studies 6(1): 65–82. Rae, Douglas W. (1975). “The Limits of Consensual Decision.” American Political Science Review 69: 1270–1294. Rawls, John (1971). A Theory of Justice. Cambridge, MA: Harvard University Press. Riker, William (1980). “Implications from the Disequilibrium of Majority Rule for the Study of Institutions.” American Political Science Review 74: 432–446. Rubin, Paul (1977). “Why is the common law efficient?” Journal of Legal Studies 6(1): 51– 63. Samuels, Warren (1976). “The Myth of Liberty and the Realities of the Corporate State: a Review Article.” Journal of Economic Issues 10: 923–942. Sen, Amartya (1970). “The Impossibility of a Paretian Liberal.” Journal of Political Econo­ my 78: 152–157. Smith, Adam (1778/1981). An Enquiry into the Nature and Causes of the Wealth of Na­ tions. Indianapolis, IN: Liberty Fund. Van den Hauwe, Ludwig (1999). “Constitutional economics,” in J. G. Backhaus, ed., The Elgar Companion to Law and Economics, 100–114. Cheltenham: Edward Elgar.

Page 19 of 24

Contractarian Perspectives in Law and Economics Vanberg, Viktor J. (1994). Rules and Choice in Economics. London and New York: Rout­ ledge. Vanberg, Viktor J. (1998). “Constitutional Political Economy,” in J. B. Davis, D. W. Hands, and U. Mäki, eds., The Handbook of Economic Methodology, 69–75. Cheltenham and Northampton, MA: Edward Elgar. Vanberg, Viktor J. (2000). “Globalization, Democracy and Citizens’ Sovereignty: Can Com­ petition Among Governments Enhance Democracy?” Constitutional Political Economy 11: 87–112. Vanberg, Viktor J. (2004). “The Status Quo in Contractarian-Constitutionalist Perspec­ tive.” Constitutional Political Economy 15: 153–170. Vanberg, Viktor J. and Wolfgang Kerber (1994). “Institutional Competition among Jurisdic­ tions: An Evolutionary Approach.” Constitutional Political Economy 5: 193–219.

Notes: (1) We should note that scholars in the contractarian tradition do not always separate these two strands in their work explicitly, and there can be ambiguity about whether par­ ticular contributions should be interpreted as positive or normative accounts. Neverthe­ less, the two endeavors are analytically distinct. (2) Hume (1777/1987b, p. 40): “ … if the chieftain possessed as much equity as prudence and valour, he became, even during peace, the arbiter of all differences, and could gradu­ ally, by a mixture of force and consent, establish his authority.” Importantly, the end state of this process—the emergence of “government”—is an unintentional byproduct: “But though this progress of human affairs may appear certain and inevitable … it cannot be expected that men should beforehand be able to discover them, or foresee their opera­ tion. Government commences more casually and more imperfectly” (p. 39). (3) “[N]o one has ever found a large society that obtained a peaceful order or other public goods through an agreement among the individuals in the society” (1993, p. 568). (4) For example, the origins of the Swiss confederation are traditionally dated to the agreement among the three founding communes to the federal charter of 1291. The con­ tractual nature of these origins is reflected in the German name of the Swiss state as an “Eidgenossenschaft” (“Eid” = oath). (5) This was, of course, Thomas Jefferson’s point in his letter to James Madison on September 6, 1789: “ … no society can make a perpetual constitution, or even a perpetual law. The earth belongs always to the living generation. They may manage it then, and what proceeds from it, as they please, during their usufruct. They are masters too of their own persons, and consequently may govern them as they please. But persons and proper­ ty make the sum of the objects of government. The constitution and the laws of their pre­

Page 20 of 24

Contractarian Perspectives in Law and Economics decessors extinguished them, in their natural course, with those whose will gave them be­ ing.” (6) We ignore here potential effects of the arrangement on “outsiders,” i.e., we assume the effects of the arrangement are limited to its members. This allows us to clearly sepa­ rate the issue of historical origin and current legitimacy. (7) For general overviews of the field, see Brennan and Hamlin, 1998; Vanberg, 1998; Van den Hauwe, 1999. (8) As Buchanan (1999/1990, pp. 379f.) has pointed out, standard economics falls largely under this rubric insofar as it is concerned with choice within constraints, which include not only budget and price constraints but also the behavioral constraints, typically left im­ plicit in economic analyses, that are defined by the norms, rules, and institutions en­ forced in the social environment within which individuals act. (9) One set of rules is preferable to another if it is possible in principle for those who gain under the first system to compensate those who lose in such a way as to leave everyone better off. A system of rules is efficient if no change in rules that satisfies this criterion is possible. (10) In his 1962 presidential address to the Southern Economic Association, “What Should Economists Do?” Buchanan (1999/1964, p. 35) argued that the question of “how coopera­ tive association of individuals” may be made “mutually beneficial to all parties … should be central” to economics as an applied science. As he puts it: “This mutuality of advan­ tage that may be secured … as a result of cooperative arrangements … is the one impor­ tant truth in our discipline” (1999/1964, p. 36). Referring to this address, Buchanan (1991, p. 31) notes: “My argument was that economics, as a social science, is or should be about trade, exchange, and the many and varied institutional forms that implement and facilitate trade, including all of the complexities of modern contracts as well as the whole realm of collective agreement on the constitutional rules of political society.” (11) Buchanan and Tullock (1962, pp. 250–252): The “advantage of an essentially econom­ ic approach to collective action lies in the implicit recognition that ‘political exchange,’ at all levels, is basically equivalent to economic exchange. By this we mean simply that mu­ tually advantageous results can be expected from individual participation in community effort … The ‘social contract’ is, of course, vastly more complex than market exchange, involving as it does many individuals simultaneously. Nevertheless, the central notion of mutuality of gain may be carried over to the political relationship.” (12) “Agreement among all individuals in the group upon change becomes the only real measure of ‘improvement’ ” (pp. 6f.).—“The only test of the mutuality of advantage is the measure of agreement reached” (p. 250).—“The only test for the presence of mutual gain is agreement” (p. 252).—“Only through the securing of unanimity can any change be judged desirable on the acceptance of the individualistic ethic” (p. 261).

Page 21 of 24

Contractarian Perspectives in Law and Economics (13) The analysis of the constitutional choice among rules—including establishing an indi­ vidual-level rationale for why individuals might choose different collective decision rules for different circumstances, including less than unanimity rules—was, of course, the cen­ tral and seminal contribution of Buchanan and Tullock’s The Calculus of Consent (1962). (14) In this context, it is useful briefly to highlight an issue that has fueled some critiques of contractarian arguments, namely the status of the status quo. The centrality of agree­ ment in contract theory readily suggests that the status quo enjoys a privileged position, a result that may be troubling if the status quo embodies illegitimate privileges (see Bar­ ry, 1965; Rae, 1975; Samuels, 1976). A full discussion of this issue is beyond the scope of the current chapter, but see Vanberg (2004) for a detailed treatment that highlights that contractarian approaches are neither committed to a normative defense of the status quo, nor require unanimity as an “in-period” decision rule for moving away from the status quo. (15) Buchanan (1999/1954, p. 94): “The social welfare function of the utilitarians was based, in this way, on components imputable to individuals. But the welfare edifice so constructed was not necessarily coincident with that resulting from the ordinary choicemaking process. It was made to appear so because the utilitarians were also individual­ ists and, in one sense, philosophically inconsistent.” (16) Buchanan (1999/1964, p. 32): “The definition of our subject [as “maximizing choice” or “allocation of scarce means to their best uses”] makes it all too easy to slip across the bridge between personal or individual units of decision and ‘social aggregates’ … The utilitarians tried to cross the bridge by summing utilities.” (17) Notably, this perspective is one that Buchanan shares with Rawls (1971, p. 27), who offered a similar critique of utilitarianism, the philosophical precursor of welfare econom­ ics: “This view of social cooperation is the consequence of extending to society the princi­ ple of choice of one man, and then, to make this extension work, conflating all persons in­ to one through the imaginative acts of the impartial sympathetic spectator. Utilitarianism does not take seriously the distinction between persons.” (18) Buchanan (1999/1954, pp. 100f.): “A necessary condition for deriving a social welfare function is that all possible social states be ordered outside or external to the decisionmaking process itself. What is necessary, in effect, is that the one erecting such a function be able to translate the individual values (which are presumably revealed to him) into so­ cial building blocks. If these values consist only of individual orderings of social states (which is all that is required for either political voting or market choice) this step cannot be taken.” (19) It should be noted that the above critique of “utility individualism” applies most di­ rectly to what is commonly called “act utilitarianism.” A rule utilitarianism that concerns itself with the question of “why ought we to agree to having certain sorts of rules” (Down­

Page 22 of 24

Contractarian Perspectives in Law and Economics ie 1987, p. 553) would clearly be much closer to the choice individualism of contractarian constitutional economics. (20) See also Arrow (1951, p. 17): “The objects of choice are social states. The most pre­ cise definition of a social state would be a complete description of the amount of each type of commodity in the hands of each individual … ” (21) The issue of preference aggregation is obviously of primary importance in the exten­ sive social choice literature derived from Arrow’s (1951) Possibility Theorem. For a recent survey and evaluation of this literature, with special reference to democratic theory, see Patty and Penn (2014). (22) See also Buchanan and Tullock (1962, p. 306), who argue that the task is to under­ stand how “man can organize his own association with his fellows in such a manner that the mutual benefit from social interdependence can be effectively maximized.” (23) Like Smith, Buchanan stresses that the promotion of “public interest” through the pursuit of self-interest depends crucially on the appropriate institutional framework: “When economists refer to ‘the market’ and to its efficiency in enhancing the well-being of persons, they are presuming that the exchange order is operative within a set of appro­ priately drawn ‘laws and institutions,’ to use the terms of Adam Smith.” Given an appro­ priate framework, self-interested behavior is channeled “in such a direction that it be­ comes beneficial rather than detrimental to the interests of all members of the communi­ ty” (Buchanan and Tullock, 1962, p. 304). (24) For a more extensive discussion surrounding the efficiency of the common law, see Rubin (1977), Priest (1977), but also Friedman (2000). See Ellickson (1989) for an argu­ ment that American whalers in the nineteenth century established wealth-maximizing norms of possession for hunted whales. (25) Posner (1980, p. 497): “The perfectly free market in which there are no third-party ef­ fects, is paradigmatic of how utility is promoted noncoercively, through voluntary transac­ tions of autonomous, utility-seeking individuals.” (26) It should be noted that in taking the “perfectly free market” as his paradigmatic benchmark Posner bypasses the issues that arise for real-world markets that operate within a particular legal-institutional framework, a framework that defines what counts as “fraud and duress” or “third-party effects” and, accordingly, what qualifies as a voluntary transaction. (27) Posner employs the example of purchasing a lottery ticket that does not win a prize, or an entrepreneur losing business as the result of the superior product produced by a competitor. (28) Posner (1980, p. 496): “Wealth maximization as an ethical norm has the property of giving weight both to preferences, though less heavily than utilitarianism does, and to consent, though less heavily than Kant himself would have done.” Page 23 of 24

Contractarian Perspectives in Law and Economics (29) It is interesting to contrast Posner’s neoclassical view of the market as a welfare- or wealth-maximizing mechanism with Buchanan’s (1999/1964, p. 38) choice-individualist outlook at the market: “The ‘market’ or market organization is not a means toward the accomplishment of anything. It is, instead, the institutional embodiment of the voluntary exchange processes that are entered into by individuals in their several capacities…. The task of the economist includes the study of all such cooperative trading arrangements which become merely extensions of markets as more restrictively defined.” (30) See also Buchanan (2001/1977, p. 113): “The observing economist can suggest ways and means through which improvements may be made by agreement among all parties, and the test of his hypothesis lies only in agreement itself.” Similarly, Buchanan (1999/1959, p. 199): “The problem for the political economist is that of searching out and locating from among the whole set of possible combinations one which will prove accept­ able to all parties.” (31) To the constitutional economist applies, in this sense, what Hayek (1960, p. 114) says about the political philosopher: “Though he must not arrogate himself the position of a ‘leader’ who determines what people ought to think, it is his duty to show possibilities and consequences of common action.” (32) David Hume’s critique of tacit consent based on residency through his analogy of a sailor carried on board a ship while asleep (Hume 1777/1987a, p. 475) focuses, in modern language, on “exit costs.” Importantly, Hume ignores the fact that political communities can and do differ significantly in the degree to which they make meaningful consent pos­ sible (including the costs of exit). The relevant object of constitutional reform is to en­ hance such possibilities, even if idealized consent is not attainable.

Georg Vanberg

Georg Vanberg is Professor of Political Science at Duke University. Viktor Vanberg

Viktor Vanberg, Walter Eucken Institut

Page 24 of 24

Austrian Perspectives in Law and Economics

Austrian Perspectives in Law and Economics   Shruti Rajagopalan and Mario J. Rizzo The Oxford Handbook of Law and Economics: Volume 1: Methodology and Concepts Edited by Francesco Parisi Print Publication Date: Apr 2017 Subject: Economics and Finance, Law and Economics, Econometrics, Experimental and Quanti­ tative Methods Online Publication Date: May 2017 DOI: 10.1093/oxfordhb/9780199684267.013.021

Abstract and Keywords This article describes and analyzes the Austrian approach to law and economics within the context of the law and economics discipline. The important and distinctive feature of the Austrian approach is the emphasis on economic and legal processes. The article fo­ cuses on four themes within the Austrian approach to law and economics: the sponta­ neous origin of legal institutions; the analysis of implications of ignorance, decentraliza­ tion of knowledge, and static and dynamic uncertainty; the interaction between the changes in legal institutions and the market process and coordination; and entrepreneur­ ship in market and non-market settings. Keywords: Austrian approach, law and economics, economic process, legal process, legal institutions

13.1 Introduction AUSTRIAN law and economics is first and foremost a framework of analysis. It is not in it­ self a set of substantive conclusions about law or legal processes. Although the various authors writing in this tradition have arrived at substantive conclusions, these can change within the broad parameters of the framework. The dual purposes of this chapter are to sketch the essential features of the Austrian framework or method of analysis, and to situate it within the field of law and economics more generally.1 The essential and dis­ tinctive feature of the Austrian school of law and economics is its emphasis on both eco­ nomic and legal processes. Standard law and economics is actually two distinct fields clubbed together. The first is the study of market behavior within the framework of legal institutions. Given a specific set of legal institutions, economic choices and outcomes are examined. The second field is an application of rational choice theory to legal rules, where legal institutions are ratio­ nalized or understood by their hypothesized function. In either case, institutions generate Page 1 of 23

Austrian Perspectives in Law and Economics socially efficient behavior by individuals who are subject to their incentives. Using the above bifurcation, the former branch is associated with Ronald Coase and can be termed “law and economics” while the latter is associated with Richard Posner and can be la­ beled the “economic analysis of law” (see Coase, 1993; Posner, 1993; Harnay and Mar­ ciano, 2009; and Leeson, 2012).2 In this chapter, we focus on the method of analysis instead the object of analysis. The Austrian emphasis on processes can be applied to both these branches of law and economics: individual behavior within institutions, as well as individual behavior leading to the emergence of institutions. Most often, the standard analysis is static in each of the branches. This is an important distinction between the standard law and economics and the Austrian approach. (p. 269)

We focus on the following themes: first, the spontaneous development of legal institu­ tions; second, the implications of decision-making in the face of ignorance, dispersed knowledge, and uncertainty; third, interaction between legal institutions and the market process, given that both orders are constantly evolving; fourth, the coordination of plans in light of constantly changing circumstances; and fifth, the role of entrepreneurs in mar­ ket and non-market settings.

13.2 Origin of Legal Institutions The Austrian law and economics is most closely associated with F. A. Hayek and his work on the origins of legal institutions. Hayek described evolved law as “conceived at first as something existing independently of human will” and distinguished it from legislation, which was “the deliberate making of law” by a few individuals (Hayek, 2011/1960, pp. 118–119; and Hayek, 1973, pp. 72–73). Hayek’s analysis is the direct outgrowth of the earliest Austrian insights as well as those of the Scottish Enlightenment of the eighteenth century. Carl Menger, the founder of the Austrian approach, stated that social scientists must explain “how can it be that institutions which serve the common welfare and are ex­ tremely significant for its development come into being without a common will directed toward establishing them” (Menger, 1963/1883, p. 146).3 Menger’s question links the Austrian approach to Scottish Enlightenment scholars, who explained the spontaneous or­ ders as “the result of human action, but not the execution of any human design” (Fergu­ son, 1782/1767, p. III, S.2). Menger argued that the emergence of money is one such example of spontaneous devel­ opment of institutions. To solve the problem of the double co-incidence of wants, individu­ als find more highly valued commodities to exchange, and therefore add an exchange val­ ue to the use value of these goods, increasing the demand. As more individuals partici­ pate in such exchange, they converge to one or two generally accepted media of ex­ change, which we call money (Menger, 1892).

Page 2 of 23

Austrian Perspectives in Law and Economics In the same spirit of Menger’s explanation for the emergence of money, Mises attempted to explain the emergence of legal rules. Mises argued that property law originally arose from recognition of simple possession, and contract law from primitive (p. 270) acts of ex­ change within localized areas. While the former may have had as its primary motive the avoidance of violence and the creation of peaceful conditions, the latter was almost bound to arise under conditions of de facto property in order to pursue the gains from exchange. But ultimately the world created by these early efforts produced institutions that could be viewed as “a settlement, an end to strife, an avoidance of strife” and thus “their result, their function” is to produce peace within a community (Mises, 1981/1922, p. 34). Mises built on the argument made much earlier by Adam Smith, though Smith’s focus is not on violence but on trade. For Smith, individuals have a natural propensity to “truck, barter and exchange” (Smith, 1981/1776, p. 15). The growth in opportunities for ex­ change is the growth of markets. So, in pursuing exchange, individuals are unintentional­ ly providing greater scope for the division of labor and more general specialization both within and across firms (Smith, 1981/1776, p. 31). However, none of this would be possi­ ble without the extension and formalization of property and contract law. In dealing with strangers in perhaps far-flung areas, rules may need to become more detailed and more definite. This extension of trade itself creates incentives to modify and adapt existing rules into new legal forms. Merchants will make rule-like agreements to facilitate their trading activity (Benson, 1989). Therefore, the individual desire to gain from exchange can be seen as the impetus for the development of property and contract law. Law, then, is the spontaneous or unintended consequence of exchange. This account of development of institutions by Smith, Menger, and Mises roots founda­ tional institutions like property and contract not to any grand design or some superior mind directing institutions, but to mundane and pragmatic origins of choices made by in­ dividuals in their daily lives. Both Scottish and Austrian scholars were careful to empha­ size that the spontaneous and emergent explanation of law does not imply that no human intelligence is involved, but simply that the system as a whole was unintended. That institutions are not pre-existing, but instead develop spontaneously, has important implications for economists. Institutions provide the constraints for individual behavior in society. The constraints for these institutions in turn come from individuals engaging in the market. What is required is a step-by-step time-segmented analysis. For example, the advanced or complex development of law presupposes, or is constrained by, some basic framework of primitive or more elementary property rights that may emerge as a result of gains from market exchange (Demsetz, 1967). Property rights and associated legal rules themselves develop through their interaction with the economic exchange process that is simultaneously taking place. As the exchange process changes the technology, and economic value of goods, modifications to the existing rule are required and generally take place. The initial development of property rights which leads to the development of basic contract rights is the consequence of resources becoming valuable where previous­

Page 3 of 23

Austrian Perspectives in Law and Economics ly they were not. Further developments build on these initial rights in conjunction with changes in economic circumstances. In the example discussed by Demsetz on property rights in the Labrador Peninsula during fur trade, property rights develop to internalize externalities, when the gains of internal­ ization become larger than the cost of internalization. This results from changes (p. 271) in economic values, which in turn stem from the development of new technology and the opening of new markets and so forth. Property rights emerge because “old property rights are poorly attuned” to the changing economic circumstances (1967, p. 350). Applying this to the Great American Plains, Anderson and Hill (1975) find that when bene­ fits increased or costs decreased, individuals in the American West devoted more re­ sources to define and enforce rights in land, water, and livestock. This demonstrates how market interactions, based on basic property constraints, can give rise to commercial (contract) law without a centralized lawmaker. The self-interested interactions of mer­ chants lead them to develop and adhere to rules that increase their trade and hence in­ crease overall social cooperation. These rules develop through a process of trial and er­ ror. A process involving entrepreneurial alertness at a higher level—that is, the level of rules of the game—doubtless plays an important role.4 Spontaneous order processes do not generate all relevant rules and institutions simulta­ neously. Furthermore, there may not be a deterministic end state for rules and institu­ tions, and therefore their evolution should be treated as a continuing process. The institu­ tions within which individuals are acting are developing incrementally, in a particular or­ der and usually over a long period of time. However, it is not necessary that all spontaneous order processes develop over a long pe­ riod of time. Drastic changes in circumstances (caused by natural or other causes) may hasten such a process. Haddock and Keisling (2002) find that during the Black Death in the middle of the fourteenth century, the abruptly changed benefits and costs became the paramount statistical event of the era. The benefits and costs of property rights changed so rapidly, in fact, that contemporaries were initially at a loss to determine a proper reac­ tion. Human resources had increased greatly in value relative to other factors of produc­ tion and eventually came to be protected much more assiduously against dissipation of their rent. This resulted in a severe weakening of serfdom and related feudal institutions, institutions that in effect treated human factors more as communal property than as pri­ vate property. In a number of studies of the organization of eighteenth-century pirate activity, Leeson (2007, 2009) shows how an outlaw sub-group of society developed rules of governance without central direction. Outside of a market context, pirates converged on a set of rules whereby their ability to steal wealth from the rest of society was enhanced. This involved rules within the pirate society itself as well as rules governing its treatment of others. Within their society, “democracy” was used; outside of it, the use of brutality was con­

Page 4 of 23

Austrian Perspectives in Law and Economics strained. This case is a good example of a spontaneous ordering process with “good” con­ sequences within the sub-group and yet negative consequences for society as a whole. Martin and Storr (2008) also agree that spontaneous orders may not always be benign— that in fact perverse orders may also spontaneously emerge. Examples of such perverse emergent orders can be found in sustained negative belief systems (p. 272) (e.g. rabbyism in Bahamian culture) and spontaneous social violence or mob behavior (e.g. the emer­ gence of the riot of 1942 in Bahamas). It is hard to both predict and prevent such orders, a consequence of orders emerging not as a product of intelligent design or will, but from individual action within a certain context and constraints. Hayek applied spontaneous order analysis not just to specific legal institutions, but to the entire legal tradition of common law. Hayek described it as “deeply entrenched tradition of a common law that was not conceived as the product of anyone’s will but rather as a barrier to all power” (Hayek, 1973, p. 84). Scholars have written about the role of liti­ gants, judges, lawyers, and others in the emergence of common law rules. Some have de­ scribed the efficiency of common law as a result of private interests of litigants to resolve the dispute. On the supply side, Zywicki (2003) describes the common law system in the Middle Ages as polycentric lawmaking and attributes the emergence of efficient rules to courts and judges competing for litigants and fees in overlapping jurisdictions. Since the evolution of legal rules and institutions is a continual process, institutional entrepreneurs have an important role in finding opportunities to resolve conflicts and form more effi­ cient, context-specific rules. This may inadvertently give rise to the development of not only the specific rule, but the entire legal tradition. In the themes addressed in Austrian analysis, this spontaneous development of legal insti­ tutions is seen in light of individuals attempting to coordinate their actions and plans in a world of limited and dispersed knowledge and uncertainty. We shall see this in the follow­ ing sections.

13.3 Time and Ignorance Karen Vaughn, while describing the overall Austrian approach, wrote that it was “impos­ sible to think of Austrian economics as anything but the economics of time and igno­ rance” (Vaughn, 1994, p. 134). Rizzo clarifies that it is the economics of individuals cop­ ing with real time and radical ignorance (O’Driscoll and Rizzo, 1996, xiii). The problem of decentralized knowledge and uncertainty is at the very core of the Austrian approach to economics, especially the Austrian approach to law and economics. If individuals were not continuously faced with uncertainty and ignorance, a system of legal rules would be quite different. If one takes seriously the fact that all knowledge is decentralized, institu­ tions must have certain characteristics to solve the problem of dispersed knowledge in so­ ciety. Legal institutions play an important role in coordinating plans and expectations of different individuals in a market economy, as they help individuals overcome problems of exchange. Page 5 of 23

Austrian Perspectives in Law and Economics Perhaps the most important contribution of Austrian economics, as exemplified especially in the work of F. A. Hayek, is the understanding that individual behavior and social coop­ eration take place in the face of decentralized knowledge. However, individuals have lim­ ited knowledge, and social knowledge is dispersed or decentralized. (p. 273) Furthermore, this knowledge may not be costless to acquire, or may not even exist in the form required for decision-making. In his essay The Use of Knowledge in Society, Hayek notes that the “economic problem of society is thus not merely a problem of how to allocate ‘given’ resource … It is rather a problem of how to secure the best use of resources known to any of the members of soci­ ety, for ends whose relative importance only those individuals know” (1945, pp. 519–520). So, for Hayek, the function of the law is to provide dispersed economic decision-makers the “additional knowledge” or, more exactly, surrogates for the knowledge necessary to rationally plan (1945, p. 521). However, the law cannot do this directly; it can only create the basis for a knowledge-disseminating and economizing price system. The problem of decentralized knowledge starts with the knowledge available to the indi­ vidual. This knowledge may be local in a geographical sense, known only to the “man on the spot.” Or the knowledge may be personal, subjective to the preferences, motives, and mental states of the individual. The individual may possess both of these types of knowl­ edge either explicitly or tacitly (Rizzo, 2005). Additionally, the knowledge problem may take a static or dynamic form. The static knowl­ edge problem is the utilization of the current stock of dispersed factual knowledge so that individuals can coordinate their actions and plans at a particular time. The dynamic knowledge problem is the growth of new knowledge, not currently in the system, so that improved forms of coordination can occur over time (Kirzner, 1973; Kirzner, 1992; Leoni, 1991/1961).5 Given static and dynamic knowledge problems, legal institutions are impor­ tant in the coordination of different individuals at a point in time, and over a period of time. Before we can understand how property and contract law facilitate solving knowl­ edge problems, we must understand the importance of avoiding endogenous uncertainty produced by the law itself. Lachmann argues that incremental changes of rules, especially in their application to par­ ticular new conflicts, must not change the predictable nature of legal rules. “If institu­ tions are to serve us as firm points of orientation their position in the social firmament must be fixed. Signposts must not be shifted” (Lachmann, 1971, p. 50). However, this does not mean that a specific legal rule cannot or should not change. It just means that the system of rules must be predictable. Lachmann’s unchanging signposts are about (p. 274) the stability of the system, rather than the stability of each particular rule. This echoes Hayek’s argument that laws “are intended to be merely instrumental in the pur­ suit of people’s various individual ends…. They could almost be described as a kind of in­ strument of production, helping people to predict the behavior of those with whom they must collaborate, rather than as efforts toward the satisfaction of particular needs” (Hayek, 1944, pp. 72–73). To maintain the stability of the system at the same time Page 6 of 23

Austrian Perspectives in Law and Economics that rules can adjust marginally to new economic circumstances requires the “decompos­ ability” of the system of rules (Simon, 1969). Decomposable systems are flexible in the sense that they can isolate the effect of changes to just one part of the overall system.6 This is exactly the genius of the common law system. In addition to the characteristics of the “system,” the avoidance of endogenous uncertain­ ty also has implications for the characteristics of the rules. In a world filled with everchanging and complex interactions, what is required is a simple system of rules to guide individual behavior. Epstein (1995) argues that simple legal rules can act as signposts in a complex world that is uncertain and dynamic. As systems go from hierarchical orders to spontaneous orders, the degree of coordination required, and the inherent complexity in mutual compatibility of plans, is magnified. Once it is appreciated that in referring to le­ gal rules we are referring to inputs into individual decision-making, it becomes evident that the more decentralized and complex a system is, the more critical it is that rules are simple.7 Zywicki (1998) compares and draws a common theme from Michael Polanyi, Richard Ep­ stein, and Hayek’s treatment of social orders. He argues that a highly decentralized sys­ tem necessitates simple rules, so that dispersed individuals performing many different tasks can understand these rules and incorporate them into their plans. Individuals have to be able to understand legal rules in order to act in accordance with them. Therefore, Epstein (1995) gives content to this claim and outlines various types or properties of sim­ ple rules in property, contract, and tort law that would help individuals prosper in a com­ plex world. Rizzo (1980a) compares the rule of strict liability with one of negligence from the per­ spectives of the knowledge required for legal decisions and “institutional” efficiency. He concludes that a strict liability system, with appropriate defenses and excuses, has fewer knowledge requirements than an efficiency-based negligence system. The more demand­ ing knowledge requirements of the latter make judicial decision-making more subject to error and to uncertainty from the point of view of individuals making decisions subject to the rules.8 The inability of the negligence system to generate sufficiently precise knowl­ edge increases endogenous legal uncertainty. More generally in the field of public policy, simply to assume that the policymaker, paternalist, or central planner has the relevant knowledge to bring out his stated goals is (p. 275)

to assume the knowledge problem away (e.g. Rizzo and Whitman, 2009, p. 905). The task is actually the opposite—to solve the problem of decentralized knowledge. When policy­ makers act on the basis of a pretense of knowledge to which they have no access, they in­ crease uncertainty relative to attainment of individuals’ goals. Viewing the issue of legal certainty somewhat more positively, we can see that legal rules also “condition the ways in which individuals pursue their various ends” (Zywicki, 1998, p. 146). For individuals to successfully coordinate plans, ex ante expectations must be consistent, not conflicting or tangential. Individuals attempting to coordinate in the fu­ ture must have consistent expectations on, among other things, legal rules and institu­ Page 7 of 23

Austrian Perspectives in Law and Economics tions. This implies that legal institutions must enable some type of certainty or pre­ dictability. Leoni defines certainty of the law as the “possibility open to individuals of making long run plans on the basis of a series of rules spontaneously adopted by people in common and eventually ascertained by judges through centuries and genera­ tions” (1991/1961, p. 70). Legal institutions can facilitate a concurrence of mutual knowl­ edge and expectations completely analogous to the mutual knowledge required for mar­ ket equilibrium (Kirzner, 1992, p. 171). The precise characteristic of these institutions through which they increase predictability is that they create constraints on individuals—not unlike prices in markets. “Nonprice constraints are as much part of decentralized economy as are the prices they help to gen­ erate. These constraints are reference frameworks and points, in terms of which actors form expectations. Prices are formed on markets composed of contracts, rules, and cus­ toms, which are part of the constraints and basis for observed behavior” (O’Driscoll and Rizzo, 1996, p. 106). Lachmann emphasizes the aspect of institutions that aids the formation of expectations and argues that institutions enable each of us to rely on the actions of thousands of anonymous others about whose individual purposes and plans we can know nothing. They are nodal points of society, coordinating the actions of millions whom they relieve of the need to ac­ quire and digest detailed knowledge about others and form detailed expectations about their future action. But even what knowledge of society they do provide in highly condensed form may not all be relevant to the achievement of our immedi­ ate purposes. (1971, p. 50) Going beyond the issues of avoiding endogenous uncertainty and providing a stable legal framework, there are more specific implications of the knowledge-problem framework for the dissemination of knowledge in both a market and a legal context. Specifically, for le­ gal institutions, Barnett (1998) discusses three levels of knowledge problems. The first-or­ der problem is to ensure that individuals can use their own knowledge in using their re­ sources, and second, incorporate the knowledge of others in using their resources. Con­ tract law makes an effective price system possible. Because (p. 276) agreements must be voluntary, that is, both the terms of agreements and whether there is an exchange at all are subject to the veto of one of the parties, the terms of trade reflect the local and per­ sonal knowledge of each of the participants. As other market agents adapt to such prices, they are incorporating not only their own knowledge into decisions but also the knowl­ edge of other agents whom they do not know. This is an example of how the law facili­ tates the dissemination of market knowledge. Barnett’s second-order knowledge problem is the need to communicate the requirements of legal rules to all participants in society. This implies that the rules of justice must be consistent and non-arbitrary, that is, they are not changed on an ad hoc basis. Barnett’s third-order knowledge problem is to deter­ mine specific laws that will fulfill the above general requirement of justice, while also ap­

Page 8 of 23

Austrian Perspectives in Law and Economics plying to particular and complex circumstances. This is accommodated by the common law system of adaptation to the circumstances in each dispute before the court. To deal with this world that is constantly in flux, individuals rely on rules. The passage of real time, and therefore uncertainty, requires individuals to look to rules in society to cre­ ate stable patterns of behavior and coordinate with one another. When rules do not aid coordination, individuals engage in entrepreneurial discovery to close gaps and coordi­ nate, thereby aiding the development of rules (see section 13.6). Therefore, legal and eco­ nomic processes interact with each other either sequentially or simultaneously. The devel­ opment of both processes is in response to individuals dealing with problems posed by re­ al time and genuine uncertainty.

13.4 Coordination in Society The emphasis on ignorance, decentralized knowledge, and uncertainty leads us to the Austrian emphasis on coordination in society. Before we distinguish the Austrian ap­ proach in law and economics from the standard neoclassical approach in this regard, it is important to understand the difference between coordination and optimality. In the Aus­ trian approach, the focus is on coordination, and not on optimality. The fundamental meaning of coordination is simply the mutual compatibility of plans. This requires two things. First, each individual must base his plans on the correct expec­ tation of what other individuals intend to do. Second, all individuals base their expecta­ tions on the same set of external events (Rizzo, 1990, p. 17). In this basic meaning, the existing dissemination of knowledge has led to a state of affairs where each party is able to implement her plans. All offers to buy are accepted by sellers. All offers to sell are ac­ cepted by buyers. This is to be distinguished from the process of coordination whereby, through trial and error learning and entrepreneurial discovery, agents are able to make their plans compatible or more nearly compatible with those of others. Coordination is analytically different from, though not incompatible with, the concept of optimality. Pareto optimality implies that individuals exhaust all the potential (p. 277) gains from trade. This is a special case of coordination.9 However, there can be coordina­ tion, or the execution of mutually compatible plans, which do not exhaust all potential gains from trade, when “ … these plans are mutually compatible and that there is conse­ quently a conceivable set of external events, which will allow all people to carry out their plans and not cause any disappointments” (Hayek, 1937, p. 39). A state of mutually com­ patible plans “represents in one sense a position of equilibrium, it is however clear that it is not an equilibrium in the special sense in which equilibrium is regarded as a sort of op­ timum position” (Hayek, 1937, p. 51). Everyone within a system may have mutually com­ patible plans and yet there may be better trading opportunities out there so that at least some parties can improve their positions by alternative trades. Thus if there is a sense in which the mutual compatibility of plans is an optimum, it is only a local optimum, that is, between the direct parties to an exchange. Page 9 of 23

Austrian Perspectives in Law and Economics In the first instance, the type of coordination that is facilitated by legal structures is the basic mutual compatibility of plans. This is because an important aspect of correct predic­ tion about the actions of others is knowledge of the abstract framework in which deci­ sions must take place. These are Lachmann’s “nodal points” discussed above. Both prop­ erty and contract law, if stable, make it possible to more accurately predict much about the behavior of others.10 In a fully or perfectly coordinated state of affairs, each individual correctly takes into ac­ count: (1) the actions being taken by everyone else in the set, and (2) the actions that the others might take if one’s own actions were to be different (Kirzner, 2000, p. 136). The latter ensures that no buyer transacts at a price higher than that which a potential seller would offer, and that no seller transacts at a price lower than that which a potential buyer would offer. In this sense, the Austrian idea of coordination is compatible with the neo­ classical concept of Pareto optimality. If each individual fully takes account of the actions (and potential actions) of every other individual, all courses of action that might be pre­ ferred by any one participant without hurting anyone else must already have been suc­ cessfully pursued. In this sense, Pareto optimality corresponds to perfect coordination (Kirzner, 2000, p. 144). The importance of law to basic and perfect coordination is indirect. Legal rules obviously cannot affect a state of affairs in which plans are mutually compatible and all arbitrage opportunities are eliminated. Nevertheless, they can, by facilitating exchange and pro­ tecting or ensuring the right to entrepreneurial (arbitrage) profit, make the discovery processes that move the system toward mutual compatibility and full coordination more likely to be unleashed. While the concept of full coordination is compatible with Pareto optimality in an “exten­ sional” sense (that is, they both point to the same thing), there is a subtle difference be­ tween the two. The Pareto optimality concept is generally used as a criterion relevant to “social efficiency” in the allocation of society’s resources. Therefore, it is not (p. 278) in the first instance about the fulfillment or frustration of individuals’ plans, but about allo­ cating all the resources in society in a particular way, that is, to maximize aggregate wealth (Kirzner, 2000, p. 145). The importance of this distinction lies in the use of the cri­ terion. To see this, we must turn our attention to the operational form of the aggregate welfare notion: the Kaldor–Hicks criterion. The Kaldor–Hicks efficiency criterion asks whether the beneficiaries from the change (in legal rule or policy) could theoretically fully compensate the losers for their losses, and still remain better off. To satisfy the Kaldor–Hicks efficiency criterion, “Resources are al­ located efficiently in a system of wealth maximization when there is no reallocation that would increase the wealth of society” (Posner, 1980, p. 243). The compensation to losers is possible in principle and need not actually occur. The Kaldor–Hicks criterion is general­ ly seen as the efficiency objective of the aggregate wealth maximizer who tries to assign legal rights so as to maximize the total value of all goods and services. This policy amounts to assigning ownership to the would-be highest bidder—the party who, in the Page 10 of 23

Austrian Perspectives in Law and Economics outside observer’s estimation, would end up owning the right if costless bargaining had taken place in a hypothetical market with zero transaction costs (Harper, 2013, p. 64). Thus in the law and economics literature this weaker form of the optimality criterion is used as a normative standard. The key requirement for conducting Kaldor–Hicks analysis is the ability to gather and sum up people’s willingness to pay attached to different outcomes. This assumes a higher level of knowledge in the hands of the policymaker, judge, or economist. For economists using the Austrian approach, this is a pivotal and problematic assumption. Kaldor–Hicks analysis requires objective data to aggregate individuals’ willingness to pay and therefore uses existing prices. However, existing prices tell us about exchanges to which individu­ als have consented, whereas Kaldor–Hicks analysis requires consideration of hypothetical exchanges that have not taken place, and which will perhaps never take place, making the discovery of the requisite information difficult (Stringham, 2001, p. 43). In real-life ex­ amples of legal cases, Kaldor–Hicks analysis is used when there is a dispute that is com­ plex and there is no willingness to exchange, and where similar cases with objective data that may be substituted is difficult to come by. This automatically limits the applicability of this type of cost–benefit analysis in traditional law and economics.11 These considerations limit, but do not entirely preclude, the use of the Kaldor–Hicks crite­ rion in some cases (Epstein, 1988). First, in cases where the transfer of wealth produced by a change in legal rules is considerable, compensation can often be paid. Compensation can be in forms besides money payments—like in-kind compensation where the parties have access to the benefits involved. This can lead to something (p. 279) approximating the mutual compatibility of plans. Second, in some cases the relative economic value of the claims at stake may be tolerably clear. For example, with the advent of commercial air travel, modification of the property rule that guarantees property ownership to the heav­ ens (the ad coelum rule) was clearly a Kaldor–Hicks improvement. Similarly, that certain resources like rivers should be left in the commons and individual ownership prohibited makes sense when applying the Kaldor–Hicks criterion. In each of these cases the gainers gain more than the losers (e.g. property owners under the path of the plane or the poten­ tial private owners of parts of the river) lose as a consequence of the rule. Obviously, in the Austrian perspective, this is not a substitute for actual market transactions where they can take place. It is rather an expedient where there are large benefits that a legal rule can effect, much in the way of creating a framework for facilitation of (further) mar­ ket exchanges. More importantly, the consistency of the limited use of the Kaldor–Hicks criterion with the Austrian emphasis on plan coordination can be seen in a contractarian framework. Behind the veil of ignorance, all parties would have it in their rational self-interest to agree to a rule that made Kaldor–Hicks improvements under the three circumstances out­ lined above: some kind of compensation where the transfer is large, limitation to cases where the Kaldor–Hicks improvement is also large, and where market transactions can­

Page 11 of 23

Austrian Perspectives in Law and Economics not be facilitated to effect the improvement.12 This satisfies the mutual compatibility of plans ex ante, that is, at the rule-making level (Brennan and Buchanan, 2000/1985).

13.5 Legal Order Economic activity takes place within the framework of a “given” legal order. However, some explanation is required to produce clarity about the meaning of such “givenness.” Something can be given in the objective sense, in which the legal rules are given to the omniscient observing economist. This is a conceptual expedient for the creation of nar­ rowly specified and limited models. More important is the subjective sense, in which they are given to the individuals whose actions we are trying to explain (Hayek, 1937, p. 39). Givenness in this second sense means that the framework, at least insofar as it affects the plans of the individual, is predictable. However, it is impossible for any system of legal rules to be completely defined, specified, unambiguous, and hence perfectly predictable either in theory or in application. If igno­ rance and genuine uncertainty is taken into account, then it is problematic to assume a completely specified set of legal rules. While legal institutions may help individuals cope with ignorance in the market, these institutions are themselves subject to (p. 280) the knowledge problem. Hayek emphasized the knowledge problem not only in the context of the market, but extended it to other orders. The language of rules and legal decisions is always characterized by some ineradicable degree of uncertainty or vagueness and there­ fore even if it were conceptually possible to define all rules clearly, it would be prohibi­ tively costly (Whitman, 2002, p. 6). Accordingly, different individuals may hold different expectations about property rights. This would lead to disputes, and legal rules are required to resolve such disputes. As new circumstances arise, even a widely accepted property rule can engender expectations that conflict with other expectations sanctioned by other widely accepted rules (Hayek, 1973, pp. 115–116). Whitman argues that it is inaccurate to see law as a process of one-way causation where a given set of exogenous legal rules resolves conflicts (2002, p. 3). In this area, an impor­ tant aspect of the Austrian approach to law and economics is to endogenize the system of legal rules. Especially in a legal system in which judges make law by establishing, modify­ ing, overturning, and reaffirming precedents, the actions of participants play a pivotal role in determining the direction of the law. Even when there is no new legal rule, or a novel application of an old legal rule, the law still changes as a result. Applications of the existing rule to a new situation will “enlarge” the rule and if the rule is applied so as to exclude a new case, the meaning of the rule “shrinks” to that extent (Whitman, 2002, p. 7). Thus there is an entrepreneurial or gap-filling process that animates legal change.13 If the legal rules are constantly evolving, then what do we mean by the “givenness” of rules to the agents in the system? If the rules are constantly challenged and modified, then how do they provide any kind of guidance for human action? And more importantly, Page 12 of 23

Austrian Perspectives in Law and Economics how does a constantly evolving system of rules act as a constraint on individual behavior? Within any legal system there is a tension between the need to produce certainty of the laws and the need for the law to evolve and be relevant to new situations. This is particu­ larly the case in common law. The question is often posed as a trade-off between certainty and flexibility of the law. In the first place, at the moment of choice, a certain framework of rules is given. Today’s market transactions must be executed within the framework of rights as given today, but that framework is itself the unintended result of the past actions of many individuals. These rules of the game are the “relics” of successful plans of earlier generations that have “gradually crystallized” into institutions (Lachmann, 1971, pp. 68–69). Second, legal rules are not being changed in their entirety, but rather the change is marginal. “A change in the law can be marginal in the sense that it is perceived as deviating only slightly from precedent” (Rizzo, 1980b, p. 651). Third, Rizzo further argues that certainty of the law and its flexibility are not incompatible and that the “law endures by chang­ ing” (1999, p. 499). The law must have a certain plasticity to survive through economic changes. A rigid or static framework would break apart. These considerations imply that the “system” of rules is relatively stable while marginal changes to specific rules adapt to new or changing circumstances. This is the idea of the decomposability of the system of rules. (p. 281)

The most important factor that enhances the predictability of law, even as it changes and adapts to new circumstances, is the nature of the process involved. To see this we must distinguish between two forms of coherence in the law. Rizzo (1999) differentiates logical coherence of the law from the praxeological coherence, or the coordination, which arises from the law. For Rizzo, and the Austrian approach more generally, it is praxeological co­ herence, or coordination in society, which is at the forefront of analysis. The logical con­ sistency of laws is neither necessary nor sufficient for such coordination. Hayek argued that common law is an order where the legal rules facilitate the “order of actions” for individuals in a society. The “order of actions” is essentially the coordination of plans of economic actors (Hayek, 1973, p. 113). The emphasis is not on the “order of the law” but on the “order of actions” in society constrained by such laws. Therefore, the question is not whether a particular law is “socially optimal” and leads to wealth maxi­ mization. The question is not even whether the various laws form a socially optimal sys­ tem. The important question is whether the system of legal rules facilitates greater coor­ dination in society by enhancing expectational certainty. When the law succeeds in en­ hancing the order of actions, it is “praxeologically coherent.” On the other hand, other approaches to legal analysis emphasize the “logical coherence” of the law. Prominent among these is the idea that common law areas (property, contract, and tort) can be understood in a unified way as the expression of social wealth maximiza­ tion. Posner made a bold claim in the first edition of the Economic Analysis of Law (1973), that common law rules are “efficient,” that is, wealth-maximizing. According to Posner, the common law provides a coherent and consistent system of incentives that induce effi­ Page 13 of 23

Austrian Perspectives in Law and Economics cient behavior, not merely in explicit markets, but in all social contexts. Posner argues that there is an economic logic to these common law rules—an economic logic that is sub­ tle, and non-obvious, but nevertheless present. As the complexity of the wealth-maximization characterization of legal rules increased to take account of many different characteristics of circumstances and agents (Shavell, 2004), the divergence between this logical coherence and the Austrian idea of praxeologi­ cal coherence has become even clearer. In the process of adaptation to new circum­ stances, the wealth-maximization approach takes a general model and adds new parame­ ters. For example, a model that previously did not incorporate the risk preferences of the agents now adds a parameter for attitude toward risk. The model has its own internal log­ ic; it need not be accessible to the agents or the judges. A coherent order of actions can­ not be produced by a logically consistent framework (model) if its theoretical or applied results cannot be predicted by agents. As Rizzo (1999) argues, the method of adaptation must be “bottom-up” rather than “topdown.” In its simplest terms this means that adjustments to new circumstances (p. 282) must appeal to and be consistent with the agents’ (implicit) moral views. Does a particu­ lar result seem intuitively right? Is it reasonable in the context of other and previous rules? These are factors that may not fit into a logically coherent system of rules. People may not be fully consistent in how they see the correct resolution of disputes. Neverthe­ less, a system of rules that is not logically consistent may still lead, at least in the short run, to a tolerably coordinated order of actions. This is what really matters in the Austri­ an perspective. Another related reason for the emphasis on praxeological coherence of rules is the recog­ nition that there is no one single correct set of legal rules. If the moral intuitions of indi­ viduals are not completely consistent or if they have gaps, then there may be more than one right answer in a particular case. All of these may be in the range of expectations of the agents. Presumably, this does not unduly disturb the order of actions as long as the acceptable range is within ordinary limits. The process of rule evolution or generation in a common law system is based on trial and error (Hayek, 2011/1960, pp. 122–125). Therefore at any given point in time some rules or application of rules will simply be wrong. In other words, they will be ripe for revision as the process continues. For Hayek, the primary focus is on the overall system. The sys­ tem is or should be the primary object of normative evaluation. Its mistakes are in a sense simply part of the process. It should be noted that Menger did not share Hayek’s presumption that common law was more conducive to the general welfare than statutory law. He argued, “For common law has also proved harmful to the common good often enough, and on the contrary, legisla­ tion has just as often changed common law in a way benefitting the common good” (1985, p. 233). Menger believed that common law could serve the common welfare without a sin­ gle mind directing its development, but this need not occur. Page 14 of 23

Austrian Perspectives in Law and Economics Neither Hayek nor Menger specified the precise conditions under which the spontaneous organizing forces would produce benign or malign institutions. Nevertheless, the contin­ ued presence of a norm or precedent, or rule, has an important function in coordination behavior (Hazlitt, 1964, pp. 70–74). Stringham and Zywicki (2011) argue that Hayek’s idea of economic discovery influenced his later ideas about legal discovery, and the process of these two types of discovery is re­ markably similar. Moreover, once this conceptual similarity is recognized, certain conclu­ sions logically follow: namely, that just as economic discovery requires the competitive process of the market to provide information and feedback to correct errors, competition in the provision of legal services is essential to the judicial discovery in law. The authors come to the conclusion that competition in legal services, as prevalent in the Middle Ages in England, would create a legal order that has all the properties that Hayek ascribed to the common law tradition.14

(p. 283)

13.6 Entrepreneurship

An important and recurring theme in Austrian economics is the role of entrepreneurial alertness in seizing profit opportunities, and thereby enhancing the level of coordination in the market. Krecké argues that the Austrian concept of entrepreneurship is, in princi­ ple, applicable to legal decision-making. Decisions on which course to follow in a given case, and on which sources to rely, can be supposed to involve entrepreneurial judgments (2002, p. 8). Legal entrepreneurs, like their counterparts in the market, are alert to the “flaws, gaps and ambiguities in the law” (Krecké, 2002, p. 10). Whitman (2002) also ex­ tends the idea of entrepreneurship to the role played by lawyers and litigants. He exam­ ines how legal entrepreneurs discover and exploit opportunities to change legal rules—ei­ ther the creation of new rules or the reinterpretation of existing ones to benefit them­ selves and their clients. Harper (2013) believes that the entrepreneurial approach lays the groundwork for explaining the open-ended and evolving nature of the legal process— it shows how the structure of property rights can undergo continuous endogenous change as a result of entrepreneurial actions within the legal system itself. The most important differentiating factor separating the entrepreneurship of the market process from legal entrepreneurship is the absence of the discipline of monetary profit and loss in the latter case. Although money may change hands in the process of legal en­ trepreneurship, its outputs may not be valued according to market prices, especially when there is a public-goods quality to the rule at issue. Whether effective feedback mechanisms exist in the contexts is therefore an open question. Martin argues that, in such structures, the feedback mechanism in polities is not as tight as feedback in the market mechanism, and therefore ideology plays a greater role in such decision-making (Martin, 2010). Legal entrepreneurship can be coordinating and yet also increase uncertainty and con­ flicts in society. It all depends on the kind of legal order in operation and the mechanism by which it is generated and maintained. Rubin (1977) and Priest (1977) originally ana­ Page 15 of 23

Austrian Perspectives in Law and Economics lyzed how the openly competitive legal process tends to promote economic efficiency. They more recently point out that the common law system has succumbed to interest group pressures and has deviated from producing efficient rules (Tullock, 2005/1980; Tul­ lock, 2005/1997; Priest, 1991). They argue that litigation efforts by private parties can ex­ plain both the common law’s historic tendency to produce efficient rules as well as its more recent evolution away from efficiency in favor of wealth redistribution through the intrusion of strong interest groups into political and legal processes. Zywicki (2003) describes the common law system in the Middle Ages as polycentric. He focuses on three institutional features of the formative years of the common law system. First, courts competed in overlapping jurisdictions and judges competed for litigants. Se­ cond, there was a weak rule of precedent instead of the present-day stare decisis rule. And third, legal rules were more default rules, which parties could contract around, in­ stead of mandatory rules. These features are missing in the present-day common law (p. 284) system, which is non-competitive, has strong rules of precedent, and is dominated by mandatory rules. The efficiency claims pertain to a social system grounded in private ordering where those who are subject to those legal rules select the rules in open compe­ tition. Rajagopalan and Wagner (2013) argue that the inefficiency claims pertaining to the cur­ rent system of common law rules are a result of the entrepreneurial action within the con­ temporary system of the “entangled political economy.” The entangled political economy is essentially a “hybrid” of a monocentric state structure interacting with polycentric or private ordering, encouraging “parasitical” entrepreneurship within the legal system (Podemska-Mikluch and Wagner, 2010). Rajagopalan (2015) provides India as a case study to discuss a system of rules incongruent to the economy consequently giving rise to “par­ asitical” entrepreneurial action and entanglement of economic and legal orders. There is also “political entrepreneurship” within a given constitutional or governance structure that seeks to create coalitions to effect specific legislation or transfers of wealth (rent seeking). Martin and Thomas (2013) describe such political entrepreneurship at dif­ ferent levels of the institutional structure, at the policy level, legislative level, or the con­ stitutional level. These non-market orders determine the precise form that entrepreneur­ ship takes (Boettke and Coyne, 2009; and Boettke and Leeson, 2009). Political entrepre­ neurship may also attempt to change higher-level rules—like property rights systems, constitutional constraints, and so forth—as a means to gain rents and transfers within an economy (Rajagopalan, 2016). It is not is not a simple matter of moving from a “parasitical” to a “good” or coordinated legal and social order. Rules of behavior that surround and define markets, constitutional systems, social and cultural systems arise out of the previous framework of rules, whether it was de facto or de jure. There is path dependency in the development of insti­ tutions. Rules that develop as “indigenously introduced endogenous institutions” are closely related to the informal practices and expectations of people that, in turn, are grounded in local knowledge and values. Other rules may be indigenously introduced but Page 16 of 23

Austrian Perspectives in Law and Economics are exogenous in the sense that they are imposed by some formal authority, and do not gradually evolve from the informal traditions of a people. Even if they are meant as a cor­ rective to defects in the current system of rules, there is a risk that these rules will not “stick” because of conflict between the current institutions and the underlying norms (Boettke, Coyne, and Leeson, 2008).

13.7 Conclusion While it is difficult to summarize a project as broad as the Austrian approach to law and economics, a few key elements can be highlighted. The Austrian tradition is distinct from all neoclassical analysis, as it emphasizes the importance of processes in time, uncertain­ ty, and ignorance of the individual. Further, it stresses that the knowledge in society is fragmented and dispersed across individuals. Therefore, the main problem (p. 285) faced in society is one of coordination and social cooperation, and here the entrepreneur as­ sumes importance. Austrian scholars extend these themes to the legal order, and this has important implica­ tions for their approach to law and economics. First, the legal rules cannot be assumed to be exogenously given; they must be evolved and discovered through a process. Second, legal systems can become important signposts enhancing expectational certainty, even if they are constantly evolving. And third, entrepreneurship is no longer restricted to within a given set of rules; entrepreneurs also operate in legal and political spheres attempting to create, change, and evolve rules. Finally, through these forces, legal institutions evolve spontaneously without a central mastermind.

References Anderson, T. L. and P. J. Hill (1975). “The evolution of property rights: a study of the American West.” Journal of Law and Economics 18(1): 163–179. Barnett, R. E. (1998). The Structure of Liberty. New York: Oxford University Press. Benson, B. (1989). “The Spontaneous Evolution of Commercial Law.” Southern Economic Journal 55: 644–661. Boettke, P. J., C. J. Coyne, and P. T. Leeson (2008). “Institutional Stickiness and the New Development Economics.” American Journal of Economics and Sociology 67(2): 331–358. Boettke, P. J. and Coyne, C. J. (2009). Context matters: Institutions and entrepreneurship. Hanover: MA, Now Publishers Inc. Brennan, G. and J. M. Buchanan (2000/1985). “The Reason of Rules: Constitutional Politi­ cal Economy,” The Collected Works of James M. Buchanan Volume 10. Indianapolis, IN: Liberty Fund.

Page 17 of 23

Austrian Perspectives in Law and Economics Coase, R. H. (1993). “Coase on Posner on Coase: Comment.” Journal of Institutional and Theoretical Economics 149: 96–98. Demsetz, H. (1967). “Toward a Theory of Property Rights.” American Economic Review 57(2): 347–359. Epstein, R. A. (1988). “Rent Control and the Theory of Efficient Regulation.” Brooklyn Law Review. 54: 741. Epstein, R. A. (1995). Simple Rules for a Complex World. Cambridge, MA: Harvard Uni­ versity Press. Ferguson, A. (1782/1767). An Essay on the History of Civil Society. 5th edition. London: T. Cadell. Haddock, D. D. and L. Kiesling (2002). “The Black death and property rights.” Journal of Legal Studies 31(S2): S545–S587. Harnay, S. and A. Marciano (2009). “Posner, economics and the law: From “law and eco­ nomics” to an economic analysis of law.” Journal of History of Economic Thought 31(2): 215–232. Harper, D. A. (2013). “Property rights, entrepreneurship and coordination.” Journal of Economic Behavior and Organization 88: 62–77. Hayek, F. A. (1937). “Economics and Knowledge.” Economica 4(13): 33–54. Hayek, F. A. (1944) The Road to Serfdom. Chicago: University of Chicago Press. Hayek, F. A. (1945). “The Use of Knowledge in Society.” American Economic Review 35: 519–530. Hayek, F. A. (1973). Law Legislation and Liberty, Vol. I, Rules and Order. Chicago: Univer­ sity of Chicago Press. Hayek, F. A. (2011/1960). Constitution of Liberty: The Definitive Edition. Chicago: University of Chicago Press. (p. 286)

Hazlitt, H. (1964). The foundations of morality. Princeton, NJ: Van Nostrand. Kirzner, I. M. (1973). Competition and Entrepreneurship. Chicago: University of Chicago Press. Kirzner, I. M. (1992). The Meaning of Market Process: Essays in the development of mod­ ern Austrian economics. New York, Routledge Press. Kirzner, I. M. (2000). The Driving Force of the Market. New York: Routledge. Koppl, R. (2015). “The Rule of Experts,” in The Oxford Handbook of Austrian Economics (eds Peter J. Boettke and Christopher J. Coyne), 343. New York, Oxford University Press. Page 18 of 23

Austrian Perspectives in Law and Economics Krecké, E. (2002). “The Role of Entrepreneurship in Shaping Legal Evolution.” Journal des Economistes et des Etudes Humaines 12(2): 1–16. Lachmann, L. M. (1971). The Legacy of Max Weber. Berkeley: The Glendessary Press. Leeson, P. T. (2007). “An-arrgh-chy: The Law and Economics of Pirate Organization.” Jour­ nal of Political Economy 115(6): 1049–1094. Leeson, P. T. (2009). The invisible hook: the hidden economics of pirates. Princeton NJ, Princeton University Press. Leeson, P. T. (2012). “An Austrian Approach to Law and Economics, with Special Refer­ ence to Superstition.” Review of Austrian Economics 25(3): 185–198. Leeson, P. T. and P. J. Boettke (2009). “Two-tiered entrepreneurship and economic devel­ opment.” International Review of Law and Economics 29(3): 252–259. Leoni, B. (1991/1961). Freedom and the Law. Princeton, NJ: D. Van Nostrand Company Inc. Loasby, B. J. (1976). Choice, complexity, and ignorance. Cambridge: Cambridge University Press. Martin, A. (2010). “Emergent Politics and the Power of Ideas.” Studies in Emergent Order 3: 212–245. Martin, A. and D. Thomas (2013). “Two-tiered political entrepreneurship and the congres­ sional committee system.” Public Choice 154(1): 21–37. Martin, N. and V. H. Storr (2008). “On perverse emergent orders.” Studies in Emergent Order 1: 73–91. Menger, C. (1963/1883). Problems of Economics and Sociology. Urbana, IL: University of Illinois Press. Menger, C. (1892). “On the origin of money.” The Economic Journal 2(6): 239–255. Menger, C. (1985). Investigations Into the Method of the Social Sciences. New York: New York University Press. Mises, L. v. (1981/1922). Socialism: An Economic and Sociological Analysis. Indianapolis, IN: Liberty Fund. O’Driscoll, G. P. and M. J. Rizzo (1996). The Economics of Time and Ignorance. New York: Routledge Press. Podemska-Mikluch, M. and R. W. Wagner (2010). “Entangled Political Economy and the Two Faces of Entrepreneurship.” Journal of Public Finance and Public Choice 28(2–3): 99– 114. Page 19 of 23

Austrian Perspectives in Law and Economics Posner, R. A. (1980). “The Ethical and Political Basis of the Efficiency Norm in Common Law Adjudication.” Hofstra Law Review 8: 487–507. Posner, R. A. (1993). “The new institutional economics meets law and economics.” Journal of Institutional and Theoretical Economics 149: 73–87. Posner, R. A. (1973). Economic Analysis of Law. First edition. New York: Aspen Publish­ ers. Priest, G. L. (1977). “The Common Law Process and the Selection of Efficient Rules.” Journal of Legal Studies 6(1): 65–77. Priest, G. L. (1991). “The modern expansion of tort liability: its sources, its ef­ fects, and its reform.” The Journal of Economic Perspectives 5(3): 31–50. (p. 287)

Rajagopalan, S. and R. Wagner (2013). “Legal Entrepreneurship within Alternative Sys­ tems of Political Economy.” American Journal of Entrepreneurship 6(1): 24–36. Rajagopalan, S. (2015). “Incompatible institutions: socialism versus constitutionalism in India.” Constitutional Political Economy 26(3): 328–355. Rajagopalan, S. (2016). “Constitutional Change: A public choice analysis,” in Sujit Choud­ hary, Pratap Bhanu Mehta, and Madhav Khosla, eds., The Oxford Handbook of the Indian Constitution. New York: Oxford University Press, pp 127–142. Rizzo, M. J. (1980a). “Law Amid Flux: The Economics of Negligence and Strict Liability in Tort.” Journal of Legal Studies 9: 291–318. Rizzo, M. J. (1980b). “The mirage of efficiency.” Hofstra Law Review 8(3): 641–658. Rizzo, M. J. (1990). “Hayek’s Four Tendencies Toward Equilibrium.” Cultural Dynamics 3(1): 12–31. Rizzo, M. J. (1999). “Which Kind of Legal Order? Logical Coherence and Praxeological Co­ herence.” Journal des Economistes et des Etudes Humaines 9(4): 497–510. Rizzo, M. J. (2005). “Problem of Moral Dirigisme: A New Argument against Moralistic Legislation.” NYU Journal of Law and Liberty 1: 790. Rizzo, M. J., & Whitman, D. G. (2009). Little brother is watching you: New paternalism on the slippery slopes. Arizona Law Review, 51: 685–739. Rizzo, M. J. (2011). “Introduction,” in M. J. Rizzo, ed., Austrian Law and Economics, Vols. I and II. Northampton, MA: Edward Elgar pp. xiii to xvi. Rubin, P. H. (1977). “Why is the Common Law Efficient?” Journal of Legal Studies 6(1): 51–63.

Page 20 of 23

Austrian Perspectives in Law and Economics Shavell, S. (2004). Foundations of Economic Analysis of Law. Cambridge, MA: The Belk­ nap Press of Harvard University Press. Simon, H. A. (1969). The sciences of the artificial. Cambridge, MA MIT Press. Smith, A. (1981/1776). An Inquiry into the Nature and Causes of the Wealth of Nations, Vols. I–V. Indianapolis, IN: Liberty Fund. Stringham, E. P. (2001). Kaldor-Hicks efficiency and the problem of central planning. Quarterly Journal of Austrian Economics 4(2): 41–50. Stringham, E. P. and T. J. Zywicki (2011). “Hayekian Anarchism.” Journal of Economic Be­ havior and Organization 78(3): 290–301. Tullock, G. (2005/1980). “Trials on Trial: The Pure Theory of Legal Procedure,” in C. Row­ ley, ed., The Selected Works of Gordon Tullock, Vol. IX. Indianapolis, IN: Liberty Fund. Tullock, G. (2005/1997). “The Case Against the Common Law,” in C. Rowley, ed., The Selected Works of Gordon Tullock, Vol. IX. Indianapolis, IN: Liberty Fund. Vaughn, K. I. (1994). Austrian Economics in America: The Migration of a Tradition. New York: Cambridge University Press. Whitman, D. G. (2002). “Legal Entrepreneurship and Institutional Change.” Journal des Economistes et des Etudes Humaines 12(2): 1–11. Whitman, D. G. (2009). The rules of abstraction. The Review of Austrian Economics, 22(1): 21–41. Zywicki, T. J. (1998). “Epstein and Polanyi on simple rules, complex systems, and decen­ tralization.” Constitutional Political Economy 9(2): 143–150. Zywicki, T. J. (2003). “The Rise and Fall of Efficiency in the Common Law: A Supply Side Analysis.” Northwestern University Law Review 97(4): 1551–1633.

Notes: (1) Our purpose is not to produce an exhaustive survey of contributions in the area. For a more detailed survey of the literature in the Austrian law and economics, see Rizzo (2011). (2) It would, however, be confusing to put a lot of emphasis on these two different labels because they are not used consistently across various authors. Nevertheless, the distinc­ tion is valid. (3) This question continues to be asked in the Austrian tradition: “How can individuals acting in the world of everyday life unintentionally produce existing institutions?” (O’Driscoll and Rizzo, 1996, p. 20). Page 21 of 23

Austrian Perspectives in Law and Economics (4) This is discussed in further detail in section 13.6. (5) Kirzner makes a further distinction between two types of knowledge problems emerg­ ing from this error and explains how the dynamic market process solves such knowledge problems. Individuals, when faced with dispersed knowledge, may overestimate or under­ estimate market information and prices of goods they wish to buy or sell. The first in­ stance is when decentralized knowledge leads to over-optimism on the part of market participants—or “Knowledge Problem A.” Market participants are over-optimistic buyers and sellers, and markets may fail to clear. The second instance is when decentralized knowledge leads to over-pessimism on the part of market participants—or “Knowledge Problem B.” Kirzner argues that “Knowledge Problem A” tends to be self-revealing—since it stimulates proposals that are bound to be disappointed. However, “Knowledge Problem B” does not result in disappointed plans, it results in failure to achieve potential gains. It is not inevitable that “Knowledge Problem B” will be self-revealing and therefore correct­ ing. “Knowledge Problem B” thereby creates an incentive for its solution by discovery by profit-alert entrepreneurs (1992, p. 174). (6) For a good discussion of decomposable systems, see Loasby (1976). (7) Whitman has argued that to forster coordination, an intermediate degree of abstrac­ tion is required. This is because both, very specific rules, and very abstract rules, fail to provide decision makers the guidance necessary to coordinate expectations (Whitman, 2009, p.21). (8) On this point, more generally, see Koppl (2015). (9) Kirzner (2000) refers to it as “perfect coordination.” (10) Of course, there is much more to affecting the mutual compatibility of plans. This will involve knowing something about the preferences of one’s potential interactors. (11) Further, Kaldor–Hicks analysis is often theoretically used to determine the wealthmaximizing and therefore efficient assignment of property rights. However, when a prop­ erty title is a large fraction of individual wealth, the identity of the highest-valuing user will be contingent upon who owns the right initially. “When A is initially assigned the right it is wealth maximizing for him to retain it; when B is initially assigned the right it is wealth maximizing for her to keep it” (Rizzo, 1980b, p. 648). (12) Simply because the Kaldor–Hicks criterion might be used in certain cases consistent with the Austrian emphasis on individual plan coordination rather than aggregate wealth is not a complete argument for the use of the criterion. There may be ethical or other is­ sues involved. (13) We shall discuss this in greater depth in section 13.6. (14) This is consistent with Zywicki’s (2003) argument that in the early development of common law there was a competitive production of precedents that weeded out prece­ Page 22 of 23

Austrian Perspectives in Law and Economics dents that did not serve the long-run interests of the agents. Therefore it was the bottomup system that generated the rules that determined their properties.

Shruti Rajagopalan

Shruti Rajagopalan is an Assistant Professor of Economics, Purchase College, State University of New York. Mario J. Rizzo

Mario J. Rizzo, Associate Professor of Economics, New York University.

Page 23 of 23

Moral Philosophy and Law and Economics

Moral Philosophy and Law and Economics   Brian H. Bix The Oxford Handbook of Law and Economics: Volume 1: Methodology and Concepts Edited by Francesco Parisi Print Publication Date: Apr 2017 Subject: Economics and Finance, Law and Economics, Econometrics, Experimental and Quanti­ tative Methods Online Publication Date: May 2017 DOI: 10.1093/oxfordhb/9780199684267.013.022

Abstract and Keywords This article explores how law and economics is viewed from the perspective of moral phi­ losophy. Given that ‘law and economics’ is a category that covers a large and diverse set of theories and theorists, and that ‘moral philosophy’ as a category is even larger and more diverse, there are a vast number of potential topics and perspectives. The article of­ fers a sample of the many intersections of those two categories. Part I offers a brief intro­ duction to some central ideas of economics and to the different schools of moral philoso­ phy. Part II summarizes some defences of law and economics from the perspective of moral philosophy, while Part III summarizes criticisms from the same source. Keywords: law, economics, moral philosophy, defences

14.1 Introduction THIS chapter will explore how law and economics is viewed from the perspective of moral philosophy. Given that “law and economics” is a category that covers a large and diverse set of theories and theorists, and that “moral philosophy” as a category is, if anything, even larger and more diverse, there are a vast number of potential topics and perspec­ tives. This chapter will offer a sample of the many intersections of those two categories. Section 14.2 offers a very brief introduction to some central ideas of economics and to the different schools of moral philosophy. Section 14.3 summarizes some defenses of law and economics from the perspective of moral philosophy, while section 14.4 summarizes criti­ cisms from the same source, before concluding.

Page 1 of 14

Moral Philosophy and Law and Economics

14.2 Efficiency and Morality 14.2.1 Law and Economics, Efficiency In very rough terms, economic analysis generally covers individual and group decisionmaking regarding the use and distribution of scarce resources, and uses some version of the rational actor model of human behavior, under which individuals try to maximize the satisfaction of their preferences. Economic analysis also tends to focus on the idea of effi­ ciency, the idea of maximizing output relative to inputs. The law and economics move­ ment (again speaking in rough terms) involves the application of economic (p. 289) con­ cepts and reasoning to law—to legal rules, legal reasoning, the reform of law, and the the­ ories purporting to describe and explain legal doctrine or the law in general. Economic analysis of law views law instrumentally, as a means to attain collective objectives (as contrasted with traditional or formalistic approaches to law, which treat it as a distinct form of reasoning to be evaluated in terms other than its consequences).1 Economic analysis often focuses on efficiency, a measure of the amount of output relative to input. Sometimes when economists speak of “efficiency,” they refer to Pareto analysis (e.g. Pareto, 1971, p. 261; Parisi, 2013, pp. 214–216). A change from one distribution of goods to another is said to be “Pareto superior” when at least one person is made better off relative to that person’s preferences, and no one is made worse off (everyone else is indifferent between the situation before and after the change). Similarly, a move is “Pare­ to inferior” when at least one person is made worse off (relative to that person’s prefer­ ences), and no one is made better off. A “Pareto optimal” situation is one where no one can be made better off without at least one person being made worse off.2 Pareto analysis is considered relatively uncontroversial, as why would anyone object to changes that (by definition) made at least one person better off and no one worse off?3 The difficulty is that basing decisions on Pareto analysis alone will not get us very far, as there are relatively few rules, actions, or events in the real world that would make some people better off without making some other people worse off, and then we are left with the problem of making controversial choices among people or among preferences. An alternative form of efficiency analysis was developed by Nicholas Kaldor and J. R. Hicks (Kaldor, 1939; Hicks, 1939). Under Kaldor–Hicks efficiency (also known as “poten­ tial Pareto” efficiency), a move is Kaldor–Hicks superior if those who have benefited by some change could compensate those left worse off in a way that, after the (hypothetical) compensation, those worse off would now be at least indifferent, and those made better off would still be left better off. (Again, it is important to remember that this is hypotheti­ cal compensation; if those better off had actually compensated those made worse off to the level of at least being indifferent, the result would be a Pareto superior move.) The ar­ gument is that a Kaldor–Hicks superior situation is one that is in some sense “equivalent to” the Pareto superior situation, and thus to be treated as an improvement over the sta­ tus quo. At the same time, it is clear that Kaldor–Hicks analysis does require a willingness to trade off some people’s gains against other people’s losses, in a (p. 290) way that Pare­ Page 2 of 14

Moral Philosophy and Law and Economics to analysis does not,4 and the problem of justifying decisions to those who are left worse off remains. Additionally, as economic analysis generally equates preferences with willingness to pay (and more generally assumes or asserts that our actions—e.g. buying or not buying some­ thing—imply our actual preferences, even when those actions are contrary to what we say, or even what we think, our values and preferences are), it is but a small step to change Kaldor–Hicks efficiency to “wealth maximization” (“wealth” here understood broadly) (Posner, 1979; 1983, pp. 60–87). It is a central position of many versions of law and economics that legal rules should maximize wealth. Generally, law and economics presents itself as an instrumental theory: Given certain ob­ jectives, law and economics offers the correct analytical tools for discovering which legal norms will best effectuate those chosen ends. On that basis, law and economics theorists sometimes dismiss or discount moral criticisms of law and economics, arguing that the criteria of success for law and economics theories are (solely or primarily) the success at predicting the effects legal rules have on behavior (e.g. Craswell, 1989, 2003). There are aspects of law and economics, including some quite controversial claims, that will not be covered at any length in this chapter, because the arguments for and against them do not come primarily from moral philosophy. For example, one can find law and economics theorists (especially much earlier in the movement’s history) making descrip­ tive claims, in particular the assertion that economic efficiency explains current legal rules in private law areas and how judges have developed legal doctrine in these areas over time (e.g. Priest, 1977; Rubin, 1977; Posner, 2014, pp. 31–33). Whether these are de­ fensible claims turns on matters of (legal) history, not moral philosophy.

14.2.2 Types of Moral Philosophy There are three main approaches to moral philosophy:5 consequentialism, deontology, and virtue ethics. Consequentialist approaches assert that morality requires a focus on consequences, usually an emphasis on maximizing some kind(s) of good outcomes (Sin­ nott-Armstrong, 2011). The best-known form of consequentialism is utilitarianism, which asserts that both individuals and communities should strive to maximize “utility” (p. 291) (a basic positive feeling roughly equated with happiness) for everyone;6 utilitarianism, in turn, is often divided into “act utilitarianism” (which directs that one do the action that maximizes utility) and “rule utilitarianism” (which prescribes instead following the rules or principles that will maximize utility). Deontology is often understood in terms of a rejection of consequentialism: it is a view that actions have their moral value independent of their consequences (Alexander and Moore, 2012). Thus, certain actions are to be done even if they might lead to bad conse­ quences, and other actions are to be avoided even if they might lead to good conse­ quences (e.g., some deontologists believe that one should always tell the truth, even when the truth might cause distress, and that one should never use torture, even if torture might lead to information that would prevent the deaths of many innocent people). Deon­ Page 3 of 14

Moral Philosophy and Law and Economics tological approaches often focus on rules and duties. Immanuel Kant’s moral philosophy is arguably the best-known form of deontological ethics. Virtue ethics emphasizes not the consequences of particular actions, nor duties regarding particular actions, but rather the virtues of the agent (Hursthouse, 2012). This approach, which is grounded in Aristotle’s ideas about morality, directs attention to the task of be­ ing a good person, which in turn involves the development of the virtues. Virtue ethics does offer guidance for how to act in individual cases, but the guidance may be contextu­ al, varying from situation to situation, and from person to person.

14.3 Defense of Law and Economics from a Moral Philosophy Perspective Richard Posner at one point (1979; 1980, pp. 92–103) argued that wealth maximization is the best approximation of justice, and has the advantage, as a matter of moral analysis, of combining the strengths of both utilitarian and deontological/Kantian theories of morality, without, he claims, carrying the distinctive weaknesses of those approaches. The distinc­ tive strength of utilitarianism is that it appears to be uncontroversial: everyone wants utility, and we all prefer more of it to less. Among the problems of utilitarianism is that it is notoriously difficult to measure utility for any individual, to compare utility across per­ sons, and to sum utility over a group. Posner pointed out that maximizing social wealth is comparably uncontroversial compared to maximizing utility (most of us prefer more to less of it for ourselves, and for our community), while wealth remains vastly easier than utility to measure, to compare, and to sum. On the Kantian side, Posner noted Kant’s emphasis on autonomy, and equated that with respect for individuals’ autonomous choices. Economic approaches in general, and (p. 292) wealth maximization in particular, give the highest value to voluntary market transactions (as such transactions, almost by definition, make all parties to the transac­ tion better off), and, in circumstances where voluntary transactions are not possible or practical, wealth maximization tries to determine what choices people would have made (“hypothetical consent”). Posner’s example of hypothetical consent was the standard to apply to accidents, where he argued (1) one could not achieve express consent for which standard of liability to apply: it was obviously impractical for everyone to enter agree­ ments regarding the appropriate standard of liability with all those whom they might hurt or who might hurt them in a future accident; and (2) if one did not know if one was more likely to be the person causing the accident or the person harmed by the accident,7 one would naturally choose whichever rule would lead to the greatest social wealth in the long term. That is, the wealth-maximizing rule is one to which all would agree if asked.8 The response from moral philosophers has been that wealth maximization in fact lacks the virtues of utilitarianism and Kantian theory, while containing additional weaknesses. Wealth may be easier to count and to sum across individuals than utility is, but utility— and its near-cognates, happiness, pleasure, and well-being—are clearly valuable in them­ Page 4 of 14

Moral Philosophy and Law and Economics selves, while wealth is, at best, a means to (valuable) ends. Because (social) wealth is not valuable in itself, it is far from obvious that it is something that should be maximized—es­ pecially if the increase in community wealth is accompanied by greater inequality, or a distribution of wealth that is in some other way unjust. Also, wealth maximization’s analyses in terms of ability to pay create well-known para­ doxical outcomes: for example, it seems that one would increase social wealth by giving the prized possessions of a poor person to her rich neighbor, even if the possessions meant everything to the poor person and almost nothing to the rich neighbor, simply be­ cause the rich person was able and willing to pay more to own those possessions than the poor person would be able to pay.9 Additionally, some commentators are suspicious of the way the argument for wealth maximization depends on hypothetical consent, these com­ mentators arguing that hypothetical consent carries little moral weight (compared to ac­ tual or express consent), at least in the ways that the concept is used by law and econom­ ics theorists (e.g., Dworkin, 1985, pp. 275–280). A moral philosopher might point out certain virtues that law and economics analyses tend to have, even if those virtues are not unique to that approach—one can find them else­ where, including within other forms of analysis. Nonetheless, these are virtues that (p. 293) seem to come easily or naturally to the kind of analysis economics uses. One ex­ ample is the way economics focuses on how legal norms can have consequences that are unintended, and may in fact be the opposite of what the lawmakers intended in enacting a rule. For example, law and economics theorists have shown the ways in which anti-dis­ crimination laws, legal rules meant to protect certain categories of tenants, or rules meant to give protections against arbitrary termination to workers or franchisees, could all end up harming the interests of the groups the rules were meant to protect.10 In a like way, economic analysis of legal rules and transactions often discloses the overlooked ef­ fects on third parties.11 This ability to consider the long-term benefits and harms to indi­ viduals is an important plus, from a moral perspective.

14.4 Criticism of Law and Economics from a Moral Philosophy Perspective 14.4.1 Justice and Efficiency As already mentioned, moral philosophers have argued strenuously that the maximization of wealth is not the same as justice, and, in fact, is not even a good proxy for any objec­ tive worth seeking for its own sake (e.g. Coleman, 1988, pp. 95–132; Dworkin, 1980a, 1980b; Kronman, 1980; Leff, 1974). A number of theorists within the law and economics movement have accepted the critique, at least in part. While they have continued to ar­ gue that wealth maximization, or some other form of efficiency, is a worthy objective, they concede that it is merely one objective among many, and it may need to be traded off

Page 5 of 14

Moral Philosophy and Law and Economics against fairness and justice and other objectives (e.g., Calabresi and Melamed, 1972; Pos­ ner, 2014, pp. 34–35). Louis Kaplow and Steven Shavell, in a series of publications (e.g. Kaplow and Shavell, 2001, 2002), presented a spirited defense of welfare economics, arguing that it is claims of justice and fairness that should give way if they conflict with the preferences of the community (especially if the choice for an outcome contrary to fairness is without dissent, as (p. 294) in the Pareto analysis, discussed above). The authors wrote: “social decisions should be based exclusively on their effects on the welfare of individuals—and, according­ ly, should not depend on notions of fairness, justice, or cognate concepts” (Kaplow and Shavell, 2002, xvii (emphasis in original)). In a sense, what Kaplow and Shavell were of­ fering was a moral argument against using moral philosophy when making policy deci­ sions. However, there are serious analytical and conceptual difficulties with the Kaplow/Shavell analysis (e.g. Kornhauser, 2011, section 5; Coleman, 2003), and few outside of law and economics (and not that many within) have been convinced by their argument. At best, their argument appears to be the tautological claim that if all choices are to be evaluated only on the basis of preference satisfaction, then any other value (including fairness, jus­ tice, etc.) is a legitimate factor in decisions only to the extent that it is preferred by indi­ viduals (Coleman, 2003, pp. 1523–1528); a structurally similar argument could be made using any evaluation standard as the starting point.12 At worst, the Kaplow/Shavell analy­ sis is grounded on a treatment of preferences, values, and welfare that is analytically in­ coherent and normatively unpersuasive (e.g. Kornhauser, 2011, section 5.1; Coleman, 2003, pp. 1538–1543).

14.4.2 Application of the Model 14.4.2.1 Modeling Human Behavior There are criticisms of law and economics that are basically methodological, but that are frequently made in the context of a moral objection. The critique is that the rational actor model of human behavior (and its slight variants used, e.g., in game theory and public choice theory) distort or misrepresent human behavior, and this defect in the model can result in morally objectionable policy prescriptions. One line of criticism, or at least of concern, is that economic analysis’s understanding of human behavior is distorted in being “flattened out,” and that this can have unfortunate consequences for what economists prescribe by way of policy. Whether economic analysis is grounded on preference satisfaction, wealth maximization, or maximizing utility, the approach involves seeing all human actions and decisions in terms of a single variable. However, many moral philosophers believe that human action and motivation involve a variety of values and goods, and a complexity of different sorts of mental states, and that this complexity cannot be reduced (commensurated) to a single metric without serious distortion (e.g. Nussbaum, 1997; Finnis, 1990).

Page 6 of 14

Moral Philosophy and Law and Economics As one example of the criticism, John Finnis argued that the economic analysis of tort law (accident law), by reducing everything to a single metric, necessarily distorts a central moral distinction in the area, between intentional harms and accidental harms (Finnis, 1990, pp. 200–203). Oliver Wendell Holmes famously observed that “even a dog (p. 295) distinguishes between being stumbled over and being kicked” (Holmes, 1887, p. 3), that is, between accidental and intentional harm. The question is: How effectively can eco­ nomic analysis recognize the same thing? After all, the harm done might be the same, whether done negligently or on purpose. The victim may subjectively feel more resent­ ment upon finding out the harm was intentional rather than accidental, and perhaps eco­ nomic analysis can incorporate that subjective resentment into its cost analysis or its evaluation of preference satisfaction, but this seems at best an intricate and imperfect way to get to a distinction both clear and basic to conventional moral analysis. A different sort of criticism of the rational actor model has been offered by a number of scholars within the economic tradition: they have argued that the model should be changed to reflect experimental evidence that people generally do not evaluate or reason in the simple preference-satisfaction-maximizing way the model suggests. These chal­ lenges, sometimes discussed under the rubrics “bounded rationality,” “behavioral eco­ nomics,” or “cognitive biases” (e.g. Simon, 1955; Kahneman, Slovic, and Tversky, 1982; Kahneman, 2011), often indirectly involve moral objections, in that the proposed revised model may place a greater descriptive and prescriptive emphasis on altruism, and the dif­ ferences between the models may determine which sort of government policies are justi­ fied (e.g. whether and when government regulation, including paternalistic intervention, is justified).

14.4.2.2 (Mis-)Understanding Law As earlier mentioned, economic analysis is frequently offered as the best way to under­ stand particular legal rules and doctrinal areas of law. However, the notion of “under­ standing” here may be ambiguous between description, prescription, or some combina­ tion of the two.13 Law and economics theorists tend to view the rules of tort law, descrip­ tively or prescriptively, as being about reducing the overall costs relating to accidents (the costs imposed by accidents combined with the costs related to accident prevention). Critics respond that this view of tort law fails, both descriptively and prescriptively, be­ cause it ignores or discounts the central role of corrective justice in tort law (Coleman, 1988; cf. Kraus, 2007). To critics, there is something morally objectionable about missing this connection between legal principles on the one hand, and ideas about right, wrong, and justice on the other. A different contentious claim about tort law appears in one of the foundational texts of law and economics, Ronald Coase’s “The Problem of Social Cost” (Coase, 1961). Part of Coase’s point in the article had been to rebut certain welfare economics positions, posi­ tions that attempted to justify government action against businesses that create pollution and other nuisances, action that would force those companies to “internalize their externalities” (to pay the equivalent of the harms they impose on nearby landowners). (p. 296) One part of this rebuttal was the argument that it was inaccurate (or at least un­ Page 7 of 14

Moral Philosophy and Law and Economics helpful) to say that in cases of pollution and other non-intentional tortious acts one party imposed harm (costs) on another; rather, Coase argued, one should say that each party imposed costs on the other, that there were costs that came from the intersection or con­ flict of two activities. This was Coase’s idea of “the reciprocity of causation,” and it has been strongly criticized by those who believe that claims of right, property, and justice (rather than efficiency) are at the core of tort law, and that we can and should speak in such cases of “one party harming the other” (e.g. Epstein, 1979, 1987; Fletcher, 1996, pp. 164–167).

14.4.2.3 Applying the Model to Non-Commercial Behavior For a long time, economic analysis was applied primarily to commercial behavior.14 Today, economic analysis (especially within law and economics scholarship) appears in areas far afield from economic transactions (e.g. Posner, 2014, pp. 159–189, 891–985). Gary Becker made important efforts to show how economic analysis (often relatively straightforward applications of ideas about incentives and disincentives) could expand our understanding of areas like criminal law and family law, including predictions about the effects of changes in rules or social norms (e.g. Becker, 1968, 1971, 1991). While most scholars now accept that economic analysis can offer valuable perspectives on matters outside com­ mercial activities, there remain debates about whether there are also significant limita­ tions on economic analysis when it deals with matters like social norms, altruism, social relationships, romantic and family ties, political ideology, and so forth (e.g. White, 1987; West, 1998; Bix, 2015). As with the previous discussion of economic analysis and justice/ fairness, the ultimate conclusion about the application of economic analysis to non-com­ mercial activities may well be that economic analysis offers insights, but insights that re­ quire supplementation by, or balancing with, other approaches (e.g. Bix, 2015).

14.4.3 Criticism of Public Choice Theory Public choice theory applies the basic principles of economics and the rational choice model of human behavior to the actions of public officials (e.g. Buchanan and Tullock, 1962; Farber and Frickey, 1991). One moral (or “moralistic”) criticism of public choice theory is that the approach denies or discounts the idea of “the public good” or “the pub­ lic interest” as a motivation for official action, or as the proper grounds for judicial review of official action. Public choice theory instead portrays officials, including legislators and judges, as being only or primarily interested in satisfying their own preferences—re-elec­ tion, promotion, gaining in status, (for judges) not being reversed, and so forth. One re­ sponse is that public choice theorists need not deny that some—and perhaps even many— officials act altruistically, for the public interest (as they see it). (p. 297) Public choice the­ orists need only argue that a model with a simplifying assumption that officials act “ratio­ nally” has significant predictive power, doing a better job of predicting official actions than models that assume that officials are generally altruistic and working for the com­ mon good (e.g. Levmore, 2002). If this is the case, then public choice theory is useful, however cynical its assumptions might appear to be.

Page 8 of 14

Moral Philosophy and Law and Economics

14.5 Conclusion While moral philosophy and law and economics are often seen as antagonists, this con­ flict is neither universal nor inevitable. While there are many criticisms of aspects of, or versions of, law and economics grounded in moral claims, one should also note how eco­ nomics analysis at its foundation is related to one major school of moral philosophy, con­ sequentialism (in particular, utilitarianism). Additionally, many theorists have found a way to try to make economic analysis compatible with moral views, by seeing both efficiency analysis and claims of justice or fairness as legitimate objectives, objectives that some­ times must be traded off. Finally, a number of theorists defend (aspects of or versions of) law and economics from a moral philosophy perspective.

References Alexander, Larry and Michael Moore (2012). “Deontological Ethics,” in Edward N. Zalta, ed., The Stanford Encyclopedia of Philosophy, available at . Becker, Gary S. (1968). “Crime and Punishment: An Economic Approach.” Journal of Polit­ ical Economy 76: 169–217. Becker, Gary S. (1971). The Economics of Discrimination. 2nd edition. Chicago: University of Chicago Press. Becker, Gary S. (1991). A Treatise on the Family. Enlarged edition. Cambridge, MA: Har­ vard University Press. Bix, Brian (2015). “Engagement with Economics: The New Hybrids of Family Law/Law & Economics Thinking,” in Aristides N. Hatzis and Nicholas Mercuro, eds., Law and Eco­ nomics: Philosophical Issues and Fundamental Questions, 245–266. London: Routledge. Buchanan, James M. and Gordon Tullock (1962). The Calculus of Consent: Logical Foun­ dations of Constitutional Democracy. Ann Arbor, MI: University of Michigan Press. Calabresi, Guido and A. D. Melamed (1972). “Property Rules, Liability Rules, and Inalien­ ability: One View of the Cathedral.” Harvard Law Review 85: 1089–1128. Coase, Ronald (1961). “The Problem of Social Cost.” Journal of Law and Economics 3: 1– 44. Coleman, Jules L. (1988). Markets, Morals and the Law. Cambridge: Cambridge Universi­ ty Press. Coleman, Jules L. (2003). “The Grounds of Welfare.” Yale Law Journal 112: 1511–1543. Craswell, Richard (1989). “Contract Law, Default Rules, and the Philosophy of Promis­ ing.” Michigan Law Review 88: 489–529.

Page 9 of 14

Moral Philosophy and Law and Economics Craswell, Richard. (2003). “In That Case, What is the Question? Economics and the Demands of Contract Theory.” Yale Law Journal 112: 903–924. (p. 298)

Dworkin, Ronald (1980a). “Is Wealth a Value?” Journal of Legal Studies 9(2): 191–226. Dworkin, Ronald (1980b). “Why Efficiency: A Response to Professors Calabresi and Pos­ ner.” Hofstra Law Review 8: 563–590. Dworkin, Ronald (1985). A Matter of Principle. Cambridge, MA: Harvard University Press. Epstein, Richard A. (1979). “Causation and Corrective Justice: A Reply to Two Critics.” Journal of Legal Studies 8: 477–504. Epstein, Richard A. (1987). “Causation—In Context: An Afterword.” Chicago-Kent Law Re­ view 63: 653–680. Farber, Daniel A. (2015). “Autonomy, Welfare, and the Pareto Principle.” in Aristides N. Hatzis and Nicholas Mercuro, eds., Law and Economics: Philosophical Issues and Funda­ mental Questions, 159–182. London: Routledge. Farber, Daniel A. and Philip P. Frickey (1991). Law and Public Choice: A Critical Introduc­ tion. Chicago: University of Chicago Press. Finnis, John (1990). “Allocating Risks and Suffering: Some Hidden Traps.” Cleveland State Law Review 38: 193–207. Fletcher, George P. (1996). Basic Concepts of Legal Thought. Oxford: Oxford University Press. Hart, H. L. A. (2012). The Concept of Law. 3rd edition. Oxford: Oxford University Press. Hicks, J. R. (1939). “The Foundations of Welfare Economics.” Economics Journal 49: 696– 712. Holmes, Oliver Wendell (1887). The Common Law. London: Macmillan. Hursthouse, Rosalind (2012). “Virtue Ethics,” in Edward N. Zalta, ed., The Stanford Ency­ clopedia of Philosophy, available at . Kahneman, Daniel (2011). Thinking Fast and Slow. New York: Farrar, Straus and Giroux. Kahneman, Daniel, Paul Slovic, and Amos Tversky, eds. (1982). Judgment Under Uncer­ tainty: Heuristics and Biases. Cambridge: Cambridge University Press. Kaldor, Nicholas (1939). “Welfare Propositions of Economics and Interpersonal Compar­ isons of Utility.” Economics Journal 49: 549–552. Kaplow, Louis and Steven Shavell (2001). “Fairness Versus Welfare.” Harvard Law Review 114: 961–1388. Page 10 of 14

Moral Philosophy and Law and Economics Kaplow, Louis and Steven Shavell (2002). Fairness versus Welfare. Cambridge, MA: Har­ vard University Press. Kornhauser, Lewis (2011). “The Economic Analysis of Law,” in Edward N. Zalta, ed., Stan­ ford Encyclopedia of Philosophy, available at . Kraus, Jody (2007). “Transparency and Determinacy in Common Law Adjudication: A Philosophical Defense of Explanatory Economic Analysis.” Virginia Law Review 93: 287– 359. Kronman, Anthony T. (1980). “Wealth Maximization as a Normative Principle.” Journal of Legal Studies 9(2): 227–242. Leff, Arthur Allen (1974). “Economic Analysis of Law: Some Realism About Nominalism.” Virginia Law Review 60: 451–482. Levmore, Saul (2002). “From Cynicism to Positive Theory in Public Choice.” Cornell Law Review 87: 375–383. Nussbaum, Martha C. (1997). “Flawed Foundations: The Philosophical Critique of (a Par­ ticular Type of) Economics.” University of Chicago Law Review 64: 1197–1214. Pareto, Vilfredo (1971). Manual of Political Economy. Trans. Ann S. Schwier, ed. Ann S. Schwier and Alfred N. Page. New York: Augustus M. Kelley. (p. 299)

Parisi, Francesco (2013). The Language of Law and Economics. Cambridge: Cambridge University Press. Posner, Richard A. (1979). “Utilitarianism, Economics, and Legal Theory.” Journal of Legal Studies 8: 103–140. Posner, Richard A. (1980). “The Ethical and Political Basis of the Efficiency Norm in Com­ mon Law Adjudication.” Hofstra Law Review 8: 487–507. Posner, Richard A. (1983). The Economics of Justice. Cambridge, MA: Harvard University Press. Posner, Richard A. (2014). Economic Analysis of Law. 9th edition. New York: Aspen Pub­ lishing. Priest, George L. (1977). “The Common Law Process and the Selection of Efficient Rules.” Journal of Legal Studies 6: 65–82. Rubin, Paul H. (1977). “Why is the Common Law Efficient?” Journal of Legal Studies 6: 51–63. Scitovsky, T. De (1941). “A Note on Welfare Propositions in Economics.” Review of Eco­ nomic Studies 9: 77–88. Page 11 of 14

Moral Philosophy and Law and Economics Sen, Amartya (1971). “The Impossibility of a Paretian Liberal.” Journal of Political Econo­ my 78: 152–157. Simon, Herbert (1955). “A Behavioral Model of Rational Choice.” Quarterly Journal of Economics 69: 99–118. Sinnott-Armstrong, Walter (2011). “Consequentialism,” in Edward N. Zalta, ed., The Stan­ ford Encyclopedia of Philosophy, available at . West, Robin (1998). “The Other Utilitarians,” in Brian Bix, ed., Analyzing Law: New Es­ says in Legal Theory, 197–222. Oxford: Clarendon Press. White, James Boyd (1987). “Economics and Law: Two Cultures in Tension.” Tennessee Law Review 54: 161–202.

Notes: (1) This is also in contrast to modern forms of analytical legal philosophy, which build the­ ories of the nature of law around the perspective of citizens, at least some of whom see legal rules not merely as costs and benefits, but as reasons for action (independent of the likely sanctions or rewards) (Kornhauser, 2011, section 2.3; Hart, 2012, pp. 82–91). (2) Despite the usually positive term “optimal,” there need not be anything particularly at­ tractive about a “Pareto optimal” distribution: e.g., if I own everything, and no one else in the group owns anything, that would be a Pareto optimal distribution, for any change in circumstances would make me worse off. (3) The Pareto principle is actually more controversial than would at first appear, especial­ ly as one disentangles the different possible understandings in terms of consent, choice, and preference. See Sen, 1971; Farber, 2015. However, that is a discussion for another time. (4) Additionally, in what has become known as “the Scitovsky Paradox,” two states of af­ fairs can each be Kaldor–Hicks superior to the other (Scitovsky, 1941; Coleman, 1988, pp. 104–105). This obviously undermines the ability of Kaldor–Hicks analysis to create a rea­ soned preference among policy alternatives in some situations. (5) Two caveats: First, “moral philosophy” here means secular moral philosophy. This chapter will not be discussing moral philosophies grounded on religion, though the views of such forms of moral philosophy generally do not vary that much from secular moral ap­ proaches, at least on the topics relevant to this chapter. Second, there are significant ap­ proaches to moral philosophy not covered by the three categories mentioned, including contractualist approaches, and the various forms of relativism, skepticism, and error the­ ories about morality.

Page 12 of 14

Moral Philosophy and Law and Economics (6) Utilitarianism is thus not to be confused with hedonism: utilitarianism is about choos­ ing the outcome that creates, in a well-known expression, “the greatest good of the great­ est number,” not just (as with hedonism) the greatest happiness for oneself. (7) Of course, if one did know ahead of time that one was, say, likely to be the person harmed by the accident, one would naturally prefer whichever standard was likely to lead to the largest and most certain compensatory payment (and one would prefer the stan­ dard creating the smallest or least likely compensation order if one knew that one would be the party causing the harm). (8) Posner assumed that this would be a negligence standard for liability. Whether negli­ gence is in fact more efficient (wealth-maximizing) than strict liability standards is be­ yond the scope of this chapter. (9) This is sometimes known as “wealth effects.” There are corrections and restrictions that could be added to efficiency analysis generally and to wealth maximization in partic­ ular (just as there are comparable corrections and restrictions to act utilitarianism) to avoid the most counterintuitive outcomes, but these alterations do not erase the doubts that the basic model evokes. (10) In very brief and rough summary: (1) with anti-discrimination laws, the effect of these laws is to create a risk of expensive litigation every time an employer decides to fire or not to promote workers in the protected group; employers thus have a financial reason not to hire protected workers; (2) similarly, if it will be difficult or expensive to evict a certain kind of tenant, then landlords will find ways to avoid leasing to such tenants; and (3) where it is difficult to fire an employee or terminate a franchisee, employers and fran­ chisors will be understandably reluctant to take on employees or franchisees who seem risky, while under an at-will termination rule, risky employees and franchisees could be given a chance as the employer and franchisor know that if things do not work out termi­ nation is possible without significant risk of expensive litigation. (11) A standard example here is rent control, where a restriction on rent increases may harm those who are not current tenants but might want to rent in the future, as the re­ strictions may reduce the incentive for builders to create additional rental housing. (12) For example: one could argue with equal legitimacy that if fairness is the sole basis for evaluating policy decisions, then one should never consider preference-satisfaction or overall welfare, except to the extent that either can be incorporated into an analysis of the fairness of an outcome. (13) In legal practice and legal scholarship, especially in common law countries, “rational reconstruction” of a line of cases or an area of law is common. Rational reconstruction in­ volves interpretation of the cases that fits those cases to a significant degree, but with a preference for readings that are normatively attractive.

Page 13 of 14

Moral Philosophy and Law and Economics (14) There were some apparent exceptions that, by some accounts, would include Cesare Beccaria’s and Jeremy Bentham’s ideas about optimal punishments for crimes (Posner, 2014, p. 29 n. 2).

Brian H. Bix

Brian H. Bix is the Frederick W. Thomas Professor of Law and Philosophy at the Uni­ versity of Minnesota. He holds a JD from Harvard University and a D.Phil. from Ox­ ford University. His publications include Jurisprudence: Theory and Context (5th edn, Sweet & Maxwell, 2009), A Dictionary of Legal Theory (Oxford, 2004), and Law, Lan­ guage, and Legal Determinacy (Oxford, 1993).

Page 14 of 14

Critiques of Law and Economics

Critiques of Law and Economics   David Driesen and Robin Paul Malloy The Oxford Handbook of Law and Economics: Volume 1: Methodology and Concepts Edited by Francesco Parisi Print Publication Date: Apr 2017 Subject: Economics and Finance, Law and Economics, Econometrics, Experimental and Quanti­ tative Methods Online Publication Date: May 2017 DOI: 10.1093/oxfordhb/9780199684267.013.024

Abstract and Keywords This article summarizes leading critiques of law and economics. These critiques are grouped into three categories. The article first addresses concerns about the normative value of economic efficiency as a leading goal for law. It then addresses methodological criticisms, which often call into question the coherence of the allocative efficiency con­ cept. Finally, it discusses the interpretive criticisms of law and microeconomics, which view law and economics as a rhetorical form and raises questions based on that view. Scholars have questioned the normative value of economic efficiency as a central goal of law. They have also challenged the coherence and objectivity of the methods used to as­ sess economic efficiency. They have questioned the claim of economics as somehow pro­ viding a scientific justification for law, arguing instead that it constitutes a rhetorical form shifting the terms of legal argument and changing its outcomes. Keywords: law, economics, normative value, economic efficiency, allocative efficiency, microeconomics

THIS chapter summarizes leading critiques of law and economics. For the most part, we put aside objections to particular applications of law and economics to distinct fields of law. Scholars in a variety of fields have voiced the concern that law and economics does not adequately address important values or information central to their specialties. We do not have the space to discuss these particularized objections except as relevant to more general critiques. We focus on rather general criticisms that properly apply to widely shared core commit­ ments within the field of law and economics. We have grouped these critiques into three categories. We first address concerns about the normative value of economic efficiency as a leading goal for law. We next address methodological criticisms, which often call into question the coherence of the allocative efficiency concept. Finally, we discuss the inter­

Page 1 of 21

Critiques of Law and Economics pretive criticisms of law and microeconomics. The interpretive critique views law and economics as a rhetorical form and raises questions based on that view.

15.1 Normative Critiques For the most part, law and economics concerns itself with allocative efficiency, the bal­ ancing of costs and benefits at the margin. Although useful as a model for a market, many scholars question economic efficiency’s value as a leading norm for law.

15.1.1 Critiques of the Normative Value of Wealth Maximization Richard Posner (1979, 1980, 1981a, 1981b) once argued that a wealth maximization norm justifies law and economics’ emphasis on efficiency. Several scholars doubt that (p. 301) wealth maximization constitutes a normative value justifying a central place for it in law (Dworkin, 1980a; Weinrib, 1980; Coleman, 1980, 1982; Kronman, 1980; Mercuro and Ryan, 1984). These critiques aim squarely, although not always explicitly, at Kaldor–Hicks efficiency. Because government decisions usually come about because of disputes among people, they rarely if ever can be Pareto optimal (Dworkin, 1980a, 1980b; Calabresi, 1991). In practice, therefore, law and economics scholars implicitly rely on a very different efficien­ cy concept than the concept used in the economic analysis of markets, that of Kaldor– Hicks efficiency, the idea that a legal decision is efficient if it increases the wealth of the people benefiting from it more than it decreases the wealth of others (Posner, 2014; Cole­ man, 1988; Kronman, 1980; e.g., Driesen, 2011a). Kaldor–Hicks efficient legal decisions increase net wealth (Kronman, 1980; Posner, 1980). Kaldor–Hicks efficiency is said to be good, because those benefiting from the decision could compensate those harmed for their losses. This criterion, however, does not require that the law’s beneficiaries actually compensate those whom the law harms and therefore may rationalize theft or govern­ ment takings without compensation, both of which may transfer assets to beneficiaries who value the good more than the current owner. In any case, scholars have sharply criti­ cized the normative value of Kaldor–Hicks efficiency and claimed that Pareto efficiency cannot generally govern legal decisions (e.g. Coleman, 1988; Calabresi, 1991). Scholars criticizing wealth maximization argue that wealth functions as a means toward other ends. Accordingly, its maximization is neither good nor bad in and of itself, but rather good insofar as it advances ends that its defenders have not identified. Moral philosophers often endorse justice as a goal for society and do not consider wealth maximization as having any normative value independent of justice (compare Klapow and Shavell, 2001). Some scholars seeking to define justice more precisely have argued for a “capabilities” approach to human flourishing, which casts doubt on wealth maximization as an ideal (Nussbaum and Sen, 1993; Nussbaum, 2000; Williams, 2002; Purdy, 2005; Chon, 2006; Alexander et al., 2009; Sen, 2009; Roesler, 2011). Proponents of this ap­ proach link human flourishing not to wealth aggregation but to the spread of capacities Page 2 of 21

Critiques of Law and Economics vital to human flourishing to all members of a society, for example, by improving public health. They criticize law and economics (and utilitarianism generally) for neglecting dis­ tribution, rights and freedoms, and the tendency of those suffering persistent harms to adapt (see Sen, 1999; Nussbaum, 2000). A related critique of efficiency as a legal goal criticizes aggregation of existing prefer­ ences as expressed in markets as a criterion for political decision-making (Baker, 1975). Prominent scholars, for example, argue that political decisions should reflect value choic­ es regarding a good society that are different in principle from aggregating the self-inter­ ested preferences of individuals within the society (Sagoff, 1988; Sen, 2009). Individuals, for example, may prefer a tax deduction for a second home while believing that the soci­ ety should not grant such deductions because it needs the revenue to relieve poverty or achieve other social goals. An individual’s own self-interested preferences may differ from what that same individual believes that society should do. Furthermore, democracy should feature debate that changes people’s positions about what a good (p. 302) society is, which may change people’s views, rather than merely aggregate each individual’s wants as expressed in market behavior. One of us has argued that law often should aim to countervail damages created by myopic market behavior, not to institutionalize it (Driesen, 2012). Hence, critics have raised serious questions about law and economics by challenging the normative value of efficiency and wealth maximization.

15.1.2 Critiques of the Notion that Efficiency Maximizes Wealth and Welfare More recent critiques of the wealth-maximization ideal as a means of defending an em­ phasis on allocative efficiency take aim at the assumption that allocative efficiency plays a very important role in maximizing wealth and human welfare (e.g. Ashford, 2004). Econo­ mists and lawyers studying antitrust and intellectual property law (as well as other areas) find that economic growth contributes much more to wealth than allocative efficiency (e.g. Driesen, 2003; Carrier, 2009; Cooter, 2013). And they find that innovation con­ tributes more to economic growth than efficiency. They sometimes buttress these claims by pointing to tensions between maximizing short-term efficiency and innovation, which are pretty well recognized in the economics literature (e.g. de Soto, 2009). Conversely, the financial crisis suggests that systemic risk constitutes a greater threat to economic wealth maximization than inefficiency, which is, after all, ever-present in any real econo­ my. Of course, one can view an economic collapse as inefficient, but that seems rather ob­ tuse. In general, economic efficiency considerations justified neglecting systemic risk in favor of deregulation in the run-up to the financial crisis, so that in practice efficiencybased policies tend to deviate from policies aiming to minimize systemic risk, which toler­ ate near-term inefficiencies in order to guard against occasional disasters. Together, these points suggest that a macroeconomic approach to law and economics may be needed (Driesen, 2012). Such an approach might treat law as a general framework aiming at avoiding systemic risk and maximizing economic opportunity, rather than as a series of transactions aimed at optimality. And it would focus on analyzing change over Page 3 of 21

Critiques of Law and Economics time under conditions of uncertainty rather than performing a simplified static optimiza­ tion analysis. Some scholars have also argued that a reasonably equal distribution of wealth matters more to economic growth than allocatively efficient transactions (e.g. Ashford, 2009). This argument depends (mostly) on the importance of economic equality to production of adequate demand and the potential of reasonably equal distribution of wealth to make more people productive than would be possible in a very unequal society. Finally, scholars seeking to measure human happiness directly (through survey instru­ ments mostly) find that once a certain minimum income is reached, increasing wealth does not bring greater happiness (Easterlin, 1995; Frank, 1999; Diener and Diener, 2002). This empirical information buttresses a longer philosophical literature (p. 303) that doubts the relationship between individual preference satisfaction and individual welfare (e.g. Baker, 1975; Sagoff, 2004; Sen, 2009; Adler and Posner, 2008; Bronsteen, Buccafusco, and Masur, 2010). Put simply, what we think we want is not always what is good for us. The tendency of businesses to actively shape preferences through advertising and other means, hoping to make us desire to make purchases whether they increase our well-be­ ing or not, casts doubts on an efficiency metric built from individual preferences (Baker, 1975). Hence, scholars have raised serious normative questions about law and economics both by challenging wealth maximization as a valid and important norm and by casting doubt on the importance of economic efficiency in maximizing wealth and human welfare.

15.2 Methodological Critiques The economic analysis of law treats a law as if it were a mere transaction and asks whether it allocates resources efficiently. This view distorts law. Law usually functions as a framework that establishes a set of rules (Rubenfeld, 2001). Since a variety of transac­ tions can arise under these rules, law does not determine resource allocation, although it may, of course, influence resource allocation. Hence, without questioning the value of the efficiency concept as a tool for analyzing markets, one may question its aptness as a goal for law. Many scholars have questioned efficiency by raising questions about the coherence of ef­ ficiency determinations and the limitations of the tools used to make them. A fundamental methodological objection to efficiency stems from the problem of starting points. People’s willingness to pay depends heavily on their ability to pay. Hence, economic efficiency tends to favor the wealthy (Bebchuk, 1980; Kronman, 1980). More generally, economists and law and economics scholars evaluate the efficiency of proposed changes against a status quo baseline. Absent a normative theory justifying the baseline, it’s hard to see why a move that efficiently departs from the baseline is desir­ able (Bebchuk, 1980).

Page 4 of 21

Critiques of Law and Economics Many scholars criticize the efficiency ideal as indeterminate for a variety of reasons. The most prominent criticism comes from a finding of behavioral economics about the dispari­ ty between willingness to pay and willingness to accept payment for a good or service. It turns out that market actors generally demand a much higher price to part with some­ thing they own than they would pay to acquire it in the first instance (Rizzo, 1980). Since these two metrics lie at the heart of efficiency determinations, this observation implies that efficiency is indeterminate (Dworkin, 1980a). These disparities fundamentally undermine the coherence of using efficiency as a criteri­ on for the assignment of legal rights. Because rights allocations influence prices, assign­ ments of legal rights based on efficiency can be inherently unstable (Scitovszky, 1941; Rizzo, 1980). A set of prices may make it appear efficient to assign a right to one person, but the assignment can change relative prices, making it efficient to reassign the (p. 304) right to somebody else. This reassignment can in turn make the second assignment ineffi­ cient, suggesting the need to reverse the reassignment. Another coherence critique for law involves problems in establishing the boundaries of systems under analysis. An analyst can characterize a legal change as efficient or ineffi­ cient depending on how she defines the system both in terms of scope and in terms of time (e.g. Driesen, 2011b). This constitutes a pretty serious problem because policymak­ ers seem to treat economic modeling as a revelation of objective truth, when its outcomes depend upon the scope of the model and its chosen assumptions. Law and economics scholars have not emphasized these limitations. This problem of scope is closely tied to the theory of the second best, which teaches that when distortions exist throughout an economy (from taxes or monopoly, for example) a move increasing the efficiency of one sector can decrease the efficiency of the economy as a whole (Lipsey and Lancaster, 1956; Rizzo, 1980). Another challenge to the coherence of efficiency, at least as defined within law and eco­ nomics as opposed to economics, involves the relationship between utility and the distrib­ ution of income. Economists have argued that transferring resources from the rich to the poor increases overall utility because poor people get more marginal utility out of the same amount of resources (Baker, 1975). This argument means that increasing equality should contribute to total utility, thereby undermining a key move in law and economics, the neat separation of distribution and efficiency. It also calls into question the validity of many analyses that assess overall efficiency without adjusting for marginal utility based on distribution. Finally, scholars have criticized the particular techniques used to predict law’s efficiency. Law and economics scholars usually invoke cost–benefit analysis (CBA) as the appropri­ ate method to use, but rarely actually use a quantified analysis of costs and benefits to justify their conclusions (Driesen, 2012). So, for example, Richard Posner and Robert Bork recommended not using CBA of mergers on the sensible ground that analysts can­ not predict the costs and benefits of proposed mergers ex ante. Instead, they assume that companies seeking to merge are rational actors possessed of good information and that Page 5 of 21

Critiques of Law and Economics therefore their proposed mergers are often likely to prove efficient. Still, CBA (meaning quantification of costs and benefits in dollar terms) plays a role in evaluating many regu­ latory proposals in practice. But law and economics scholars often use CBA only as a trope lending their project prestige and an image of rigor. Moreover, CBA has significant limitations, especially in the regulatory fields where it is actually used. Often it analyzes effects (e.g. climate disruption) that have a wide range of uncertainty with respect to magnitude if not with respect to basic facts (Driesen, 2012). Indeed, it is common for the range of uncertainty to be so wide that an analysis incorpo­ rating the full range of uncertainty would not provide concrete policy advice, except per­ haps in very extreme cases. Because of this policymakers often ignore scientists’ pleas to express benefits as a range and construct a CBA based on a very misleading point esti­ mate (McGarity, 1998, p. 24). Such point estimates depend heavily on a host of rather ar­ bitrary assumptions (e.g. Ackerman and Heinzerling, 2004). Furthermore, (p. 305) nonquantifiable benefits usually make such analysis radically incomplete and dangerously misleading. Law and economics scholars, however, more often assume that market actors are rational and possessed of perfect information and purport to reach conclusions about what an ac­ tual CBA would show based on logical deductions from these assumptions. Such an analy­ sis tends to ignore a lot of pertinent information, sometimes the most important informa­ tion we have. So, for example, Richard Posner continued to assume that many mergers would prove efficient based on this set of assumptions long after many economists had become skeptical because of data showing that mergers had often proven inefficient (Pitofsky, 2008). And in the run-up to the financial crisis, law and economics scholars and policymakers recommending the deregulatory measures that helped trigger the crisis did not take into account the likelihood that financial firms would tend toward optimism and hubris during a bubble and that investors would panic after a bubble burst (Driesen, 2014). Although Richard Posner has argued that these tendencies are rational, law and economics scholars recommending deregulation did not take these particular deviations from standard definitions of rationality into account in “predicting” that deregulation would prove efficient. Although perfect information and rational actor assumptions can sometimes yield significant insights, these assumptions can also serve as a means of glo­ rifying markets and assuming away the most significant problems meriting legal reme­ dies. This criticism of scholarly work and policymaking based on rational actor and perfect in­ formation assumptions, unlike the criticism of CBA, focuses on Chicago school law and economics. Although the Chicago school has proven influential enough to make this a valid criticism of law and economics, proponents of law and behavioral economics and law and institutional economics often ally themselves with critics of law and economics generally on this point. In addition to challenging the normative value of efficiency, scholars have challenged law and economics by questioning the coherence of the efficiency concepts and the objectivi­ Page 6 of 21

Critiques of Law and Economics ty of efficiency determinations. These normative and methodological critiques also inform the interpretive critique of law and economics that we explore below.

15.3 Interpretive Critique of an Economic Analysis of Law A key reference point for the interpretive critique of an economic analysis of law is the philosopher Charles S. Peirce (Houser and Kloesel, 1992; Peirce, 1998). Peirce was a cofounder of American Pragmatism along with James and Dewey (Menand, 2001), and de­ veloped a philosophical approach that he referred to as semiotics (Apel, 1995; Hausman, 1997; Hookway, 1992; Ketner, 1992; Liska, 1996; Merrell, 1997; Potter, 1997). The first sustained analysis of law from a semiotic perspective was done by Roberta Kevelson (p. 306) (Kevelson, 1986, 1988, 1991, 1993). In her seminal work, Law as a System of Signs, she covered many aspects of law, including the use of economic analysis in law (Kevelson, 1986; Malloy, 1990a). The interpretive critique of law and economics was fur­ ther developed by Robin Paul Malloy, an active participant in Kevelson’s semiotic round tables held at the Pennsylvania State University, and for several years an affiliated Re­ search Fellow of her Center for Semiotic Research in Law, Government, and Economics (Brigham, 1999; Malloy, 1999). Malloy used the semiotic approach developed by Kevelson to advance a critique of an economic analysis of law focused on economic analysis as a rhetorical and cultural-interpretive device (Malloy, 1989, 1990a, 1990b-a, 1991a, 2000, 2003, 2004, 2009). Malloy’s work paralleled and benefited from the work of others writ­ ing on legal semiotics (Balkin, 1991; Brion, 1991; Goodrich, 1986, 1987; Jackson, 1985; Kennedy, 1991; Paul, 1991), and on such related topics as: the rhetoric of economics (Mc­ Closkey, 1985, 1990, 1994); economics and culture (Throsby, 2001); economics and aes­ thetics (Gagnier, 2000); volitional economics (Bromley, 2009); feminist economics (Ferber and Nelson, 1993); behavioral economics (Sunstein, 2000; Sunstein and Thaler, 2009); and humanistic economics (Lutz and Lux, 1988). The interpretive critique is primarily focused on economics as a system for understanding markets as a dynamic process of human interactions and exchange. It does not equate economics with the market but instead understands economics as one of several ways of interpreting the market. Other ways of understanding market relationships may be grounded in sociology, political science, and feminist or critical legal theory. Even within economics there can be variations in how market activity is interpreted based on one’s commitment to one of several schools of thought (Malloy, 1990a, 1991). Some of these variations include behavioral, institutional, Marxist, neoclassical, and Austrian economics (Mercuro and Medema, 2006). The interpretive critique is not about the formal doctrines, theorems, and assumptions that define any of these approaches. It is focused on the way that each or any of these approaches may be used to influence and transform legal mean­ ing when used to analyze legal relationships.

Page 7 of 21

Critiques of Law and Economics The interpretive critique recognizes that the primary task of lawyers and judges involves interpreting and translating texts (Garner and Scalia, 2012; Goodrich, 1986). In tradition­ al jurisprudence, interpretation is guided by case precedent under the doctrine of stare decisis, and by the norms and customs of the legal community. Law is interpreted by close and careful examination of cases, and by studying the way that lawyers and judges ana­ lyze competing claims. In general, traditional legal interpretation seeks to achieve rea­ sonable, consistent, and predictable outcomes that advance a sense of fairness and jus­ tice within the community. With the rise of the law and economics movement, legal inter­ pretation began to include references to and borrowing from economics. This included a goal of advancing efficient and wealth-maximizing outcomes in addition to or as the equivalent of fair and just ones (Posner, 1981a, 1990, 1999). The issue then became one of thinking about the ways that law and legal meaning were being transformed by the use of economics as a key interpretive reference in deciding legal disputes and promoting legal reform. From the outset, the economic analysis of law made law the subject of economic inquiry. Therefore, the interpretive frame of reference was one of an economist examin­ (p. 307)

ing law as a subject of interest. Consequently, inquiry focused on questions that an econo­ mist might ask regarding the application of economic ideas and concepts to law. These questions included such things as testing particular legal rules and doctrines to assess their implications for achieving economic efficiency and wealth maximization (Cooter and Ulen, 2007; Posner, 2014). In this respect, legal economists simply treated law as “raw material” for the economist to evaluate (Cooter and Ulen, 2007; Posner, 2014). This inter­ pretive framework privileged economics and the economic point of view over that of law; a perfectly fine perspective for professional economists but hardly ideal for lawyers and judges. An interpretive critique of an economic analysis of law, therefore, begins with the recognition of the interpretive hegemony of economics in making law the subject of its in­ quiry. Thus, instead of treating law as a subject of economic analysis, the interpretive ap­ proach argues that legal professionals should be treating economics as a subject of law. In making an “interpretive turn” to evaluate economic criteria, assumptions, and doc­ trines from the perspective of law, one no longer asks if a legal rule is efficient but rather if an efficiency criterion promotes justice and fairness in a given situation. One no longer thinks in terms of a well-functioning and efficient market as an end in itself, but in terms of the market as a means to advance important normative values. As a result of this turn, the lawyer’s interpretive point of reference is shifted back to law and to the legal culture and tradition that gives law its meanings and values. The questions then become ones of asking how rearranging economic relationships, altering income distributions, resetting discount rates, changing initial allocations, and adjusting other market variables can make legal outcomes more just and more fair. Instead of privileging efficiency and seek­ ing to maximize wealth, the goals shift to promoting legal values such as fairness, equali­ ty, and access. With the interpretive turn, the lawyer’s task is one of identifying the best way to use market mechanisms to facilitate legal goals and values, and not one of reshap­ ing law to fit economic criteria and doctrines. Thus, the interpretive turn positions eco­ nomics as a strategic “tool kit” for advancing particular normative goals. For legal profes­ Page 8 of 21

Critiques of Law and Economics sionals, determining how and when to invoke particular economic terms and concepts fa­ cilitates the achievement of pragmatic results while giving them a “gloss” of neutrality and objectivity. Importing economics into law, however, does not make law objective because interpreta­ tion is always influenced by a subjective point of view. Moreover, law is concerned with being fair, just, and reasonable, and with avoiding arbitrary and capricious decision-mak­ ing. This is a value-based undertaking and not an amoral or “objectively value-free” process in the sense that some legal economists might suggest. In practice, therefore, economics provides grammatical and rhetorical devices for framing legal inquiry and for advancing certain value-based approaches to the allocation and distribution of scare re­ sources (Malloy, 1987, 1992, 2003, 2004). For example, assuming that resources should move to the highest bidder discounts the preferences of those with little or nothing to use for bidding. This simple and fundamental assumption in economics (p. 308) expresses a so­ cio-economic value. While one may or may not agree that this is a good value to import in­ to law, the resolution of the matter is one grounded in persuasion and opinion and not in an objective calculus of economic analysis. Central to an interpretive critique of an economic analysis of law are concerns regarding the ways that economics implicitly and opaquely works to shift the underlying meanings and values of law (Malloy, 1987, 1992, 2003, 2004). Economic terms such as efficiency and wealth maximization are frequently used in substitution for traditional legal values such as fairness and justice, or are offered as equivalent ways of expressing similar val­ ues; but this is not the case. Efficiency and wealth maximization have particular mean­ ings to an economist and these meanings are typically different and narrower than what lawyers mean when they discuss fairness and justice. When these economic terms are substituted for the legal norms of justice and fairness they facilitate particular meanings limited and defined by economists, and not lawyers. A fair price in law is not necessarily the same as an efficient or wealth-maximizing price in economics. The difference can be important, as when the law seeks to protect a fair market value, a fair rate of return, and reasonable investment-backed expectations. Substituting one word for another word and importing economics into law always results in a change in meaning and a shift in values. The interpretive critique does not necessarily opine as to whether the changes in mean­ ing and the shifts in value are substantively good or bad; the critique is one of pointing out that the economic analysis of law is not a neutral and objective undertaking. The eco­ nomic analysis of law imports the values and assumptions of economics into law, and lawyers must make informed and normative judgments concerning when and how to use economic devices to advance particular strategic objectives (Malloy, 1986, 1987, 1991a, 1991b). Some of the more common terms and concepts borrowed from economics and incorporat­ ed into law include making references to efficiency, free markets, competition, rational actors, costs and benefits, the tragedy of the commons, the prisoner’s dilemma, the pres­ ence of transaction costs, and the implications of the wealth effect (Cooter and Ulen, 2007; Malloy, 2004; Posner, 2014). Each of these devices functions as an interpretive Page 9 of 21

Critiques of Law and Economics “screen” to filter and direct legal reasoning in favor of meanings and values embedded within the language and assumptions of the economist’s trade. These devices may work to favor people with resources and to affirm the validity of prior distributions. They may create an assumption that individuals can easily coordinate interdependent activities and solve complex problems through private contract without government oversight and regu­ lation. In the alternative, the use of some of these devices may suggest a need to inter­ vene in private arrangements because of the difficulty of individuals achieving socially ef­ ficient outcomes due to such variables as transactions costs, externalities, and poorly de­ fined or incomplete property rights. Some of the devices may undervalue the role of histo­ ry and culture in the forming of legal meaning, while others may undermine a commit­ ment to altruism and to the idea of the “public” interest. Moreover, most of these econom­ ic devices function to redirect legal conversation away from considerations of morality, virtue, sympathy, fairness, and justice. In practice, these devices function to change the structure of legal discourse and shift the underlying meanings and (p. 309) values of law. These shifts have implications for outcomes and they can therefore be used strategically to expand the opportunities for developing legal arguments that favor a particular result. In addition to the use of specific economic terms and concepts in law, there are several thematic assumptions from economics that have implications for the structure and sub­ stance of legal reasoning and meaning. The most commonly used thematic assumptions include: (1) objective point of “viewlessness”; (2) invariance; (3) calculus of choice; and (4) the fact–value distinction. Each of these thematic devices is discussed below.

15.3.1 Point of Viewlessness The point of viewlessness critique challenges the assumption that economic inquiry has no particular interpretive point of view (MacKinnon, 1989). This critique focuses on the idea of the economist as working with complete objectivity and rationality, and being un­ informed by personal bias and cultural influences. A point of viewlessness is amoral, ster­ ile, and painfully scientific. In contrast to this scientific approach, interpretation theory focuses on an interpretive community—a shared culture and set of values that make inter­ pretation possible. From this perspective, a person must be part of a community; part of a culture in which certain signs, gestures, words, and actions have a shared meaning and are therefore capable of interpretation. Interpretation is not simply the product of isolat­ ed, detached, and objective thought; interpretation always embodies a point of view. From an interpretive perspective, economic analysis has a point of view; it has a bias grounded in its assumptions and its norms. The economist might believe that she is en­ gaged in mere factual “observation” and not in interpretation but observations are influ­ enced by cultural context, and ascribing meaning to that which is observed requires ref­ erence to an interpretive community (Hom and Malloy, 1994). Likewise, the legal econo­ mist may believe that by using economic analysis in evaluating law, law is made more ob­ jective, more neutral, and more scientific (Hackney, 2006). In so doing, the legal econo­ mist may seek to reclaim a sense of the impersonal in law; to reassert the claim that “lady justice” is blind and makes decisions without regard to the identities of the parties before Page 10 of 21

Critiques of Law and Economics the court. In this way, law may be made to look more like economics by reaffirming a point of viewlessness. Nonetheless, economics assumes a point of view. For example, in economics it is assumed that moving resources to the people most willing and able to pay is efficient and desirable, but there are other ways of allocating scare resources that em­ anate from different assumptions. Resources might be allocated on a first-in-time rule, on a lottery system, or in terms of a guiding principle, such as from each according to his ability and to each in accordance with her needs (Malloy, 2004). There are supporters and critics of each of the many ways that resources might be distributed and redistributed over time. The critique is not that we do not have reasons for favoring one system of mov­ ing resources relative to another, but that the choices reflect a point of view. When eco­ nomic analysis starts from a foundation of assumptions such (p. 310) as the desirability of moving resources to the highest bidder, it incorporates a value-based preference. Conse­ quently, when economic analysis is imported into law, law is given the same value-based preference. Reinforcing the interpretive function of point of viewlessness is the related assumption of methodological individualism. As a primary interpretive assumption in economics, methodological individualism means that the economist focuses on the decision-making process and the actions of individual actors. In essence, market activity is assumed to be driven by individual actors who make choices based on the pursuit of self-interest. Group activity is simply the aggregation of individual decision-making and individual action. This focus on the individual reinforces the assumption of a point of viewlessness in that it posi­ tions the unit of measure in terms of culturally unanchored individuals. In a culturally unanchored world, a world without a point of view, one might think that fairness and jus­ tice would be enhanced; but in such a world, there is no interpretive reference point for one to evaluate the meaning of fairness and justice, and thus, one may measure or calcu­ late differences among observations and at the same time be unable to ascribe meaning and value to these observations. Consequently, fairness and justice may be less certain in such a world precisely because fairness and justice only have meaning in the context of an interpretive community with a shared cultural-interpretive point of view.

15.3.2 Invariance The invariance principle centers on an idea originating with Adam Smith that people are led by an “invisible hand” such that their self-interested pursuits simultaneously advance the interest of the public, even though a regard for the public is no part of their original intention (Evensky, 2005; Evensky and Malloy, 1994; Malloy, 2004, 2010). In economic terms, Smith can be understood to be saying that private marginal costs and benefits equal public margin costs and benefits. There is, in other words, equivalence and invari­ ance between the pursuit of self-interest and the advancement of the public interest. In critiquing the idea of invariance in economics, it is understood that modern economics has “softened” Adam Smith’s original assumption in terms of what has been learned re­ specting such things as transaction costs, the wealth effect, the prisoner’s dilemma, the tragedy of the commons, and behavioral science. Nonetheless, the interpretive founda­ tion of economic analysis still incorporates a version of the assumption of invariance, and Page 11 of 21

Critiques of Law and Economics the invariance idea drives legal outcomes more likely to be unfavorable to government regulation and redistribution. As an implicit part of the assumption of invariance, economists often assume when mak­ ing policy recommendations that markets are self-correcting. This means that markets are assumed to be capable of automatically adjusting to continuously changing circum­ stances and doing so in a way that simultaneously maximizes efficiency and wealth for in­ dividuals and the public alike. All of this happens as if the market were being directed by an invisible hand, and the upshot of this is a legal narrative supporting a belief that gov­ ernment regulation is unnecessary; or, at the very least, that the need for (p. 311) regula­ tion is not the norm but an aberration and an exception to the rule. Consequently, those who might seek to promote environmental or land use regulations, for instance, must de­ vote time and effort to demonstrating conditions that give rise to the breakdown of the norm and that justify limited and targeted corrective action. The idea of invariance in economics favors the desirability of limited government, individ­ ual choice, and minimal redistribution. It also implies that efficient and desirable re­ source distributions occur naturally and without a need for intentional coordination. When applied to law, the economic idea of invariance is inconsistent with the interpretive view that markets are the product of volitional human action (Bromley, 2009), and that market mechanisms are approved or disapproved of with reference to the way that they privilege particular patterns and hierarchies of resource distribution. From an interpre­ tive perspective one might also agree that government should be limited, decision-making decentralized, and redistribution minimized, but this would be a volitional choice made for normative reasons and not a choice grounded in an uncritical application of an as­ sumption of invariance.

15.3.3 Calculus of Choice The calculus of choice addresses assumptions about the way that people make decisions (Buchanan and Tullock, 1962). In economic analysis it is generally assumed that people engage in a calculus of choice based on a rational and self-interested assessment of costs and benefits. While economists may suggest that multiple factors go into this calculus, they often end up using price as a proxy. Price functions as a useful proxy for economic analysis because it makes a mathematical calculus possible, and it permits a cost and benefit computation to determine an “optimal” course of action. The problem with using price as a proxy is that price is simply an incomplete interpretation of value and not value itself. Prices simply reflect value differences among competing options, and computations based on price inherently privilege options that are easily quantifiable. For example, the cost of buying and installing an air filter to reduce pollution is easier to quantify than the value of an additional unit of clean air. Requiring legal decisions to be based on a cost and benefit analysis, therefore, results in a bias toward the quantifiable. The larger problem with the calculus of choice is not that it favors the things that are readily quantifiable over those that are more abstract, but that it does this while implying that people make choices on the basis of a calculus rather than as a matter of interpretation. This in turn Page 12 of 21

Critiques of Law and Economics can lead legal economists to focus on attributing too much force to the power of law to adjust the cost and benefit calculations of innumerable individuals. Behavioral analysis reveals that people often have difficulty making cost and benefit calculations (Schwartz, 2005; Sunstein, 2000, 2009). In contrast to the economist, a person applying interpretation theory would suggest that choice is about interpretation and meaning, and this involves more than a calculus of costs and benefits; it implicates values derived from a shared cultural-interpretive com­ munity. Choice reflects differences based on culture, race, ethnicity, religion, (p. 312) gen­ der, age, education, income, and a variety of other factors. Concepts such as justice, fair­ ness, equity, reasonableness, and equality are not the subject of mathematical calculus; they are values formed from the human experience of living in a community with others. If such concepts as justice and fairness are simply translated into the economic equiva­ lent of efficiency and wealth maximization, they lose much of their social and cultural meaning. Therefore, treating legal decision-making and law reform as a calculus of choice undermines the humanistic values and norms of our legal tradition.

15.3.4 Fact–Value Distinction In an economic analysis of law, disputes and conflicts between parties are often framed as disagreements as to facts. When facts are in dispute the parties can undertake further in­ vestigation and they can recalculate their choices and reassess their optimal course of ac­ tion. The focus on factual disagreement lends itself to the objective and rational point of viewlessness that grounds the claim that economics is a science. In law, however, dis­ putes are frequently about something more than a disagreement as to facts; they involve disagreements as to values (Putnam, 2002). These value-based disagreements shape the facts as people understand them, and influence the relative importance attributed to any given fact by any particular party. Value disputes are not easily resolved by appeal to eco­ nomic analysis. At best, economics can only offer some indirect input on factors to consid­ er in a given situation, but in the end law must operate to make a judgment—a value choice between and among competing claims that are often based on emotion, culture, and other human characteristics that are not easily subject to an economic calculus. Con­ sequently, when economic analysis is applied to law, it often functions to redirect atten­ tion away from a conflict involving deeply held values and translates the disagreement in­ to one of competing facts. The problem with this move is that it may function to “mask” what the law is really doing and can undermine the traditional role of law in working to mediate tensions among competing and deeply held values in our system of democratic governance (Noonan, 1976). Understanding the way that economics transforms legal meanings and values from an in­ terpretive point of view is of substantive and strategic importance because it has implica­ tions for the distribution and allocation of resources in society. The way we talk about law, the language we use in writing about law, and the metaphors we invoke in making law, af­ fects outcomes; therefore, lawyers and judges should borrow from economics in an inten­

Page 13 of 21

Critiques of Law and Economics tional and informed way, and with an understanding of the normative implications of their actions.

Conclusion Scholars have questioned the normative value of economic efficiency as a central goal of law. They have also challenged the coherence and objectivity of the methods used (p. 313) to assess economic efficiency. Finally, they have questioned the claim of economics as somehow providing a scientific justification for law, arguing instead that it constitutes a rhetorical form shifting the terms of legal argument and changing its outcomes.

References Ackerman, Frank and Lisa Heinzerling (2004). Priceless: On Knowing the Price of Every­ thing and the Value of Nothing. New York: The New Press. Adler, Matthew and Eric A. Posner (2008). “Happiness Research and Cost–Benefit Analy­ sis.” Journal of Legal Studies 37: 253–292. Alexander, Gregory S., Eduardo M. Penalver, Joseph William Singer, and Laura S. Under­ kuffler (2009). “A Statement of Progressive Property.” Cornell Law Review 94: 743–745. Apel, Karl-Otto (1995). Charles S. Peirce: From Pragmatism to Pragmatism. Atlantic High­ lands, NJ: Humanities Press. Ashford, Robert (2004). “Socioeconomics and Professional Responsibilities in Teaching Law-Related Economic Issues.” San Diego Law Review 41: 133–176. Ashford, Robert (2009). “Using Socio-Economics and Binary Economics to Serve the In­ terests of the Poor and Working People: What Critical Scholars can do to Help.” Seattle Journal for Social Justice 8: 173–227. Baker, C. E. (1975). “The Ideology of the Economic Analysis of Law.” Philosophy and Pub­ lic Affairs 5: 3–48. Balkin, J. M. (1991). “The Promise of Legal Semiotics.” Texas Law Review 69: 1831–1852. Bebchuk, Lucian A. (1980). “The Pursuit of a Bigger Pie: Can Everyone Expect a Bigger Slice?” Hofstra Law Review 8: 671–709. Brigham, John (1999). “Millennium Reflections: Roberta Kevelson and the Law and Semi­ otic Round Table.” International Journal for the Semiotics of Law 12: 333–342. Brion, Denis J. (1991). “Rhetoric and the Law of Enterprise.” Syracuse Law Review 42: 117–161. Bromley, Daniel (2009). Sufficient Reason: Volitional Pragmatism and the Meaning of Eco­ nomic Institutions. Princeton, NJ: Princeton University Press. Page 14 of 21

Critiques of Law and Economics Bronsteen, John, Christopher Buccafusco, and Jonathan S. Masur (2010). “Welfare as Happiness.” Georgetown Law Journal 98: 1583–1641. Buchanan, James M. and Gordon Tullock (1962). The Calculus of Consent: Logical Foun­ dations of Constitutional Democracy. Ann Arbor, MI: Ann Arbor Paperbacks/Michigan. Calabresi, Guido (1991). “The Pointlessness of Pareto: Carrying Coase Further.” Yale Law Journal 100: 1211–1238. Carrier, Michael A. (2009). Innovation for the 21st Century: Harnessing the Power of In­ tellectual Property and Antitrust Law. New York: Oxford University Press. Chon, Margaret (2006). “Intellectual Property and the Development Divide.” Cardozo Law Review 27: 2821–2912. Coleman, Jules (1980). “Efficiency, Utility and Wealth Maximization.” Hofstra Law Review 8: 509–551. Coleman, Jules (1982). “The Normative Basis of Economic Analysis: A Critical Review of Richard Posner’s The Economics of Justice (Book Review).” Stanford Law Review 34: 1105–1132. Coleman, Jules (1988). Markets, Morals, and the Law. New York: Cambridge University Press. Cooter, Robert D. (2013). The Falcon’s Gyre: Legal Foundations of Economic Innovation and Growth. Berkeley, CA: Berkeley Law Books. Cooter, Robert and Thomas Ulen (2007). Law and Economics. 5th edition. Boston, MA: Addison-Wesley. (p. 314)

de Soto, Jesus H. (2009). The Theory of Dynamic Efficiency. New York: Routledge. Diener, Ed and Robert Biswas-Diener (2002). “Will Money Increase Subjective Well-Be­ ing?: A Literature Review and Guide to Needed Research.” Social Science Indicators Re­ search Series 37: 119–154. Driesen, David M. (2003). The Economic Dynamics of Environmental Law. Cambridge, MA: MIT Press. Driesen, David M. (2011a). “Contract Law’s Inefficiency.” Virginia Law and Business Re­ view 6: 301–340. Driesen, David M. (2011b). “Two Cheers for Feasible Regulation: A Modest Response to Masur and Posner.” Harvard Environmental Law Review 35: 313–341. Driesen, David M. (2012). The Economic Dynamics of Law. New York: Cambridge Univer­ sity Press.

Page 15 of 21

Critiques of Law and Economics Driesen, David M. (2014). “Legal Theory Lessons from the Financial Crisis.” Journal of Corporation Law 40: 55–97. Dworkin, Ronald (1980a). “Is Wealth a Value?” Journal of Legal Studies 9: 191–226. Dworkin, Ronald (1980b). “Why Efficiency?: A Response to Calabresi and Posner.” Hofstra Law Review 8: 563–589. Easterlin, Richard A. (1995). “Will Raising the Incomes of All Increase the Happiness of All?” Journal of Economic Behavior and Organization 27: 35–47. Evensky, Jerry (2005). Adam Smith’s Moral Philosophy: A Historical and Contemporary Perspective on Markets, Law, Ethics, and Culture. New York: Cambridge University Press. Evensky, Jerry and Robin Paul Malloy, eds. (1994). Adam Smith and the Philosophy of Law and Economics. Dordrecht: Kluwer. Ferber, Marianne A. and Julie A. Nelson, eds. (1993). Beyond Economic Man: Feminist Theory and Economics. Chicago: University of Chicago Press. Frank, Robert H. (1999). Luxury Fever: Why Money Fails to Satisfy in an Era of Excess. New York: The Free Press. Gagnier, Regenia (2000). The Insatiability of Human Wants: Economics and Aesthetics in Market Society. Chicago: University of Chicago Press. Garner, Bryan A. and Antonin Scalia (2012). Reading Law: The Interpretation of Legal Texts. St. Paul, MN: West. Goodrich, Peter (1986). Reading the Law: A Critical Introduction to Legal Method and Techniques. Oxford: Blackwell. Goodrich, Peter (1987). Legal Discourse: Studies in Linguistics, Rhetoric, and Legal Analysis. Basingstoke: Palgrave Macmillan. Hackney, James R. (2006). Under Cover of Science: American Legal-Economic Theory and the Quest for Objectivity. Durham, NC: Duke University Press. Hausman, Carl R. (1997). Charles S. Peirce’s Evolutionary Philosophy. Cambridge: Cam­ bridge University Press. Hom, Sharon and Robin Paul Malloy (1994). “China’s Market Economy: A Cross Boundary Discourse Between Law and Economics And Feminist Jurisprudence.” Syracuse Law Re­ view 45: 815–851. Hookway, Christopher (1992). Peirce. London: Routledge. Houser, Nathan and Christian Kloesel, eds. (1992). The Essential Peirce, Vol. I. Indianapo­ lis, IN: Indiana University Press. Page 16 of 21

Critiques of Law and Economics (p. 315)

Jackson, Bernard (1985). Semiotics and Legal Theory. London: Routledge.

Kennedy, Duncan (1991). “A Semiotics of Legal Argument.” Syracuse Law Review 42: 75– 116. Ketner, Kenneth Laiane, ed. (1992). Reasoning and the Logic of Things: Charles Sanders Peirce. Cambridge, MA: Harvard University Press. Kevelson, Roberta (1986). Charles S. Peirce’s Method of Methods. Philadelphia, PA: John Benjamins Publishing Co. Kevelson, Roberta (1988). The Law as a System of Signs. New York: Plenum. Kevelson, Roberta (1991). Peirce and Law: Issues in Pragmatism, Legal Realism, and Semiotics. New York: Peter Lang. Kevelson, Roberta (1993). Peirce’s Esthetics of Freedom: Possibility, Complexity, and Emergent Value. New York: Peter Lang. Klapow, Louis and Steven M. Shavell (2001). “Fairness Versus Welfare.” Harvard Law Re­ view 114: 961–1388. Kronman, Tony (1980). “Wealth Maximization as a Normative Principle.” Journal of Legal Studies 9: 227–242. Lipsey, R. G. and Kelvin Lancaster (1956). “The General Theory of Second Best.” Review of Economic Studies 24: 11–32. Liska, James Jakob (1996). A General Introduction to the Semeiotic of Charles Sanders Peirce. Indianapolis, IN: Indiana University Press. Lutz, Mark A. and Kenneth Lux (1988). Humanistic Economics: The New Challenge. New York: Bootstrap Press. McCloskey, Deirdre (Donald) N. (1985). The Rhetoric of Economics. Madison, WI: Univer­ sity of Wisconsin Press. McCloskey, Deirdre (Donald) N. (1990). If You’re So Smart: The Narrative of Economic Expertise. Chicago: University of Chicago Press. McCloskey, Deirdre (Donald) N. (1994). Knowledge and Persuasion in Economics. Cambridge: Cambridge University Press. McGarity, Thomas O. (1998). “A Cost–Benefit State.” University of Chicago Law Review 50: 7–79. Mackinnon, Catherine A. (1989). Towards a Feminist Theory of the State. Cambridge, MA: Harvard University Press.

Page 17 of 21

Critiques of Law and Economics Malloy, Robin Paul (1986). “Equating Human Rights and Property Rights—The Need for Moral Judgment in Economic Analysis of Law and Social Policy.” Ohio State Law Journal 47: 163–177. Malloy, Robin Paul (1987). “The Political Economy of Co-Financing America’s Urban Re­ naissance.” Vanderbilt Law Review 40: 67–134. Malloy, Robin Paul (1989). “Of Icons, Metaphors, and Private Property: The Recognition of ‘Welfare’ Claims in the Philosophy of Adam Smith,” in Roberta Kevelson, Law and Semiotics vol. 3, 241–254. New York: Plenum. Malloy, Robin Paul (1990a). “A Sign of the Times—Law and Semiotics (Book Review of Kevelson, Law As A System of Signs).” Tulane Law Review 65: 211–219. Malloy, Robin Paul (1990b). Law and Economics: A Comparative Approach to Theory and Practice. St. Paul, MN: West. Malloy, Robin Paul (1991a). “Toward a New Discourse of Law and Economics.” Syracuse Law Review 42: 27–73. Malloy, Robin Paul (1991b). Planning for Serfdom, Legal Economic Discourse and Down­ town Development. Philadelphia, PA: University of Pennsylvania Press. Malloy, Robin Paul (1992). “Letters from the Longhouse: Law, Economics And Na­ tive American Values.” Wisconsin Law Review 1992: 1569–1642. (p. 316)

Malloy, Robin Paul (1999). “Law and Market Economy: The Triadic Linking of Law, Eco­ nomics, and Semiotics.” International Journal for the Semiotics of Law 12: 285–307. Malloy, Robin Paul (2000). Law and Market Economy: Reinterpreting the Values of Law and Economics. Cambridge: Cambridge University Press. Malloy, Robin Paul (2003). “Framing the Market: Representations of Meaning and Value in Law, Markets, and Culture.” Buffalo Law Review 51: 1–126. Malloy, Robin Paul (2004). Law in a Market Context; An Introduction to Market Concepts in Legal Reasoning. Cambridge: Cambridge University Press. Malloy, Robin Paul (2009). “Place, Space, and Time in the Sign of Property.” International Journal for the Semiotics of Law 22: 265–277. Malloy, Robin Paul (2010). “Adam Smith in the Courts of the United States.” Loyola Law Review 56: 33–94. Menand, Louis (2001). The Metaphysical Club: A Story of Ideas in America. New York: Farrar, Straus and Giroux. Mercuro, Nicholas and Steven G. Medema (2006). 2nd edition. Economics and the Law: From Posner to Post-Modernism. Princeton, NJ: Princeton University Press. Page 18 of 21

Critiques of Law and Economics Mercuro, Nicholas and Timothy P. Ryan (1984). Law, Economics and Public Policy. Green­ wich, CT: JAI Press. Merrell, Floyd (1997). Peirce, Signs, and Meaning. Toronto: University of Toronto Press. Noonan, Jr., John T. (1976). Persons and Masks of the Law. Berkeley: University of Califor­ nia Press. Nussbaum, Martha C. (2000). Women and Human Development: The Capabilities Ap­ proach. Cambridge: Cambridge University Press. Nussbaum, Martha C. and Amartya Sen, eds. (1993). The Quality of Life. Oxford: Claren­ don Press. Paul, Jeremy (1991). “The Politics of Legal Semiotics.” Texas Law Review 69: 1779–1830. Peirce edition project, ed. (1998). The Essential Peirce, Vol. II. Indianapolis, IN: Indiana University Press. Pitofsky, Robert, ed. (2008). How the Chicago School Overshot the Mark: The Effect of Conservative Analysis on U.S. Antitrust. New York: Oxford University Press. Posner, Richard A. (1979). “Utilitarianism, Economics, and Legal Theory.” Journal of Legal Studies 8: 103–140. Posner, Richard A. (1980). “The Ethical and Political Basis of the Efficiency Norm in Com­ mon Law Adjudication.” Hofstra Law Review 8: 487–507. Posner, Richard A. (1981a). The Economics of Justice. Cambridge, MA: Harvard Universi­ ty Press. Posner, Richard A. (1981b). “A Reply to Some Recent Criticisms of the Economic Theory of the Common Law.” Hofstra Law Review 9: 775–794. Posner, Richard A. (1990). The Problems of Jurisprudence. Cambridge, MA: Harvard Uni­ versity Press. Posner, Richard A. (1999). The Problematics of Moral and Legal Theory. Cambridge, MA: Belknap Press of Harvard University Press. Posner, Richard A. (2014). Economic Analysis of Law. 9th edition. Austin, TX: Wolters Kluwer. Potter, S. J. and G. Vincent (1997). Charles S. Peirce On Norms & Ideals. New York: Ford­ ham University Press. (p. 317)

Purdy, Jedediah (2005). “A Freedom-Promoting Approach to Property: A Renewed

Tradition for New Debates.” University of Chicago Law Review 72: 1237–1298.

Page 19 of 21

Critiques of Law and Economics Putnam, Hilary (2002). The Collapse of the Fact/Value Dichotomy. Cambridge, MA: Har­ vard University Press. Rizzo, Mario J. (1980). “The Mirage of Efficiency.” Hofstra Law Review 8: 641–658. Roesler, Shannon M. (2011). “Addressing Environmental Injustices: A Capability Ap­ proach to Rulemaking.” West Virginia Law Review 114: 49–107. Rubenfeld, Jed (2001). Freedom and Time: A Theory of Constitutional Self-Government. New Haven, CT: Yale University Press. Sagoff, Mark (1988). The Economy of the Earth: Philosophy, Law, and the Environment. New York: Cambridge University Press. Sagoff, Mark (2004). Price Principle, and the Environment. Cambridge: Cambridge Uni­ versity Press. Schwartz, Barry (2005). The Paradox of Choice: Why More is Less. New York: Harper Perennial. Scitovszky, Tibor (1941). “A Note on Welfare Propositions in Economics.” Review of Eco­ nomic Studies 9: 77–88. Sen, Amartya (1999). Development As Freedom. New York: Alfred A. Knopf. Sen, Amartya (2009). The Idea of Justice. Cambridge, MA: Belknap Press of Harvard Uni­ versity Press. Sunstein, Cass R. (2000). Behavioral Law and Economics. Cambridge: Cambridge Univer­ sity Press. Sunstein, Cass and Richard Thaler (2009). Nudge: Improving Decisions About Health, Wealth, and Happiness. New York: Penguin. Thorsby, David (2001). Economics and Culture. Cambridge: Cambridge University Press. Weinrib, Ernest J. (1980). “Utilitarianism, Economics, and Legal Theory.” University of Toronto Law Journal 30: 307–332. Williams, Cynthia A. (2002). “Corporate Social Responsibility in an Era of Economic Glob­ alization.” U.C. Davis Law Review, 35: 705–778.

(p. 318)

David Driesen

David Driesen is a University Professor at Syracuse University. Robin Paul Malloy

Page 20 of 21

Critiques of Law and Economics Robin Paul Malloy, E.I. White Chair, and Distinguished Professor of Law, Syracuse University College of Law.

Page 21 of 21

Optimal Redistributional Instruments in Law and Economics

Optimal Redistributional Instruments in Law and Eco­ nomics   Chris William Sanchirico The Oxford Handbook of Law and Economics: Volume 1: Methodology and Concepts Edited by Francesco Parisi Print Publication Date: Apr 2017 Subject: Economics and Finance, Law and Economics, Econometrics, Experimental and Quanti­ tative Methods Online Publication Date: May 2017 DOI: 10.1093/oxfordhb/9780199684267.013.026

Abstract and Keywords This article discusses three strands of the literature on optimal redistributional instru­ ments. The first strand concerns what is sometimes referred to as the ‘tax substitution ar­ gument’, which supports the proposition that distributional goals should generally be pur­ sued exclusively through taxes (and subsidies) on labour earnings. The argument rests the controversial assumption that, controlling for labour earnings, all individuals are identical. The second strand, a response to the first, attempts to counter the view that labour earnings exclusivity for redistributional policy is the natural lesson of the econom­ ic literature on ‘optimal taxation’. This second subliterature examines the consequences of removing the tax substitution assumption while retaining, arguendo, the optimal tax and policy framework. The third strand of the literature, concerning ‘policy uncertainty’, steps outside the bounds of the conventional model of optimal tax and policy. It first aris­ es as a kind of alternative defence of labour earnings exclusivity in light of the challenge posed by policy eclecticism. Even if the conventional optimal tax and policy model sans tax substitution assumption points toward eclecticism, the defence proceeds, policymak­ ers lack adequate information about the proper direction and scale of distributionally mo­ tivated adjustments to other policy instruments. Given the risk of perverse unintended consequences, policymakers should refrain from distributionally motivated tinkering. Keywords: optimal redistributional instruments, tax substitution, labour earnings, redistributional policy, taxation, policy uncertainty

16.1 Introduction MOST systematic normative approaches to social welfare take into account not just the total amount of resources available to society, but also, to varying extents, how that total is divided among society’s members. The literature on optimal redistributional instru­ Page 1 of 37

Optimal Redistributional Instruments in Law and Economics ments thus begins with the assumption that society has some preference for equality, leaving the precise degree unspecified. It then asks: How should society pursue that pref­ erence? More specifically, what kinds of policy instruments—whether categorized as “tax­ es,” “transfers,” “public goods,” “government programs,” “regulations,” or “legal rules”— should be informed by society’s distributional objectives? This chapter focuses on three strands of the literature on optimal redistributional instru­ ments, reflecting the bulk of contributions to date. The first strand of the literature, dis­ cussed in section 16.2, concerns what is sometimes referred to as the “tax substitution ar­ gument.” The tax substitution argument is offered in support of the proposition that dis­ tributional goals should generally be pursued exclusively through taxes (and subsidies) on labor earnings. The argument rests on a controversial assumption: very roughly, that con­ trolling for labor earnings, all individuals are identical. As discussed in detail in section 16.2, the full, formal tax substitution assumption accomplishes two tasks. It ensures that any information about individuals provided to the government by attributes other than la­ bor earnings is provided as well by labor earnings. And it ensures that labor earnings tax­ ation is a relatively efficient means of utilizing such information. Section 16.3 discusses the second strand of the literature on optimal redistribu­ tional instruments. This second portion of the literature, a response to the first, attempts to counter the view that labor earnings exclusivity for redistributional policy is the natur­ al lesson of the economic literature on “optimal taxation”—that is, the large literature growing out of Mirrlees (1971) concerning optimal tax and policy with an information-lim­ ited, potentially distribution-minded government. This second subliterature examines the consequences of removing the tax substitution assumption while retaining, arguendo, the optimal tax and policy framework.1 Contributors to this literature argue that, without the tax substitution assumption, the conventional optimal tax and policy model—so far as it goes—actually prescribes a strong form of “policy eclecticism” with regard to redistribu­ tional instruments. (p. 322)

Section 16.4 of the chapter discusses a third strand of the literature, concerning “policy uncertainty.” This subliterature steps outside the bounds of the conventional model of op­ timal tax and policy. It centers on a kind of alternative defense of labor earnings exclusivi­ ty in light of the challenge posed by policy eclecticism. Even if the conventional optimal tax and policy model sans tax substitution assumption points toward eclecticism, the de­ fense proceeds, policymakers lack adequate information about the proper direction and scale of distributionally motivated adjustments to other policy instruments. Given the risk of perverse unintended consequences, policymakers should refrain from distributionally motivated tinkering. Critics of this policy uncertainty defense argue that it neglects the comparable informational difficulties of labor earnings taxation. Moreover, say critics, the connection that this defense implicitly draws between relative uncertainty and relative in­ action—a kind of comparative “precautionary principle”—finds little support in systematic thinking about uncertainty and choice.

Page 2 of 37

Optimal Redistributional Instruments in Law and Economics Three final notes regarding the chapter’s general approach: First, the foregoing three topic areas are merely the major components of the literature on optimal redistributional instruments. The literature, like the topic itself, is expansive, and many other important issues have been raised, some systematically studied. Such issues include relative admin­ istrative competence, administrative costs (Yitzhaki, 1979; Slemrod, 1990), responsive bargaining and market price adjustments (Polinsky, 1989), uncertainty faced by individu­ als surrounding the consequences of their private choices (contrast: policy uncertainty), the role of insurance markets (Logue and Avraham, 2003), and the implications of cogni­ tive biases (Jolls, 1998). In light of space constraints, and given the intricacy of the literature’s central results, the strategy for compiling this chapter has been to focus on providing a relatively comprehensive explanation of a short list of key points, rather than a more comprehensive, but necessarily more discursive review. The hope is that this chapter will provide the reader with a template against which extensions and (p. 323) un­ covered issues can be efficiently evaluated. Moreover, the reader will find that many is­ sues not explicitly addressed in this chapter are treated in the articles it cites. Second, the literature on optimal redistributional instruments resides both in law and economics and in public finance economics. Early results were developed in the literature on optimal taxation within public finance and then imported into law and economics. Many later points were developed within law and economics. While legal applications are the ultimate target of this chapter, the chapter’s structure reflects the view that it is not possible to understand legal applications without placing them in the broader context of optimal tax and policy. Lastly, while this chapter is an attempt to provide a balanced assessment of the literature, it necessarily reflects the author’s view of where the balancing point is located. Fortu­ nately, encyclopedic entries and surveys from alternative perspectives are in abundant supply.

16.2 The Tax Substitution Argument and its Key Assumption The literature on optimal redistributional instruments—both in law and economics and in public finance—is dominated by the tax substitution argument.2 The tax substitution argu­ ment shows that any redistribution effected outside the labor earnings tax system can be better accomplished by making appropriate adjustments to labor earnings taxation.3 The implication is that the government should pursue distributional goals exclusively through the labor earnings tax system, while setting other policies without regard to their distrib­ utional impact. Both proponents and detractors of the tax substitution argument recognize that, like any argument, it requires assumptions. Controversy surrounds the nature of these assump­ tions. Proponents of the tax substitution argument refer to its key assumption as a “quali­ fication” (Kaplow and Shavell, 2000, p. 822). Adopting the assumption is said to generate Page 3 of 37

Optimal Redistributional Instruments in Law and Economics the “natural model” (Kaplow and Shavell, 2000, p. 821). Conversely, (p. 324) prescriptions of the model sans assumption are viewed as “exotic” (Bankman and Weisbach, 2007, p. 793), or as “theoretical curiosities” (Kaplow and Shavell, 2000, p. 822). Critics of the tax substitution argument regard its key assumption as result-driving, empirically unground­ ed, and indeed facially implausible once clearly revealed (Sanchirico, 1997; 2000, p. 813; 2001, p. 1058; 2010a, pp. 874–875, 940; 2011a). Perhaps the best way to develop an informed opinion of the tax substitution assumption is to understand the role it plays in the tax substitution argument. The primary object of this section is to facilitate such an understanding. Thus, the first part of this section lays out the tax substitution argument—verbally, but systematically—highlighting the work done by its key assumption. Collaterally, this first part reviews the chief policy questions to which the tax substitution argument has been applied, drawing connections where appropriate, and highlighting how law and economic applications fit into the larger literature. Building on this explanation of tax substitution logic, the second part of this section makes some general observations about the tax substitution argument and its key as­ sumption, identifying some potential sources of confusion regarding both.

16.2.1 Explanation of the Tax Substitution Argument Highlighting the Role of the Tax Substitution Assumption 16.2.1.1 An Inefficient, Distributionally Motivated Adjustment to “Policy X” Suppose that the government is considering modifying some “policy X”—some tax, law, regulation, or rule whose individual-by-individual application is not based solely on indi­ viduals’ labor earnings—in a way that redistributes among individuals in a manner that the government regards as favorable. Imagine, however, that this distributionally motivated adjustment to policy X is “ineffi­ cient.” That is, while some individuals “lose” and some “win” under the policy X modifica­ tion, losers lose more in aggregate than winners win in aggregate. More precisely, imagine the hypothetical configuration of individually tailored lump sum transfers to or from the government that precisely mimics the impact of the policy X mod­ ification on every individual’s utility (hereinafter the “utility-equivalent lump sum scheme”). Assume that this scheme takes in, in total, from those who are burdened by the policy X modification more than it sends out, in total, to those who are benefited by the policy X modification.4 The question then arises, why bother with the policy X adjustment? Why not im­ plement the utility-equivalent lump sum scheme instead, thus arriving at the same config­ uration of individual utilities with revenue to spare—revenue that can then be used for general benefit? (p. 325)

Page 4 of 37

Optimal Redistributional Instruments in Law and Economics Two obstacles prevent this lump sum trump. But before considering these difficulties, let us be more concrete about the kinds of policy modifications that have played the role of policy X in the literature and how inefficiency manifests in each case.

16.2.1.2 “Policy X” in the Literature 16.2.1.2.1 Differential Commodity Taxation Imagine that the government is considering taxing purchases of certain luxury items and uniformly distributing the revenue thus raised. A distribution-minded government might regard this (two-part) policy modification as beneficial from the standpoint of equity. Relatively large purchasers of luxury goods will pay more in tax than they receive in the uniform transfer-back of aggregate revenue, and vice versa. Yet the policy would be inefficient in the sense described above. That is, a lump sum scheme that was utility-equivalent to the luxury tax cum uniform transfer would actually raise revenue. The reasoning behind this point is referenced at several points in this chapter and so is worth spelling out. Begin by additively decomposing such a utility-equivalent lump sum scheme into a utilityequivalent lump sum scheme for the luxury tax taken alone and a utility-equivalent lump sum scheme for the uniform transfer-back of luxury tax revenue taken alone. The latter, the utility-equivalent lump sum scheme for the uniform transfer-back of luxury tax rev­ enue, is simply the uniform transfer-back of luxury tax revenue. Thus, this second compo­ nent expends revenue equal to that collected by the luxury tax. The result targeted in the preceding paragraph then follows from the fact, not entirely obvious, that the former, the utility-equivalent lump sum scheme for the luxury tax taken alone, raises more revenue than the luxury tax. The way to see that the utility-equivalent lump sum scheme for the luxury tax taken alone raises more revenue than the luxury tax is to establish that a lump sum scheme that raised the same revenue—from each individual—as the luxury tax taken alone would pro­ vide each individual with greater utility than under the luxury tax taken alone. Charging each individual in lump sum merely what she would have paid in luxury tax would leave that individual able to afford the consumption bundle she would have chosen under the luxury tax. Yet, relative prices under the luxury tax differ from relative prices under such a revenue-equivalent lump sum charge. Hence, the consumption bundle that the individ­ ual would have chosen under the luxury tax would not maximize her utility under the rev­ enue-equivalent lump sum charge. In other words, the individual could attain greater util­ ity under the revenue-equivalent lump sum charge than under the luxury tax. It follows that a lump sum tax that is utility-equivalent to the luxury tax (taken alone) (p. 326) must collect more from each individual than would be collected from her by the luxury tax.5 Any differential taxation of consumption spending, broadly defined, would be similarly in­ efficient.6 This includes subsidies (or tax rate reductions) for milk or other staples. It also Page 5 of 37

Optimal Redistributional Instruments in Law and Economics includes taxing capital income, which in effect differentially taxes current and future con­ sumption.7 Differential commodity taxation is the earliest application of the tax substitu­ tion argument (Atkinson and Stiglitz, 1976; Deaton, 1979,8 1981; Deaton and Stern, 1986).

16.2.1.2.2 Externality Correction Inclusive of “Legal Rules” Alternatively, the government might be considering a distributionally motivated modifica­ tion to a policy that is designed to internalize external costs or benefits. Consider the fol­ lowing examples. 16.2.1.2.2.1 Prohibitory Regulations The policy X modification might consist of lifting regulations prohibiting a type of conduct that imposes costs on other individuals in excess of the benefits to the actor herself. The regulation might be viewed as regressive, and its removal thus pro-distributional. Yet lift­ ing the regulation would be inefficient. A lump sum tax and transfer scheme that was util­ ity-equivalent to lifting the regulation would (1) pay each individual a lump sum transfer equal to the benefit she would accrue were she herself allowed to take the prohibited ac­ tion; (2) charge each individual a lump sum tax equal to the burden she would incur were everyone else allowed to take the prohibited action.9 By hypothesis the payout attribut­ able to each individual’s prohibited action in (2) would exceed the pay-in by such individ­ ual in (1). Notice that in an externality setting, any given individual’s utility-equivalent lump sum amount has two determinants: the importance to her of her own action level for the externality; and how she is positioned with regard to the externality action levels of oth­ ers. (p. 327)

16.2.1.2.2.2 Pigovian Taxation The policy X modification might consist of a distributionally motivated increase in what begins as a precisely calibrated externality-internalizing tax or subsidy (a “Pigovian tax”) (Sandmo, 1975). Imagine that this Pigovian tax is coupled with a uniform lump sum trans­ fer or tax of all the net revenue it collects. Increasing the Pigovian tax (cum uniform transfer) beyond its efficient level might be re­ garded as distributionally favorable for reasons similar to those discussed with regard to a luxury tax. However, such a policy adjustment would be inefficient. The precise explana­ tion for this inefficiency in terms of utility-equivalent lump sum amounts is too involved to be included in this chapter. However, it may be roughly regarded as a combination of the logic for differential commodity taxation and regulatory prohibition, the two preceding cases. 16.2.1.2.2.3 Private Law Rules

Page 6 of 37

Optimal Redistributional Instruments in Law and Economics The policy X modification might be an increase in the damages rule in tort law above the amount that corrects for the external cost accidents (Shavell, 1981; Kaplow and Shavell, 1994). In this case the analysis would be the same as for the increased Pigovian tax, ex­ cept that the increased “revenue” collected from increasing damages would be specifical­ ly paid out to accident victims, rather than used for a uniform demogrant.

16.2.1.2.3 Public Goods and Government Programs The policy X modification might concern the provision of public goods (Hylland and Zeck­ hauser, 1979; Kaplow, 1996). Thus, the government might be considering imposing a uni­ form lump sum tax to “oversupply” parks and community centers in poorer neighbor­ hoods. To oversupply the public good would mean to supply it beyond the point where dollar-denominated benefits exceed dollar-denominated costs. Thus, the portion of the utility-equivalent lump sum scheme consisting of the payout attributable to the benefit to public good users would fall short of the portion consisting of the pay-in attributable to the financing tax.

16.2.1.3 Two Problems with Swapping in the Utility-Equivalent Lump Sum Scheme As noted, there are two problems with simply swapping the revenue-superior, utilityequivalent lump sum scheme for the distributionally motivated modification to policy X. The first concerns the sufficiency of government information; the second concerns the collateral impact of such a swap on labor earnings choices and the pre-existing labor earnings tax system. (p. 328)

16.2.1.3.1 The Informational Problem

In the conventional optimal tax and policy model, in which the current discussion is im­ plicitly framed, the government is assumed to know quite a bit. It knows the population distribution of all relevant parameters and functions, and so can determine the popula­ tion distribution of individual choices under any configuration of policies. It observes la­ bor earnings on an individual-by-individual basis; this is why it is able to tax labor earn­ ings. It also observes on an individual-by-individual basis whatever attributes determine the individual-specific application of policy X. However, in the conventional optimal tax and policy model the government is not able to observe on an individual-by-individual basis the utility-equivalent lump sum amounts for various policy modifications. It cannot, for example, observe lump sum equivalents for the pre-existing labor earnings tax—else there would be no reason to have such a tax in the first place. And similarly, it cannot observe—at least not directly—a given individual’s util­ ity-equivalent lump sum amount for the policy X modification.

16.2.1.3.2 The “Theory of the Second Best” Problem Even if the government could observe each individual’s utility-equivalent lump sum amount for the policy X modification, swapping the utility-equivalent lump scheme for the policy X modification might not in fact deliver surplus revenue if, after the swap, there re­

Page 7 of 37

Optimal Redistributional Instruments in Law and Economics main in place policies whose revenue implications are sensitive to individuals’ behavioral responses.10 Suppose, for example, that there is a pre-existing labor earnings tax system. Swapping the utility-equivalent lump scheme for the policy X modification might well cause individ­ uals to alter their labor earnings. A potential consequence is that the labor earnings tax will raise less revenue. This revenue loss might overwhelm the local revenue surplus cre­ ated by swapping in the utility-equivalent lump sum scheme. The possibility of background labor earnings taxation is particularly important for the tax substitution argument. The argument’s conclusion is that distributional objectives should be pursued solely through the labor earnings tax system. Were the argument only able to establish the inferiority of distributionally motivated policy X modifications under the con­ dition that no labor earnings tax system were in place, it could not rule out the optimality of a system that used both labor earnings taxation and policy X to pursue distributive goals.

16.2.1.4 How the Tax Substitution Assumption Solves Both Problems The role of the tax substitution assumption is to solve both of the foregoing problems. The next two subsections build up to the full tax substitution assumption in two steps corre­ sponding to the two problems, respectively. (p. 329)

16.2.1.4.1 Solving the Informational Problem

To isolate the information problem, imagine for the moment that were the utility-equiva­ lent lump sum scheme for the policy X modification swapped for the policy X modifica­ tion, no individual would alter her labor earnings.11 Under this temporary hypothesis, swapping the utility-equivalent lump sum scheme for the policy X modification would have no impact on the pre-existing labor earnings tax system, and thus would indeed gen­ erate surplus revenue overall. This leaves only the first problem: the government’s lack of sufficient individual-by-individual information to implement the utility-equivalent lump sum scheme. The first part of the tax substitution assumption implies that the utility-equivalent lump sum scheme can indeed be implemented—through adjusting the labor earnings tax. Al­ though the government cannot see individuals’ utility-equivalent lump sum amounts di­ rectly, it can see individuals’ labor earnings levels. The tax substitution assumption im­ plies that individuals’ labor earnings levels fully reveal their lump sum utility equivalents for the policy X modification. This implication follows from the part of the tax substitution assumption requiring that all individuals with the same level of labor earnings have the same utility-equivalent lump sum amounts for the policy X modification. In the conventional optimal tax and policy framework, the government already knows precisely the correspondence in the popula­ tion between labor earnings and these lump sum equivalents. It also knows labor earn­ ings on an individual-by-individual basis. If it is further assumed that the known corre­ Page 8 of 37

Optimal Redistributional Instruments in Law and Economics spondence between labor earnings and lump sum equivalents is such that each level of la­ bor earnings is associated with one and only one lump sum equivalent, then (p. 330) the government is able to deduce each individual’s lump sum equivalent from her labor earn­ ings. This first portion of the tax substitution assumption is sometimes referred to as the ab­ sence of “within income group differences.”

16.2.1.4.2 Solving the “Theory of the Second Best” Problem To additionally solve the second best problem noted in section 16.2.1.3.2, the tax substitu­ tion assumption must be further strengthened. The second problem arises because swap­ ping the utility-equivalent lump sum scheme for the policy X modification potentially af­ fects labor earnings choices and so labor earnings tax revenues. The strengthened, full tax substitution assumption guarantees not only that labor earnings provides sufficient in­ formation to identify utility-equivalent lump sum amounts on an individual-by-individual basis, but also that it is possible to structure a labor earnings surtax that delivers such amounts without causing individuals to alter their labor earnings. How such labor earnings responses are prevented is the most ingenious portion of tax substitution reasoning. The idea is to expand the notion of utility-equivalent lump sum amounts to labor earnings-contingent utility-equivalent lump sum amounts. 16.2.1.4.2.1 The Full Tax Substitution Assumption So far, in the process of moving toward the full tax substitution assumption, we have as­ sumed that if two individuals choose the same labor earnings (post policy X modification), they have the same utility-equivalent lump sum amount for the policy X modification. As­ sume now that any two individuals, were they hypothetically forced to choose any given level of labor earnings, whether or not either would actually choose this level under the policy X modification, would, under this forced choice, have the same utility-equivalent lump sum amount for the policy X modification. Thus, take any two individuals. Imagine that they would choose L’ and L” labor earnings respectively under the policy X modification. These labor earnings levels may not be the same. Then take any level of labor earnings L, which may equal neither L’ nor L”. Hypo­ thetically force both individuals to generate this third level of labor earnings L. Then de­ termine each individual’s L-conditional utility-equivalent lump sum amount for the policy X modification. The tax substitution assumption requires that these L-conditional lump sum equivalents be the same for the two individuals—that is, for any two individuals. Note that this stronger form of the tax substitution assumption, which is the full tax sub­ stitution assumption, does indeed subsume the weaker form described in section 16.2.1.4.1. The weaker assumption also in effect requires equal L-conditional utilityequivalent lump sum amounts for policy X modifications, but only in the special case that L is what both individuals would actually choose under the policy X modification.

Page 9 of 37

Optimal Redistributional Instruments in Law and Economics 16.2.1.4.2.2 Preventing a Behavioral Response in Labor Earnings When the full tax substitution assumption holds, a more intricate labor earnings tax sub­ stitution becomes possible—one that generates the same labor earnings choices as the policy X modification. Consider the following labor earnings “surtax,” to be charged in addition to any pre-existing labor earnings tax, and in lieu of the policy X modification: At each level of la­ bor earnings L, charge in surtax the assumed-to-be-uniform-across-individuals L-condi­ tional utility-equivalent lump sum amount for the policy X modification. (p. 331)

Now imagine any given individual is deciding whether to choose any level L of labor earn­ ings either under the policy X modification or under the substitute labor earnings surtax just defined. Under the policy X modification, given this choice of L, the individual is able to attain a particular level of maximal utility. With the labor earnings surtax in place in­ stead of the policy X modification, this choice of L enables the individual to attain precise­ ly the same level of maximal utility: the labor earnings surtax was specifically designed to make this so. It puts this L-choosing individual in precisely the same place as she would be were the policy X modification adopted. The same equivalence holds for all possible levels of labor earnings for this individual. Therefore, the labor earnings choice that leads to the greatest utility for this individual is the same under the labor earnings surtax as under the policy X modification. Thus, this individual chooses the same level of labor earnings under the labor earnings surtax as un­ der the policy X modification. Because the L-conditional utility-equivalent lump sum amount for the policy X modifica­ tion is the same for all individuals, the substitution labor earnings surtax has the same impact—which is to say, no impact—on all individuals’ labor earnings choices. This is what the tax substitution assumption is and what it does. I now turn to some gen­ eral comments based on this analysis.

16.2.2 Potential Misperceptions Regarding the Tax Substitution Argu­ ment and Its Key Assumption The tax substitution argument and its assumption have been the subject of some appar­ ent confusion—within both the legal and economics literatures. This section attempts to provide a series of clarifying notes in light of the preceding section’s description of how the argument and the assumption actually fit together. In general, some degree of confusion appears to have resulted from attempts to render an intuitive and compact account of the tax substitution argument without also rendering an intuitive and compact account of the tax substitution assumption. If the argument’s as­ sumption is fully grasped, the argument’s conclusion seems at least plausible: it makes some sense that non-labor earnings policies would be inferior to labor earnings taxation as tools for redistribution were individuals observationally identically controlling for labor Page 10 of 37

Optimal Redistributional Instruments in Law and Economics earnings. Difficulties arise when one attempts to intuitively excavate the argument while leaving the assumption buried in technical argot. (p. 332)

16.2.2.1 Theoretical Versatility, Empirical Agnosticism

The bare logic of the tax substitution argument does not, in and of itself, favor labor earn­ ings over other policies. This is merely how the logic has been applied to date. The elemental structure of the tax substitution argument is that it is better to adjust “this policy” in lieu of imposing “that policy” if individuals are “this policy”-contingently identi­ cal with respect to modifications of “that policy.” Were “this policy” taken to be some­ thing other than labor earnings, and were “that policy” taken to be labor earnings taxa­ tion, the tax substitution argument would prove the superiority of something other than labor earnings taxation over labor earnings taxation. More generally, the tax substitution argument calls for a preliminary partitioning of all policy instruments into two mutually exclusive and exhaustive cells: “these policies” and “those policies” (Mirrlees, 1976, p. 338; Laroque, 2005, p. 143; Sanchirico, 2010a, pp. 952–954). “These policies” might be a singleton subset containing only labor earnings taxation or only wealth taxation, or it might be a subset containing several elements, such as “taxation” (however defined) or “regulation” (however defined), or it might be the sub­ set of all policies save one, such as tort rules or labor earnings taxation or environmental regulation. The tax substitution assumption in this general setting amounts to the “these policies”-contingent identity with respect to “those policies.” The conclusion accordingly is that redistributional efforts are optimally contained within “these policies” and not ex­ tended to “those.” Given that the theory itself is agnostic about the pre-sorting of policies into “these” and “those,” the question arises whether there is any empirical evidence that the tax substitu­ tion assumption holds for labor earnings or some subset containing labor earnings, or is at least more likely to hold for labor earnings than anything else. No such evidence has been offered. And it would appear that no such evidence exists.

16.2.2.2 The Partial Role of Information It is sometimes presumed that the chief issue for the tax substitution argument is whether individuals’ labor earnings give a fair indication of how individuals are affected by a distributionally motivated modification of a given policy X. However, the information­ al content of labor earnings is only part of the story. This was discussed in sections 16.2.1.3.2 and 16.2.1.4.2. But two additional comments are in order. First, on a general level it is fairly intuitive that assuming merely the informational suffi­ ciency of labor earnings is inadequate. The assumption that the labor earnings tax system can deliver the same redistribution as the policy X modification does not imply that it can always deliver it more efficiently. The ability to do something does not imply the ability to do it best.

Page 11 of 37

Optimal Redistributional Instruments in Law and Economics Second, more concretely, the inadequacy of looking only at labor earnings’ information content is starkly revealed in the following extreme example. Suppose that individuals’ la­ bor productivity as well as their marginal utility from consuming some commodity Z are both determined by a single underlying, fixed parameter p. Suppose, further, that individ­ uals differ only according to the value of this parameter. In that case, (p. 333) from an individual’s labor earnings level the government will generally be able to back out the individual’s parameter value.12 From this parameter the government may in turn deter­ mine the individual’s preferences for commodity Z. Transitively, therefore, an individual’s labor earnings will reveal how she values a tax on Z. Thus, labor earnings are fully infor­ mative in this respect. Yet the tax substitution assumption is definitively not satisfied in this example. And, ac­ cordingly, the tax substitution argument will not necessarily go through. Thus, take any two individuals with differing values for parameter p. Force them to earn the same amount of labor earnings. Because the individuals have different values for pa­ rameter p, and p affects their preferences for commodity Z, they will still differently value policy X modifications in the form of taxes on commodity Z. Given that individuals differ­ ently value the tax on commodity Z at any given choice of L, it will not generally be possi­ ble to design a single labor earnings surtax that simultaneously induces every individual to make the labor earnings choice that she would make under the commodity Z tax (see section 16.2.1.4.2.2). There is thus no guarantee that a utility-equivalent labor earnings surtax will increase revenue when revenue under the pre-existing labor earnings tax is al­ so taken into account. Consistent with this, the policy eclecticism results discussed in section 16.3 do not rely on the assumption that other policies supply incremental information relative to labor earn­ ings (Sanchirico, 1997; 2000, p. 815).

16.2.2.3 A Mechanical Equivalence, Not to Be Confused with the Tax Substi­ tution Argument There is another argument used in comparing taxes and policies that is more basic than the tax substitution argument and may be potentially confused for it. This argument es­ tablishes the equivalence—not the relative superiority—of seemingly different taxes in special settings.13 To understand this mechanical equivalence, imagine a simple world in which individuals chose how much to earn from labor and then spend all their earnings on a single com­ modity. In this world, taxing commodity spending is precisely equivalent to taxing labor earnings. A 50% tax on labor earnings is equivalent to a 100% tax on consumption spend­ ing. It does not matter to the consumer, or to government revenue, whether the govern­ ment takes half her labor earnings or, instead, doubles the after-tax price she faces for the single consumption good.

Page 12 of 37

Optimal Redistributional Instruments in Law and Economics Thus, in this two-choice-variable setting whatever could be accomplished with a tax on commodity spending could indeed be accomplished with a tax on labor earnings. (p. 334) And, symmetrically, whatever could be accomplished with a tax on labor earnings could be accomplished with a tax on commodity spending. The taxes are equivalent; neither is superior. The reasoning is unrelated to the tax substitution argument; the tax substitution assumption is not required. By the same logic, in a world in which individuals have income from labor earnings and spend on many commodities, taxing total consumption expenditure (or, what is the same, taxing expenditure on each commodity at a uniform rate) is also the same as taxing labor earnings. This equivalence explains why the first application of tax substitution argument described in section 16.2.1.2.1 specifically concerned differential commodity taxation.

16.2.2.4 Problematic Descriptions of the Tax Substitution Argument 16.2.2.4.1 Double Distortion Argument(s) The tax substitution argument is often explained intuitively in a form that has been dubbed the “double distortion argument” (Sanchirico, 1997; 2000, p. 799). There are in fact two distinct variants of the double distortion argument. Both are problematic. 16.2.2.4.1.1 Labor Earnings-Contingent Policy X Adjustments Suppose that damage awards in tort cases are adjusted upward, above their efficient ex­ ternality-corrective level, when the tortfeasor is determined to have a relatively large quantity of labor earnings. Then, it is argued, the adjustment to the damages rule acts as an additional tax on labor income while distorting precautionary behavior to boot. Thus, the labor earnings-contingent damages adjustment is doubly distorting. If the same distri­ butional objective were instead accomplished by taxing labor earnings directly, such re­ distribution would cost but one distortion, to labor earnings. One problem (not the only problem) with this first version of double distortion reasoning is that it implicitly assumes that the only possible adjustment to the damages rule makes damages contingent on labor earnings (Sanchirico, 2000, pp. 799–800; 2001, pp. 1016– 1018; Liscow, 2014, pp. 2483, 2486–2502). Yet the damages rule might be adjusted in a way that is not a function of individuals’ labor earnings—the way a luxury tax might be imposed irrespective of individuals’ labor earnings. For example, the level of damages for a given kind of tort could be raised across the board—on the basis of a finding that the well-off tend to be tortfeasors in these circumstances and the less well-off tend to be acci­ dent victims. If so, then even assuming arguendo that distortion counting is a valid exer­ cise (it is not), the adjustment to the damages rule produces one distortion just like the la­ bor earnings tax. 16.2.2.4.1.2 The Missing Third Margin The second version of the double distortion argument does not rely on a policy-created connection between labor earnings and the distributionally motivated policy modification.

Page 13 of 37

Optimal Redistributional Instruments in Law and Economics Rather, the connection runs through individuals’ budget constraints (Bankman and Weis­ bach, 2006, 2007). Consider, for example, a tax on commodity Z—perhaps Z is a luxury item, or per­ haps it is future consumption for which savings is required, the earnings on which might be taxed as capital income. Double distortion reasoning argues that to tax commodity Z is to tax both the consumption of Z and the labor earnings that are spent on Z. To tax what labor earnings are spent on, that is, is to effectively tax labor earnings itself. Therefore, the tax on commodity Z produces two distortions—even if the Z-tax is not itself a function of labor earnings. (p. 335)

This version of double distortion reasoning is problematic because it ignores a requisite third good (Sanchirico, 2010a, pp. 875–888). Recall from section 16.2.2.3 that in a simple two-choice-variable world in which individuals choose how much to earn from labor and then spend all their earnings on a single commodity Z, taxing labor earnings is not superi­ or to, but equivalent to, taxing spending on commodity Z. The reason is unrelated to the tax substitution argument and its assumption. In order for labor earnings taxation to be superior—as opposed to mechanically equiva­ lent—to taxing commodity expenditure, there must be at least two spending options for labor earnings, two options that may be differently taxed. Thus, suppose, for example, that individuals could spend their labor earnings on Z or Y and that a tax on Z, and not also Y, is under consideration. What changes? First, a tax on Z is not now equivalent to a tax on labor earnings. A tax on Z-spending changes the relative price of the two commodities Y and Z. A tax on labor earnings does not. Second, distortion counting is made more complex and problematic. Removing the tax on Z would indeed seem to reduce the distortion as between Z and Y; both would be untaxed. And replacing the tax on Z with a labor earnings tax might seem to leave intact the level of distortion between leisure and Z, with the new labor earnings tax counteracting the re­ moval of the tax on Z. But what about distortion as between leisure and Y? Presumably this would worsen as Y taxation remains fixed (at zero) while labor earnings taxation in­ creases. In the end, is there less distortion? The ZY margin is less distorted, the LZ distortion is un­ changed, and the LY margin is more distorted. The net effect seems indeterminate. In fact, distortion counting itself is a fraught exercise with uncertain connection to sys­ tematic thinking about optimal tax and policy (Sanchirico, 2001, pp. 1017–1018; 2010a, pp. 886–887).14 Two indirect clues that double distortion reasoning is problematic are as follows:

Page 14 of 37

Optimal Redistributional Instruments in Law and Economics First, double distortion reasoning appears to apply without regard to whether the tax sub­ stitution assumption is in place. Yet the actual tax substitution argument itself, which the double distortion reasoning is meant to illuminate, appears highly reliant on this assump­ tion, as discussed in section 16.2. Second, double distortion reasoning seems to apply just as well in a hypothetical economy consisting of a single individual. This is in fact how double distortion reasoning is usually presented, implicitly or explicitly. Thus, it would appear to establish the superi­ ority of labor earnings taxation in this single-individual setting. Yet as discussed in the next subsection, it is evident that labor earnings taxation is not in fact superior in a sin­ gle-individual economy (even with the tax substitution assumption in place). (p. 336)

16.2.2.4.2 Single-Individual Illustrations On occasion an exposition of the tax substitution argument will be presented in stepwise fashion. First, the exposition will consider the tax substitution argument in the simplified setting in which there is only a single individual in the economy, showing that labor earn­ ings taxation is superior even when there are no distributional issues—that is, when the only issue is efficiency. Second, the exposition will extrapolate to the case of many nonidentical individuals (Bankman and Weisbach, 2006, 2007). This approach, which is helpful in many other settings, turns out to be problematic here (Sanchirico, 2010a, pp. 902–929; 2011c). The problem resides in the first step. In a sin­ gle-agent model, the tax substitution assumption holds vacuously. The assumption re­ quires a kind of identity across members of the population. A single individual is identical with herself in all respects. On the other hand, in a single-agent model there is no differ­ ence between assuming (as in the conventional optimal tax and policy framework) that the government has knowledge of population distributions and assuming the government has knowledge of particular individuals in the population: the single individual’s attribut­ es and the population distribution are one and the same. Accordingly, a uniform lump sum tax is the same as an individually tailored lump sum tax. Lump sum taxation is therefore optimal (by the argument in section 16.2.1.2.1). Contrapositively, any tax affecting rela­ tive prices, including any labor earnings tax, is suboptimal. More than this—and this is the key point—the closer any tax is to the lump sum optimum—that is, the closer its mar­ ginal rates are to zero, and the larger its lump sum component—the better it is. It follows that any given labor earnings tax schedule is dominated by any tax on any other taxable attribute that, by virtue of having shallower margins, looks more like the lump sum opti­ mum. To be sure, because the tax substitution assumption holds (vacuously) one can indeed construct a labor earnings tax that is superior to any given tax on some other base (Sanchirico, 2010a, pp. 917–924). However, the tax substitution assumption also holds (just as vacuously) with respect to this other base. Accordingly, it is just as possible to turn around and construct a tax on this other base that would be superior to the just-con­

Page 15 of 37

Optimal Redistributional Instruments in Law and Economics structed labor earnings tax. This back and forth can be continued ad infinitum. No tax or policy base is superior to any other. Thus, in the single-agent case, the tax substitution assumption holds vacuously, but the implications of the tax substitution argument are of a piece. (p. 337)

16.2.2.5 Descriptions of the Tax Substitution Assumption

16.2.2.5.1 Two Ways to Defeat the Tax Substitution Assumption The tax substitution assumption—for a tax substitution argument directed at any distribu­ tionally motivated adjustment to policy X—is again this: all individuals, hypothetically forced to choose any level of labor earnings, would have the same utility-equivalent lump sum amount for any modification to policy X. One approach to understanding the scope of this assumption is to ask: what would it take to defeat it? There are actually two ways to defeat the tax substitution assumption, one of which might be regarded as the “easy way,” the other as the “hard way.” The hard way has received more attention in the litera­ ture. Indeed the easy way occasionally escapes mention. 16.2.2.5.1.1 The Hard Way: Differences Due to Differing Amounts of Residual Leisure The hard way is to allow the possibility that the amount of an individual’s “leisure”—a term of art used to describe the amount of time not sold in the formal labor market, which may well be time spent in strenuous exertion—affects her valuation of policy X modifications. Recall that the tax substitution assumption concerns labor earnings-contin­ gent identity, not leisure hour-contingent identity. This is because, by assumption, labor earnings, but not leisure hours, are assumed to be government-observable. Thus labor earnings, but not leisure hours, may be subject to utility-equivalent sur-taxation. Now, any two individuals, forced to earn the same amount of labor income, will typically be left with differing amounts of leisure. One individual is likely to be more productive than the other and so be paid at a higher hourly wage rate. The tax substitution assump­ tion requires that, forced to generate this amount of labor earnings, the two individuals value the policy X modification the same despite their differing amounts of leisure. Yet the quantity of leisure time is likely to affect individuals’ consumption and activity pat­ terns. Therefore, the quantity of leisure affects how an individual values policy X modifi­ cations. Allowing this in the model implies that the tax substitution assumption fails to hold across differently productive individuals. 16.2.2.5.1.2 The Easy Way: Differences, Period The easy way to break the tax substitution assumption is to allow that individuals differ directly with respect to policy X, rather than through leisure differences. To isolate this easier way, imagine for the moment that individuals’ respective quantities of leisure per se have no impact on how they value any given policy X modification. Might two individu­ als, forced to generate a given level of labor earnings, still differently value a given policy modification? Even though they are forced to generate the same labor earnings, and even though their potentially different amounts of residual leisure are assumed to be irrele­ Page 16 of 37

Optimal Redistributional Instruments in Law and Economics vant, the individuals might still have different sets of feasible consumption patterns or ac­ tivities, or different preferences. Consider that labor productivity is likely to be the outward economic manifestation of a constellation of more basic traits. Whatever underlying characteristics make (p. 338) indi­ viduals differ in their labor productivity—intelligence, interpersonal skills, ambition, dri­ ve, judgment, luck, upbringing—would presumably also affect their posture with regard to precautionary measures, intentional torts, illegal behavior, environmental nuisances, savings, wealth accumulation, firm ownership, employment, education level, occupation, consumption pattern, physical and financial inheritance, and so forth. (And recall from section 16.2.2.2 that there is more to the issue than whether these other observables add incremental information to labor earnings.) Why does the easy way to break the tax substitution assumption receive less attention than the hard way? Historical accident may be one explanation. The earliest optimal tax models were specifically studies of optimal labor earnings taxation (Mirrlees, 1971). Oth­ er kinds of taxes and policies were not considered. These studies made a simplifying as­ sumption that was arguably natural and justifiable given their targeted research agenda —that individuals differed only with regard to their labor productivity, and that control­ ling for differences in labor earnings they were identical. Subsequent contributions to the optimal tax literature moved on to a broader set of questions—including whether other taxes and policies are justifiable in the presence of labor earnings taxation—without in most cases adjusting the original assumption to fit the new set of questions (Atkinson and Stiglitz, 1976).

16.2.2.5.2 Quasi-Technical Descriptions of the Tax Substitution Assumption The literature contains two near-formal descriptions of the tax substitution assumption that are, if not inaccurate, then incomplete. 16.2.2.5.2.1 Absence of Strong Complements for Leisure The tax substitution assumption is also often stated in terms of the complementarity/sub­ stitutability between leisure and other taxable attributes. Although various species of complementarity and substitutability can be precisely defined, this version of the assump­ tion is often difficult to pinpoint mathematically. In any event, this description of the as­ sumption appears to be exclusively focused on the “hard way” of defeating the tax substi­ tution assumption—as laid out in section 16.2.2.5.1.1. 16.2.2.5.2.2 “No Within Income Group Differences” The tax substitution assumption is sometimes described as the absence of “within income group differences” (where “income” means labor earnings).15 This description seems to imply that identity is required only among individuals that actually choose the same level of labor earnings. However, as discussed in detail in sections 16.2.1.4.2 and 16.2.2.2, the tax substitution assumption requires more than that individuals that choose the same la­ bor earnings place the same value on the policy modification. The tax substitution as­ Page 17 of 37

Optimal Redistributional Instruments in Law and Economics sumption requires, rather, that all individuals—across all earnings groups—value the poli­ cy X modification the same, at all possible hypothetically fixed labor earnings levels.

16.2.2.5.3 Technical Descriptions of the Tax Substitution Assumption: the “Weak Separability of Leisure” (p. 339)

The tax substitution assumption is often equivalently stated in terms of the structure of individual utility functions. This statement of the assumption—dubbed the “weak separa­ bility of leisure”—is logically complete. Critics have argued, however, that casting the as­ sumption in this way is potentially misleading. 16.2.2.5.3.1 The Restatement (The following discussion is somewhat more technical. Some readers may wish to skip to the next section.) This restatement of the tax substitution assumption has an implicit first step. Preliminari­ ly, individual choice problems are restated so that individuals are regarded as choosing taxable attributes rather than direct sources of utility.16 “Taxable attributes” are govern­ ment-observable individual characteristics, possibly determined by individual choice, up­ on which policy impact may be individually tailored. For instance, individuals are regarded as choosing labor earnings, a “taxable attribute,” rather than leisure, a direct source of utility. Wage rate thus becomes a utility function parameter implicitly converting labor earnings into leisure time. Moreover, differences in individuals’ budget constraints that were due to their differences in wage rates are elimi­ nated: in place of the product of individual wage and work hours (total hours less leisure) there appears, identically for all individuals, simply dollars earned from labor. In general the restatement of individual choice problems in terms of taxable attributes moves any individual differences in budget constraints into the utility function, leaving in­ dividual budget constraints identical and concentrating all individual differences in the utility function. Given this first step, it is then assumed that non-labor earnings taxable attributes enter individual utility functions (so transformed) solely through a “sub-utility function” that is identical for all individuals. This sub-utility function may be a function of labor earnings. Thus utility for each individual i has the form ui(L, s(L,A)), where L is labor earnings, s(⋅,⋅) is the individual-uniform sub-utility function, and A is the list of all other taxable attribut­ es. The implication is that any two individuals, hypothetically forced to earn the same level of labor earnings L’, have utility functions (so transformed) that, although not necessarily precisely the same, have the same ranking of taxable attributes A. Given that their bud­ get constraints (so transformed) are the same, the individuals thus choose the same array of taxable attributes. They thus identically value (in terms of lump sum equivalents) poli­ cies affecting those attributes. Page 18 of 37

Optimal Redistributional Instruments in Law and Economics As an aside, it is often emphasized that “weak separability of leisure” means that individ­ uals (forced to earn the same L) have identical marginal rates of substitution between non-labor earnings taxable attributes in A. This is a clear implication of the form ui(L, (p. 340) s(L,A)). Individual-differing marginal u ’s appear in both the numerator and de­ i nominator of the marginal rate of substitution and so cancel out, leaving marginal s(L,⋅)’s that are individual-identical (when individuals are forced to earn the same L). 16.2.2.5.3.2 Potentially Misleading Aspects Critics contend that stating the tax substitution in terms of the “weak separability of leisure” is problematic (apart from its ambiguous use of the adjective “weak”) (Sanchiri­ co, 2010a, pp. 940–946). First, stating the tax substitution assumption in terms of sub-utility functions— conceptu­ al tools without natural existence—arguably renders the assumption more opaque than it needs to be. The favored alternative is to state the assumption in terms of lower-altitude concepts like choice and willingness to pay, as attempted in section 16.2.1.4. Second, casting the tax substitution assumption as a matter of the “separability” of leisure encourages an unjustified emphasis on the hard way to defeat the tax substitution assumption, as discussed in section 16.2.2.5.1. As there noted, separability, as that con­ cept is commonly and naturally conceived, is insufficient. Suppose, for example, that utili­ ty has the form ui(L) + si(A), so that labor earnings (and so leisure) is isolated in its own additive component. If si(⋅) differs across individuals, ui(L) + si(A) does not exhibit “weak separability of leisure,” despite the fact that leisure appears to be separated—indeed “strongly.” Third, stating the tax substitution assumption solely in terms of utility functions may lead to the impression that the assumption is solely concerned with individuals’ preferences, rather than their budget constraints and market interactions. Technically, the assumption can be viewed as solely concerning utility functions. But, as discussed, this is only after utility functions have been transformed so as to remove other potential sources of individ­ ual difference that would otherwise lie outside the scope of utility and preferences. This is important if one is inclined to view preference heterogeneity as more elusive or less im­ portant than other more concrete sources of individual difference.

16.2.2.6 “Indicator Goods” It is sometimes said that the tax substitution assumption will fail to hold when other gov­ ernment-observable attributes are “indicator goods.” Different authors use the term “in­ dicator good” to mean different things. Some mean that observable consumption of the good is a signal of labor productivity. In this case, the indicator goods intuition would ap­ pear to neglect the possibility that labor productivity is not the only underlying source of difference among individuals—a la the “easy way” to break the tax substitution assump­ tion, as discussed in section 16.2.2.5.1. By contrast, other uses of the indicator good intu­ ition seem to have in mind the condition that a given non-labor earnings attribute pro­ vides information about underlying characteristics over and above what can be gleaned Page 19 of 37

Optimal Redistributional Instruments in Law and Economics from labor earnings. This version of indicator good reasoning appears to neglect the fact that information supply is only part of the issue, per the discussion in section 16.2.2.2.

(p. 341)

16.3 Policy Eclecticism

The preceding section discussed the portion of the literature on optimal redistributional instruments that explores a particular condition (the tax substitution assumption) under which it would not be optimal to account for distributional considerations in policies oth­ er than labor earnings taxation. This section considers a second portion of the literature that approaches the issue from the opposite direction, yet still within the same general framework. Retaining the optimal tax and policy model—but eschewing the tax substitu­ tion assumption—this strand of the literature asks: under what conditions would it be op­ timal to inform a given policy with distributional considerations? This inverse approach is important for two reasons. First, the tax substitution argument is but one argument for labor earnings exclusivity in distributional policy. The rejection of one argument for a proposition does not imply that there are no others. Ruling out the ex­ istence of possibly unknown others requires a valid argument for the inverse. Second, the inverse approach sheds some light on whether the conditions under which non-labor earnings policies should be distributionally informed are merely “exotic” “theoretical cu­ riosities,” as has been asserted, or rather are more closely tied to the core of the optimal tax and policy framework. Papers taking this inverse approach generally assert that the conditions for some distributionally motivated adjustments are quite general, indeed nearly universal. (The difficulty of empirically determining precisely what these adjustments are, and the signifi­ cance of this difficulty, are discussed in section 16.4.) According to these contributions, the unadulterated implication of the optimal tax and policy model is not labor earnings exclusivity, but a strong form of eclecticism that contradicts labor earnings exclusivity a fortiori. This section of the chapter discusses, and draws a connection between, three sublitera­ tures that have made this point.

16.3.1 An Arbitrage Approach to the Optimal Choice of Redistribu­ tional Instruments Certain complications arise in the policy eclecticism argument. Most notably, any distrib­ utionally motivated policy adjustment must be evaluated in light of its ancillary revenue effects on a pre-existing labor earnings tax system. Adding to the complexity, certain ap­ plications of the policy eclecticism argument assume these ancillary revenue effects away, while others do not. The following framework is systematic enough to handle these com­ plications directly. It also has the added intuitional benefit of linking to the relatively fa­ miliar concept of price arbitrage.17

Page 20 of 37

Optimal Redistributional Instruments in Law and Economics (p. 342)

16.3.1.1 Policy X’s “Revenue Price of Redistribution”

Return to the framework introduced in the opening paragraphs of section 16.2.1, wherein a labor earnings tax system is assumed to be in place and the possibility of adjusting some policy X for distributional purpose is under consideration. One way to understand the policy eclecticism argument is to think in terms of each policy X adjustment’s “revenue price of redistribution.” This is the marginal dollar of revenue lost per “unit” of favorable redistribution effected by a “compensated” “pro-distribution­ al” marginal adjustment to policy X. The adjustment is “marginal” in the usual economic sense. Thus the focus is on directions of change, rather than discrete changes. The argument establishes only that there is a distributionally motivated direction of adjustment to policy X that is social welfare im­ proving. This implies, but does not determine, an optimal discrete adjustment. However, the analytical use of infinitesimal changes does not imply that the optimal adjustment is also infinitesimal, or even small. The policy adjustment is “compensated” in the special sense that the utility-equivalent lump sum scheme (as defined in section 16.2.1) for the adjustment is calculated, aggre­ gated across individuals, and then distributed uniformly to all individuals. Accordingly, the compensated policy X adjustment has no impact on total well-being, but is rather purely distributional. The marginal policy adjustment is “pro-distributional” in the sense that on average it transfers from individuals with low “marginal social welfare weight,” to individuals with high social welfare weight. An individual’s “marginal social welfare weight” is the margin­ al impact on social welfare from transferring resources to such individual. It accounts for both the individual’s own marginal utility—which may be diminishing—and the marginal social welfare of the individual’s utility—which may depend on the individual’s level of utility relative to the rest of the population. How do we know that a pro-distributional adjustment exists? Two points here: First, if one direction of adjustment is anti-distributional, the opposite direction will be pro-distri­ butional. For example, if a compensated tax increase on commodity Z tends to redistrib­ ute in the “wrong” direction, a compensated tax decrease (possibly a subsidy) will redis­ tribute in the right direction. Second, we may essentially rule out the possibility that policy X has no distributional im­ pact whatsoever in either direction. Of course, if the policy X adjustment is itself a uni­ form lump sum tax, then a compensated policy X adjustment gives back to each individual the same amount it takes, and there is no distributional impact. But let us assume that policy X is not a non-starter redistributionally—that it has winners and losers. In fact, in a heterogeneous population, this will be true of essentially every policy save those specifi­ cally designed to be otherwise.

Page 21 of 37

Optimal Redistributional Instruments in Law and Economics The policy X adjustment would also have no distributional impact if all individuals possessed the same marginal social welfare weight. But this is also essentially a nonevent in a heterogeneous population. Note in this regard that the pre-existence of the la­ bor earnings tax does not change this. A labor earnings tax system will generally be inca­ pable of equalizing marginal social welfare weights. And even it were possible, it would not be optimal except in very special cases. (p. 343)

Given that the policy has differential impact across individuals and that individuals have different marginal social welfare weights, it is, to be sure, conceivable that these two dif­ ferentials align in such way that the policy still has zero distributional impact—that is, that the covariance between policy impact and marginal social welfare weight is precisely zero. But given two variables each with some variance, true zero covariance, however conceivable, is a knife edge phenomenon. It vanishes with the slightest variation in mod­ eling parameters. What is a “unit” of “pro-distributional impact”? The pro-distributional impact of the com­ pensated policy X adjustment is the adjustment’s marginal social welfare—which, given that the adjustment is “compensated,” will come from a zero-sum transfer of burden across individuals, and thus be purely attributable to the social welfare impact of redistri­ bution. The choice of social welfare function determines the units of pro-distributional im­ pact and insures that such units are uniform across policy instruments, which is all that matters, as will become clear. Most of the discussion so far has concerned the denominator of the revenue price of dis­ tribution, the pro-distributional impact. Now consider the numerator, the revenue cost. Importantly, the revenue impact of the compensated pro-distributional adjustment to poli­ cy X will depend on policies already in place. Thus, even a policy X modification that in and of itself has no revenue implications may well have global revenue implications if there is already in place a behaviorally sensitive tax system and the policy X modification affects revenue-relevant behavior. Recognizing this point helps to avoid some potential sources of confusion with respect to “tag” taxation and the “taxes versus legal rules” de­ bate, as discussed in sections 16.3.2–16.3.4. If revenue is gained from the compensated pro-distributional adjustment to policy X, then the revenue price of redistribution for the policy X adjustment is negative. If there are no revenue effects, such price is zero.

16.3.1.2 Arbitrage-Like Conditions for Eclecticism Imagine that an optimal labor earnings tax system is already in place—optimal, that is, conditional on its exclusivity—and the question under consideration is whether to effect a marginal compensated pro-distributional adjustment to policy X. Under what conditions is this warranted? The adjustment to policy X is certainly warranted if the policy X adjustment has a nega­ tive or zero revenue price of redistribution. The case of a zero revenue price of redistribu­

Page 22 of 37

Optimal Redistributional Instruments in Law and Economics tion is important for understanding how the eclecticism point manifests in the debate over “distributionally informed legal rules,” as discussed in section 16.3.4. But what if policy X’s revenue price of redistribution were positive? That is, what if favorably redistributing via policy X expended revenue on the margin? It might in this case seem that the propriety of adjusting policy X for distributional purpose would de­ pend on whether its revenue price of redistribution were lower than that of the labor earnings tax.18 But this is incorrect. (p. 344)

Suppose that the revenue price of redistribution for policy X is greater than for labor earnings. Now imagine reversing the policy X adjustment. By hypothesis this would raise a relatively large amount of revenue at the cost of unfavorable redistribution. This unfa­ vorable redistribution could then be offset by a pro-redistributional adjustment in the la­ bor earnings tax. By hypothesis this offsetting pro-distributional labor earnings tax ad­ justment will expend less revenue than was raised from the anti-distributional policy X adjustment. In the end, therefore, the same distribution would be accomplished, but with revenue to spare. Note that in the particular case in which policy X’s revenue price of redistribution is greater than for labor earnings, the modification to policy X would not itself favorably re­ distribute. However, policy X would still be included in the redistributional program. It would not be determined solely on the basis of efficiency. Were it a tax, its determinants would be in the tax base. Thus, the logic of policy eclecticism is like the logic of arbitrage. Any price difference is exploitable. Moreover, in the case of redistributional instruments, some price difference is essentially inevitable.

16.3.1.3 Pro- or Anti-Distributional? It is worth reiterating that the optimal distributionally motivated adjustment to policy X may not itself be pro-distributional when viewed in isolation. Rather, the role of such ad­ justment may be to raise revenue that otherwise would be raised by the labor earnings tax so that the latter can be made more pro-distributional. The same can be said of labor earnings taxation itself. Optimal adjustments to the tax rate at any given level of labor earnings may be locally anti-distributional, enabling pro-distri­ butional adjustments in other policy areas. In general, the optimal distributional adjustment in any policy area must be evaluated globally. This further complicates the practical determination of optimal distributional ad­ justments, as considered in section 16.4.

16.3.2 Taxing Immutable “Tags” The literature on policy eclecticism begins with the subliterature on “tags.” A tag is an (effectively) immutable, government-observable marker displayed by each individual, Page 23 of 37

Optimal Redistributional Instruments in Law and Economics according to which an individual’s tax liability or transfer entitlement may be de­ termined. Race, ethnicity, gender, height, and eye color are examples of tags.19 (p. 345)

The main finding in the literature on tags is that, even in the presence of optimal labor earnings taxation, social welfare can be improved by redistributing resources on the ba­ sis of how individuals manifest the tag. Such redistribution can be accomplished with a revenue-neutral tax-cum-transfer based on the tag (Akerlof, 1978; Allingham, 1975; Stern, 1982; Logue and Slemrod, 2008; Mankiw and Weinzierl, 2010). It is tempting to analyze “tags” as non-distortionary hooks for favorable redistribution based on their covariance with social welfare weight: the next best thing to the hypotheti­ cal fully individualized lump sum taxes that lead to first-best outcomes, as discussed in section 16.2.1.1. But this is somewhat misleading. Tags are only imperfect proxies, and thus the fully optimal system may include some policies, like labor earnings taxation, that are not tags. Thus to argue that tags are part of the optimal system, one must allow that tag-based taxes and transfers are being imposed against the backdrop of pre-existing be­ haviorally sensitive taxes. Through its impact on choices such as labor earnings, then, the tag tax-cum-transfer-back is likely to have some effect on global revenue, even though it is itself revenue-neutral. Therefore there is more to the tag story then finding the proper direction of correlation between the tag and marginal social welfare weight. The revenue price of redistribution framework—based as it is on the ratio of revenue effects and distributional effects— presents a more complete picture. To be sure, taxing or subsidizing tags like eye color and height may seem ludicrous. And taxing or subsidizing gender, ethnicity, or race is likely to be controversial. Yet the propri­ ety of such a tag taxation does indeed precipitate from the conventional optimal tax and policy model. Mankiw and Weinzierl (2010) suggest that this reflects a flaw in the model itself, or perhaps more generally in the desire to conduct redistribution. Sanchirico (2013) argues that the problem may lie partly in the conventional model’s assumption, as discussed in section 16.2.1.3.1, that the government knows for certain relevant popula­ tion covariances. He argues that relaxing this assumption (as is considered in section 16.4) sensibly narrows the range of optimal tag taxes.

16.3.3 Behaviorally Responsive Attributes Given the possibility of pre-existing policies with behaviorally responsive revenue effects, there is not so great a distance as might be supposed between the analysis of tag taxation and the analysis of “distortionary” taxes and policies. Such taxes and policies are consid­ ered in Saez (2002) and Sanchirico (2011a), both of which focus on capital income taxa­ tion as a species of differential commodity taxation.

Page 24 of 37

Optimal Redistributional Instruments in Law and Economics (p. 346)

16.3.4 “Legal Rules” in an Artificially Separated Legal System

Sanchirico (1997, 2000, 2001) offers the eclecticism argument in the context of “legal rules,” which are essentially externality-correcting policies. (See also Avrahan, Fortus, and Logue, 2004.) These papers specifically address the question of whether redistribu­ tive goals should be pursued solely through the labor income tax, or also through adjust­ ing “legal rules” away from what is dictated for such rules by efficiency (i.e. by externali­ ty internalization). The condition for the optimality of redistributional eclecticism provid­ ed in these papers is not stated in terms of the revenue price of redistribution. Rather the condition is that the distribution in the population of the net burden of the legal rule ad­ justment has a non-zero covariance with marginal social welfare weight. Within the spe­ cial model that these papers employ, however, this condition is equivalent to the condition that the revenue price of redistribution be zero, so that inclusion of the legal rule in the distributional program provides marginal benefit without marginal cost. What is special about these models is that they rule out ancillary revenue effects through pre-existing labor earnings taxation. The models in Sanchirico (1997, 2000, 2001) are de­ rived from the model in Kaplow and Shavell (1994), to which Sanchirico responds. That model has two key features. First, the legal rule is locally revenue neutral: all money that the rule itself collects (damages) is paid out (as compensation) within the policy. Second­ ly, individual decisions pertaining to the legal system, specifically precautionary effort/ spending, are housed in a separate choice problem that is divorced (save potentially through the structure of the tax system) from individual decisions pertaining to labor earnings and consumption. Sanchirico (2001, pp. 1062–1064) provides a detailed descrip­ tion of how this separation is achieved. The upshot of these two modeling features is that adjustments to the legal rule have no revenue impact, locally or globally. Thus, so long as the legal rule has non-zero distribu­ tional impact, the revenue price of redistribution via adjustments to the legal rule is zero.20

16.4 Policy Uncertainty The literature considered thus far in this chapter remains largely within the general con­ fines of the conventional optimal tax and policy model. This reflects the intellectual histo­ ry on the issue of optimal redistributive instruments. Of late, however, the debate’s cen­ ter of gravity—at least within law and economics—appears to have shifted to an issue that arises outside the conventional model. This issue first arises in the literature as a (p. 347) kind of extra-model alternative defense of the proposition that distributional goals are best pursued through labor earnings taxation. An appeal to pragmatism rather than math­ ematical reasoning, this alternative defense turns on the readily cognizable possibility that the government may be uncertain about even the population distributions of key pa­ rameters and so also uncertain regarding which policies are optimal.

Page 25 of 37

Optimal Redistributional Instruments in Law and Economics

16.4.1 Policy Certainty in the Conventional Framework The information assumption of the conventional optimal tax and policy model was men­ tioned in section 16.2.1.3.1. In more detail, although the government in this model cannot directly discern the underlying characteristics of a particular individual that may be placed before it, the government is assumed to be certain and correct regarding popula­ tion distributions. That is, it knows how many individuals have each possible manifesta­ tion of every relevant characteristic, including utility functions and budget constraints. For example, in Mirrlees’ (1971) original optimal labor earnings tax model, the govern­ ment is assumed to know the number of individuals with each possible level of labor pro­ ductivity, even though it is unable to observe the labor productivity of any given individ­ ual. Given that the government is certain and correct at the population level, it is essentially certain and correct regarding the impact of policy choices. It cannot tell how any particu­ lar individual has been affected by any given policy alternative. But it can tell how many individuals have been affected in each possible manner. Since such “anonymous” knowl­ edge is all it cares about (an assumption of the model), the policymaking process is for it essentially a matter of decision-making under certainty.

16.4.2 The Alternative Defense Based in Policy Uncertainty The alternative defense of labor earnings exclusivity in distributional policy implicitly— and justifiably—steps outside the conventional model by removing the assumption of poli­ cy certainty (at least partially). Even if it is technically true that the optimal tax and poli­ cy framework points towards policy eclecticism, the argument goes, it would still be un­ wise for the government to expand its deployment of distributional instruments beyond labor earnings taxation because the impact of alternative instruments is too uncertain. Theoretical arguments for policy eclecticism, even if correct, are of little practical use, the argument continues, because they indicate only that—and not how—non-labor earn­ ings policies should be distributionally informed. Indeed, they fail even to prescribe whether non-labor earnings attributes should be taxed or subsidized, let alone to what de­ gree. Further, the data is simply not available for reliable empirical estimates. (p. 348) As a consequence, adding non-labor-earnings attributes to distributional policy in an attempt to improve that policy may well backfire. Given the additional policy adjustment costs of expanding the policy base, such risks are not advisably assumed (Kaplow and Shavell, 2000, p. 832; Bankman and Weisbach, 2006, pp. 1447, 1445; 2007, p. 794; 2011, p. 550).

16.4.3 Premises of the Policy Uncertainty Defense The policy uncertainty defense of labor earnings exclusivity appears to rest on two premises.

Page 26 of 37

Optimal Redistributional Instruments in Law and Economics The first premise concerns the quantity and character of the existing information avail­ able to the government. The extra-model alternative defense is, of course, explicit regard­ ing its premise that the government does not know enough about the impact of non-laborearnings-based policies. A necessary, corresponding, yet often implicit premise is that the government does know enough about the impact of taxing or subsidizing labor earnings. The second premise concerns not the quantity or character of existing information, but the proper way to deal with limited information in policymaking. This second premise ap­ pears to be a kind of comparative “precautionary principle” for policy: Given two policy instruments, one with relatively uncertain impact and one with relatively certain impact, policymakers should favor use of the instrument with more certain impact. Critics of the policy uncertainty defense have questioned both of the foregoing premises, arguing first that there is no reason to believe that the government does or could know any more about labor earnings than other major contenders for distributional policy, and second that there is no real basis for a relative precautionary principle for distributional policy.

16.4.4 How Much Do We Know About Distributing Through the Labor Earnings Tax? With regard to the first premise, critics have argued labor earnings taxation is itself sub­ ject to great policy uncertainty, and that no coherent argument has been, or perhaps could be, made that the problem is less severe for labor earnings taxation than for other distributional instruments (Sanchirico, 2011b, pp. 10–37). The pure theory of optimal la­ bor earnings taxation offers only the barest minimum by way of prescription (Mirrlees, 1976; Stiglitz, 1982, pp. 239–240; Slemrod, 1987). Its sharpest useful prescription, ar­ guably, is that the tax rate should lie somewhere between 0% and 100%. And even this re­ lies on untenable assumptions regarding how individuals differ (Mirrlees, 1976, pp. 333, 341–344; Choné and Laroque, 2010, pp. 2532–2533; Judd and Su, 2006; (p. 349) Boadway et al., 2002; Bošković, 2008, pp. 13–14). Moreover, it applies only when—in somewhat conclusory fashion—the only taxes considered in the optimization problem are those on labor earnings (Mirrlees, 1976, pp. 333–340, 344–352; Nava, Schroyen, and Marchand, 1996). Further, computer simulations of optimal tax models—which input empirically determined parameters into the theoretical model in an attempt to narrow down its outputted pre­ scriptions—have not in fact been able to tighten the range of possibilities for optimal poli­ cy (Diamond, 1998, p. 84; Kanbur and Tuomala, 1994, p. 281; Myles, 2000, p. 114). This appears to have two causes. The first is the extreme dependence of such simulations on portions of the hypothesized theoretical model (such as the curvature of utility functions) that do not submit to empirically grounded parameterization. The second is the substan­ tial width of the range of plausible values for parameters that do submit to empirical in­ vestigation. Regarding the latter, a case in point is the surprisingly wide spread of esti­ mates for the after-tax price elasticity of work hours—the percentage change in work Page 27 of 37

Optimal Redistributional Instruments in Law and Economics hours per percentage change in take-home hourly wage, an elemental parameter in opti­ mal labor earnings tax design (Joint Comm. on Tax’n, 2003, pp. 15–16; Tuomala, 1990, p. 14; Ashenfelter, Doran, and Schaller, 2010, pp. 1–2; Blundell and MaCurdy, 1999, pp. 1684–1685; Evers, de Mooij, and van Vuuren, 2008, p. 26; Pencavel, 2002, p. 252).

16.4.5 A Relative Precautionary Principle for Distributional Policy? With regard to the second premise of the policy uncertainty defense, critics argue that a relative precautionary principle for distributional policy finds little if any support in sys­ tematic thinking about decision-making under uncertainty (Sanchirico, 2010b; 2011b, pp. 37–62). Indeed, it has been suggested that the view that uncertainty counsels caution may spring from a conflation of two logically distinct questions. First, is one averse to un­ certainty? Second, does uncertainty make one averse to a given choice? Aversion to uncertainty does not imply aversion to any given choice—including choices that might be regarded as “actions” as opposed to omissions. To be sure, it is reasonable to suppose that uncertainty will decrease the (expected) pay-off from each choice. This is a manifestation of uncertainty aversion and/or risk aversion. But the fact that the pay-offs from each choice decrease says nothing about how uncertainty affects the relative attractiveness of one choice versus another. The effect of uncertainty on the relative at­ tractiveness of such choices turns on which choice’s pay-offs decrease more by virtue of uncertainty. In the context of optimal redistribution, uncertainty reduces social pay-offs whether or not policy X is included in the distributional program. And it is not possible to make gen­ eral statements about whether uncertainty’s negative impact is larger or smaller in (p. 350) either case. Sanchirico (2010b; 2011b, pp. 37–62) describes one reason why un­ certainty may well increase the relative attractiveness of redistributing via a positive tax on a given non-labor earnings attribute: a positive tax rate may dampen the negative im­ pact on expected social welfare of uncertainty regarding the population distribution of the attribute. This point resonates with a precedent economics literature on the effect of uncertainty on business investment. The lesson of this literature is that uncertainty regarding future market prices and wages is as likely to increase as to decrease a rational firm’s invest­ ment in business capital. The effect on such investment turns on the effect of a “meanpreserving spread” in the probability distribution of prices or wages (representing in­ creased uncertainty therein) on the marginal pay-offs from investment. This is in turn a matter of whether the marginal pay-off function from investment—not the pay-off func­ tion itself—is concave or convex in prices and wages over the relevant range. And this is, in turn, a matter of the sign taken by third-order cross-derivatives (Hartmann, 1973). High-altitude notions like third-order cross-derivatives are difficult to grasp conceptually (the change in the change due to x, in the change due to y?) let alone observe and mea­ sure in practical applications.

Page 28 of 37

Optimal Redistributional Instruments in Law and Economics However, when a model does prescribe caution based on uncertainty—whether the model concerns business investment, precautionary savings, or “robust control” in macroeco­ nomic policy—some assumption of this nature is likely to be lurking in the background. In some cases the assumption will be warranted by the structure of the problem; in other cases it will be arbitrarily imposed.

16.5 Conclusion Despite several decades of discourse on optimal redistributional instruments, the litera­ ture in this important area is still largely incubatory. Most of the literature has been occu­ pied with understanding the lessons of the conventional optimal tax and policy model. The work has been difficult because the model is subtle. But the product of those efforts seems disappointing after all. Without special assumptions, the model is too undiscern­ ing. And the assumptions that have produced sharp results, once fully understood, seem uncomfortably close to the results themselves. The conventional model may well have more to say. But it seems more likely that the time has come to break the shell and step outside the usual framework. Further advances most probably lie in the systematic analysis of factors not fully encompassed by the language of “distortion,” “deadweight loss,” and “welfare weights”—factors like cognitive limitations, political feasibility, the absence of perfectly competitive markets (De Geest, 2013), policy uncertainty (Sanchirico, 2010b, 2011b, 2013), administrability, evasion and avoidance ac­ tivities (Gamage, 2014, 2015), and relative administrative competence.

References Akerlof, George A. (1978). “The Economics of ‘Tagging’ as Applied to the Optimal Income Tax, Welfare Programs, and Manpower Planning.” American Economic Review 68: 8–19. Allingham, M. G. (1975). “Towards an Ability Tax.” Journal of Public Economics 4: 361– 376. Ashenfelter, Orley, Kirk Doran, and Bruce Schaller (2010). “A Shred of Credible Evidence on the Long-Run Elasticity of Labour Supply.” Economica 77: 637–650. Atkinson, Anthony B. and Joseph E. Stiglitz (1976). “The Design of Tax Structure: Direct Versus Indirect Taxation.” Journal of Public Economics 6: 55–75. Avraham, Ronen, David Fortus, and Kyle Logue (2004). “Revisiting the Roles of Legal Rules and Tax, Rules in Income Redistribution: A Response to Kaplow & Shavell.” Iowa Law Review 89: 1125–1158. Bankman, Joseph and David A. Weisbach (2006). “The Superiority of an Ideal Consump­ tion Tax over an Ideal Income Tax.” Stanford Law Review 58: 1413–1456.

Page 29 of 37

Optimal Redistributional Instruments in Law and Economics Bankman, Joseph and David A. Weisbach (2007). “Reply, Consumption Taxation Is Still Su­ perior to Income Taxation.” Stanford Law Review 60: 789–802. Bankman, Joseph and David A. Weisbach (2011). “A Critical Look at A Critical Look—Re­ ply to Sanchirico.” Tax Law Review 64: 539–550. Blundell, Richard and Thomas MaCurdy (1999). “Labor Supply: A Review of Alternative Approaches,” in O. C. Ashenfelter and D. Card, eds., Handbook of Labor Economics 3A. Amsterdam: Elsevier, pp. 1559–1695. Boadway, Robin, Maurice Marchand, and Pierre Pestieau (2000). “Redistribution with Un­ observable Bequests: A Case for Taxing Capital Income.” Scandinavian Journal of Eco­ nomics 102: 253–267. Boadway, Robin, Maurice Marchand, Pierre Pestieau, and Maria del Mar Racionero (2002). “Optimal Redistribution with Heterogeneous Preferences for Leisure.” Journal of Public Economic Theory 4: 475–498. Bošković, Branko (2008). “Optimal Income Taxation with Heterogeneous Preferences.” (September 2, 2008). Manuscript, University of Toronto Department of Eco­ nomics. Choné, Philippe and Guy Laroque (2010). “Negative Marginal Tax Rates and Heterogene­ ity.” American Economic Review 100: 2532–2547. Deaton, Angus (1979). “Optimally Uniform Commodity Taxes.” Economics Letters 2: 357– 361, corrected in Deaton and Stern (1986). Deaton, Angus (1981). “Optimal Taxes and the Structure of Preferences.” Econometrica 49: 1245–1260. Deaton, Angus and Nicholas Stern (1986). “Optimally Uniform Commodity Taxes, Taste Differences and Lump-Sum Grants.” Economics Letters 20: 263–266. De Geest, Gerrit (2013). “Removing Rents: Why the Legal System is Superior to the In­ come Tax at Reducing Income Inequality” (October 8, 2013). Washington University in St. Louis Legal Studies Research Paper No. 13-10-02, available at . Diamond, Peter A. (1998). “Optimal Income Taxation: An Example with a U-Shaped Pat­ tern of Optimal Marginal Tax Rates.” American Economic Review 88: 83–95. Evers, Michiel, Ruud de Mooij, and Daniel van Vuuren (2008). “The Wage Elasticity of Labour Supply: A Synthesis of Empirical Estimates.” De Economist 156: 24–43. Gamage, David (2014). “How Should Governments Promote Distributive Justice?: A Framework for Analyzing the Optimal Choice of Tax Instruments,” Tax Law Review 68: 1– 87.

Page 30 of 37

Optimal Redistributional Instruments in Law and Economics Gamage, David (2015). “The Case for Taxing (All of) Labor Income, Consumption, Capital Income, and Wealth.” Tax Law Review 68: 355–442. (p. 352)

Hartman, Richard (1973). “Adjustment Costs, Price and Wage Uncertainty, and Invest­ ment.” Review of Economic Studies 40: 259–267. Hylland, Aanund and Richard Zeckhauser (1979). “Distributional Objectives Should Affect Taxes but Not Program Choice or Design.” Scandinavian Journal of Economics 81: 264– 284. Joint Committee on Taxation (2003). 108th Cong., Overview of Work of the Staff of the Joint Committee on Taxation to Model the Macroeconomic Effects of Proposed Tax Legis­ lation to Comply with House Rule XIII.3.(h)(2) (Comm. Print). Jolls, Christine (1998). “Behavioral Economic Analysis of Redistributive Legal Rules.” Van­ derbilt Law Review 51: 1653–1677. Judd, Kenneth L. and Che-Lin Su (2006). “Optimal Income Taxation with Multidimensional Taxpayer Types.” Working Paper No. 471. Society for Computational Economics, available at . Kanbur, Ravi and Matti Tuomala (1994). “Inherent Inequality and the Optimal Graduation of Marginal Tax Rates.” Scandinavian Journal of Economics 96: 275–282. Kaplow, Louis (1996). “The Optimal Supply of Public Goods and the Distortionary Cost of Taxation.” National Tax Journal 49: 513–533. Kaplow, Louis (2004). “On the Undesirability of Commodity Taxation Even When Income Taxation Is Not Optimal.” Discussion Paper, vol. 10407. NBER, published as noted below. Kaplow, Louis (2006). “On the Undesirability of Commodity Taxation Even When Income Taxation Is Not Optimal.” Journal of Public Economics 90: 1235–1250. Kaplow, Louis and Steven Shavell (1994). “Why the Legal System is Less Efficient than the Income Tax in Redistributing Income.” Journal of Legal Studies 23: 667–681. Kaplow, Louis and Steven Shavell (2000). “Should Legal Rules Favor the Poor? Clarifying the Role of Legal Rules and the Income Tax in Redistributing Income.” Journal of Legal Studies 29: 821–835. Laroque, Guy (2005). “Indirect Taxation Is Superfluous under Separability and Taste Ho­ mogeneity: A Simple Proof.” Economics Letters 87: 141–144. Lipsey, Richard G. and Kelvin Lancaster (1956–57). “The General Theory of the Second Best.” Review of Economic Studies 24: 11–32. Liscow, Zachary (2014). “Reducing Inequality on the Cheap: When Legal Rule Design Should Incorporate Equity as Well as Efficiency.” Yale Law Journal 123: 2478–2510. Page 31 of 37

Optimal Redistributional Instruments in Law and Economics Logue, Kyle and Ronen Avraham (2003). “Redistributing Optimally: Of Tax Rules, Legal Rules, and Insurance.” Tax Law Review 56: 157–258. Logue, Kyle and Joel Slemrod (2008). “Genes as Tags: The Tax Implications of Widely Available Genetic Information.” National Tax Journal 61: 843–863. Mankiw, N. Gregory and Matthew Weinzierl (2010). “The Optimal Taxation of Height: A Case Study of Utilitarian Income Redistribution.” American Economic Journal: Economic Policy 2: 155–176. Mirrlees, James A. (1971). “An Exploration in the Theory of Optimum Income Taxation.” Review of Economic Studies 38: 175–208. Mirrlees, James A. (1976). “Optimal Tax Theory: A Synthesis.” Journal of Public Econom­ ics 6: 327–358. Myles, Gareth D. (2000). “On the Optimal Marginal Rate of Income Tax.” Economics Let­ ters 66: 113–119. Nava, Mario, Fred Schroyen, and Maurice Marchand (1996). “Optimal Fiscal and Public Expenditure Policy in a Two-Class Economy.” Journal of Public Economics 61: 119– 137. (p. 353)

Pencavel, John (2002). “A Cohort Analysis of the Association Between Work Hours and Wages Among Men.” Journal of Human Resources 37: 251–274. Polinksy, A. Mitchell (1989). An Introduction to Law and Economics. 2nd edition. Chap­ ters 2, 15. Boston, MA: Little, Brown and Company. Saez, Emmanuel (2002). “The Desirability of Commodity Taxation under Non-Linear In­ come Taxation and Heterogeneous Tastes.” Journal of Public Economics 83: 217–230. Sanchirico, Chris William (1997). “Taxes Versus Legal Rules as Instruments for Equity: A More Equitable View.” Discussion Paper No. 9798-04, Columbia Economics Department, available at . Sanchirico, Chris William (2000). “Taxes Versus Legal Rules as Instruments for Equity: A More Equitable View.” Journal of Legal Studies 29: 797–820. Sanchirico, Chris William (2001). “Deconstructing the New Efficiency Rationale.” Cornell Law Review 86: 1003–1089. Sanchirico, Chris William (2010a). “A Critical Look at the Economic Argument for Taxing Only Labor Income.” Tax Law Review 63: 867–956. Web appendix available at . Sanchirico, Chris William (2010b). “Policy Uncertainty and Optimal Taxation.” Research Paper No. 10-23, University of Pennsylvania, Institute for Law & Economics, available at . Page 32 of 37

Optimal Redistributional Instruments in Law and Economics Sanchirico, Chris William (2011a). “Tax Eclecticism.” Tax Law Review 64: 149–228. Web appendix available at . Sanchirico, Chris William (2011b). “Optimal Tax Policy and the Symmetries of Igno­ rance.” Tax Law Review 66: 1–62. Web appendix available at . Sanchirico, Chris William (2011c). “A Counter-Reply to Professors Bankman and Weis­ bach (June 20, 2011).” Tax Law Review 64: 551–561. Sanchirico, Chris William (2013). “Good Tags, Bad Tags.” Research Paper No. 13-24, Uni­ versity of Pennsylvania, Institute for Law & Economics. Available at SSRN: . Sandmo, Agnar (1975). “Optimal Taxation in the Presence of Externalities.” Swedish Jour­ nal of Economics 77: 86–98. Shavell, Steven (1981). “A Note on Efficiency vs. Distributional Equity in Legal Rulemak­ ing: Should Distributional Equity Matter Given Optimal Income Taxation?” American Eco­ nomic Review: Papers and Proceedings 71: 414–418. Shaviro, Daniel (2007). “Beyond the Pro Consumption Tax Consensus.” Stanford Law Re­ view 60: 745–788. Slemrod, Joel (1983). “Do We Know How Progressive the Income Tax System Should Be?” National Tax Journal 36: 361–369. Slemrod, Joel (1990). “Optimal Taxation and Optimal Tax Systems.” Journal of Economic Perspectives 4(1): 157–178. Stern, Nicholas (1982). “Optimum Taxation with Errors in Administration.” Journal of Public Economics 17: 181–211. Stern, Nicholas (1987). “The Theory of Optimal Commodity and Income Taxation: An In­ troduction,” in D. Newbery and Nicholas Stern, eds., The Theory of TaXation for Develop­ ing Countries. New York: Oxford University Press, pp. 22–59. Stiglitz, Joseph (1982). “Self-Selection and Pareto Efficient Taxation.” Journal of Public Economics 17: 213–240. (p. 354)

Tuomala, Matti (1990). Optimal Income Tax and Redistribution. New York: Oxford Univer­ sity Press. Yitzhaki, Shlomo (1979). “A Note on Optimal Taxation and Administrative Costs.” Ameri­ can Economic Review 69: 475–480.

Page 33 of 37

Optimal Redistributional Instruments in Law and Economics

Notes: (1) I use this phrase throughout the chapter to refer to the framework first laid out by Mirrlees (1971) to study optimal labor earnings taxation, and later applied more general­ ly. Broadly speaking, the framework is characterized by social welfare maximization con­ strained by limited direct government observation of relevant individual traits and self-in­ terested individual behavior. See Stern (1987) for an unusually clear introduction. (2) The argument was put forward by Atkinson and Stiglitz (1976) and refined and extend­ ed by Hylland and Zeckhauser (1979), Kaplow (2004, published as 2006), and Laroque (2005). (3) The phrases “labor earnings tax system” and “labor earnings taxation” are synony­ mous in this chapter and are meant to encompass all transfers to or from individuals, on the one hand, and the government, on the other, whose direction and magnitude are a function solely of the individual’s assumed-to-be-government-observable level of labor earnings. The labor earnings tax system may involve graduated rates and regions of sub­ sidy (such as in the U.S. Earned Income Credit). Many papers employing the tax substitution argument use the term “income tax system” in place of “labor earnings tax system.” Among other differences, an income tax is typical­ ly based on capital income as well as labor earnings. A closer look at papers using the term “income tax” reveals that they are using the term as a shorthand for labor earnings taxation. Indeed some papers use the tax substitution argument specifically to argue against including capital income in the tax base. (4) In formal microeconomic parlance, the aggregate of individuals’ “equivalent varia­ tions” for implementing the policy X modification is negative. (Note that “winners” are those with positive equivalent variations, and “losers” are those with negative.) Equiva­ lently, the aggregate “compensating variation” for removing the policy X modification is positive. (5) For readers who wish to delve more deeply into this argument, it may be helpful to point out that there are four different lump sum taxes that need to be distinguished. In the order they are introduced in the text in this section 16.2.1.2.1 (Differential Commodi­ ty Taxation) these are: (1) the uniform lump sum transfer of luxury tax revenues, which is part of the policy X under consideration, as discussed in the second text paragraph of this section; (2) the hypothetical utility-equivalent lump sum scheme for the luxury tax cum retransfer of uniform luxury tax revenues, as referenced in the third text paragraph; (3) The hypothetical utility-equivalent lump sum scheme for the luxury tax taken alone, as referenced in the first sentence of the current paragraph; (4) The hypothetical revenue-equivalent lump sum scheme for the luxury tax taken alone—this is the lump sum tax considered in the current text sentence and used to determine the size of (3) later in the current para­ graph.

Page 34 of 37

Optimal Redistributional Instruments in Law and Economics (6) The importance of the adjective “differential” is described in section 16.2.2.3, which discusses the mechanical equivalence, in certain models, of uniformly taxing commodities (more precisely, expenditures on commodities) and taxing labor earnings. (7) Saez (2002), Bankman and Weisbach (2006, 2007), Shaviro (2007), Sanchirico (2010a, 2011a, 2011b), Bankman and Weisbach (2011), and Sanchirico (2011c) consider capital income taxation in light of applications of the tax substitution argument to differential commodity taxation. (8) Deaton (1979) contains a material error that is corrected in Deaton and Stern (1986). (9) More precisely, item (1) would be calculated holding everyone else’s actions constant at the levels they choose before the regulation is removed, and item (2) would be calculat­ ed holding the actor’s own action constant at the level she chooses after the regulation is removed. The ordering could also be reversed. (10) Lipsey and Lancaster (1956–57) is the seminal paper on the “Theory of the Second Best.” The central lesson of this literature is that eliminating some distortions does not necessarily improve welfare when other distortions remain. A terminological note: the phrase “second best” without “theory of” affixed to it is often used to refer to the afore­ mentioned informational constraints. (11) Assume, additionally, that after the swap, no other policy—besides labor earnings tax­ ation—whose revenue implications are behaviorally responsive remains in place. That is, after the swap is effected, were individuals’ labor earnings held hypothetically constant, government net revenue would not be affected by behavioral response. The assumption is trivially fulfilled if the labor earnings tax schedule is the only policy in place besides possibly the policy X modification and lump sum schemes. In some applica­ tions this is a helpful simplifying assumption, and the reader may wish to adopt it here. In other settings, however, it is not possible to make this assumption. For instance, in evaluating distributionally motivated deviations from externality-correcting Pigovian tax­ es (per section 16.2.1.2.2.2), the starting point for policy is by definition the efficient Pigovian tax. This tax will generally have revenue implications that are behaviorally sensi­ tive. In such cases, however, one can adopt the slightly more complicated assumption that the problematic initial policy, such as the efficient Pigovian tax, is automatically coupled with some transfer-back or other use of all the revenue it collects, so that the impact of behavioral adjustments on tax revenues are automatically offset within the expanded sphere of the policy, and the global revenue impact is nil. In some applications this cou­ pling is natural in light of existing institutions: for example, the externality-correcting “tax” collected from losing tort defendants in the form of damages is immediately paid out as compensation to plaintiffs. The assumption described in this note is distinct from the tax substitution assumption highlighted in the text. It also has different implications for the viability of the argument. The assumption can always be fulfilled by sufficiently widening the scope of policy X. Page 35 of 37

Optimal Redistributional Instruments in Law and Economics Thus, the assumption’s necessity signifies that the tax substitution argument must be ap­ plied at once to all distributionally motivated policy adjustments outside the labor earn­ ings tax system. The argument cannot not be applied sequentially to each such adjust­ ment. The endpoint—redistribution solely through labor earnings taxation—is the same. (12) This assumes that individuals with different p values choose different labor earnings, which will generally be the case in these models. It also maintains the conventional as­ sumption that the government knows the population distribution of all relevant parame­ ters and functions, as discussed in section 16.2.1.3.1. (13) The existence of this argument explains why the conventional optimal tax analyses hold the taxation of one potential taxable attribute fixed at zero, designating this at­ tribute the “numeraire.” Doing so eliminates indeterminacies caused by these equiva­ lences. (14) Recently Gamage (2014, 2015) has provided, in effect, a new critique of distortion counting. He emphasizes that when administration and compliance are taken into ac­ count other distortionary margins are added to the analysis and the very enterprise of counting becomes problematic. (15) See note 3. (16) This may entail pre-solving some suboptimization problems, as when a given taxable attribute is determined by more than one configuration of direct utility sources. (17) The analysis in this section is based on the framework laid out in Sanchirico (2011a), which is in turn merely a particular combination/rearrangement of optimal-tax first-order conditions. The web appendix to Sanchirico (2011a) discusses the precise relationship to such first-order conditions—which are in effect the recipe-like manifestation of what is es­ sentially an arbitrage argument. (18) The revenue price of redistribution for the labor earnings tax will also be positive, given the hypothesis that the labor earnings tax system is optimal (conditional on its ex­ clusivity). (19) In fact, “tag” is not always defined to be immutable in the literature. I adopt this defi­ nition here. Behaviorally responsive taxable attributes are considered in the next section. (20) A paper by De Geest (2013) argues for the superiority of “legal rule” redistribution in the presence of economic rents (i.e. pure profits). De Geest argues that certain legal rules are better than labor earnings taxes at targeting the deleterious efficiency and equity ef­ fects of such rents.

Chris William Sanchirico

Page 36 of 37

Optimal Redistributional Instruments in Law and Economics Chris William Sanchirico, Samuel A. Blank Professor of Law, Business, and Public Policy; Co-Director, Center for Tax Law and Policy, University of Pennsylvania Law School.

Page 37 of 37

Cost–Benefit Analysis in Legal Decision-Making

Cost–Benefit Analysis in Legal Decision-Making   Richard O. Zerbe Jr. The Oxford Handbook of Law and Economics: Volume 1: Methodology and Concepts Edited by Francesco Parisi Print Publication Date: Apr 2017 Subject: Economics and Finance, Law and Economics, Econometrics, Experimental and Quanti­ tative Methods Online Publication Date: May 2017 DOI: 10.1093/oxfordhb/9780199684267.013.027

Abstract and Keywords Benefit-cost analysis (BCA), or cost-benefit analysis, is important in policy and law. This article introduces the nature and history of BCA to provide an understanding of the devel­ opment of the benefit-cost concepts, objections to the concepts, and their actual use in le­ gal and economic practice. The term ‘benefit-cost’ is used to differentiate from the term ‘cost-benefit’ used by engineers whose approach is more mechanical than terms of effi­ ciency used in law and economics. BCA is seen as a useful tool with some positive predic­ tive ability in determining judge's decisions. It also appears to contribute to greater effi­ ciency in government investment spending. Keywords: benefit-cost analysis, law, economics, legal practice, economic practice

17.1 The Legal Landscape BENEFIT–COST analysis (BCA) or cost–benefit analysis is important in policy and law. A LexisNexis Academic search finds the following results, arranged by decade, for legal cases that mention cost–benefit analysis or benefit–cost analysis.1 The salient points are that there were no references in cases before the decade of the 1960s; the jump from the 1960s to the 1970s was over 6000%; the next-largest jump was from the first decade of this century to the second decade, assuming a pro-rata extension of the number of cases in the first two years of the second decade (see Table 17.1). This corroborates my sense of the substantial growth of the use of BCA. BCA appears not to be a trivial matter in the case law at least judged by frequency of mention, but rather a growing presence. A similar growth pattern is shown for articles in law reviews and relat­ ed journals as shown in Table 17.2.2 A HeinOnline search of their law journal

(p. 356)

li­

brary shows 211 articles with “cost–benefit analysis” or “benefit–cost analysis” in the ti­

Page 1 of 32

Cost–Benefit Analysis in Legal Decision-Making tle. This includes traditional law reviews as well as some additional journals such as Reg­ ulation and the Natural Resources Journal. Table 17.1 Mention of “Benefit–Cost” or “Cost Benefit” in Legal Cases Dates by Decade

References in Federal Cases Search: “cost benefit analysis” or “benefit cost analysis” everywhere

References in State Cases Search: “cost benefit analysis” or “benefit cost analysis” everywhere

2011– 2013

259 (1295 est. for decade)

130 (650 est. for decade)

2001– 2010

635

298

1991–

393

181

1981– 1990

261

115

1971– 1980

136

21

1961– 1970

2

1

1951– 1960

0

0

1901– 1950

0

0

2000

Table 17.2 Law Review and Related Journals: Number of Articles with the Term “Cost Benefit” or “Benefit Cost” Date

Number of Articles

2001–2010

12645

1991–2000

8775

Page 2 of 32

Cost–Benefit Analysis in Legal Decision-Making 1981–1990

6324

1971–1980

3463

1961–1970

540

1951–1960

43

1941–1950

11

The earliest law review article using the terms benefit–cost is from 1950 (Harding, 1950, p. 547), and it states: By standards of feasibility is meant the ratio of the benefits produced by a product to the costs of producing such benefits. To be feasible the benefit cost ratio must exceed unity.” (Emphasis added.) It is notable that the earlier use of the term was benefit cost analysis, not cost–benefit. In the same year Arthur Maass, the economist most responsible for the early promotion of (p. 357) BCA, used the following expression, inter alia: “[T]he Corp gives principle empha­ sis to the cost–benefit ratio of the individual water projects which the Congress has re­ quested it to investigate” (Maass, 1950). In what follows I introduce the nature and history of benefit–cost analysis, BCA, or cost– benefit analysis, CBA, to use the more common term. The aim is to give an understanding of the development of the benefit–cost concepts, objections to the concepts, and their ac­ tual use in legal and economic practice. I use the term benefit–cost to differentiate the term cost–benefit used by engineers, whose approach is more mechanical than the terms of efficiency used in law and economics.

17.2 Benefit–Cost Analysis Defined BCA and cost–benefit analysis (CBA), which are generally regarded as equivalent terms, are accounting frameworks used to evaluate the consequences of government decisions. In this sense, BCA is similar to analyses that corporations conduct in order to evaluate in­ vestment decisions. However, it differs in two significant ways: first, the objective of BCA is to increase public welfare, not profit. This serves to increase both the technical and po­ litical complexity of the exercise, because the exact definition of “public welfare” is elu­ sive. Second, because BCA often encompasses non-market goods, such as the value of a recreational visit or wildlife viewing at a park, the required data is often not found in market transactions. Such data must instead be generated through resource-intensive surveys or other statistical analyses. For instance, for the value of a park that has no en­ try fee, analysts might instead estimate the value of a recreational visit by asking park

Page 3 of 32

Cost–Benefit Analysis in Legal Decision-Making users how much they would be willing to pay to use the park or determine the costs visi­ tors incur to travel to the park.3 The idea is simple enough. Gains or benefits are measured against costs to determine whether or not a project is worthwhile for those given economic standing. The need to de­ termine standing arises as no benefit–cost analysis can include all values. In practice, a BCA does not include everyone affected by a policy alternative. Most analyses are done in a particular context that determines standing. For instance, a municipality might only consider effects on municipal revenues or limit effects to those within its borders. The state government might not consider effects of its pollution policy on neighboring states and the federal government might not consider effects on foreign countries. Many projects will have only minor effects on certain groups and it may be considered too ex­ pensive to include these effects. The standing decision will be affected by the choice between a partial equilibrium analy­ sis and a general equilibrium analysis. When the number of markets included in (p. 358) the BCA is highly restricted, the analysis is called a “partial equilibrium” analysis; when the number of markets included is large, it is called a “general equilibrium” analysis. A small town that conducts a BCA to determine whether to purchase a garbage truck or continue to outsource garbage collection would use a partial equilibrium analysis. On the other hand, a national oil tax will affect many markets: transportation markets, plastic goods and other petroleum product markets, labor markets, and so forth. When the ef­ fects in several markets are taken into account, the analysis is said to be a general equi­ librium analysis. An analysis of a national oil tax should be done from a general equilibri­ um perspective in which these major markets and their interactions are considered. Benefit and costs are determined in principle by the measure developed by J. Hicks (1939) known as the “compensating variation” (CV). In practice the measures are usually “consumer and producer surpluses” which are approximations of the Hicksian measure. Consumer surplus is approximately the amount one would pay minus the amount actually paid. Thus if a price is a cup of coffee is $1.00 and the willingness to pay for it is $2.50 (absent other vendors), the consumer surplus is $1.50. Producer surplus is the economic rent or the amount that could be taken away without affecting supply. For example, con­ sidered as a whole, most of the wages of college football coaches is economic rent since, were their salaries lowered en masse, they would be unlikely to leave for other jobs. The measures correspond to willingness to pay (WTP) for gains and willingness to accept (WTA) payment to bear costs, which measures are consistent with the psychological (utili­ ty) effects associated with prospect theory, which is steeper for losses than gains, convex in losses, and concave in gains (Tversky and Kahneman, 1992). They thus have an intu­ itive appeal and are consistent with the goal of BCA, which is to give those same values markets would give were prices available. The differences in values between the WTP and the WTA payment can be large, and are larger the more expensive the good (income ef­ fects) and the more unique the good (substitution effects). The differences would be very large, then, in estimating the value of the Grand Canyon, which is unique and also highly valuable, or in determining the value of a right to an abortion. For goods of less value or Page 4 of 32

Cost–Benefit Analysis in Legal Decision-Making uniqueness whose value is captured by markets, the WTP and WTA values are the same for small changes in quantities. In practice both benefits and costs are often measured solely by the WTP on grounds that the WTA is more difficult to measure. In general prac­ tice the WTP is usually approximated by the change in consumer or producer surpluses. This can be an important source of error for goods in which the discrepancy between the WTA and WTP measures are large. The relation of benefits and costs to the WTP and the WTA is summarized in Table 17.3. The basic BCA test has been the Kaldor–Hicks (KH) test, which uses the sum of the CVs. This test sums benefits as measured by WTP for them and subtracts from them the sum of costs, which is the WTA compensating payment for losses. KH is satisfied when the gains are sufficient to hypothetically compensate the losses so that it is also referred to as the “potential compensation test” (PCT).4 The economic worth of a good to an (p. 359) individ­ ual is determined by the intensity of desire for it and the income and wealth of the per­ son. These features are all captured by the WTP and WTA measures. Willingness to pay is limited by income. WTA is potentially unlimited. In general the WTP is less than the WTA for a good. These measures attempt to duplicate what market results would be were there a market for the particular project. Table 17.3 The Measurement of Benefits and Costs in Terms of Gains and Losses The Compensating Variation (KH Measure) Ben­ efits

GAIN: WTP—the sum of CVs for a positive change—is finite. LOSS RESTORED: WTA—the sum of CVs for a loss restored—could be infinite.

Costs

LOSS: WTA—the sum of CVs for a negative change—could be infinite. GAIN FORGONE: depends on legal status of right to gain; where no legal right to gain is WTP.

(*) CVs are the compensating variations which measure movements to a new indiffer­ ence curve by price tangencies at the new price. They are distinguished from the EVs, equivalent variations which measure differences at the original price. Law becomes relevant in determining the base for deciding what is a gain or a loss. Gains and losses are psychological states that are in part determined by law. Gains are normally defined from a position that recognizes legal rights; gains occur when a new right (or good) is obtained. Gains are measured by the WTP for them and costs, measured by the WTA, that is the compensation required to accept the loss. This is based on the assump­ tion that law is the determinant of a psychological reference point from which one experi­ ences a gain or loss. Where the right is divisible, such as in deciding how to divide a piece of property, it is BCA efficient for the right to go to the person to whom it would go if Page 5 of 32

Cost–Benefit Analysis in Legal Decision-Making transactions costs were zero (Posner, 1972, p.18) That is, as long as person A has a WTP that exceeds person B’s WTA, the right would go to A and vice versa. In dividing a piece of land, then, some might go to A and some to B determined by their respective WTP and WTA. This would in general leave some of the land unclaimed, as neither individual’s WTP would exceed the other’s WTA. The allocation of this remainder would then be deter­ mined by auction—by the WTP, as any additional land would be a gain for either party, which gain is to be measured by the WTP. BCA evaluates policies that have benefits and costs that will normally occur both at project outset and in the future, that is, over time. Because benefits and costs are cash flows that occur over time, the analyst must take the “time value of money” into account (i.e. the idea that $100 today is not worth the same as $100 a year from now because in­ vesting the $100 earned today would yield more a year from now). To make the money value of costs and benefits commensurate over time, cash flows in each year must be dis­ counted to their “present value.” The present value of a future sum is the amount that, if invested today, at the discount rate, would yield the expected cash flows. The (p. 360) “net present value” is the common BCA test criterion and is the present value of benefits mi­ nus the present value of costs.5 It is given by:

where: Bt = benefits at time t; Ct = costs at time t; r = discount rate; t = each year; and T = the number of future years (or other time units). Table 17.4 Present Values of One Million 30 years

100 years

Present Value at a 2% discount rate

$552,070.89

$13,803.30

Present Values at a 7% discount rate

$131,367.12

$115.25

The effect of the discount rate on present values is very powerful. Consider Table 17.4, which gives the present values of one million dollars using a discount rate of 2% and 7% respectively and the million dollars in value appearing in thirty or 100 years respectively. When conducting a BCA, all monetary amounts must be in comparable units—either all in “constant dollars” or all in “nominal dollars.” Constant dollars (also called “real dollars”) take inflation into account, adjusting the value of future benefits and costs to reflect ex­ pected inflation. If constant dollars are chosen, then the inflation component must be sub­ tracted to the discount rate. Thus, if the market interest rate is 8% and expected inflation is 5%, the real interest rate (by which future benefits and costs are discounted) would be

Page 6 of 32

Cost–Benefit Analysis in Legal Decision-Making about 8% -5%, or 3%. These considerations and others are frequently capsulated in steps for a BCA. A typical but somewhat abbreviated statement of steps is shown in Table 17.5.

17.3 Economic Origins of BCA: the Develop­ ment of Kaldor–Hicks Kaldor–Hicks is the standard for BCA.6 It arose during the late 1930s out of discussions among prominent British economists about repealing the Corn Laws (Kaldor, 1939; Har­ rod, 1938; Robbins, 1938; Hicks, 1939). Before that time it was generally assumed that (p. 361) each individual had an “equal capacity for enjoyment,” and that gains and losses among different individuals could be directly compared.7 By 1939, however, leading British economists, including the future Nobel Prize winner Sir John Hicks, were raising questions about such policy prescriptions because they involved interpersonal compar­ isons of utility (Hicks, 1939).8 Kaldor provided a solution: he acknowledged the inability of economists to establish a scientific basis for making interpersonal comparisons of utili­ ty but suggested that this difficulty could be made irrelevant (Kaldor, 1939). He argued that policies that led to an increase in aggregate real income are always desirable be­ cause the potential exists to make everyone better off: [T]he economist’s case for the policy is quite unaffected by the question of the comparability of individual satisfaction, since in all such cases it is possible to make everybody better off than before, or at any rate to make some people better off without making anybody worse off. (Kaldor, 1939, pp. 549–550) Table 17.5 Basic Steps in Benefit–Cost Analysis 1) Define the scope and assumptions of the BCA 2) Identify outcomes and quantify costs and benefits for each alternative 3) Choose a discount rate and calculate the present value of costs and benefits 4) Choose a measure for comparing alternatives and carry out the necessary calcula­ tions 5) Discuss and treat risk and uncertainty According to Kaldor, a project is desirable if the money measure of gains exceeds the money measure of losses. With regard to the potential compensation that could turn losers into winners in such situations, Kaldor notes that whether actual compensation Page 7 of 32

Cost–Benefit Analysis in Legal Decision-Making should take place “is a political question on which the economist, qua economist, could hardly pronounce an opinion” (Kaldor, 1939).9 Hicks, perhaps the most prominent (p. 362) economist of the time, accepted the Kaldor approach, which eventually became known as the KH criterion (Hicks, 1939, 712). KH attempts to avoid interpersonal utility comparisons by separating equity from efficien­ cy. Kaldor proposed that decision-makers address ethical values regarding equity outside the purview of BCA.10 The change in aggregate gains was to be the measure of efficiency, so according to KH there is a separation of efficiency and distributional effects (Kaldor, 1939, 551). Kaldor endorsed the procedure adopted by Pigou, which Kaldor describes as “dividing welfare effects into two parts: the first relating to production, and the second to distribution.”11 The KH approach produces outcomes that are equivalent to those pro­ duced by the assumption that the marginal utility of income is the same across all individ­ uals, that is, that each dollar of benefit or cost is treated the same regardless of who re­ ceived it (Kaldor, 1939). Hicks agreed also with this separation and noted that “if mea­ sures making for efficiency are to have a fair chance, it is extremely desirable that they should be freed from distributive complication as much as possible” (Hicks, 1939, p. 697). A full description of the KH assumptions is reasonably characterized by: 1) the use of WTP for gains and for losses;12 (2) a reliance on potential compensation tests so that a project is KH efficient only when winners could hypothetically compensate losers, i.e. it passes a potential compensation test (PCT); (3) an emphasis on efficiency that is separat­ ed from equity; (4) an assumption that a dollar is to be treated the same regardless of who receives it, so that a dollar is assumed to have the same value to each person (equal and constant marginal utility of income); (5) a recognition and inclusion of non-pecuniary effects; (6) the omission of values represented by ethical considerations; (7) a reliance on externalities and market failure to determine where BCA might be useful in making cor­ rections; (8) an assumption that transactions costs are zero;13 (9) the treatment of BCA as a mechanism to provide the answer rather than as an approach providing information as part of an ongoing discussion.

(p. 363)

17.4 History of BCA Use

Originally BCA was conceived as simply a measure to determine whether a water project, typically a dam, should be built.14 The Army Corps of Engineers introduced benefit–cost methods into the United States (borrowing from the French) at least as early as the Rivers and Harbor Act of 1902, and its use was explicitly mandated in the 1920 amend­ ment to the Act (Porter, 1995, p. 150; Hammond, 1966, p. 195; Holmes, 1972).15 Before the creation of the Corps, evaluations of public investments were almost completely ad hoc (Porter, 1995, p. 150). By the 1920s, the Corps required that its recommended projects have expected benefits in excess of costs. Throughout the 1930s, the numbers put forward by the Corps were generally accepted without question (Porter, 1995, p. 150). Congress recognized the Corps as a relatively neutral and respected arbiter in con­ gressional fights over water projects (Porter, 1995, p. 153). The creation of the Corps, Page 8 of 32

Cost–Benefit Analysis in Legal Decision-Making then, represented not only the creation of an agency to build projects, but an agency to increase congressional and public efficiency. After 1940 Corps decisions became the subject of rather bitter controversy as the Corps was challenged first by powerful electric and railroad utilities, by shipping interests, and then by rival federal agencies, especially the Bureau of Reclamation and the Department of Agriculture (Porter, 1995, pp. 161–175). The further development of BCA and its in­ creasing quantification was not the product of technical elites but of disagreement, suspi­ cion, and conflict, particularly bureaucratic conflict (Porter, 1995). Rival techniques or standards for BCA became the norm, although an attempt was made to resolve differ­ ences by relying on first principles of economics. The attempt closest to reaching agree­ ment was the “Green Book.”16 Although agreement was significantly incomplete, the grounds for decision-making were reasonably well established as rooted in economic the­ ory. BCA was thereby transformed by conflict into a set of rationalized economic princi­ ples building on work by British economists in the late 1930s. The incorporation of economic principles into BCA “began in earnest in the mid-1950’s” (Porter, 1995, p. 188). Economists agreed with budget officials that the stan­ dards for passing a benefit–cost test had not been set strictly enough. In particular, the use of BCA at the federal level increased in 1981 when President Ronald Reagan issued an Executive Order declaring that Regulatory Impact Analyses (RIA) be conducted for major initiatives. Executive Order (EO) 12,866, issued in 1993 by President Bill Clinton, (p. 364) required executive-branch agencies, including the Environmental Protection Agency (EPA), to prepare cost–benefit analyses for any regulatory proposals having annu­ al economic effects of $100 million or more. Subsequently President Clinton issued an ad­ ditional Executive Order in 1994, confirming the government’s commitment to BCA and highlighting the bipartisan support for BCA in federal regulatory decision-making (Exec. Order No. 12,866, 3 C.F.R. 638, 1993–2000, reprinted as 5 U.S.C. § 601, 1994). The 1995 Unfunded Mandates Reform Act requires federal agencies to conduct cost–benefit analy­ ses of all regulations entailing annual economic costs of $100 million or more. The Act al­ so requires the agencies to adopt the economically least burdensome regulatory alterna­ tive that accomplishes the regulatory purpose or to explain why they chose a different op­ tion. While comprehensive legislation requiring the broader use of formal BCA has yet to be approved by Congress, the presence of BCA is nonetheless apparent within various levels of governmental decision-making (Zerbe, 2007; Graham, 2008). The Clinton E. O. did not require that the benefits of proposed regulations exceed the costs. It did, however, re­ quire the EPA to prepare CBAs, even when it could not legally consider them in making regulatory decisions, as when setting national ambient air-quality standards under the Clean Air Act (Cole, p. 72). Since 2000 BCA has expanded in its use and in the attention paid to it. In 2001 the Coali­ tion for Evidence Based Policy was started,17 in 2007 the Society for Benefit–Cost Analy­ sis was formed, and in 2009 the Journal of Benefit–Cost Analysis was created. A set of Page 9 of 32

Cost–Benefit Analysis in Legal Decision-Making Principles and Standards (see ), funded by the MacArthur Foundation, was produced in 2010 and interest in ex­ panding this effort is shown by proposals from the National Science Foundation and the National Institutes of Health. In addition, centers for BCA study have been formed at the University of Washington’s Daniel J. Evans School of Public Affairs and at the New York University School of Law, and perhaps elsewhere.

17.5 Expanded Kaldor–Hicks It is reasonably clear that KH as historically constituted is not likely to remain the main­ stream view.18 The modern view contains KH as a limiting case (Adler and Posner, 2001; Adler, 2003; Sunstein, 2001; Thaler and Sunstein, 2009; Knetsch, 1995; Zerbe, 2007). Zerbe and Scott (2014) call the new view the Expanded Kaldor–Hicks test (EKH). The (p. 365) new view can be reasonably characterized by the following assumptions: (1) the use of the Pareto requirement itself furnishes a justification for the use of BCA so that (2) the PCT test can be dropped and replaced with the simple requirement that the net present value of a project be positive, where (3) the definition of benefits and losses are grounded in law and in the WTP and WTA; where (4) there is recognition that all goods for which there is a WTP are to be included in principle in any BCA analysis, so that EKH adds to KH the requirements that all ethical values for which there is a WTP or WTA are included in the analysis, including those concerning distributional and ethical considera­ tions; (5) the understanding that proper use of ethical BCA is to furnish information and predictions and not to furnish the decision; and (6) that transactions cost economics rather than market failure is the basis for a justification that government intervention might be useful. The rationale for the judgment that the Pareto approach itself applies directly to a justifi­ cation for EKH is found in a demonstration by Zerbe and Scott (2014), showing that un­ der reasonable and even somewhat extreme assumptions biased against their hypotheses, almost all individuals gain from the use of BCA across a portfolio of projects. The striking feature of these results is how few are the number of projects to produce the “almost all win” results. Thus when considered as a decision rule, EKH shows that it is likely that al­ most all gain from using BCA and that losers arise from the tax structure. Zerbe and Scott (2014) introduce a better-grounded version of Kaldor–Hicks itself, EKH, based directly on Pareto superiority that includes all economic goods, including moral sentiments, for which there is a willingness to pay. EKH adopts a portfolio perspective and is built from a BCA rule for individual projects similar to the PCT but without its excess baggage. This project acceptance rule under EKH is: BCA Project Rule: adopt all projects for which the WTP exceeds the WTA, where all goods for which there is a WTP are economic goods. This approach differs from the PCT in five respects. First, it considers a portfolio perspec­ tive as well as a project perspective. Second, the Expanded KH rule (or EKH) includes all goods, including equity goods, for which there is a WTP. Third, hypothetical compensation Page 10 of 32

Cost–Benefit Analysis in Legal Decision-Making is not required; the rule finds its justification in actual, not hypothetical, compensation. Fourth, the rule recognizes that net gains as measured by the Kaldor–Hicks test does not mean that a project will pass the PCT, even if compensation is hypothetically costless (Broadway and Bruce, 1984). Fifth, such a rule, absent the PCT, is not subject to reversals such as Scitovsky’s (1941). This view abandons the PCT, includes moral sentiments and ethical values as their values are reflected in WTA. This has the following advantages. By dropping the PCT, technical criticisms such as Scitovsky reversals and Baker’s criticism are obviated. By including moral and ethical values, the scope of BCA is legitimately ex­ tended in a manner that is consistent with economic theory. By recognizing the grounding of BCA in law, it recognizes the bases for the differing use of WTP and WTA. By regarding BCA as furnishing information to the decision process, it recognizes the necessary limita­ tions by any normative standard.

(p. 366)

17.6 Issues with BCA

There are three issues concerning the use of BCA that have been especially controversial. These are (1) what the discount rate should be, (2) the possibility of Scitovsky reversals, and (3) the role of moral sentiments in BCA. In addition, an issue about the inability to use BCA in legal cases deserves mention here.

17.6.1 Moral Sentiments and BCA This separation of efficiency and equity has remained the common, though not universal, basis of normative economic analysis almost to this day.19 The most widespread criticisms of BCA rest on missing values, such as concern for distributional changes, care for the welfare of others, the value of life, and the like. With respect to distributional changes, the modern justification might be found in the notion that changes in the income distribu­ tion are usually better effected through macroeconomic and tax policy rather than through individual projects (Polinsky, 1989). This is a practical or expedient reason in practice but not as a matter of principle. The best formal expression of this expediency is by Shavell (1981). He notes that there are two crucial substitutes for distributional re­ sults embedded in a BCA project, the income tax and the distributional effects of other projects. Shavell (1981) and Kaplow and Shavell (1994) speak of an efficient legal system state, noting that any regime with an inefficient legal rule can be replaced by a regime with an efficient legal rule and a modified tax system designed so that every person is made better off. The idea is that a change in a legal rule to accomplish a distributional change will induce the same distortion as an income tax change with equivalent distribu­ tional effects, plus the additional distortion of an inefficient legal rule. But the distribu­ tional effects cannot, contrary to some claims, be ignored altogether. This would claim too much, as Shavell (1981) points out: Now of course, no one would really expect the income tax structure to be adjusted in response to each and every change in legal rules (much less to individual changes in other domains), for this would be impractical. Therefore, one’s attitude Page 11 of 32

Cost–Benefit Analysis in Legal Decision-Making toward the result under discussion will depend on his expectation that the income tax would be or could be altered in response to changes in legal rules whenever these changes resulted in a “sufficiently important” shift in the distribution of in­ come. Shavell’s proof assumes that the correct adjustment for distributional effects would be made. This may not be the case even for changes that are important as he rec­ ognizes as shown by Zerbe (2014). Moreover, this defense leaves unaddressed matters of equities for identifiable peoples or groups, or ethical values attached to particular projects that cannot be handled by macroeconomic policy; equity and justice are particu­ lar as well as general.20 (p. 367)

The missing values criticism arises from the fact that historically BCA avoided moral val­ ues as a matter of expediency. Missing values are not intrinsic to BCA. They disappear when applied to a modestly expanded KH that recognizes moral sentiments. This is dis­ cussed at length by Zerbe (2007). However, other criticisms resist the notion that values should be measured by the willing­ ness to pay for them. One expression focuses on the fact that money is not the currency for many values, for example friendship or love or moral responsibility. This is true, but money is used in BCA not exactly as a measure of value but simply as a means of ranking preferences. This reduces the power of this criticism. Some critics hold certain values to be priceless. This implies that no trade-off for them is possible so that no amount of the priceless good is worth giving up by a WTA measure to gain any amount of some other good—that the value of certain goods is infinite. There are few goods for which this is the case. The trenchant response, however, is that BCA is a practical tool for assessing pref­ erences; it does not, or should not, claim to be the language in which the most fundamen­ tal moral values should be wholly discussed. The absence of any perfect rule for decisions suggests that imperfect rules should be improved and their defects made apparent, not that they should be abandoned. Most of the critical analyses, however, are concerned not with improving BCA but with killing it. As Robert Verchick (2005) notes, “Ackerman and Heinzerling make clear from the beginning that they have come to bury BCA, not to save it … ” The criticisms are gen­ erally devoid of empirical analysis and, where they are not, the empirical analysis is high­ ly suspect. In general the criticisms are weak on offering alternatives to BCA. These criti­ cisms have been explicated and addressed by Zerbe (2007). A sense of the nature of the non-technical criticisms can be gained from the following fairly recent quote from Shapiro and Schroeder (2008, p. 434): CBA has become a one-size fits all technique applied to policy problems as varied as regulating mercury emissions from power plants to the roof strength for new (p. 368) automobiles. Its foundation rests on a positivist approach to knowledge, facts can be pursed independently of values, only information subject to empirical verification counts as fact, and the goal of policy research is to discover universal laws that can then be applied to all policy problems-that have been discredited in Page 12 of 32

Cost–Benefit Analysis in Legal Decision-Making a wide-ranging literature … These criticisms, however, have remained a dissenting tradition, as CBA has only strengthened its dominance in the past twenty-five years [emphasis added]. The authors picture a commitment to discussion and de­ bate as the solution which would be fine in “le meilleur de tous les mondes possi­ bles, but perhaps not in the real world.” They note: The unifying theme of this diverse (Lasswellian) literature has been its commit­ ment to broadening rather than narrowing the “theories, issues and processes” relevant to public decisionmaking, as well as its aspiration to an analytic process that is “problem-oriented, contextual, eclectic, and process-sensitive.” These sorts of criticisms, of which there are many, contemplate solving collective actions problems by structured discussion. The suggestion is not convincing as a substitute for BCA absent a showing that such a discussion would work well and especially as it re­ mains quite unclear why the use of BCA should not be part of the discussion process as indeed it is in many real cases. Such structured discussions are not a substitute for BCA but rather a structure for making use of it.

17.6.2 Scitovsky Reversals There have also been strong statements of BCA inadequacy on technical grounds, mainly on the possibility of Scitovsky reversals. The Scitovsky reversal paradox arises when, starting from state of the world A, position B appears superior to A, but when starting from B, A appears superior to B. However, reversal paradoxes such as Scitovsky’s are purely creatures of the PCT. They arise from the PCT assumption that compensation is costless and that potential compensation is the correct measure. The legal philosopher Coleman (1980, pp. 519f) uses an example of reversibility to argue that the Kaldor–Hicks potential compensation test is not a useful criterion for decision-making. Table 17.6 outlines Coleman’s hypothetical example, demonstrating the before-project and afterproject states of the world. Both Mr. 1 and Ms. 2 are assumed to prefer having one unit each of good X and good Y to having two units of either good. The situations before and after the project are shown in Table 17.6. Note that the project production possibility frontier (PPF) transforms one unit of X into one unit of Y. The proposed project passes the KH test as, in the new state of the world, Ms. 2 could give one unit of Y to Mr. 1, leaving him better off with one unit of X and one unit of Y, and Ms. 2 no worse off, having one unit of Y as in the original state. However, in the status quo situation Mr. 1 could give one unit of X to Ms. 2, leaving her better off than she would be after the proposed project and Mr. 1 no worse off than he would be following the pro­ posed project. This example does not, in fact, work (1) when compensation is costly, or (2) when both situations are initially first best; (3) when compensation is actually paid so ei­ ther Mr. 1 or Ms. 2 loses in moving between states A and B; and usually reversals cannot occur (4) when goods X and Y are normal goods. The Coleman (p. 369) example does not characterize a practical case of potential reversals. The assumption of costless compensa­ tion is consistent with the standard KH approach, but it loses salience in the real world. Page 13 of 32

Cost–Benefit Analysis in Legal Decision-Making Without costless compensation a reversal may not occur (e.g. if having one unit of each good in the example is not preferred over two units of one good by more than the cost of compensation). A complete explanation is furnished by a series of papers by Schmitz and Zerbe (2009) and by Just, Schmitz, and Zerbe (2012).21

Page 14 of 32

Cost–Benefit Analysis in Legal Decision-Making Table 17.6 Coleman’s Preference Reversal Example Status Quo (State A)

Good X

Proposed Project (State B) (without compen­ sation) Good Y

Good X

Good Y

Mr. 1

2

0

1

0

Ms. 2

0

1

0

2

Page 15 of 32

Cost–Benefit Analysis in Legal Decision-Making Critics of BCA have relied on the paradox to suggest abandoning BCA. The Scitovsky re­ versal paradox (Scitovsky, 1941) is often invoked as a reason to drop the PCT. For exam­ ple, in a recent book, Markovits (2008) writes, This Scitovsky Paradox invalidates the Kaldor–Hicks test because it implies that, if the test were accurate and a Scitovsky paradox arose, both the policy and its re­ versal would be economically efficient and, hence, the policy would simultaneous­ ly be economically efficient and economically inefficient. (p. 53) Coleman (1980) offers the same sentiment and an example that relies on the PCT. Even in an article advocating for the usage of benefit–cost analysis, Adler and Posner (1999) write “even if the reversal will not occur, its possibility haunts the entire project of CBA [BCA]” (p. 186). Just et al. (2011, 2013), however, have shown that the conditions un­ der which reversal might occur are quite unlikely. More importantly, Scitovsky reversals are solely a feature of the PCT, which is eliminated by the Expanded Kaldor–Hicks test that I recommend below, so that to drop the PCT is to eliminate the possibility of such re­ versals (Schmitz and Zerbe, 2009). (p. 370)

17.6.3 Baker’s Critique Baker cleverly points out a practical objection to the PCT that applies to the failure of the PCT to provide guidance in legal cases where potential compensation is not possible. Since common law is held to rest, at least in significant part, on attempts by judges to adopt economically efficient legal rules, this is potentially a major failing. Considerable literature holds that BCA is a mainstay of common law in the sense that judges use BCA reasoning in making new law (Adler and Posner, 1999, p. 186). Baker (1980, p. 939) points out that in lawsuits it is the usual case that the sum of parties’ expectations re­ garding the ownership of a legal right (or good) exceeds the actual value of the legal right, so that the PCT test cannot be satisfied regardless of who wins the case. Baker con­ siders a situation in which A and B each believe with, say, 75% probability that the value of the property (right) is theirs. Thus the value of the good to each over the status quo is $100 − $75, or $25 dollars. If the value of the good in question is $100, and goes to per­ son A at law, person A gains $25 and person B loses $75 and vice versa, so that no matter who wins, the other cannot even potentially be compensated. Baker concludes on this ba­ sis that the use of the PCT criterion is not useful for legal analysis in such cases. Baker’s analysis is incorrect. With no ownership, the net loss is $150 ($75 + $75) as com­ pared with the loss of $50 ($75 − $25) under assignment of the right, in this case to ei­ ther party so that legal assignment is efficient. The winners by this are the members of the society whom the legal system serves.

17.6.4 The Discount Rate Perhaps the most contentious technical (or perhaps semi-technical) issue in BCA is the question of what discount rate to use. The rate is important as it materially affects the size of the public sector and individual welfare (e.g. Moore et al., 2004; Palmon, 1992; An­ Page 16 of 32

Cost–Benefit Analysis in Legal Decision-Making dreoni and Sprenger, 2012). The discount rate is the easiest and most common parameter used to influence the outcome of the BCA. Consider again the example in Table 17.4 of the difference between using a rate of 2% and 7% for the present value of a one million dollar benefit or cost incurred either thirty years or 100 years in the future. (p. 371) The differences are large and increasing the further out the benefits or costs are received. Since benefits are normally felt in the future and cost in the present, it is easy to see why those in support of a project will often use lower rates to justify it and why those opposed will use higher rates, and why those with particular concern for far distant effects will want to use lower discount rates.

17.6.4.1 Constant Rates There are three main schools of thought: (1) the Social Opportunity Cost of Capital (SOC); (2) the Social Rate of Time Preference (STP), also called the Shadow Price of Capital (SPC); and (3) the pure rate of time preference (TP). Little serious attention is paid to cat­ egory (3), the pure rate of time preference, as there is no agreement about what number represents it. The SOC rate is based on a simple but fundamental notion: one should not invest in a project unless its rate of return is at least equal to that of other available projects. This rate is reckoned at between 6% and 8% (Burgess and Zerbe, 2011). Cole (2012), in a recent article in the Alabama Law Review, incorrectly characterizes the SOC as an approach that “simply estimates the before-tax rate of return to private capital in the U.S. economy over a number of years.” Such a rate would be about 8.5% but is not the SOC. The SOC instead is a blend of the return to private capital and the consumption rate of time preference, which is taken to be the after-tax rate of return; nor is its calcula­ tion simple. The common and understandable impulse is to look at market interest rates to determine a rate for government. This impulse should be resisted (Burgess and Zerbe, 2011). The return to private capital is not determined by momentary market rates. The idea of the STP or SPC is also a blend of consumption and private capital rates in which all returns are reduced to consumptions streams by adjusting for the capital penal­ ty produced by the displacement of private capital. Proponents of the STP discount rate argue that concerns about the higher rate of return on private investment be reflected not in the discount rate but in the shadow price of capital. Critics contend that the STP fails to take into account fully the opportunity cost of capital and is based on assumptions that do not reasonably hold. Disagreements between SOC and STP advocates center around proper weights to be given to the consumption rate of interest (the rate of substi­ tution) and the marginal rates of transformation (see, for example, Burgess and Zerbe). The crucial (but not the only) error made by STP is shown in the following statement by a leading STP advocate, Bradford, as follows: How … can it possibly make sense for the government to invest in a project with a yield of 5 percent when there are projects available in the private sector yielding 10? The answer is that this can make sense only if the private sector investment is not also an investment opportunity for the government.

Page 17 of 32

Cost–Benefit Analysis in Legal Decision-Making But the answer Bradford offers to this question is incorrect: In fact, Bradford notwith­ standing, it doesn’t make sense to invest at 5% even where the government (p. 372) can­ not invest in the private sector, for, whenever there is public debt outstanding, debt re­ duction is always an option and the rate of return on debt reduction is the SOC rate. That is, the government always has available the SOC (or higher-yielding) projects. The capital penalty for government investment cannot be avoided, as the SOC recognizes. A usable time preference rate will not be found any time soon by looking directly at indi­ vidual rates; individual rates are too diverse to be usable or too high to be practical.22 In a broad survey of empirically elicited discount rates, Frederick et al. (2002) found “spec­ tacular disagreement among dozens of studies that all purport to be measuring time pref­ erence”—from annual discount rates of −6% to infinity (see table 1 in Frederick et al., 2002).23 They found a median value of 24%, with an interquartile range of 8% to 158%. Frederick et al. note (p. 389): This lack of agreement likely reflects the fact that the various elicitation proce­ dures used to measure time preference consistently fail to isolate time preference, and instead reflect, to varying degrees, a blend of both pure time preference and other theoretically distinct considerations. A variation on the time preference approach is suggested by Ramsey. Here the discount rate is r = p + πg, where r is the discount rate, p is the time preference rate, π is the elas­ ticity of marginal utility, which measures the amount of consumption society is willing to sacrifice today to ensure against some expected loss in the future, and g is the expected rate of growth in per capita consumption. Those deriving rates in this matter usually find rates at or below, often substantially below, 3.5% real. As Cole (p. 64) correctly notes, “None of the elements comprising the Ramsey equation can be specified objectively. They are all substantially subjective and subject to dispute, as illustrated by the controversy that followed the U.K. Treasury’s 2006 publication of The Stern Review on the Economics of Climate Change.” In an attempt to standardize use, the Office of Management and Budget (OMB) began is­ suing discount rate guidelines in 1972 (Morrison, 1998, p. 1336) and continues to do so on an annual basis. In 1992, just prior to Clinton taking office, the guidelines recommend­ ed a reduction in discount rates from 10% real to 7% real (OMB, 1992). (Real rates ex­ clude an inflation component and are to be used with constant values (inflation free) of goods (Farrow and Zerbe, 2013). The 7% real rate has remained the recommended rate through to today. The OMB has noted, however, that its guidelines have had little effect on rates that agencies actually use (OMB, 1992; Morrison, 1998, p. 1336). The OMB also suggests rates of about 3% in some cases of long-term projects, but the suggestion is not well motivated. (p. 373)

The usual result of these differences is, as Cole notes, captured by Portney and

Weyant, who note that “those looking for guidance on the choice of a discount rate could find justification for a rate at or near zero, as high as 20% and all values in between.” I Page 18 of 32

Cost–Benefit Analysis in Legal Decision-Making would only amend this to include negative rates and rates at least as high as 30% (An­ dreoni and Sprengler, 2011). In sum, rates that conform to the norms of BCA and thus to KH or expanded KH need to conform. If, however, the exercise is regarded primarily as an economic one, rather than an ideological one, must conform to the first fundamental rule of discount rates, which is that no rate should be below the rate of return for alternative projects. This means the SOC rate.

17.6.4.2 Time Declining Rates Palmon (1992), Weitzman (1998, 2001, 2012), Gollier and Zeckhauser (2005), and Arrow et al. (2013) find time declining rates. The only trenchant, practical argument for time de­ clining rates is given by Palmon. In his analysis, the rate declines based on the ability to better adjust consumption with longer lead times for knowledge of the project. These rates would decline with time towards the social rate of time preference, which Burgess and Zerbe find to be about 3.5%. Palmon derives estimates of the SOC that vary accord­ ing to the displacement of capital and the time between project announcement and com­ pletion. Arrow et al. (a group of prominent economists) present an example in which they calcu­ late the effect of the discount factor associated with each rate on the present value of fu­ ture sums. They consider the range of equally likely future rates to be between 1% and 7%, with equal likelihood (a uniform distribution). They take the average impact and then calculate the certainty equivalent rates associated with this impact. The certainty equiva­ lent rate is that found by giving all preferences an equal weight so that each rate from 1% to 7% is found to be equally likely to represent preferences. Since the discount factor is the reciprocal of one plus the rate, time decreases the impact of higher rates more as higher time preferences give smaller present values than lower rates.24 This leads to cer­ tainty equivalent rates between 1% and 3.94%. Why should these rates be used? They should not. First, the Arrow et al. examples of rates are suspect because they make no reference to rates of return to government in­ vestment that actually prevail or that have prevailed. Second, uncertainty about current and future rates may be, and probably is, a much more narrow range than assumed by Arrow et al. Third, many risk analysts regard it as unacceptable to simply assume a distri­ bution, such as the uniform distribution assumed here, where the distribution is not known. The SOC calculation of rates falls between 8% and 6% real. If, following (p. 374) Arrow et al., a uniform distribution is assumed for this range and certainty equivalent rates calculated, we have the rates in Table 17.7. Table 17.7 Certainty Equivalent Rates Within the Burgess–Zerbe Range Time in Years

Certainty Equivalent Rates

1

7.00%

Page 19 of 32

Cost–Benefit Analysis in Legal Decision-Making 10

6.97%

50

6.84%

100

6.69%

1000

6.11%

This gives a result that varies from 6.11% to 7%, a range that is very substantially differ­ ent from the result of Arrow et al.’s, which varies from 1% at 200 years to 3.94% at 1 year. Rates below about 3.5% should be considered as a lower bound, not 1%. Of course the actual uncertainty in considering rates for 100 years later, let alone 1000 years, must necessarily have a large uncertainty band as it is sensible to imagine that the uncertainty over rates increases with time. Weitzman (1998, 2001) also assumes a uniform distribution among heterogeneous rates. In this case Weitzman infers rates from a survey of 2160 Ph.D.-level economists. From the survey responses, Weitzman gives each discount factor (not the discount rate) equal weight and finds that the distribution of preferred discount rates roughly corresponds to a gamma distribution with a mean of 3.96% and standard deviation of 2.94%. Assuming a gamma distribution, Weitzman derives an implied effective time varying discount rate of µ /(1 + tσ2/µ), where µ is the mean, σ2 is the variance, and t is the number of years in the fu­ ture when benefits or costs are experienced. If there is variance in the preferred discount rates, then the implied effective discount rate falls to zero as t goes to infinity.25 Based on the mean and standard deviation from his survey results, and assuming a gam­ ma distribution, Weitzman concludes that the government should use a discount rate of “about” 4% for benefits/costs incurred one to five years hence and lower rates further in­ to the future, such that the rate is “about” 2% for benefits/costs incurred twenty-six to seventy-five years hence, “about” 1% for benefits/costs incurred seventy-six to 300 years hence, and “about” 0% for benefits/costs incurred 301+ years hence (p. 261).26 Weitzman (2001, p. 270) notes that “the decline in effective social discount rates is sufficiently pro­ nounced, and comes on line early enough, to warrant inclusion of this sliding (p. 375) scale feature in any serious benefit–cost analysis of long-term environmental projects, like activity to mitigate the effects of global climate change.” There is little or no reason to accept this approach and the values it produces. One rea­ son to disregard it is that the use of a poll, even of economists, is not the best way to de­ termine the discount rate, as there is no basis to assume they are experts, nor that they deserve equal weight. Second, there is no assurance that such estimates are grounded in the performance of the economy rather than floating in the air.27 Third, this method of ar­ riving at rates for individuals will result in projects with long time horizons, which are dominated by the valuation of a minority of individuals with low time preference rates. Consequently, programs that have benefits in the far future that would be approved by a Page 20 of 32

Cost–Benefit Analysis in Legal Decision-Making benefit–cost analyst will be overwhelmingly rejected by the public in a referendum, as Long et al. (2013) show. Their results illustrate that heterogeneity in individual time pref­ erence rates, or in experts’ opinions, is going to lead to a divergence between the median voter’s preferences and the decision made by a benefit–cost analyst who evaluates gains in aggregate wealth using disparate discount factors, particularly for projects with bene­ fits further in the future. Fourth, there is no reasonable expectation that the approach satisfies the first of the fundamental rules for a discounting determination. Suppose that in fact we know the distribution of time preference and incorporate this into discount rates using the certainty equivalent method. At base this approach incorporates an ethi­ cal rule that all rates are to be treated equally. Even if we know and use the probability of different rates, projects in the far future will be evaluated using the lowest reported rate. Suppose, as seems likely, this rate and indeed rates for earlier projects are lower than an opportunity cost rate. The use of declining rates has violated the fundamental rule of dis­ counting. We are poorer as a result.

17.7 Principles and Standards Currently there are no agreed-upon principles and standards for the use of BCA.28 There is no overarching framework of interagency principles and standards beyond general guidance as provided by OMB and the executive branch. The use of BCA in federal deci­ sion-making was formalized beyond simply referencing benefits and costs in 1981, when President Reagan issued Executive Order (EO) 12291. EO 12291 required that Regulato­ ry Impact Analyses be conducted for major government initiatives (Reagan, 1981). As a result of this order, the Office of Information and Regulatory Affairs (OIRA) within the White House Office of Management and Budget (OMB) became the central (p. 376) clear­ inghouse for all substantive agency rulemaking, at the time reviewing between 2000 and 3000 rules per year. President Reagan’s subsequent EO 12498, issued in January 1985, furthered this commitment, requiring each federal agency to submit a regulatory plan to OMB discussing all significant current or proposed regulatory activity for yearly review (Reagan, 1985). In 1993, President Clinton introduced EO 12866, at once revoking both EO 12291 and 12498 and establishing a new format for OIRA reviews. In compliance with EO 12866, OIRA organized an interagency review committee with representatives from every major federal regulatory agency to examine the state of the art for the economic analysis of reg­ ulatory action. Three years later, in 1996, the group released a “Best Practices” document detailing the appropriate standards and methodology under which to conduct analysis of significant regulatory action as mandated by Clinton’s Executive Order (US OMB, 1996). In 2003, the OMB under President George W. Bush issued Circular A-4 (US OMB, 2003), which details methods for identifying benefits and costs as well as informs agencies what should be included in a BCA. The document replaced the 1996 “Best Practices” publica­ tion and a subsequent guidance form issued in 2000 (US OMB, 2000). In Circular A-4, OMB refers to the combination of BCA and other information as “regulatory analysis.” Page 21 of 32

Cost–Benefit Analysis in Legal Decision-Making The document gives details regarding required aspects of a regulatory analysis, including a statement of need for a rulemaking, an identification of regulatory alternatives, and an identification of benefits and costs. In January 2009, President Barack Obama issued EO 13497, which formally revoked EO 13258 and EO 12866 and installed an updated version of President Clinton’s EO 12866 (Obama, 2009a). The Obama Administration’s October 2009 Executive Order (13514), ti­ tled “Federal Leadership in Environmental, Energy, and Economic Performance,” signifi­ cantly addressed regulatory review and cost–benefit analysis. EO 13514 mandates retro­ spective BCA policy analysis through annual performance evaluation to enhance account­ ability and extend or expand projects that have net benefits and reassess or discontinue underperforming projects (Obama, 2009b). In keeping with this emphasis, a December 2009 proposal submitted by the White House Council on Environmental Quality to the National Academy of Sciences (NAS) for review seeks to update federal principles and standards for water resources planning and deci­ sion-making. The document seeks to update and expand the established principles and guidelines in place for water resource projects, currently collected as the Economic and Environmental Principles and Guidelines for Water and Related Land Resources Imple­ mentation Studies (US Water Resources Council, 1983). The proposed revisions aim to modernize national water resources development by increasing the role of science in de­ cision-making, incorporating both monetary and non-monetary benefits in the calculation of net benefits, and ensuring the transparency of analyses and determinations. Recent Principles and Standards for BCA developed with support from the MacArthur Founda­ tion can be found at the following website: and in Farrow and Zerbe (2012). (p. 377) This is currently the most thorough set of guide­ lines available. In addition, recent years have seen the creation of several good textbooks, of which the most used currently is by Boardman et al. (2010). These efforts have appar­ ently helped contribute to a currently ongoing process begun in 2013 supported by the National Research Council (NRC) and the National Institutes for Health (NIH) to further develop endpoint for scientific research.

17.8 Conclusion BCA provided the foundation for the law and economics movement beginning around 1960. The concept of wealth maximization only provided a confusing gloss on it. The ex­ treme positions for and against BCA use in law have faded. Correctly it is seen as a useful tool with some positive predictive ability in determining judges’ decisions. It appears to contribute to greater efficiency in government investment spending. John Graham (2008), for example, presents many instances in which bad policy was avoided due to the use of BCA. Cole (2012) in a recent article considers two examples in which BCA led to desirable outcomes and one attempt to manipulate a BCA in connection with George W. Bush’s Clear Skies initiative. As an example of the positive role of BCA, Cole looks at the Clinton Administration changes to Clean Air Act air-quality standards for ozone and particulate Page 22 of 32

Cost–Benefit Analysis in Legal Decision-Making matter and President Obama’s recent decision to suspend the EPA’s reconsideration of the Bush Administration’s air-quality standard for ozone. The empirical evidence present­ ed to bolster the view that BCA is unsatisfactory has not been found in general to be cred­ ible. Porter (1995) notes that in the 1920s and 1930s BCA served as a powerful tool to eliminate bad projects, as Congress relied on the Army Corps of Engineers to provide in­ formation. Congressional reliance on the Corps was due both to the prestige of the Corps and to Congress’s own recognition that its spending was out of control. Of course some BCAs are not well done, or are fraudulent or ignored. The Tellico Dam story (Zerbe and Dively, 1994) shows this well on all counts, and Cole examines the George W. Bush Administration’s Clear Skies initiative as a complex example of how BCA can be manipulated to impede welfare-enhancing regulations. Perhaps the great advan­ tage of BCA is that it furnishes a record that can be examined, criticized, and amended (Zerbe and Dively, 1994). Cole calls this a virtuous transparency.

References Ackerman, Frank et al. (2002). “Pricing the Priceless: Cost Benefit Analysis of Environ­ mental Protection.” University of Pennsylvania Law Review 150: 1553. Ackerman, Frank and Lisa Heinzerling (2004). Priceless: On Knowing the Price of Every­ thing and the Value of Nothing. The New Press. Adler, Matthew (2008). “Risk Equity: A New Proposal.” Harvard Environmental Law Review 32. (p. 378)

Adler, Matthew and Eric Posner (2001). Cost–Benefit Analysis: Economic, Philosophical, And Legal Perspectives. University of Chicago Press. Allen, Douglas (1991). “What are Transaction Costs?” RES. L. & ECON. 14 (Zerbe & Goldberg eds.). Anderson, C. Leigh and Mary Kay Gugerty (2009). “Intertemporal Choice and Develop­ ment Policy: Cross-National Evidence on Discount Rates.” Developing Economics (June). Andreoni, James and Charles Sprenger (2012). “Estimating Time Preferences from Con­ vex Budgets.” American Economic Review 102(7). Arrow et al. (2013). “Determining Benefits and Costs for Future Generations.” Science 341. Baker, Edwin (1980). “Starting Points in the Economic Analysis of Law.” Hofstra Law Re­ view 8: 939. Boardman, Anthony, David Greenberg, Aidan Vining, and David Weimer (2011). Cost Ben­ efit Analysis: Concepts and Practice. 4th edition. Prentice Hall. Broadway, Robin and Neil Bruce (1984). Welfare Economics. New York: Basil Blackwell. Page 23 of 32

Cost–Benefit Analysis in Legal Decision-Making Burgess, David F. and Richard O. Zerbe (2011). “Appropriate Discounting for Benefit–Cost Analysis.” Journal of Benefit–Cost Analysis 2(2). Chipman, John and James Moore (1978). “The New Welfare Economics 1939–1974.” Inter­ national Economic Review 19(3): 578. Coase, Ronald H. (1991). The Firm, the Market, and the Law. University of Chicago Press. Cole, Daniel H. (2012). “Law, Politics, and Cost–Benefit Analysis.” 64 ALA. L. REV. 55. Coleman, Jules L. (1980). “Efficiency, Utility, and Wealth Maximization.” Hofstra Law Re­ view 8: 509. Exec. Order No. 12,291, 46 Fed. Reg. 13,193 (1981). Farrow, Scott and Richard Zerbe (2013). Principles and Standards for Benefit–Cost Analy­ sis. Cheltenham: Edward Elgar. Frederick, S. et al. (2002). “Time Discounting and Time Preference: A Critical Review.” Journal of Economic Literature 40: 351–401. Friedman, Lee (1984). Microeconomic Policy Analysis. Gollier, Christian and Richard Zeckhauser (2005). “Aggregation of Heterogeneous Time Preferences.” Journal of Political Economy 113(4). Goulder, Lawrence H. (2007). Benefit–Cost Analysis, Individual Differences, and Third Parties 23 RES. L. & ECON. (Richard O Zerbe Jr. & John B. Kirkwood eds., Elsevier 2007). Graham, John D. (2008). “Saving Lives through Administrative Law and Economics.” Uni­ versity of Pennsylvania Law Review 157: 395. Hammond, Peter (1985). “Welfare Economics,” in George Feiwel, ed., Issues in Contempo­ rary Microeconomics and Welfare, 406. Hammond, Richard J. (1966). “Convention and Limitation in Benefit–Cost Analysis.” Nat­ ural Resources Journal 6: 195–222. Harberger, Arnold C. (1971). “Three Basic Postulates for Applied Welfare Economics: An Interpretive Essay.” Journal of Economic Literature 9 (3): 785–797. Harding, S. (1950). “Background of California Water and Power Problems.” California Law Review 38: 547. Harrod, Roy F. (1938). “Scope and Method of Economics.” Economic Journal 48: 383. Hicks, John R. (1939). “The Foundations of Welfare Economics.” Economic Journal 49: 696. Page 24 of 32

Cost–Benefit Analysis in Legal Decision-Making Holmes, B. H. (1972). A History of Federal Water Resources Programs, 1800–1960. U.S. Dept. of Agriculture, Economic Research Service. Just, Richard, Andrew Schmitz, and Richard O. Zerbe (2012). “Scitovsky Reversals and Practical Benefit Cost Analysis.” J. COST-BENEFIT ANALYSIS 3(2). (p. 379)

Just, Richard, Andrew Schmitz, and Richard O. Zerbe (2013). “Scitovsky Reversals in Ben­ efit–Cost Analysis with Normal Goods.” Journal of Benefit–Cost Analysis 4(3): 411–413. Kaldor, Nicholas (1939). “Welfare Propositions of Economics and Interpersonal Compar­ isons of Utility.” Economic Journal 49: 549. Kalpow, Louis and Steven Shavell (1994). “Why the Legal System is Less Efficient than the Income Tax in Redistributing Income.” Journal of Legal Studies 23(2): 667–681. Knetsch, Jack L. (1995). “Assumptions, Behavior Findings, and Policy Analysis.” Journal of Policy Analysis and Management 14(1): 68–78. Lasswell, Harold D. (1936). Politics: Who Gets What, When, How. Long, Mark et al. (2013). “The Social Discount Rate as a Function of Individuals’ Time Preferences: Elusive Theoretical Constructs and Practical Advice for Benefit–Cost Ana­ lysts.” Working paper. Maass, Arthur (1950). “Administering the CVP.” California Law Review 38: 666. McCloskey, Donald (1982). The Applied Theory of Price. Markovits, Richard S. (2008). Truth or Economics: On the Definition, Prediction, and Rel­ evance of Economic Efficiency. New Haven, CT: Yale University Press. Mishan, Ezra J. (1981). An Introduction to Normative Economics 120–121. Moore, Mark A. et al. (2004). “Just Give Me A Number! Practical Values for the Social Discount Rate.” Journal of Policy Analysis and Management. 23(4): 789–812. Morrison, Edward (1998). “Judicial Review of Discount Rates Used in Regulatory Cost– Benefit Analysis.” University of Chicago Law Review 65: 1333. Palmon, Oded (1992). “Anticipating Costs and Benefits and the Social Discount Rate.” Canadian Journal of Economics 25(2). Pareto, Vilfredo (1896). 2 Cours D’Economie Politique. G.H. Bousquet & G. Busino eds., F. Rouge. Pigou, A. C. (1932). The Economics of Welfare. 4th edition. MacMillan. Polinsky, Mitchell A. (1989). An Introduction to Law and Economics. 2nd edition. Wolters Kluwer. Page 25 of 32

Cost–Benefit Analysis in Legal Decision-Making Porter, Theodore M. (1995). Trust In Numbers: The Pursuit Of Objectivity In Science and Public Life 187. Princeton University Press. Posner, Richard (1972). “A Theory of Negligence.” Journal of Legal Studies 29. Posner, Richard (1985). “Wealth Maximization Revisited.” Notre Dame Journal of Law, Ethics and Public Policy 2: 85. Posner, Richard (1986). Economic Analysis of Law. 3rd edition. Wolters Kluwer. Posner, Richard (1987). “The Justice of Economics.” Journal of Public Finance and Public Choice 15: 23. Ramsey, F. P. (1938). “A Mathematical Theory of Savings.” Economic Journal 38: 543–559. Robbins, Lionel (1938). “Interpersonal Comparisons of Utility: A Comment.” Economic Journal 48: 635. Schmitz, Andrew and Richard O. Zerbe (2009). “The Relevance of the Scitovsky Principle for Benefit–Cost Analysis.” Journal of Agricultural and Food Industrial Organization 6(2). Scitovsky, Tibor. “A Note on Welfare Propositions in Economics.” Review of Economic Studies 9: 77–88. Shapiro, Sidney and Christopher Schroeder (2008). “Beyond Cost–Benefit Analysis: A Pragmatic Reorientation.” Harvard Environmental Law Review 32: 433. Shavell, Steve (1981). “A Note on Efficiency vs. Distributional Equity in Legal Rulemaking: Should Distributional Equity Matter Given Optimal Income Taxation.” Ameri­ can Economic Review 71: 414. (p. 380)

Sunstein, Cass R. (2001). “Cost–Benefit Default Principles.” Michigan Law Review 99: 1651, 1663–1666. Thaler, Richard and Cass Sunstein (2009). Nudge. New York: Penguin Press. Tversky, A. and D. Kahneman (1992). Journal of Risk and Uncertainty 29. Verchick, Robert R. M. (2005). The Case Against Cost Benefit Analysis. Paper in Social Science Research Network, available at . Weimer, David L. and A. R. Vining (1992). Policy Analysis: Concepts and Practice. 2nd edi­ tion. Weitzman, Martin L. (1998). “Why the Far-Distant Future Should be Discounted at its Lowest Possible Rate.” Journal of Environmental Economics and Management 36(3). Weitzman, Martin L. (2001). “Gamma Discounting.” American Economic Review 91(1).

Page 26 of 32

Cost–Benefit Analysis in Legal Decision-Making Weitzman, Martin L. (2012). “The Ramsey Discounting Formula for a Hidden-State Sto­ chastic Growth Process.” Journal of Environmental and Resource Economics 53(3). Whittington, Dale and Duncan MacRae, Jr. (1986). “The Issue of Standing in Benefit–Cost Analysis.” 5(4) Journal of Policy Analysis and Management 5(4): 665. Zerbe, Richard O. (2007). “The Legal Foundations of Cost–Benefit Analysis.” 2 Charleston Law Rev. 2: 93. Zerbe, Richard O. (2014). “Welfare Economics for Law with Costly Markets,” in Richard O. Zerbe, ed., Efficiency in Law and Economics. Series editors Richard Posner and Francesco Parisi. Northampton, MA: Edward Elgar. Zerbe, Richard O. and Dwight Dively (1994). Benefit–Cost Analysis. New York: Harper Collins. Zerbe, Richard O. (with David Burgess) (2011). “Appropriate Discounting for Benefit–Cost Analysis.” Journal of Benefit–Cost Analysis 2(2): 1–20. Zerbe. Richard O. (with David Burgess) (2011). “Calculating the Social Opportunity Cost Discount Rate.” Journal of Benefit–Cost Analysis 2(3). Zerbe, Richard O., ed., (2013). Efficiency in Law and Economics. Series editors Richard Posner and Francesco Parisi. Northampton, MA: Edward Elgar.

Notes: (1) The searches resulting in Tables 17.1 and 17.2 were conducted at my request by refer­ ence librarians Anna Endter, Cheryl Nyberg, and Mary Whisner and by reference interns Matthew Flyntz, Heather Joy, and Taryn Marks of the University of Washington School of Law. (2) For Table 17.2 Cheryl Nyberg, law librarian, and Tara Marks, reference intern, used HeinOnline’s Law Journal Library to get data on the use of the phrase cost–benefit (and two variations: “cost benefit” (no hyphen) and “benefit cost,” by decade. She excluded from the search bar journals and other periodicals on HeinOnline that were not in the Law Journal Library. These numbers should be considered approximate. The HeinOnline search feature for limiting to dates doesn’t necessarily retrieve exact dates. For instance, an issue of a law review that was published in Nov. 1949 is retrieved in the 1950–59 search because the volume in which it was published covers 1949–50. The search engine picks up the 1950 in the search. Also, some of the titles that were retrieved are from out­ side the United States, but international coverage of HeinOnline is not comprehensive. The search also picks up titles like the Monthly Labor Review, the SEC Docket, and Crimi­ nal Justice (a magazine). (3) This paragraph is redacted from Zerbe and Scott (2010), a paper prepared for Dennis Culhane, Professor of Social Policy at the University of Pennsylvania. Page 27 of 32

Cost–Benefit Analysis in Legal Decision-Making (4) Though in fact KH and the PCT are not quite the same (Broadway and Bruce, 1984). (5) The net present value is not the only summary measure. Others are the benefit–cost ratio, the internal rate of return and its modified version, and payback periods. See Zerbe and Dively (1994) for an extended consideration of the use of these. (6) As envisioned by Kaldor (1939), non-pecuniary effects were to be included in benefit– cost analysis. (7) For example, Harrod (1938) argued that the net social benefit from a policy could be established on the assumption that the individuals affected were equal in their capacity to enjoy income. That is, an improvement can be assumed by looking at changes in income as long as, in modern terminology, the marginal utility of income with respect to income changes is the same for all individuals. Harrod used this reasoning to justify the 1846 re­ peal of the English Corn Laws, a classic test case for British economists. In response, Li­ onel Robbins pointed out that interpersonal comparisons of utility couldn’t rest on a sci­ entific foundation since utility cannot be measured, and that the justification for such comparisons is more ethical than scientific. Harrod complained that in the absence of comparability of utility of different individuals, “the economist as an advisor is completely stultified.” (8) This debate about whether or not prescriptions of economics were scientific is paral­ leled by the 1980s debate, mostly in the legal literature, about the normative foundations of wealth maximization. For example, see Baker (1980) and Coleman (1980). (9) It was thought that politicians or non-economists should make judgments and deci­ sions about income distribution effects. (10) It cannot be said that this second assumption of equal marginal utility of income avoids interpersonal comparisons; indeed it embraces them in a very particular way: all people are treated equally in terms of the value they place on changes in income. (11) The eagerness of economists to separate considerations of efficiency from those of distribution arose from a desire to put economics on a firm base as a policy instrument. Kaldor suggests, “the economist should not be concerned with prescriptions at all … For, it is quite impossible to decide on economic grounds what particular pattern of incomedistribution maximizes social welfare” (Kaldor, 1939). See also Pigou (1932). (12) Although it is recognized that the willingness to accept is the correct measure for losses, traditional opinion has held that there is little difference between the two mea­ sures so that WTP may be used in practice. This is, however, untrue in many important cases. (13) For an explanation of why this leads to difficulties, see Coase (1991). Coase’s view is that the major failing of welfare economics lies in its assumption of zero transactions

Page 28 of 32

Cost–Benefit Analysis in Legal Decision-Making costs. By transactions costs I mean the costs necessary to transfer, establish, and main­ tain property rights (Allen, 1991). (14) For a more extensive review of history, see Zerbe (2007). (15) According to Hammond (1966), the use of formal cost–benefit ratios goes back at least as far as the Rivers and Harbor Act of 1902. They were explicitly mandated in the amendment to the Act in 1920. (16) Economic and Environmental Principles and Guidelines for Water and Related Re­ sources Implementation Studies. These were developed in accordance with Section 103 of the Water Resources Planning Act as amended (42 U.S. C. 1962a-2). New standards were developed from September 9, 1982 when the Water Council voted to repeal the ex­ isting Principles and Standard and Procedures (18 CFR, Parts 711, 713, 714) and to es­ tablish new guidelines. President Nixon approved these on February 3, 1983 in accor­ dance with Executive Order 11747 (38 FR 30993, Nov 7, 1973). (17) To be exact, it was September 10, 2001, a day before 9/11. (18) A concept much used in the law is that of wealth maximization developed by Richard Posner. The concept is murky and the extent to which it differs from KH is unclear. It ap­ pears from the current vantage point to be without legs and we do not consider it further here. Adler has developed an approach centered around social welfare functions but this has yet to achieve meaningful application by others, and is not considered here. (19) Posner’s term “wealth maximization” appears identical to KH except that he would al­ low altruistic concerns where there is a WTP for them consistent with Expanded Kaldor– Hicks. A strong theory of wealth maximization is said to have three crucial features. First, wealth maximization is an aggregate concept. That is, it is more concerned with societal well-being than with individual welfare. For example, a wealth maximizer is not con­ cerned with the distribution of wealth among citizens, and any coerced payment to effect distribution is presumed unproductive. See Posner (1985, p. 251). (20) For example, as Boardman et al. (1992) note, “Strict use of the Kaldor–Hicks test means that information on how benefits and costs are distributed among groups is ig­ nored in decision making.” Friedman also notes, “Some analysts would like to ignore eq­ uity altogether and use the compensation test as the decisive analytic test…. [A] second rationale for relying on the compensation test is the belief that concern for equity is sim­ ply unfounded” (Friedman, 1984). Additionally, Posner notes that wealth maximization is simply the Kaldor–Hicks test and that wealth maximization ignores distributional effects (Posner, 1986; Posner, 1987). McCloskey incorrectly contends that the consumer surplus measure of social happiness is the same as the national income measure (McCloskey, 1982). Of course, the national income measure contains no measure of income distribu­ tion. Posner’s claim is, however, at variance with his acceptance of valuing ethical consid­ erations by the WTP.

Page 29 of 32

Cost–Benefit Analysis in Legal Decision-Making (21) For example, consider Coleman’s use of a second-best situation. A second-best state means that there is one that is Pareto superior to it. The obvious question is: Why not move to a first-best situation? A full array of possibilities is shown in the footnote table. To make the example concrete, suppose the two goods X and Y represent wheat and cot­ ton, and that one acre of land produces one unit of either wheat or cotton. Welfare opti­ mization does not lead to either states A or B because a state A′ is possible that is Pareto superior to state A and a state B′ is possible that is Pareto superior to state B (recall that the PPF can transform one unit of one good into one unit of the other good). Because states A′ and B′ are first-best states, they are Pareto non-comparable. Neither the move from A′ to B′ nor the move from B′ to A′ passes the KH test a reversal cannot occur. Footnote table Absence of Reversals When Comparing to First-Best States

Page 30 of 32

Cost–Benefit Analysis in Legal Decision-Making A Status Quo

Wheat

Cotton

B Uncompensated State After the Proposed Project

A′ Pareto Superior State Compared to Status Quo

B′ Pareto Superior State Compared to State Af­ ter the Proposed Project

Wheat

Wheat

Wheat

Cotton

Cotton

Cotton

Mr. 1

2

0

1

0

1

1

1

0

Ms. 2

0

1

0

2

0

1

1

1

Page 31 of 32

Cost–Benefit Analysis in Legal Decision-Making (22) As Long et al. (2013) note, “It is very difficult to isolate individual’s time preference rates, let alone the joint distribution of project valuations and time.” (23) See also Anderson and Gugerty (2009). (24) For example, calculate the net present values (NPVs) of a future amount at both 1% and 7%. Take the average of these two NPV. Now find the rate that gives the average fu­ ture value. This rate is the certainty equivalent. (25) This implied effective discount rate asymptotes towards the minimum value support­ ed by the gamma distribution, which is zero. It is worth noting that three of the respon­ dents to his survey gave negative time discount rates (with the minimum value being −3%). By assuming a gamma distribution, Weitzman’s implied effective discount rate as­ ymptotes towards zero rather than −3%. (26) Note that this declining social discount rate is not the result of individual preferences for hyperbolic discounting. (27) The gamma distribution Weitzman uses to fit the data is not a close fit for lower dis­ count rates chosen. (28) The Benefit–Cost Analysis Center at the University of Washington Evans School of Public Affairs is currently in the process of assessing these various documents. Draft re­ views are available by request online at .

Richard O. Zerbe Jr.

Richard O. Zerbe Jr., Daniel J. Evans Professor of Public Affairs, University of Wash­ ington.

Page 32 of 32

Well-Being and Public Policy

Well-Being and Public Policy   John Bronsteen, Christopher Buccafusco, and Jonathan Masur The Oxford Handbook of Law and Economics: Volume 1: Methodology and Concepts Edited by Francesco Parisi Print Publication Date: Apr 2017 Subject: Economics and Finance, Law and Economics, Econometrics, Experimental and Quanti­ tative Methods Online Publication Date: May 2017 DOI: 10.1093/oxfordhb/9780199684267.013.028

Abstract and Keywords Governments rely on certain basic metrics and tools to analyze prospective laws and poli­ cies and to monitor how well their countries are doing. In the United States, cost-benefit analysis (CBA) is the primary tool for analyzing prospective policies, especially with re­ spect to administrative regulations. Similarly, Gross Domestic Product (GDP) is perhaps the most prominent metric for monitoring a country's progress. In recent years, one of the most important developments in social science has been the emergence of psychologi­ cal research measuring subjective well-being (SWB) or ‘happiness’. This article first ex­ plains the way in which SWB is measured and how those measurements have been vali­ dated. It then discusses well-being analysis (WBA), which uses happiness data to analyze prospective policies more accurately than does CBA. Next, it covers the ways in which SWB data have been used to generate prices that can be used by traditional economic analysis. This is followed by a discussion of attempts to revise CBA to deal with the limita­ tions stemming from the fact that it uses wealth to assess the effects of policy on quality of life. Finally, the article lays out the progress made towards creating an SWB-based al­ ternative to GDP. Keywords: cost-benefit analysis, economic analysis, GDP, subjective well-being, happiness, well-being analysis, policymaking, quality of life

18.1 Introduction GOVERNMENTS rely on certain basic metrics and tools to analyze prospective laws and policies and to monitor how well their countries are doing. In the United States, cost–ben­ efit analysis (CBA) is the primary tool for analyzing prospective policies, especially with respect to administrative regulations. Similarly, Gross Domestic Product (GDP) is perhaps the most prominent metric for monitoring a country’s progress. Both CBA and GDP are

Page 1 of 24

Well-Being and Public Policy economic measures: CBA values outcomes by trying to approximate how they would be valued by the market, and GDP focuses on total economic output. For decades, critics of such economic measures have argued that they ignore important aspects of value that are not fully reflected by output or by willingness to pay (WTP). This first manifested itself in the social indicator movement of the 1960s and 1970s (Stiglitz, Sen, and Fitoussi, 2009, p. 208 and n. 86; Veenhoven, 1996, p. 2) that aimed to produce rival measures based on objective values such as health and education. The most famous such measure is the Human Development Index (HDI). In recent years, one of the most important developments in social science has been the emergence of psychological research measuring subjective well-being (SWB) or “happi­ ness.” Researchers have made great strides in replicating and validating their findings about happiness, and world leaders have called for this research to be used to supply SWB-based metrics and tools as alternatives to the existing economic ones (Cohen, 2011; Samuel, 2009; General Assembly Resolution, 2011). In response to these calls, early efforts have been made to use SWB research to create new social indicators. The efforts have been sensitive to a number of important obstacles. For example, subjective well-being cannot be validated via objective proof in the (p. 382) same way that economic output or longevity can be. In addition, quality of life may have components other than SWB, and SWB itself may have different components that cannot be easily combined. But despite these concerns, scholars and policymakers understand the political power of a simple, unitary metric such as GDP or the HDI. And they also see the value of SWB measures in assessing the quality of human life more directly and perhaps more accurate­ ly than can be done via traditional economic measures, and more comprehensively than via other social indicators. As a result, they have shown intense interest in using SWB da­ ta to create alternatives to the existing tools and measurement devices. In this chapter, we discuss some of the efforts that have been made in this regard.1 We first briefly explain the way that SWB is measured and the way those measurements have been validated. We then explain our own contribution—well-being analysis (WBA)—which uses happiness data to analyze prospective policies more accurately than does CBA. Next, we cover the ways in which SWB data have been used to generate prices that can be used by traditional economic analysis. We then discuss attempts to revise cost–benefit analysis to deal with the limitations stemming from the fact that it uses wealth to assess the ef­ fects of policy on quality of life. Finally, we lay out the strides that have been made to­ ward creating an SWB-based alternative to GDP.

18.2 SWB Data Since the days of Jeremy Bentham, economics has focused on utility. Although Bentham and his followers believed that utility could plausibly be measured directly, most modern Page 2 of 24

Well-Being and Public Policy economists have sought proxies for utility (Colander, 2007). For the most part, modern economics has assumed that the choices people make are the best available proxy for their well-being. Daniel Kahneman has called this “decision utility.” More recently, howev­ er, a new wave of social scientists has called for a return to Bentham and, accordingly, the measurement and analysis of “experienced utility” (Kahneman et al., 1997). These social scientists have attempted to develop more direct and valid proxies for individuals’ wellbeing that rely on self-reported happiness. Over the past quarter-century, researchers in the field of hedonic psychology have devel­ oped a variety of new tools for studying SWB. We will describe some of these tools, dis­ cuss issues about their validity and reliability, and briefly summarize a few of the key find­ ings from the literature. Hedonic psychologists believe that the best way to figure out how an experience makes a person feel is to ask her about it while she is experiencing it. The “gold standard” of such measures is the experience sampling method (ESM), which uses handheld computers or mobile phones to survey people about their experiences (Kahneman et al., 1997). (p. 383) Subjects are beeped randomly throughout the day and asked to record what they are do­ ing and how they feel about it. The data that emerge from such studies provide a detailed picture of how people spend their time and how their experiences affect them. The data can also be combined with socio-economic and demographic data via regression analyses for even greater insight (e.g. do the unemployed spend more time in leisure activities than the employed, and do they enjoy these activities as much?). The oldest method of measuring SWB is the life satisfaction survey. These surveys ask in­ dividuals to respond to a question such as, “All things considered, how satisfied with your life are you these days?” (Pavot and Diener, 1993). Respondents answer on a scale that ranges from “not very happy” to “very happy.” Life satisfaction surveys have been includ­ ed in the U.S. General Social Survey since the 1970s; as a result, we now have substantial quantities of longitudinal data on thousands of individuals. The principal value in such surveys is the ability to correlate SWB data with a variety of other facts about people’s lives. Using multivariate regression analyses that control for different circumstances, re­ searchers are able to estimate the strength of the correlations between SWB and factors such as income, divorce, unemployment, disability, and the death of family members (Lu­ cas et al., 2003; Clark et al., 2008). ESM and life satisfaction surveys have different advantages and disadvantages. ESM re­ lies less on difficult cognitive processes like remembering and aggregating experiences, so it avoids certain kinds of errors. But while ESM can give a fine-grained picture of people’s lives, it tends to be very expensive to run for large samples.2 Life satisfaction surveys are relatively inexpensive to administer and can be easily included in a variety of larger survey instruments. Accordingly, they are valuable as sources of large-scale data about many subjects and of longitudinal data about changes in SWB over time.

Page 3 of 24

Well-Being and Public Policy The various SWB metrics have generally scored well in terms of their reliability and valid­ ity, often performing at least as well as more established economic measures of utility (Bronsteen, Buccafusco, and Masur, 2015). Meta-analyses of different well-being tools have found high levels of reliability for both life satisfaction and experience sampling methods. This is especially true of more advanced multi-item measures (Diener et al., 2009). In addition, SWB measures tend to correlate well with a variety of other relevant indicators, including other SWB data, third-party well-being estimates, the amount of time someone spends smiling, and neurological activity (Diener et al., 2009; Bronsteen, Buccafusco, and Masur, 2015). Finally, SWB metrics seem appropriately sensitive to ex­ ternal factors in peoples’ lives, including unemployment, health, social relationships, and income. Despite these findings, there are still reasons to be cautious about some uses of SWB da­ ta. The major concern about hedonic data is the one that first dissuaded economists from attempting to measure utility directly—anxieties about interpersonal cardinality. (p. 384) How can we know, for example, that when two people both rate their SWB as “5 out of 10” they mean the same thing? While this is an important objection, it is one to which there are answers. First, if different people use the happiness scales differently, this will not necessarily lead to systematic errors if SWB data is used to formulate policy. Instead, it may simply produce noise. Different scale usage will only be a problem if it correlates with differences between two populations that are being compared.3 Moreover, the use of monetization as a proxy for utility suffers from the same interpersonal cardinality prob­ lem. Whereas a reported “5.0” might reflect a different level of well-being for one person than it does for another, there are definitely situations in which a particular sum of money will affect one person’s well-being differently from another’s. One thousand dollars might mean everything to a homeless person and nothing to Bill Gates. Because of the diminish­ ing marginal value of money, two individuals with differing levels of personal wealth can obtain vastly different amounts of welfare from the same gain (or loss) of income (Frank, 1997). And even two people with identical wealth could derive radically different amounts of utility from the same amount of additional money. In assessing the value of these data, the question should not be whether they perfectly measure utility. SWB data are not and cannot be a perfect proxy for human welfare. The important question is how they compare with the available alternatives. As we explain in more detail below, we think that in many cases they will perform better than traditional economic tools.

18.3 Well-Being Analysis Virtually every law makes people’s lives better in some ways but worse in others. For ex­ ample, an air-quality law could make people healthier, but it could also force them to pay more money for the products they buy. Every proposed law thus raises the question: Would its benefits outweigh its costs?

Page 4 of 24

Well-Being and Public Policy To answer that question, there needs to be a way of comparing seemingly incommensu­ rable things like health and consumer consumption. The most widely adopted method, which has now become standard within the American regulatory state, is cost–benefit analysis (CBA). CBA involves “monetizing” costs and benefits—translating real-world ef­ fects into their equivalent dollar values—and then comparing them in monetary terms. Thus, conducting a CBA of an air-quality law would involve estimating the monetary costs of the law to consumers, estimating the monetary value they would place on cleaner air, and then comparing the two quantities. Every economically significant regulation from executive-branch agencies must, by law, be evaluated via CBA or an analogue. This has been the case since 1981, when President Ronald Reagan mandated (p. 385) it by execu­ tive order. That order has been reaffirmed by every president since, including Presidents Bill Clinton and Barack Obama (Bronsteen, Buccafusco, and Masur, 2013).4 Yet this monetization process is hardly straightforward. The benefits produced by laws and regulations are often nonmarket goods such as health and safety, and thus there are no direct means for determining what value individuals place on them. CBA must thus turn to indirect mechanisms: Either researchers must conduct contingent valuation sur­ veys in which they ask, hypothetically, how much an individual would pay for a particular benefit, or they must attempt to determine revealed valuations by studying individuals’ market decisions such as the wage premium that workers demand to take a more danger­ ous job. If an agency is considering regulating some aspect of workplace safety, for example, it will have to compare the costs of that regulation (including implementation costs, in­ creased prices, and unemployment) with the benefit of improved employee health. The data for these estimates are derived from revealed preference studies that demonstrate the implicit value that workers attach to their lives by comparing the salaries that work­ ers are paid in workplaces with different levels of risk (Hammitt, 2007). Alternatively, if an agency is calculating the benefits of environmental regulation, it will typically attempt to estimate the value of, for example, an increase in protected wetlands by asking people to indicate how much they are willing to pay for the benefit. In these stated preference or contingent valuation studies, a representative population will be surveyed and asked hy­ pothetical questions about their willingness to pay for various benefits (Desvousges et al., 1993). Revealed and stated preference studies provide a method for converting non-mar­ ket benefits into monetizable and comparable units. Both of these methods are flawed in a variety of ways, and numerous studies have demon­ strated their shortcomings. The validity and reliability of the data produced by revealed and stated preference studies are questionable. These studies place considerable cogni­ tive demands on subjects that undermine the trustworthiness of the data they generate. In the workplace safety example, employees are assumed to understand the level of risk that different jobs pose, and they are assumed to make rational trade-offs between in­ creased wages and potential health risks. A wealth of psychological research suggests that people fare very poorly at these sorts of tasks. People’s minds are not designed to

Page 5 of 24

Well-Being and Public Policy differentiate between exceedingly small risks, and when asked to do so rationally, they frequently fail (Sunstein, 2002). Most importantly, there is one insuperable flaw, common to both of CBA’s methods of monetization, which looms particularly large: both mechanisms require individuals to pre­ dict how much they will enjoy some good, or how much they will suffer from some harm, if they experience it in the future. In the vast majority of cases, the individuals in question will never have experienced the harm or benefit. Thus, CBA will attempt to put a value on, for instance, avoiding a case of cancer by looking to how much people who have never had cancer are willing to pay to avoid it. Psychologists have amassed mountains of evidence demonstrating that individu­ als very frequently make significant mistakes when attempting to estimate how they will feel about some future experience (e.g. Wilson and Gilbert, 2005). These mistakes, called “affective forecasting errors,” are extraordinarily difficult to avoid. In particular, multiple studies indicate that people struggle with attempting to gauge the magnitude and dura­ tion of the effects of disabilities (Gilbert and Wilson, 2007; Wilson and Gilbert, 2005). If people misestimate the impact of disabilities, then inferring perfect judgment from their (p. 386)

behaviors about health risks will systematically bias estimates about the welfare effects of regulations. Similarly, in the environmental protection example, there is good reason to doubt that people will be able to accurately predict how much their lives will be improved by things such as increased wetlands protection (Desvousges et al., 1993). Importantly, these errors are not random; they exhibit systematic biases that do not just create noise but also produce more intractable distortions in valuation. Affective forecasting errors re­ sult in some goods being overvalued and others undervalued relative to their actual ef­ fects on people’s lives. Accordingly, there is every reason to believe that CBA—which re­ lies on forward-looking estimations—is making significant errors in pricing costs and ben­ efits (Sunstein, 2007; Sunstein, 2016). Policymakers who rely upon CBA are likely arriv­ ing at flawed conclusions regarding which laws and regulations they should implement. With CBA’s critical flaw in mind, we designed an alternative method for comparing the positive and negative consequences of a law (Bronsteen, Buccafusco, and Masur, 2013). This method, which we termed “well-being analysis” (“WBA,” for short), would directly analyze the effect of various costs and benefits on people’s subjective well-being or quali­ ty of life. For example, a policymaker would assess an air-quality law by comparing how much more people would enjoy their lives if they became healthier with how much less they would enjoy their lives if their consumption were reduced. This is the most natural and direct way to put seemingly incommensurable things on the same scale. And it yields the specific answer that is needed: whether a law will make people’s actual experience of life better or worse on the whole. WBA relies on the same basic cost–benefit weighing principle that undergirds CBA: all else equal, regulations whose benefits exceed their costs are valuable because they en­ hance overall welfare. The main difference between the two techniques involves the way in which costs and benefits are calculated and compared. Instead of monetizing the ef­ Page 6 of 24

Well-Being and Public Policy fects of regulation, WBA “hedonizes” them. That is, it measures how much a regulation raises or lowers people’s enjoyment of life. To do that, it relies on hedonic psychology da­ ta that measure how different factors affect people’s enjoyment of their lives. The positive and negative hedonic impacts can then be compared with one another. They are the rele­ vant costs and benefits. Instead of converting regulatory effects into monetary values, WBA converts them into well-being units (WBUs). WBUs are intended to be subjective, hedonic, cardinal, and in­ terpersonally comparable units that indicate the degree of a person’s happiness for a giv­ en period of time. They are, in some respects, similar to the quality-adjusted life years (QALYs) that are increasingly popular in health economics. WBA maps a person’s subjective well-being onto a scale that runs from −10 to 10, in which 10 indicates perfect happiness (subjectively defined), −10 indicates perfect mis­ ery, and 0 indicates neutrality or the absence of experience. This type of scale would al­ low individuals to register experiences that are worse than non-experience (undergoing torture, for instance) and would simplify the comparison between experience and non-ex­ perience. Each decile of the scale is equivalent and indicates a 10% change in the (p. 387)

person’s subjective well-being. One well-being unit is equivalent to 1.0 on the scale for a period of one year. Thus, if a person lives to the age of 100 and has a subjective well-be­ ing of 7.0 out of 10.0 for each year, that person has experienced 700 WBUs (7.0 WBU/ year × 100 years). If an event such as illness causes a person’s SWB to drop from 7.0 to 5.5 for a period of ten years, that person loses 15 WBUs (1.5 WBU/year × 10 years). This type of scale has significant benefits for any type of decision analysis, particularly regulatory analysis, because it enables the direct comparison of the hedonic impacts of proposed policy changes. Imagine, for example, that the Occupational Safety and Health Administration (OSHA) is contemplating a simple regulation of workplace safety that will prevent 100 workers from each losing an arm while on the job. Implementing such a mea­ sure, however, will increase the costs of production and force factories to fire 300 work­ ers in the affected industry. CBA would attempt to calculate the value of the regulation by monetizing the costs and benefits it generates. With respect to the costs, CBA could estimate the lost wages of the 300 unemployed people (although not the lost well-being entailed by those lost wages). The benefits, however, are trickier. Establishing a market price for the loss of an arm is a fraught enterprise. Given these shortcomings, the value CBA applies to the loss of an arm will be beset by a number of systematic errors, especially individuals’ poor ability to pre­ dict how events like losing an arm will affect them. Accordingly, CBA may substantially and systematically misstate the benefits of the regulation. WBA would approach the measure in the same general fashion but with different analyti­ cal data. Like CBA, WBA would attempt to quantify the cost of unemployment. But in­ stead of looking solely to the workers’ lost wages, it would calculate the hedonic cost of being unemployed by surveying individuals who have actually become unemployed and comparing their responses pre- and post-unemployment. Some data suggest that unem­ Page 7 of 24

Well-Being and Public Policy ployment has a significant effect on well-being (Lucas et al., 2004). Thus, the welfare costs of unemployment may be much greater than CBA predicts. On the other side of the ledger, WBA is well positioned to hedonize the benefits of the regulation. Studies of peo­ ple who have lost limbs provide credible information on the hedonic loss associated with losing an arm, and thus the benefits of avoiding such a loss (Frederick and Loewenstein, 1999). Again, the results are likely to be different from those determined by CBA. Studies show that individuals who lose limbs often adapt substantially to their new condition, re­ covering most of their lost happiness within a few years (Frederick and Loewenstein, 1999). This result is contrary to the predictions of healthy people, who typically assume that such disabilities will be devastating and discount the possibility that they will adapt to the loss (Ubel, Loewenstein, and Jepson, 2005; Ubel et al., 2001). Accordingly, WBA holds the potential to deliver results that are far more accurate (p. 388) in welfare terms than CBA, despite the fact that WBA is a much “younger” procedure (Graham, 2016). WBA thus holds two primary advantages over CBA. The first is that it will provide more accurate measures of benefits and costs because it relies on actual human experiences, rather than speculation about unknown futures. The second is that in relying upon subjec­ tive well-being rather than money, it employs a closer proxy for welfare than does CBA. Importantly, these two advantages are severable. Even if agencies should be attempting to maximize wealth rather than welfare, as some commentators believe (Weisbach, 2015), WBA will still yield more accurate “prices” for the consequences of regulation than CBA.

18.4 Happiness Data as Prices Although we believe that WBA offers the best method for estimating the effects of laws and policies, researchers have developed other approaches that improve upon traditional CBA. The biggest challenge for CBA involves estimating the value of non-market regulato­ ry benefits, especially health and safety improvements. Often, CBA will assess the value of a regulation in terms of the number of lives it saves multiplied by an estimate of the value of a statistical life (VSL) that has been computed from revealed preferences data (Sunstein, 2004). In addition to suffering from the standard flaws common to all revealed preferences data, these estimates have been criticized for ignoring the length of the lives that get saved. VSL treats equally saving the lives of hospice patients and saving the lives of teenagers. To remedy this, some scholars have proposed analyzing regulatory costs and benefits using the number of years of life that are saved and the value of a statistical life year (VSLY), which take into account the estimated life expectancy of those lives saved by regulation. But just as VSL ignores the length of lives saved, VSLY ignores their quality: five years spent in perfect health would be treated equivalently to five years spent in poor health. Recently, some scholars have recommended instead using quality-adjusted life years in cost–benefit analysis. QALYs rely on survey responses from the public, patients, and med­ ical professionals to estimate the impact of various health states on a person’s quality of life (Weinstein et al., 2009). QALYs were initially developed in the field of cost-effective­ Page 8 of 24

Well-Being and Public Policy ness analysis to provide data on the efficient use of scarce resources in medical decisionmaking, but some commentators—including courts and agencies—see value in the use of QALYs in CBA. As yet, however, QALY analysis faces a number of methodological hurdles before it can be successfully incorporated into CBA. The first difficulty with adopting QALY analysis as part of traditional CBA is determining how to monetize QALYs. When QALYs are used in cost-effectiveness analysis in healthcare decision-making, no effort is made to quantify the value of a QALY. But to incorporate QALYs into CBA, policymakers must place a dollar value on QALYs. As yet, however, no acceptable number has emerged (Hirth et al., 2004). More problematic is the method that researchers use to elicit QALY values. The most typical methods for valuing QALYs are surveys, in which people are asked how much they would pay for a year of life at a certain level of health. Just as with contingent valua­ tion and WTP studies, this forces people to guess how they will feel in the future, raising many of the same problems with affective forecasting errors that undermine traditional CBA. This problem is compounded by the fact that QALY studies often require healthy in­ dividuals to make value judgments about health states that they have never experienced. (p. 389)

To be valuable in welfare analysis, QALYs should reflect how people feel in various states of health. Instead, when healthy people are asked about states of poor health they will tend to provide answers about how they feel about those health states. These data do not provide a sound basis for regulatory policymaking. Because of these problems, social scientists have increasingly turned to data from hedo­ nic psychology to estimate the monetary effects of non-market goods. Unlike the studies typically used to inform CBA, happiness measures do not impose significant cognitive de­ mands on subjects. Instead of having to estimate the probability of a risk, the effect of that risk, and the amount of money that would compensate for it, people only have to an­ swer simple questions about how they are feeling and how satisfied they are with their lives (Clark and Oswald, 2002). Researchers can use hedonic data and income data from cross-sectional and longitudinal panel surveys in regression analyses that control for oth­ er demographic variables to estimate the “shadow price” of a non-market good (Powdthavee and van den Berg, 2011). For example, if researchers know the degree to which increased income improves well-being and how much an individual’s well-being is typically reduced by disability, they can estimate how much additional income a disabled person would need to make her as happy as a non-disabled person. Using this method, researchers can establish “exchange rates” between income and a wide variety of non-market goods (Dolan, Fujiwara, and Metcalfe, 2011). Multiple studies have estimated shadow prices for health states using happiness data (Ferrer-i-Carbonell and Van Praag, 2002; Groot and Van den Brink, 2004; Oswald and Powdthavee, 2008; Powdthavee and van den Berg, 2011). For example, Powdthavee and van den Berg (2011) use different well-being measures to estimate the amount of additional income necessary to compensate an individual for a variety of health conditions. They estimate that, for a person with average household income, skin conditions reduce well-being by approxi­ mately £5000 per year, while experiencing migraine headaches would require an addi­ Page 9 of 24

Well-Being and Public Policy tional £3.2 million per year to make sufferers as happy as healthy people. In another pa­ per, Oswald and Powthavee (2008) estimate the compensating income associated with the death of a family member. They range from £39,000 per year for the death of a sibling to £286,000 per year for the death of a partner. Numbers such as these could inform both CBA and torts damages awards (Sunstein, 2008). Over the past decade, researchers have conducted pricing studies of a wide variety of non-market goods (and bads) based on happiness data (van Praag and Baarsma, 2005). These include marriage and divorce (Blanchflower and Oswald, 2004; Clark and Oswald, 2002), crime (Powdthavee, 2005), terrorism (Frey et al., 2004), natural disasters (Luechinger and Raschly, 2009), and the environment (Welsch and Kühling, 2009). (p. 390) Valuations of environmental goods are particularly helpful in the context of CBA. The environment is a major subject of regulation, yet it is especially difficult to value us­ ing revealed and stated preference studies. People may not appreciate the effect of the environment on their welfare, and they may be resistant to placing explicit monetary val­ ues on things like clean air. Accordingly, hedonically derived estimates about the compen­ sating values of water, air, and noise pollution are especially promising. As with other uses of happiness data for public policy, attempts to calculate implicit prices from changes in income and subjective well-being face difficulties. Questions arise over whether the happiness data can be treated cardinally, which would produce the most helpful and easily interpretable results. In general, however, issues of cardinality are no greater for happiness data than they are for money (Bronsteen, Buccafusco, and Masur, 2013). Another issue with studies that attempt to determine implicit prices of goods based on their hedonic impact is the potential endogeneity of income and happiness. In­ come and happiness may exhibit collinearity, and increased income may produce indirect as well as direct effects on well-being. More recent research, however, has attempted to measure and control for this endogeneity (Dolan, Fujiwara, and Metcalfe, 2011). WBA and versions of CBA that use SWB data as prices are not mutually exclusive, and governments might alternatively prefer one or the other in different circumstances. Un­ like WBA, in which all of the inputs are converted into hedonically measured WBUs, CBA with SWB-based prices retain monetary figures for all goods that are generally priced in markets, including wages, consumption, and the direct costs of implementing regulations. The relative merits of the two different metrics will thus depend upon whether CBA’s use of monetization is a sufficiently accurate proxy for welfare in some cases and whether it can be done more inexpensively than WBA.

18.5 Distributionally Weighted CBA Another type of welfarist decision procedure that bears a family resemblance to WBA is distributionally weighted CBA. Like WBA, distributionally weighted CBA attempts to cure problems that arise from traditional CBA’s use of wealth as a proxy for welfare. As origi­ nally conceived by Robin Boadway (1974, 1975, 1976), distributionally weighted CBA springs from two separate insights about wealth distribution and welfare, one economic Page 10 of 24

Well-Being and Public Policy and one philosophical. The first is the well-understood idea of diminishing marginal utility of money. The more wealth an individual possesses, the less each marginal dollar means to that individual. As wealth increases, changes in wealth (increases or decreases) will have smaller effects on the individual’s welfare. That is, a millionaire should not care if she gains or loses $100, but for someone earning $10/day that sum of money could have a significant impact on welfare. Thus, equivalent amounts of money do not necessarily pro­ vide the same amount of welfare to different people. The second insight is that in a fair or just society, all changes to individual welfare might not be equal in moral terms and should not be treated equivalently (Samuelson, 1947, p. 221; Harsanyi, 1977). Imagine a society with just two individuals: Rich, who has a great deal of welfare, and Poor, who has very little welfare. A classical utilitarian would measure the welfare of that society—that is, compute the social welfare function—as the sum of the individual welfares of Rich and Poor. A gain in welfare for Rich would increase the welfare of the society as much as an equivalent gain in welfare for Poor—no more and no less. However, this strictly utilitarian formulation ignores considerations of equity and distribution of welfare. If Rich gains welfare, that increases the welfare gap between rich (p. 391)

and poor and makes the society even more inequitable. On the other hand, if Poor gains welfare, the society’s inequality diminishes. Depending on one’s normative view of social welfare, this might make the society better off. Boadway’s original insights have spawned a significant literature on the technical complexities of distributional weighting (e.g. Adler, 2013; Cowell and Gardiner, 1999; Creedy, 2006; Dasgupta and Pearce, 1972; Das­ gupta, Sen, and Marglin, 1972; Dreze and Stern, 1987), as well as a number of important critiques (Weisbach, 2015). To remedy the disconnect between wealth and welfare, distributionally weighted CBA typ­ ically incorporates two separate weighting terms: one to adjust for the different welfare values of money to richer and poorer individuals and a second to adjust for the different contributions to the social welfare function of changes in the welfare of different individu­ als. The inputs to the cost–benefit calculus, which are derived using traditional methods (WTP and contingent valuation studies), are then multiplied by these weighting terms. Ideally, economists would measure the first weighting term empirically, by determining the rate at which the marginal utility of money declines for the typical individual. The sec­ ond term is a matter of normative philosophy (Adler, 2013). A society that valued equality would use a weighting factor that was smaller for better-off individuals and larger for less well-off individuals, to reflect the judgment that changes in the welfare of individuals who are already doing very well are less important than changes in the welfare of individuals who are faring poorly. A society that valued inequality would use a weighting factor that placed greater weight on the welfare of individuals who were already doing well. A strict­ ly utilitarian society would use no weighting factor whatsoever. These two terms are often combined into a single term for any given individual, and this single term is typically re­ ferred to as the distributional weight. However, they are best understood separately, and we will consider them separately here.

Page 11 of 24

Well-Being and Public Policy What is the relationship between distributionally weighted CBA and WBA? Both tools arise from concerns about using wealth as a proxy for welfare, but their solutions to the problem are different. As we explained above, WBA incorporates two elements that are absent from traditional CBA. First, it uses hedonic data, rather than willingness-to-pay studies or contingent-valuation surveys, to value goods. And second, it measures a close proxy for actual welfare, rather than wealth. Distributionally weighted CBA does not in­ corporate either of those elements. Proponents of distributionally weighted CBA would continue to rely upon the same willingness-to-pay studies that traditional CBA (p. 392) us­ es to translate goods into monetary terms and would then add weights to correct for wealth effects. Similarly, there is no analog in WBA to the fact that distributionally weighted CBA might place different values on the welfare of particular individuals when calculating overall so­ cial welfare. Recall that a strictly utilitarian decision procedure would value the welfare of all individuals equally, while a more egalitarian society would value increases in the welfare of less well-off people more than increases in the welfare of better-off people. WBA was explicitly constructed as a utilitarian decision procedure in order to mimic CBA (Bronsteen, Buccafusco, and Masur, 2013). Yet this is not a necessary feature of WBA. That is, a decision-maker could employ WBA in conjunction with any type of social wel­ fare function, whether it is weighted to favor egalitarian outcomes or not. WBA as we originally conceived of it is not designed to be distributionally weighted in this manner. But if a decision-maker were able to resolve the difficult normative questions involved in deciding upon a weighting factor, WBA could be easily adjusted to fit the desired norma­ tive outcome. Indeed, even the original version of WBA overlaps considerably with distributionally weighted CBA in that each of them values changes in wealth differentially depending on the wealth of the affected individual. In this way, distributionally weighted CBA at least aims to capture actual changes in welfare more closely than does traditional CBA—the same goal that WBA seeks to achieve. Rather than simply measuring changes in total wealth, distributionally weighted CBA makes more of an attempt to measure changes in actual welfare. But where distributionally weighted CBA attempts to correct for flawed welfare proxies, WBA simply abandons them in favor of better ones. Distributionally weighted CBA has also been criticized on the ground that it will promote welfare less efficiently than could be done through alternative means. Critics note that the tax system is generally believed to be a more efficient mechanism of redistributing wealth than legal rules or regulations (Kaplow and Shavell, 1994). Accordingly, they ar­ gue, legal regulations should be wealth-maximizing, rather than welfare-maximizing. The tax system should then be used to redistribute wealth in whatever welfare-enhancing manner the decision-maker desires. Such critics believe that the overall effect of wealthefficient regulation plus welfare-enhancing taxes and transfers will lead to greater wealth and welfare gains than welfare-maximizing regulation alone (Weisbach, 2015). This criti­ cism implicates WBA as well, because WBA, like distributionally weighted CBA, is a wel­ farist decision procedure. However, Weisbach (2015) acknowledges that WBA might nev­ Page 12 of 24

Well-Being and Public Policy ertheless produce more accurate measures of the welfare value of various goods than CBA. As we noted above, happiness data can be used to provide more accurate prices than WTP or contingent valuation studies. Even if critics of weighted CBA are correct, WBA and its hedonic approach to welfare may well be superior to CBA. What then of this criticism? As a theoretical matter, critics of weighted CBA are likely cor­ rect to claim that it is more efficient to redistribute wealth via the tax system than (p. 393) through legal regulation. But CBA is not designed or implemented in a theoretical vacuum. The power to regulate and the power to tax and transfer do not reside within the same institution. Executive-branch agencies, under the control of the President, are the institutions with primary regulatory authority. Tax policy, by contrast, is made through legislation passed by Congress. The critical pragmatic question is thus whether Congress will actually respond to regulation by taxing and transferring in such a way as to increase social welfare (Fennell and McAdams, 2016). Critics of distributionally weighted CBA point to the vast number of changes that are made to the tax code every year (Weisbach, 2015). Yet the rate of change, by itself, pro­ vides no information on whether those changes are social welfare enhancing or diminish­ ing, and whether they redistribute wealth in an equitable or inequitable fashion. To our knowledge there has been no systematic study of the multitude of tax changes. But it is at least suggestive that overall inequality in the United States has been steadily increasing for much of the twentieth century, including most of the past four decades. If Congress is attempting to redistribute through the tax system, it is doing a noticeably poor job. If Congress cannot necessarily be counted upon to enact welfare-maximizing taxes, this leaves open the question of what an executive-branch regulator should seek to maximize. She could employ standard CBA and attempt to maximize social wealth. She could em­ ploy WBA and attempt to maximize social welfare. Or she could employ distributionally weighted CBA and attempt to maximize some combination of wealth, welfare, and equali­ ty. Absent any reason to believe that Congress is more likely than not to tax and transfer effectively, it is hard to find a rationale for preferring unweighted CBA. At minimum, it seems obvious that any regulatory agency should consider not only the welfare effects of its own regulations but the combined welfare effects of its regulations and whatever tax policy it expects the taxing authority to implement (Adler, 2013). Weisbach (2015) pursues this critique one step further in arguing that proponents of weighted CBA need a theory of government to explain why the President, and not just Congress, should be involved in maximizing social welfare. Weisbach argues that there is no reason to believe that the President is any likelier to arrive at the welfare-maximizing outcomes than Congress, and thus no reason for the President to assume this authority. But this critique misunderstands both the existing structure of government and the role of welfare maximization in regulation. When the executive branch regulates, it is expect­ ed to do so in the public interest. The statutes empowering executive-branch agencies to regulate the environment, health, and safety certainly do not prohibit those agencies from considering the welfare effects of their regulations. To the contrary, they are generally ex­ Page 13 of 24

Well-Being and Public Policy pected to do so. It would not serve the public interest—and thus not comport with the executive’s congressionally mandated mission—to regulate in a way that diminished so­ cial welfare. An agency decision to maximize welfare is more, not less, in keeping with its statutory mission and the theory that underpins its delegation.

18.6 Happiness-Based Alternatives to Gross Domestic Product (p. 394)

Economic tools are used not only to analyze policies but also to measure the success and quality of nations. And just as SWB data can be used to replace monetary prices in the context of policy analysis, it can also be so used in the context of monitoring national well-being. As we will explain, this process relies on precisely the same principles as does WBA. The difference is merely that calculating national well-being is simpler in certain respects than is calculating the effect of policies on well-being. For example, policy analy­ sis requires comparing things that are difficult to commensurate, whereas national moni­ toring requires only a summing of individual well-being. This summing is traditionally achieved via the proxies of production and consumption, but it can be achieved more di­ rectly via data on happiness. The most common traditional economic measure of national success is Gross Domestic Product (GDP), the total market value of the goods and services produced within a nation’s borders in one year. Although it is not calculated for the specific purpose of ana­ lyzing or crafting public policy, it is nonetheless watched closely so that if it drops, “spe­ cific policies may then be developed to ensure it rises again” (Dolan, Layard, and Met­ calfe, 2011, p. 2). Despite its ubiquity and importance, GDP has always been controver­ sial because, among other things, it measures economic productivity rather than quality of life. Nearly fifty years ago, Robert Kennedy said that it “measures everything … except that which makes life worthwhile” (Kennedy, 1968).5 Not long after Kennedy’s speech, the nation of Bhutan announced that it would pursue a measure of Gross National Happiness in order to address the concerns with GDP’s limita­ tions (Krueger, 2009, p. 1). But outside Bhutan, only recently has the idea been discussed with the serious goal of implementation. The first notable effort in this direction came from the Dutch social scientist Ruut Veenhoven (1996). Veenhoven argued that economic indicators like GDP, and also social indicators like the Human Development Index, mea­ sure primarily means of achieving quality of life instead of life’s quality itself. To remedy the problem, Veenhoven proposed an alternative to GDP called Happy Life Expectancy (HLE). Specifically, the average quality of life in a nation would be calculated by multiply­ ing the average life expectancy of its citizens by their average happiness as measured by survey data on subjective well-being. The survey data could be either people’s responses to a single question about their happiness, a single question about their life satisfaction, or their “Affect Balance.” Affect Balance, which Veenhoven calls 1996 (p. 395) “the best

Page 14 of 24

Well-Being and Public Policy choice” (1996, p. 32), is measured by the “occurrence of specific positive and negative af­ fects in the last few weeks” (1996, p. 22). Veenhoven’s idea of using subjective well-being to create an alternative to GDP gained lit­ tle political traction until 2009, when France’s then-President Nicolas Sarkozy endorsed it, saying: “It is time for our statistics system to put more emphasis on measuring the well-being of the population than on economic production” (Samuel, 2009). Prompted by Sarkozy’s support, a group of renowned economists led by Joseph Stiglitz were commis­ sioned to assess whether alternatives to GDP are desirable and feasible (Stiglitz, Sen, and Fitoussi, 2009). The Stiglitz Commission produced an influential report that made three main contributions. The first was to suggest ways to improve GDP itself; the second was to propose methods for measuring quality of life more directly; and the third was to dis­ cuss sustainable development and environmental concerns. Our focus here is on the sec­ ond of these contributions: measuring quality of life. The Commission noted that quality of life could be measured by data on subjective wellbeing; by capabilities such as objective measures of health and education, among many other things; or by “fair allocations,” an approach that asks people to place their own val­ ues on different non-market elements of the quality of life (Stiglitz, Sen, and Fitoussi, 2009, pp. 42, 145–155). Within the category of subjective well-being, the Commission stressed the importance of separating three distinct ways of measuring SWB. The first is life satisfaction, the second is the presence of positive moment-by-moment affect, and the third is the absence of negative moment-by-moment affect (Stiglitz, Sen, and Fitoussi, 2009, p. 146). The Commission recognized that pursuing a single-component measure of life quality would involve value judgments and might well miss important nuances (Stiglitz, Sen, and Fitoussi, 2009, p. 207). But it also acknowledged that “the influence of GDP” may be “proof that such indices are essential” (Stiglitz, Sen, and Fitoussi, 2009). One option it gave for creating such a measure focuses on hedonic experience: “While this approach may still be considered as putting a domain—hedonic experiences—at the forefront, it may also be regarded as providing a way of weighting different experiences through a common yardstick: the intensity of the hedonic experiences that they generate” (Stiglitz, Sen, and Fitoussi, 2009, p. 211). Specifically, the Commission suggested that a metric known as the U-Index could be used to create a GDP-style measure of a nation’s hedonic well-being (Stiglitz, Sen, and Fitous­ si, 2009, p. 212). The U-Index is a metric created by Alan Krueger, Daniel Kahneman, and several co-authors that “measures the proportion of time an individual spends in an un­ pleasant state” (Krueger et al., 2009, pp. 9, 18). Using self-assessments of affect, the in­ dex divides a person’s time into two categories—positive and negative—based on the strongest emotion a person reports while doing the activity she was engaged in during that time.

Page 15 of 24

Well-Being and Public Policy The binary nature of the U-Index addresses a major concern about subjective well-being: the concern that different people may use differently the numerical scale by which they report how they feel. One person’s reported “8” out of 10 may not mean the same thing as another person’s 8. Because the U-Index relies on an ordinal measure rather than a (p. 396) cardinal one, such a problem is eliminated. Still, this strength is also a weakness. As Richard Layard has noted (2009, pp. 145, 146–151), and as the creators of the U-Index have acknowledged (Krueger et al., 2009, p. 20), well-being itself is cardinal and non-bi­ nary: two activities may each be unpleasant (or pleasant), but one may be substantially more unpleasant (or pleasant) than the other one. In our view, this makes the U-Index sig­ nificantly less desirable for use in WBA than simpler SWB measures such as cardinal data on affect and life satisfaction. Moreover, Layard argues that the danger of differential scale usage is overstated because the survey data behave, when subjected to statistical analysis, in an orderly way that would be “inconceivable with a … scale that varies widely across individuals” (2009, p. 149). Nevertheless, the U-Index may still make sense for use in GDP-style measures. The rea­ son is that such measures are often used for comparing different countries (Krueger et al., 2009, pp. 69–77), and there is evidence that scale usage may vary across countries, perhaps due to differences in culture or even language. For example, the French appear to use the top end of the scale less than do Americans (Krueger et al., 2009, pp. 69–77), a problem that the U-Index is uniquely suited to overcoming. In addition, despite the fact that the U-Index does not measure degrees of pleasantness or unpleasantness, it does “take[] on useful cardinal properties” due to its aggregation of time (Krueger et al., 2009, p. 20). In other words, although it does not say how unpleasant an episode is, it says something meaningful about how unpleasant a day or a life is by showing how much of the day or the life is spent in experienced unpleasantness. The upshot of these strengths and weaknesses is that the U-Index is a useful tool that may be most fitting for measures such as GDP that involve inter-country comparisons. The Stiglitz Commission praised the U-Index not only because it solves the problem of dif­ ferential scale usage, but also because it “naturally focuses attention on the most miser­ able in society” (Stiglitz, Sen, and Fitoussi, 2009, p. 212). This is a theme in the literature on using subjective well-being measures to create a GDP-like indicator for the quality of life. Veenhoven, the first to try to create such a measure, more recently suggested a way to make the measure sensitive to inequality in happiness (Veenhoven and Kalmijn, 2005). His measure, Inequality-Adjusted Happiness (IAH), gives equal weight to the amount of happiness in a country and to the extent to which that happiness is evenly distributed (Veenhoven and Kalmijn, 2005, p. 428), resulting in a composite score for each country between 0 and 100. Malta, Denmark, and Switzerland scored very highly, whereas sever­ al African and Eastern European countries scored the lowest, and the United States was toward the high-middle of the list (Veenhoven and Kalmijn, 2005, p. 430). Interestingly, these results have many similarities to—but also notable differences from—the ranking of countries by Gini coefficients, which measure income inequality (CIA, 2014). Countries like Malta, Denmark, and Switzerland do very well by either measure and several African countries do very poorly, but the United States does much better on the IAH measure Page 16 of 24

Well-Being and Public Policy than on the measure of pure income inequality, whereas Eastern European countries tend to have the opposite results. In 2011, Paul Dolan and Robert Metcalfe followed up on the Stiglitz Commission report by surveying British citizens’ opinions of the extent to which government policy should aim to improve subjective well-being (2011). What they found was that people (p. 397) have “a strong preference for targeting the SWB of the worst-off as opposed to maximis­ ing the sum total of SWB in society” (2011, p. 50). Reducing suffering was considered the most important aim of policy (2011, pp. 48–49). Dolan and Metcalfe thus recommended, among other things, that Britain “measure more about the experiences of daily lives, es­ pecially negative affect” (2011, p. 49), and that it “estimate efficiency-equality tradeoffs” (2011, p. 50) so as to compare the value of helping the least happy people with the value of maximizing overall happiness. The current state of the literature on this matter is that more data, especially on negative affect and (perhaps to a slightly lesser degree) on positive affect, should be collected. As the Stiglitz Commission wrote: [T]he overwhelming conclusion that we derive from existing analyses of various aspects of subjective well-being is that these measures tap into QoL [quality of life] in meaningful ways …. The type of questions that have proved their value within small-scale, unofficial surveys should start being included in larger-scale surveys undertaken by official statistical offices. (Stiglitz, Sen, and Fitoussi, 2009, p. 151) Once collected, the data can be used to create GDP alternatives based either on the U-In­ dex, Veenhoven’s HLE or IAH, or other measures along those lines. We think that the best measure of a nation’s aggregate welfare—putting to the side normative issues about the possible wrongfulness of inequality—would be one like HLE that simply adds together all individuals’ subjective well-being throughout their lifetimes, preferably defining well-be­ ing as affect and measuring it via ESM or DRM. This is the same way that WBA calculates the costs and benefits of a regulation.

18.7 Conclusion The main targets of policy analysis and of assessments of national success are the overall quality of human life and the way in which that quality is distributed among people. Tra­ ditional economic measures such as Gross Domestic Product and cost–benefit analysis have been criticized for ignoring distributional considerations and for being only weak proxies for quality of life. New advances in the study of subjective well-being or happi­ ness have made it possible to create alternative devices for monitoring national success and for analyzing prospective policies.

Page 17 of 24

Well-Being and Public Policy We proposed well-being analysis as a method of policy assessment that holds several ad­ vantages over traditional economic measures. The chief advantage of WBA is that its data come from asking people a question they can answer relatively easily and credibly: how do you feel right now?6 By contrast, CBA relies on the (p. 398) assumption—contradicted by psychological evidence—that people accurately predict how much something will af­ fect their well-being when they say how much money they would pay or accept for it. WBA is the latest step in a progression toward using SWB data in policy analysis. Inter­ mediate steps included using SWB data to determine “prices” for non-market goods, and weighting CBA to account for the distribution of wealth on welfare and equality. Although we see both of these as improvements upon traditional economic analysis, neither pos­ sesses all the virtues of WBA. Finally, the idea behind WBA is the same as the idea that has lately been animating the project of replacing GDP with a measure based on SWB data. The simplest approach— adding every citizen’s lifetime SWB— mimics one of the steps of how WBA assesses prospective policies. Self-reports of happiness are the most valid proxies for the quality of human life. As such, it makes sense to use them as replacements for traditional economic indicators in both policy analysis and monitoring of national success. WBA is our contribution to that project.

Bibliography Adler, Matthew (2013). “Cost–Benefit Analysis and Distributional Weights: An Overview.” Review of Environmental Economics & Policy (2016) 10(2): 246–285. Blanchflower, Douglas G. and Andrew J. Oswald (2004). “Well-being over time in Britain and the USA.” Journal of Public Economics 88: 1359–1386. Boadway, Robin (1974). “The Welfare Foundations of Cost–Benefit Analysis.” Economic Journal 84: 926–939. Boadway, Robin (1975). “Cost–Benefit Rules in General Equilibrium.” Review of Economic Studies 42: 361–374. Boadway, Robin (1976). “Integrating Equity and Efficiency in Applied Welfare Econom­ ics.” Quarterly Journal of Economics 90: 541–556. Boadway, Robin and Neil Bruce (1984). Welfare Economics. Oxford: Basil Blackwell. Bronsteen, John, Christopher Buccafusco, and Jonathan S. Masur (2013). “Well-Being Analysis vs. Cost–Benefit Analysis.” Duke Law Journal 62: 1603–1689. Bronsteen, John, Christopher Buccafusco, and Jonathan S. Masur (2015). Happiness and the Law. Chicago: University of Chicago Press.

Page 18 of 24

Well-Being and Public Policy Central Intelligence Agency (CIA). The World Factbook: Country Comparison, Distribution of Family Income—Gini Index, available at . Clark, Andrew E. and Andrew J. Oswald (2002). “A Simple Statistical Method for Measur­ ing How Life Events Affect Happiness.” International Journal of Epidemiology 31: 1139– 1144. Clark, Andrew E., Ed Diener, Yannis Georgellis, and Richard Lucas (2008). “Lags and Leads in Life Satisfaction: A Test of the Baseline Hypothesis.” Economic Journal 118: F222–F243. Cohen, Roger (2011). “The Happynomics of Life.” New York Times, Mar. 13, 2011, at 12. Colander, David (2007). “Edgeworth’s Hedonimeter and the Quest to Measure Utility.” Journal of Economic Perspectives 21: 215–226. Cowell, Frank A. and Karen Gardiner (1999). “Welfare Weights.” STICERD, Lon­ don School of Economics, available at . Creedy, John (2006). “Evaluating Policy: Welfare Weights and Value Judgments.” Universi­ ty of Melbourne Dept. of Economics Working Paper No. 971, available at . Dasgupta, Ajit K. and D. W. Pearce (1972). Cost–Benefit Analysis: Theory and Practice. New York: Harper and Row. Dasgupta, Partha, Amartya Sen, and Stephen Marglin (1972). Guidelines for Project Eval­ uation. Vienna: United Nations Industrial Development Organization. Desvousges, William H., F. Reed Johnson, Richard W. Dunford, Sara P. Hudson, Nicole Wil­ son, and Kevin J. Boyle (1993). “Measuring Natural Resource Damages with Contingent Valuation: Tests of Validity and Reliability,” in Jerry A. Hausman, ed., Contingent Valua­ tion: A Critical Assessment, 91–164. Amsterdam: North-Holland. Diener, Ed, Richard Lucas, Ulrich Schimmack, and John Helliwell (2009). Well-Being for Public Policy. New York: Oxford University Press. Dolan, Paul, Daniel Fujiwara, and Robert Metcalfe (2011). “A Step Towards Valuing Utility the Marginal and Cardinal Way.” CEP Discussion Paper No. 1062. Dolan, Paul, Richard Layard, and Robert Metcalfe (2011). Measuring Subjective Well-Be­ ing for Public Policy: Recommendations on Measures. London: Center for Economic Per­ formance.

Page 19 of 24

Well-Being and Public Policy Dolan, Paul and Robert Metcalfe (2011). “Comparing Measures of Subjective Well-Being and Views About the Role They Should Play in Policy.” London: Office for National Statis­ tics. Available at . Dreze, Jean and Nicholas Stern (1987). “The Theory of Cost–Benefit Analysis,” in Alan J. Auerbach and Martin Feldstein, eds., Handbook of Public Economics, Vol. II, 909–989. Amsterdam: North-Holland. Fennell, Lee Anne and Richard H. McAdams (2016). “The Distributive Deficit in Law and Economics.” Minnesota Law Review 100: 1051–1127. Ferrer-i-Carbonell, Ada and Bernard M. S. Van Praag (2002). “The Subjective Costs of Health Losses due to Chronic Diseases: An Alternative Model for Monetary Appraisal.” Health Economics 11: 709–722. Frank, Robert H. (1997). “The Frame of Reference as a Public Good.” Economic Journal 107: 1832–1847. Frederick, Shane and George Loewenstein (1999). “Hedonic Adaptation,” in Daniel Kah­ neman, Ed Diener, and Norbert Schwarz, eds., Well-Being: the Foundations of Hedonic Psychology, 302–329. New York: Russell Sage Foundation. Frey, Bruno S., Simon Luechinger, and Alois Stutzer (2004). “The life satisfaction ap­ proach to valuing public goods: the case of terrorism.” Public Choice 138: 317–345. General Assembly Resolution 65/309, at 1, United Nations Document A/RES/65/309 (July 19, 2011). Gilbert, Daniel T. and Timothy D. Wilson (2007). “Prospection: Experiencing the Future.” Science 317: 1351–1354. Graham, Carol (2016). “Subjective Well-Being in Economics,” in Matthew D. Adler and Marc Fleurbaey, eds., The Oxford Handbook of Well-Being and Public Policy. Oxford: Ox­ ford University Press. Groot, Wim and Henriëtte Maassen van den Brink (2004). “Money for health: the equiva­ lent variation of cardiovascular diseases.” Health Economics 13: 859–872. (p. 400)

Hammitt, James K. (2007). “Valuing Changes in Mortality Risk: Lives Saved Ver­

sus Life Years Saved.” Review Environmental Economics and Policy 1: 228–240. Harsanyi, John (1977). Rational Behavior and Bargaining Equilibrium in Games and So­ cial Situations. Cambridge: Cambridge University Press. Hirth, Richard A., Michael E. Chernew, Edward Miller, A. Mark Fendrick, and William G. Weissert (2004). “Willingness To Pay for a Quality-Adjusted Life Year: In Search of a Stan­ dard.” Medical Decision Making 20: 332–342.

Page 20 of 24

Well-Being and Public Policy Kahneman, Daniel, Alan B. Krueger, David A. Schkade, Norbert Schwarz, and Arthur A. Stone (2004). “A Survey Method for Characterizing Daily Life Experience: The Day Recon­ struction Method.” Science 306: 1776–1780. Kahneman, Daniel, Peter P. Walker, and Rakesh Sarin (1997). “Back to Bentham?: Explori­ ations of Experienced Utility.” Quarterly Journal of Eocnomics 112: 375–405. Kaplow, Louis and Steven Shavell (1994). “Why the Legal System Is Less Efficient Than the Income Tax in Redistributing Income.” Journal of Legal Studies 23: 667–681. Kennedy, Robert (1968). “Remarks of Robert F. Kennedy at the University of Kansas,” available at . Krueger, Alan B. (2009). “Introduction and Overview,” in Alan B. Krueger, ed., Measuring the Subjective Well-being of Nations, 1–7. Chicago: University of Chicago Press. Krueger, Alan B., Daniel Kahneman, David Schkade, Norbert Schwarz, and Arthur A. Stone (2009). “National Time Accounting: The Currency of Life,” in Alan B. Krueger, ed., Measuring the Subjective Well-Being of Nations, 9–81. Chicago: University of Chicago Press. Lucas, Richard. E., Yannis Georgellis, Andrew C. Clark, and Ed Diener (2003). “Reexamin­ ing Adaptation and the Set Point Model of Happiness: Reactions to Changes in Marital Status.” Journal of Personality and Social Psychology 84: 527–539. Lucas, Richard E., Andrew E. Clark, Yannis Georgellis, and Ed Diener (2004). “Unemploy­ ment Alters the Set Point for Life Satisfaction.” Psychological Science 15: 8–13. Luechinger, Simon and Paul A. Raschly (2009). “Valuing flood disasters using the life sat­ isfaction approach.” Economic Journal 119: 482–515. Oswald, Andrew J. and Nattavudh Powdthavee (2005). “Does Happiness Adapt? A Longi­ tudinal Study of Disability with Implications for Economists and Judges.” Journal of Public Economics 92: 1061–1077. Oswald, Andrew J. and Nattavudh Powdthavee (2008). “Death, Happiness, and the Calcu­ lation of Compensatory Damages.” Journal of Legal Studies 37: S217–S251. Pavot, William and Ed Diener (1993). “Review of the Satisfaction with Life Scale.” Psycho­ logical Assessment 5: 164–172. Powdthavee, Nattavudh (2005). “Unhappiness and Crime: Evidence from South Africa.” Economica 72: 531–547.

Page 21 of 24

Well-Being and Public Policy Powdthavee, Nattavudh and Bernard van den Berg (2011). “Putting Different Price Tags on the Same Health Condition: Re-evaluating the Well-being Valuation Approach.” Journal of Health Economics 30: 1032–1043. Samuel, Henry (2009). “Nicolas Sarkozy Wants to Measure Economic Success in ‘Happi­ ness.’” Telegraph, Sept. 14, 2009, available at . Samuelson, Paul A. (1947). Foundations of Economic Analysis. Cambridge, MA: Harvard University Press. Stiglitz, Joseph E., Amartya Sen, and Jean-Paul Fitoussi (2009). Report by the Commission on the Measurement of Economic Performances and Social Progress. Paris: The Commission. (p. 401)

Sunstein, Cass R. (2002). “Probability Neglect: Emotions, Worst Cases, and Law.” Yale Law Journal 112: 61–107. Sunstein, Cass R. (2004). “Lives, Life-Years, and Willingness To Pay.” Columbia Law Re­ view 104: 205–252. Sunstein, Cass R. (2007). “Willingness to Pay vs. Welfare.” Harvard Law and Policy Re­ view 1: 303–330. Sunstein, Cass R. (2008). “Illusory Losses.” Journal of Legal Studies 37: S157–S194. Sunstein, Cass R. (2016). “Cost-Benefit Analysis, Who’s Your Daddy?” Journal of Cost-Ben­ efit Analysis 7: 107–120. Ubel, Peter A., George Loewenstein, John Hershey, Jonathan Baron, Tara Mohr, David A. Asch, and Christopher Jepson (2001). “Do Nonpatients Underestimate the Quality of Life Associated with Chronic Health Conditions Because of a Focusing Illusion?” Medical Deci­ sion Making 21: 190–199. Ubel, Peter A., George Loewenstein, and Christopher Jepson (2005). “Disability and Sun­ shine: Can Hedonic Predictions Be Improved by Drawing Attention to Focusing Illusions or Emotional Adaptation?” Journal of Experimental Psychology: Applied 11: 111–123. van Praag, Bernard M. S. and Barbara E. Baarsma (2005). “Using Happiness Surveys to Value Intangibles.” Economic Journal 115: 224–246. Veenhoven, Ruut (1996). “Happy Life Expectancy: A Comprehensive Measure of Qualityof-Life in Nations.” Social Indicators Research 39: 1–58. Veenhoven, Ruut and Wim Kalmijn (2005). “Inequality-Adjusted Happiness in Nations.” Journal of Happiness Studies 6: 421–455. Weinstein, Milton C. et al. (2009). “QALYs: The Basics.” Value in Health 12: S5–S9. Page 22 of 24

Well-Being and Public Policy Weisbach, David (2015). “Distributionally-Weighted Cost Benefit Analysis: Welfare Eco­ nomics Meets Organizational Design.” Journal of Legal Analysis 7: 151–182. Welsch, Heinz and Jan Kühling (2009). “Using Happiness Data for Environmental Valua­ tion: Issues and Applications.” Journal of Economic Surveys 23: 385–406. Wilson, Timothy D. and Daniel T. Gilbert (2005). “Affective Forecasting: Knowing What to Want.” Current Directions in Psychological Science 14: 131–134.

Notes: (1) This chapter borrows directly in some instances from other recent work of ours, espe­ cially Bronsteen, Buccafusco, and Masur, 2013 and 2015. (2) Some of these difficulties can be minimized with the day reconstruction method (DRM) pioneered by Daniel Kahneman and his colleagues (Kahneman et al., 2004). DRM uses daily diary entries about each day’s experiences to reconstruct an account of subjects’ emotional lives. DRM studies correlate strongly with ESM studies and can be run at low­ er cost. (3) In addition, the U-Index proposed by Krueger et al. is designed to mitigate differences in scale usage (Krueger et al., 2009). (4) Exec. Order No. 12,866, 3 C.F.R. 638. (5) Kennedy was referring to the similar measure of Gross National Product (GNP), but the points he makes in his speech are clearly meant to apply equivalently to GDP. The dif­ ference is that GNP counts all goods and services owned by a nation or its citizens, in­ cluding those produced outside the nation’s borders; whereas GDP measures all goods and services produced within a nation’s borders, including those that are foreign-owned. (6) Even the evaluative question of “How satisfied are you with your life?” is a far more valid gauge of well-being than is the assumption that someone uses money efficiently to increase his or her well-being.

John Bronsteen

John Bronsteen, Associate Professor, Loyola University Chicago School of Law. Christopher Buccafusco

Christopher Buccafusco, Assistant Professor of Law and Co-Director of the Center for Empirical Studies of Intellectual Property, IIT Chicago-Kent College of Law. Jonathan Masur

Jonathan Masur, Assistant Professor of Law, University of Chicago Law School.

Page 23 of 24

Value-Driven Behavior and the Law

Value-Driven Behavior and the Law   Tom R. Tyler The Oxford Handbook of Law and Economics: Volume 1: Methodology and Concepts Edited by Francesco Parisi Print Publication Date: Apr 2017 Subject: Economics and Finance, Law and Economics, Econometrics, Experimental and Quanti­ tative Methods Online Publication Date: May 2017 DOI: 10.1093/oxfordhb/9780199684267.013.030

Abstract and Keywords This article discusses an alternative approach to gaining compliance with the law. The ap­ proach involves motivating people through appeals to their values. Values reflect people's assessments of what is right or appropriate to do in a given situation; this involves people's feelings of obligation and responsibility to others. There are two arguments for value-based motivation. First, we gain the benefits of a value-based approach, e.g. in­ creasing voluntary cooperation. Second, we avoid the problems associated with instru­ mental approaches. To gain these advantages we need to move to a system in which val­ ue-based motivations are the primary motivation tapped, and instrumental motivations are the backup for a small group that have to be dealt with instrumentally because they are unable or unwilling to act on their values. Keywords: law, legal authority, values, value-based motivations, obligation, self-regulation

19.1 Introduction THE classic focus of the law is upon obtaining compliance with the law and with the deci­ sions of legal authorities through the threat or use of sanctions. This fits well into a framework of law in which the rules are created by legal authorities and their primary goal when dealing with the community is to secure public compliance. Traditional exami­ nations of the relationship between communities and legal authorities have emphasized the importance of having public compliance with laws and the decisions of duly constitut­ ed legal authorities. This view is based upon a model of having centralized authorities de­ termine legal policies and practices using their training and expertise, that is, a profes­ sionalized model for the administration of the police and the courts. Emphasis is on pro­ fessionalism in legal authority that is “insular, homogenous, and largely autonomous” and “purposely distanced” from communities (Sklansky, 2011).

Page 1 of 23

Value-Driven Behavior and the Law The goal in a professionalized model of legal authority is to motivate the type of public be­ havior most closely associated with a command and control model: namely, compliance. In the case of the law, Fuller argues that legal authorities “must be able to anticipate that the citizenry as a whole will … generally observe the body of rules” created by judicial au­ thorities (1969, p. 201). In the case of the police, Kelling and Moore (1988) note: “The proper role of citizens in crime control [is as] relatively passive recipients of professional crime control services,” with which they are expected to comply. While control over resources can produce changes in behavior based solely on the posses­ sion of power, this model of authority is costly and inefficient. The use of power, particu­ larly coercive power, is shown by research to require a large expenditure of resources to obtain modest and limited amounts of influence over others. Those resources are re­ quired to create a credible system of surveillance through which authorities monitor pub­ lic behavior to punish rule violators. In addition, when incentives are (p. 403) being used to reward people for acting in ways that benefit the authorities and the community, re­ sources must be available. Recent empirical research suggests that these strategies of governance can be success­ ful. For example, recent research suggests that deterrence strategies can shape crime-re­ lated behavior (Blumstein, Cohen, and Nagin, 1978; Nagin, 1998), but that its influence is usually small in magnitude and in many cases non-existent (Bottoms and Von Hirsch, 2010; Paternoster, 2006; Wright et al., 2004). In a review of the literature on American drug use, for example, a study of existing re­ search found that only approximately 5% of the variance in citizen drug use can be ex­ plained by citizen judgments of the likelihood of being caught and punished by the police and courts (MacCoun, 1993). This conclusion is typical of the findings of studies of com­ pliance with the law—deterrence is found to have, at best, a small influence on people’s behavior. General reviews of empirical deterrence research conclude that the relationship between risk judgments and crime is “modest to negligible” (Pratt et al., 2008) and that the “perceived certainty [of punishment] plays virtually no role in explaining deviant/ criminal conduct (Paternoster, 1987).” According to Piquero et al., a review of the litera­ ture results in “ … some studies finding that punishment weakens compliance, some find­ ing that sanctions have no effect on compliance, and some finding that the effect of sanc­ tions depends on moderating factors” (2011, p. 1; see also Paternoster, 2006). Even stud­ ies on the most severe form of punishment—the death penalty—suggest that the argu­ ment that capital punishment deters crime “still lacks clear proof” (Weisberg, 2005; see also The National Academy of Science report, 2012). Studies on punishment suggest that it is not only an ineffective deterrent for society at large, but it also minimally, if at all, effective in deterring the future criminal conduct of those being punished. The widespread punishment for minor crimes does not generally lower the rate of subsequent criminal behavior, as models of specific deterrence would predict (Harcourt, 2001; Harcourt and Ludwig, 2006). Similarly, studies of more severe punishment like imprisonment report that more severe punishments are unrelated to low­ Page 2 of 23

Value-Driven Behavior and the Law er rates of future criminality (Lipsey and Cullen, 2007). In fact, studies of juveniles sug­ gest that incarceration actually increases the likelihood of later criminality (McCord, Widom, and Crowell, 2001). While the previously mentioned studies refer to the effectiveness of deterrence and sanc­ tioning generally, studies specifically focused on issues of white-collar crime yield similar results. Studies in work settings, for example, also support this argument (Huselid, 1995; Jenkins et al., 1998). Braithwaite and Makkai studied compliance among nursing home executives and concluded that there were no significant deterrence effects (1991). Exper­ iments on decisions by businesses about whether to engage in fraud similarly did not find deterrence effects (Jesilow, Geis, and O’Brien, 1985, 1986). Tyler examined the influence of deterrence on the rule-following behavior of employees and found minimal deterrence effects(2011; Tyler and Blader, 2000, 2005). Reviewing this evidence, Simpson described evidence supporting the existence of deterrent effects in corporate settings as “equivo­ cal” (2002, p. 42). As noted, sanctions do not have a particularly (p. 404) powerful impact, although researchers often note that it is difficult to actually monitor behavior (Langevoort, 2002). In addition to punishment, researchers have also considered the role of incentives in or­ ganizational contexts (Tyler and Blader, 2005). Based upon their workplace-based study, Tyler and Blader estimate that around 10% of the variance in employee behavior is shaped by incentives in the work environment (2000) (see also Podsakoff et al., 2006). These results suggest that, while they are somewhat effective, incentive systems also only have a limited impact on employee behavior. In recent years, the limits of the utilitarian command and control model, which seeks to implement regulations through sanctions and incentives, have been emphasized in work settings (Katyal, 1997; Markell, 2000; Sutinen and Kuperan, 1999). In legal literature on government regulation, skepticism surrounding command and control strategies has led to the flourishing of market-based models of regulation that emphasize economic incen­ tive systems (Stewart, 2003). The same research shows that such instrumental influences come at high material costs. This leaves societies vulnerable, because disruptions in the control of resources brought on by periods of scarcity or conflict quickly lead to the collapse of effective social order when that social order is only based upon power. It is precisely in times of economic crisis that legal authorities both most need the support of those they seek to regulate and are least able to either provide incentives or effectively enforce sanctions. The concept of a company “too big to fail” illustrates the problems of punishing organizations that are vital to resolving an economic crisis. In addition to resource concerns, there are several social costs to a society that relies on punishment to deter non-compliance. First, if people comply with the law only in re­ sponse to coercive power, they will be less likely to obey the law in the future because acting in response to external pressures diminishes internal motivations to engage in so­ cially desirable behavior (Tyler and Blader, 2000). This follows from the well-known dis­ Page 3 of 23

Value-Driven Behavior and the Law tinction in social psychology between intrinsic and extrinsic motivation. Research on in­ trinsic versus extrinsic motivation shows that when people are motivated solely by the prospect of obtaining external rewards and punishments (i.e. extrinsic motivation), they become less likely to perform the desired behavior in the absence of these environmental reinforcements (e.g. Deci and Ryan, 1975). Economists refer to this phenomenon as “crowding out” non-instrumental motivations. Second, the use of sanctions undermines value-based motivations because it sends a mes­ sage to the potential targets of the sanctions that the authorities view them as untrust­ worthy and suspect. As a result, people become more suspicious and less trusting of the law and legal authorities. This is a particular issue with the internal cultures of organiza­ tions. If a company communicates an atmosphere of surveillance and sanctioning, they are communicating mistrust, which undermines identification with the company and the willingness to engage in self-regulation (Tyler and Blader, 2000, 2005). At the same time the authorities, whether within a company or in government, never de­ velop any basis for trusting the people over whom they exercise authority. The act of mon­ itoring in and of itself prevents authorities from having confidence that people have internal values that would lead them to follow rules when not being watched. Hence, a surveillance strategy sows the seeds of long-term surveillance because it does not provide a way to create the basis for trust. The way to determine if a person can be trusted to self-regulate is to allow him or her to act in unmonitored situations and see what pattern the behavior reveals. (p. 405)

The instrumental model has been the dominant model of legal authority for the last sever­ al decades. Over time its limits have become more apparent. However, authorities fre­ quently regard it as the only available approach and hence use it with a full awareness of the issues raised above. The key question is whether there is a viable alternative model.

19.2 Values An alternative approach to gaining compliance involves motivating people through ap­ peals to their values. Values reflect people’s assessments of what is right or appropriate to do in a given situation. This involves people’s feelings of obligation and responsibility to others. One example of an obligation-based judgment is the assessment of the legitima­ cy of a set of rules or authorities (Tyler, 2006a; Tyler and Huo, 2002) and work organiza­ tions (Tyler, 2005; Tyler and Blader, 2005). Obligation is often studied in the context of rule following, but people can also be motivated by their feelings of obligation to perform well on the job, as well as to be loyal to their firm. A second value is morality, for example the motivation to behave in accord with one’s sense of what is appropriate and right to do in a given situation. If people believe that the rules are consistent with their moral values, they are more likely to voluntarily defer to them. Studies find that both legitimacy and morality influence law-related behavior. Hence, legitimacy and morality are two parallel motivational forces shaping people’s relationship to rules and authorities. To the degree that authorities are legitimate and also to the extent that they act in ways that are consis­ Page 4 of 23

Value-Driven Behavior and the Law tent with people’s moral values, those authorities are better able to secure compliance from people within groups, organizations, and societies.

19.2.1 What Is a Value-Based Motivation? This volume distinguishes “value-based” motivations from instrumental motivations. As has been noted, instrumental motivations reflect people’s desire to gain material re­ sources and to avoid material losses. Value-based motives, as discussed by psychologists, differ in that they are motivations that flow from within the person. There are four ways to distinguish between instrumental and value-based motivations. The first is by the content of the concerns that people express within each domain. Instru­ mental concerns focus on the potential for material gains and/or the possibility of materi­ al losses. Such gains and losses involve gains in terms of rewards, or losses in terms of costs or punishments. In contrast, value-based motivations are linked to gains (p. 406) and losses of a non-material nature. Such gains and losses are linked to issues such as person­ al identity and consistency with ethical/moral values. Second, indicators of value-based motivations should be found to be empirically distinct from indicators of material gain or loss. For example, in the literature on social justice, it is found that people distinguish between receiving a favorable outcome and receiving fair treatment (Tyler et al., 1997). Hence, judgments about justice are found to be distinct from the favorability of one’s outcomes. This distinction is clear in the literature on dis­ tributive justice, a literature in which the fairness of outcomes is distinguished from their desirability (Walster, Walster, and Berscheid, 1978). It is even clearer with the literature on procedural justice, which focuses on the fairness of the procedures by which allocation decisions are made (Lind and Tyler, 1988). If people simply viewed a favorable outcome as fair, for example, value-based motivations would not be distinct from material judg­ ments. However, this is not the case. Similarly, people distinguish between material gains and losses and behavior that is consistent with or differs from their feelings of obligation and responsibility and their sense of right and wrong. Third, value-based motivations should have a distinct influence on law-related behavior. Again, the justice literature finds that the degree to which people are willing to accept an outcome from an authority is linked, first, to the favorability of that outcome. In addition, however, people are more willing to accept an outcome that they evaluate as being fair and fairly arrived at. Hence, their fairness judgments exercise an independent influence upon their acceptance behavior that cannot be explained by outcome favorability. Similar­ ly, doing what is right and wrong is empirically distinguishable from doing what leads to favorable or unfavorable outcomes. Fourth, value-based motivations should produce a consistency of behavior across situa­ tions and over time. If, for example, someone feels an obligation to obey rules, their be­ havior should consistently reflect higher levels of compliance across settings that vary in their reward and cost characteristics. Further, they should show the same consistency in the same situation across time. This does not mean that situational forces will not influ­ Page 5 of 23

Value-Driven Behavior and the Law ence behavior, but it will be possible to see constancies in behavior that are not linked to those forces.

19.3 Legitimacy As noted above, one value-based mechanism that might be used to enhance compliance is to activate people’s feelings of responsibility and obligation to obey (“legitimacy”). Legiti­ macy is a property of an authority or institution that leads people to feel that that authori­ ty or institution is entitled to be deferred to and obeyed (Tyler, 2006b). It represents an “acceptance by people of the need to bring their behavior into line with the dictates of an external authority” (Tyler, 2006a, p. 25). This feeling of obligation is linked not only to the authorities’ possession of instruments of reward or coercion, but also to properties of the authority that lead people to (p. 407) feel it is entitled to be obeyed (Beetham, 1991). Since the classic writing of Weber (1968) social scientists have recognized that legitimacy is a property that is not simply instru­ mental, but reflects a social value orientation toward authority and institutions—that is, a normative, moral, or ethical feeling of responsibility to defer (Beetham, 1991; Kelman and Hamilton, 1989; Sparks, Bottoms, and Hay, 1996; Tyler, 2006a). This discussion will ex­ plore the importance of legitimacy, beyond the influence of instrumental factors shaping reactions to the law. These findings show that people cooperate when they believe that agents of the law are rightful holders of authority, and when they view the legal system as conferring upon them an appropriate and reasonable duty to obey. Feelings of trust and confidence in the police and courts—and a willingness to defer to their instructions—generate the belief that authorities have the right to dictate appropriate behavior, that authorities are justi­ fied in expecting feelings of obligation and responsibility from citizens. Legitimacy acti­ vates self-regulatory mechanisms. People defer to, and cooperate with, legitimate authori­ ties because they feel it is appropriate to do so. And, they are also influenced by other so­ cial factors, including their bonds with others and their motivation to do what they feel is morally right. The legitimacy of the law and of legal authorities resides most fundamentally in the recognition that the criminal justice system has the right to exist and to define appropri­ ate behaviour—to use its authority to determine the law, to use force when necessary, and to punish those who act illegally. While definitions of legitimacy vary widely, a key feature of many is that it confers the right to command and to direct the behavior of others and that it promotes the corresponding duty to obey (Weber, 1968; Tyler, 2006b). Modern dis­ cussions of legitimacy are usually traced to the writings of Weber (1968) on authority and the social dynamics of authority. Weber, like Machiavelli and others before him, argued that successful leaders and institutions use more than brute force to execute their will. They strive to win the consent of the governed so that their commands will be voluntarily obeyed (Tyler, 2006a). Page 6 of 23

Value-Driven Behavior and the Law Kelman and Hamilton (1989) refer to legitimacy as “authorization” to reflect the idea that a person authorizes an authority to determine appropriate behavior within some situa­ tion, and then feels obligated to follow the directives or rules that authority establishes. As they indicate, the authorization of actions by authorities “seem[s] to carry automatic justification for them. Behaviorally, authorization obviates the necessity of making judg­ ments or choices. Not only do normal moral principles become inoperative, but—particu­ larly when the actions are explicitly ordered—a different type of morality, linked to the duty to obey superior orders, tends to take over” (Kelman and Hamilton, 1989, p. 16). Legitimacy, according to this general view, is a quality that is possessed by an authority, law, or institution that leads others to feel obligated to accept its directives. It is “a quali­ ty attributed to a regime by a population” (Merelman, 1966, p. 548). When people ascribe legitimacy to the system that governs them, they become willing subjects whose behavior is strongly influenced by official (and unofficial) doctrine. They also internalize a set of moral values that is consonant with the aims of the system. And—for better (p. 408) or for worse—they take on the ideological task of justifying the system and its particulars (see also Jost and Major, 2001). As Kelman (1969) puts it: “It is essential to the effective functioning of the nation-state that the basic tenets of its ideology be widely accepted within the population …. This means that the average citizen is prepared to meet the expectations of the citizen role and to comply with the demands that the state makes upon him, even when this requires considerable personal sacrifice” (p. 278). Widespread voluntary cooperation with the state and the social system allows authorities to concentrate their resources most effec­ tively on pursuing the long-term goals of society. The authorities do not need to provide incentives or sanctions to all citizens to get them to support every rule or policy they en­ act. Perceptions of legitimacy are based on feelings of obligation that are disconnected from substance and material interest. Legitimacy is linked not to the authorities’ possession of instruments of reward or coercion, but to properties of the authority that lead people to feel it is entitled to make decisions and be obeyed. People will abide by the law and follow police directives even if they disagree with specific guidelines and instructions. In an in­ creasingly pluralistic and diverse society, it is important that legal authorities enjoy a sort of social influence that transcends particular moral values. Remaining salient in situa­ tions where authorities and subordinates disagree, legitimacy is an important source of governance, especially in complex societies. For example, the moral beliefs of anti-abor­ tion activists directly conflict with the views of the Supreme Court. But the legitimacy of a Supreme Court ruling on abortion is still conceded. Legitimacy is based upon the fairness of the procedures used by authorities to govern, rather than upon the substance of their decisions. This allows authorities to make decisions that are widely accepted even by those who disagree with them. What is legitimacy? Researchers typically identify three issues as central to definitions of legitimacy. The first is trust and confidence in the police. This involves the belief that po­ Page 7 of 23

Value-Driven Behavior and the Law lice officers are honest and unbiased; that they make decisions and enforce rules for the good of the people in the community; and that they try to follow the law. A second aspect of legitimacy is the perceived obligation to follow the law and accept the decisions made by legal authorities. Finally, people feel that the police share their values, that is, they want the same things for the community as do members of the public. Research findings support several general arguments. First, legitimacy matters. Studies show that people are more likely to defer to the police and the courts concerning conflict management and rule enforcement if they believe the police, the courts, and the law are legitimate (Haas, Keijser, and Bruinsma, 2014; Jackson et al., in press; Sunshine and Tyler, 2003; Tankebe, 2009; Tyler and Jackson, 2013). A second concern is with behavior that undermines state institutions or authorities such as riots and rebellions. Legitimacy also lessens willingness to engage in such actions (Fischer et al., 2008; Hohl, Stanko, and Newburn, 2012; Jackson et al., in press; LaFree and Morris, 2012; Tyler and Jackson, 2013). Further, those people who view the law as legitimate are more likely to follow the law in their everyday lives. This includes the widespread variety of laws that shape people’s behavior: traffic laws, laws against stealing, regulations against buying illegal items, laws against drug use, laws against robbery, murder, and assault, and so forth. In addition to the general influence of legitimacy on rule adherence, an additional concern is how people respond when they have personal interactions with the police. People can ei­ ther comply with police decisions and directives or they can resist and avoid them. A par­ ticular problem is that people change their behavior in the presence of the police and then revert to their original behavior when the police leave, requiring the police to deal repeatedly with the same people and problems. And, hostility and active resistance can occur, leading to the use of force and injury to both civilians and police officers. (p. 409)

Studies indicate that people are both more likely to obey law and to accept decisions when they view the police as legitimate. This includes ordinary citizens following the laws and accepting decisions related to rule breaking, disputes, and misdemeanors (Chory-As­ sad and Paulsel, 2004; Fagan and Tyler, 2005; Jackson et al., 2012; Jackson et al., 2013; Levi, Tyler, and Sacks, 2012; Martinson et al., 2006; Murphy, 2004, 2005; Murphy, Tyler, and Curtis, 2009; Reisig, Tankebe, and Mesko, 2013; Stuart et al., 2008; Sunshine and Tyler, 2003; Tyler, 2006; Tyler and Huo, 2002; Tyler and Jackson, 2013; Tyler et al., 2007; van Dijke, M. and Varboon, 2010; Wenzel, 2006). Legitimacy also matters to criminals involved in felony-level behaviors (Fagan and Pi­ quero, 2007; Kane, 2005; Papachristos, Meares, and Fagan, 2007, 2012; Reisig, 1998; Reisig and Mesko, 2009; Sparks, Bottoms, and Hay, 1996). And finally, these value-based influences are at least sometimes, and I would argue are often, more important than are instrumental concerns about punishment. By now none of these suggestions is controver­ sial and all have been supported by subsequent independent studies of compliance.

Page 8 of 23

Value-Driven Behavior and the Law Consider a specific example of the influence of legitimacy. Building on the work of Tyler (2006a), Sunshine and Tyler (2003) examined the antecedents of compliance and coopera­ tion with the police among people living in New York City. The results of this analysis show that police legitimacy influences people’s compliance with law and their willingness to cooperate with and assist the police. These findings also support the argument that le­ gitimacy is a social value that is distinct from performance evaluations. They show that such values have both an important and a distinct influence on people’s support for the police. This finding supports the arguments of Weber (1968) about the normative basis of public reactions to authority. It extends prior research findings (Tyler, 2006a) by showing that cooperation and empowerment, in addition to compliance, are influenced by legiti­ macy. This finding has been replicated in subsequent analyses of legal authority (Tyler and Fagan, 2008).

19.4 Moral Values A second form of obligation is the person’s feeling of obligation to their own moral values —their desire to do what they feel is right. A large psychological literature argues that the motivation to act in accord with their values about what is morally right and (p. 410) wrong is an important human motivation. The distinction between legitimacy and morali­ ty is that legitimacy is an obligation to obey authorities and institutions, while morality in­ volves an obligation to behave in ways consistent with personal values concerning right and wrong. People’s rule-following behavior is influenced by their internal motivation to uphold moral values. For example, Tyler (2006) showed that compliance with the law was shaped by judgments about whether the law is consistent with a person’s moral values. In his study, morality was more central to compliance than legitimacy, although both had an independent influence. As a consequence, an important question for the law is the degree to which it is congruent with public moral values. If people correctly understand the law, and if the law truly reflects the moral standards of the community, then the internalized sense of morality acts as a force for law-abidingness. Robinson and Darley (1995) similarly argue for the importance of congruence between moral values and laws. They suggest that people are more likely to accept laws if they view those laws as consistent with their own moral values. This is illustrated in their stud­ ies of people’s views about punishment for rule breaking which they show is linked to judgments about whether transgressors “deserve” punishment. Studies demonstrate that people’s views about appropriate sentencing decisions in criminal cases are driven by their judgments about deservingness rather than by instrumental judgments concerning how to deter future criminal conduct (Carlsmith, Darley, and Robinson, 2002; Darley, Carlsmith, and Robinson, 2000). People accept that a punishment is appropriate when it accords with their sense of what is deserved in response to the level and type of wrong committed, as well as to the intentions of the actor and their expressions of apology, re­ morse, or regret. People are also found to be willing to accept personal costs to punish of­

Page 9 of 23

Value-Driven Behavior and the Law fenders even when they themselves are not the victims of the crimes for which punish­ ment is occurring.

19.5 Self-Regulation What unites the study of legitimacy and morality? In both cases, the key is that people ac­ cept as their own feelings of responsibility and obligation for their actions in society. A core element of moral values is that people feel a personal responsibility to follow those values, and feel guilty when they fail to do so. Hence, moral values, once they exist, are self-regulatory in character, and those who have such values are personally motivated to bring their conduct into line with their moral standards. Internalized moral values are self-regulating and people accept and act on the basis of values that produce respect for societal institutions, authorities, and rules (Robinson and Darley, 1995; Tyler and Darley, 2000). The distinction between legitimacy and morality is that, in the case of morality, legal au­ thorities gain support for particular laws or decisions when those laws or decisions are in accord with people’s personal morality. Hence, the motivation to behave in ways that are moral does not lead to support of the rule of law when the public thinks that the (p. 411) law is inconsistent with their morality—when moral values and legal rules are congruent. To activate the motivation force of morality, legal authorities must be pursuing policies that are consistent with people’s moral values (Sunshine and Tyler, 2003). Robinson and Darley (1995), for example, show gaps between law and public morality. To the extent that such gaps are widely known, they would undermine public compliance with the law. The law can enlist people’s moral values as a motivational force supporting deference to the law by pursuing ends that people view as moral. They argue that the law is less likely to be able to call upon people’s moral motivations to support the legal sys­ tem when its values are viewed as discrepant from those of the public. Hence, the law can engage moral values when and if the law is consistent with the moral values held by the public. Of course, morality and legitimacy can be in conflict. A conflict between legitimacy and morality can occur with mundane and everyday practices, as when the government seems to criminalize drug use or certain sexual practices without the support of public morality (Darley, Tyler, and Bilz, 2003), or it can involve dramatic and high-stakes conflicts, as when the government seeks to compel people to serve in wars they think are unjust, or to pay taxes to support policies they view as immoral. Unlike legitimacy, morality is not linked to the role of the authority, and its independent roots in personal ethical values mean that, while morality usually supports following laws (Tyler, 2006a), the two internal forces do not always support one another.

Page 10 of 23

Value-Driven Behavior and the Law

19.6 The Need for Value-Based Motivation The importance of focusing upon self-regulation comes first from the widespread recogni­ tion of the value of voluntary deference to authorities. Social psychologists have long viewed the ability to motivate voluntary deference as a central attribute of authorities. Because such voluntary deference is important, a change is needed in the type of motiva­ tion that is the focus of regulatory efforts. It is important to have the type of motivation that leads to voluntary deference, and that motivation is value-based. Value-based motiva­ tions not only lead to voluntary deference. The problems of instrumental motivations are twofold. First, they do not motivate volun­ tary actions. Second, the use of sanctions undermines values. When sanctions become the focus of attention, the role of values is crowded out, or diminished. Further, the relation­ ship between authorities and people is undermined, since authorities become associated with punishment. As noted earlier, the problems of instrumentally based authority come to the fore during periods of scarcity or emergency. In those situations authorities must call upon the public to support their policies, but the prospects of gain or loss from non-compliance are low. Stressed regimes are less able to conduct surveillance and therefore cannot (p. 412) effec­ tively sanction. They also lack the resources to reward actions. If the public views law and legal authorities as legitimate, however, social order has an alternative basis for sup­ port during difficult times. People are motivated by intrinsic reasons to behave in socially desired ways and their compliance becomes more self-motivated and less context-depen­ dent.

19.7 The Goals of the Legal System The importance of values first emerges from the recognition that voluntary deference has advantages over sanction- or incentive-based compliance. This reflects a broadening of the conception of the ideal relationship between communities and legal authorities. When I wrote Why People Obey the Law in the 1980s, the ideal of a good citizen was very reac­ tive. A good citizen complied with the law. Today we have a much more active role in mind for citizens. First, we want them to not just comply with the law, but to be motivated by internal values to willingly obey the law. A force-based strategy creates long-term problems because “citizens who acquiesce at the scene can renege” (Mastrofski et al., 1996, p. 283). If citizens fail to fully agree with legal restrictions, further police intervention will eventually be required. Hence, the legal system is also concerned with its ability to gain long-term compliance, not just immediate compliance. Willing deference leads to long-term acceptance, rather than short-term com­ pliance.

Page 11 of 23

Value-Driven Behavior and the Law To the degree that people do this, the costs of surveillance and punishment diminish. The benefit of self-regulatory models is that they are linked to the actions that people will take responsibility for without fear of sanctioning or promise of rewards. The legal system benefits when people voluntarily defer to decisions and continue to defer over time. In the context of personal experiences with regulatory authorities, the legal system is more ef­ fective if people voluntarily accept the decisions made by legal authorities. Absent such acceptance, legal authorities must engage in a continuing effort to create a credible threat of punishment to assure long-term rule following/decision acceptance. Over time the goals of legal authorities have broadened in two ways. First, they increas­ ingly include the desire to motivate willing cooperation, with legal authorities and mem­ bers of the public working together to produce social order. Second, conceptions of the goals of the legal system have broadened to include the importance of promoting public engagement in communities in efforts to build social, political, and economic vitality. Legal authorities recognize the value of active voluntary public cooperation with the po­ lice and the courts. Cooperation includes willing acceptance of legal authority, deference to the decisions made by judges and police officers, everyday rule adherence, and willing­ ness to aid the police in identifying crime and criminals and the judicial system in prose­ cuting it by serving as a witness or a juror. Studies have demonstrated that legitimacy is an important antecedent of all of these forms of cooperation, shaping the (p. 413) willing­ ness to accept legal authority (Jackson et al., in press), deference (Tyler and Huo, 2002), everyday rule adherence, and aiding the police and criminal courts (Sunshine and Tyler, 2003; Tyler and Fagan, 2008). Hence the importance of legitimacy becomes even more central to the degree that cooperation is the goal of the legal system. Studies of legitimacy support the argument that traditional conceptions of legitimacy de­ fined in terms of the obligation to obey and trust and confidence captures an important element of its influence upon cooperation (Tyler and Fagan, 2008). But they also point to the potential value of expanding the framework of legitimacy. Normative alignment—the belief that police officers seem to share the purposes, goals, and values of the communi­ ty– has been found to be distinctly related to cooperation (Bradford, 2012; Jackson et al., 2012).

19.8 Engagement Studies of long-term approaches to social order point to the importance of creating viable communities. Recognizing that “you cannot arrest your way out of crime,” the police and courts have increasingly focused upon the objective of building economic, political, and social development as a means of long-term order maintenance (Geller and Belsky, 2009). This argument parallels the scholarly literature on creating viable communities, which emphasizes the importance of developing the shared attitudes which motivate engage­ ment (Loader and Walker, 2006). Sampson, Raudenbush, and Earls (1997), for example, argue that the collective willingness of neighbors to intervene for the common good sup­ ports lower levels of crime and violence. Recent studies suggest that such feelings of effi­ Page 12 of 23

Value-Driven Behavior and the Law cacy are encouraged by police legitimacy (Kochel, 2012; Sargeant, Wickes, and Maze­ rolle, 2013). The goal of engagement fits well with the recent literature within work organizations, which emphasizes the goal of engaging employees in work through building their identifi­ cation with their organization (Blader and Tyler, 2009; Tyler and Blader, 2000, 2003). Re­ search indicates that when employees identify with their organization and its leaders, they take on the values of the group, develop favorable attitudes and feelings toward their work, and engage in voluntary actions motivated by the desire to help their group be viable and effective (Tyler and Blader, 2000). This is the type of engagement that is al­ so the goal of community authorities seeking to motivate their members to be concerned about the viability of their communities. Shared feelings of obligation and responsibility to obey rules and trust and confidence in authorities encourage compliance and cooperation in fighting crime (Sunshine and Tyler, 2003; Tyler and Fagan, 2008), but whether or not they have a role in shaping engagement has not been examined. Legitimacy defined as shared goals, purposes, and values has been linked to identification with a group and to a broader willingness to actively and willingly engage in working with others in the group to address collective issues (Tyler, 2011). It is this broader sense of legitimacy that is central to legitimacy and (p. 414) en­ gagement (Bradford, 2011; Hough et al., 2013; Jackson, 2012); the increasing importance of this goal suggests the need to consider which elements of legitimacy are important in promoting engagement. It is clear that the actions of legal authorities have an impact on people’s views about so­ ciety and government (Tyler, Casper, and Fisher, 1989). Because the actions of legal au­ thorities generalize to views about society and government, it should be possible to devel­ op strategies of law enforcement that are socially beneficial because they help to build identification with government and society, as well as creating feelings of obligation. For example, the police can help give government a broader legitimacy that would lead peo­ ple to engage in economic and social activities in their own cities. They can build the type of psychological connections that lead people to work willingly and enthusiastically in their communities in many other ways, ranging from shopping in stores to going to local restaurants. In other words, rather than being viewed as a (necessary) cost, the legal sys­ tem can develop policies and practices that generate supportive attitudes and values that enhance communities. While many types of government authority could potentially shape views about one’s self and one’s community (Katz and Gutek, 1976; Lipsky, 1980), the assumption underlying the engagement model is that people are more likely to live in and visit communities in which they feel that they will be well treated by the legal authorities they are most likely to encounter—the police. This benefits communities economically because people more willingly come to them to work, to shop, as tourists, and for entertainment and sporting events. Hence, the police play a central role in creating the reassurance that makes a community inviting and desirable to the general public. Page 13 of 23

Value-Driven Behavior and the Law More generally the law provides a framework for building vibrant, successful communi­ ties. If people feel reassured by the presence of the police, and believe that they will be protected and, if they need it, helped, then they will be encouraged to engage in their communities socially and economically. When people engage in such behaviors, they build social capital and the sense of efficacy that has broad social value. If people engage in their communities, they will come to know others and know how to work with them when problems arise in the community. They build trust in others and develop the belief that others can and will join together to address issues when they arise. By providing a frame­ work of reassurance, the police can potentially create the climate that allows the commu­ nity to develop valuable psychological and sociological characteristics. Such engagement may be further facilitated when there are functioning courts that can resolve conflicts and enforce rules (Breyer, 2011). The goal of this study is to examine whether in fact the legal system can, as argued, play a role in encouraging engagement. Legitimate institutions help foster identification with collectivities and the willingness to act on their behalf (i.e. collective efficacy; see Kochel, 2012). Tyler and Blader (2000, 2003) explore a similar relationship between people and the collective in the context of work organizations. They demonstrate that identification with authorities and institutions is central to motivating the development of supportive attitudes and values, for example legitimacy, as well as to motivating engaged cooperative behavior. Hence, to the degree that the police and courts can build identification with legal authorities (p. 415) and with the community itself, they promote supportive public attitudes and voluntary cooperative behaviors. The police and the courts can similarly build identification with society and so­ cial institutions, and through that identification can motivate members of the community to more actively work on its behalf. In this study we examine the utility of disentangling “obligation to obey,” “trust and confi­ dence,” and “normative alignment.” We consider the value of treating obligation to obey as a separate dimension of legitimacy, reflecting the internalization of the value that it is appropriate to obey the police and the law. We also consider the value of treating trust and confidence as a separate dimension of legitimacy. If people feel that the authorities are sincere, benevolent, and concerned about their welfare, then they trust them to act in ways that benefit the people over whom they exercise authority. Finally, we add an exami­ nation to the role of normative alignment, the belief that power-holders have values, goals, and purposes that align with their own (Jackson et al., 2011, 2012). Normative alignment can be seen as a constitutive (and separate) dimension of legitimacy because it embodies a sense of normative justifiability of power and authority in the eyes of the citizens (Bradford et al., 2013a, 2013). When officers are viewed as having appro­ priate purposes, goals, and values in the eyes of citizens, they are generating and sustain­ ing the normative validity of the power and authority of the role and institution (Euro­ pean Social Survey, 2010, 2012). People believe that their values and the goals and pur­ poses of the police are similar, which validates power possession in the eyes of citizens (Hough et al., 2013a, 2013b).

Page 14 of 23

Value-Driven Behavior and the Law An analysis of these approaches identified three target behaviors (compliance, coopera­ tion, and engagement) and outlined three elements of legitimacy (obligation, trust and confidence, and normative alignment)(Tyler and Jackson, in press). Their approach was to examine the distinct contribution of each aspect of legitimacy to predicting these three behavioral goals. Their results indicated that legitimacy always influenced the relevant behavior. However, obligation was most important with compliance and normative align­ ment with engagement. These results support a value-based approach because each of these aspects of legitima­ cy was distinguished by its normative content. In other words, it involved a judgment about what was appropriate. For example, obligation was linked to the perceived respon­ sibility to accept authority, not to the costs or rewards of deference; trust and confidence was linked to the character and intentions of the authorities, not their competence or ability to deliver services or safety; and normative alignment was an alignment in values and purposes. While everyone in the community may share the goal of stopping crime, normative alignment is about shared values and purposes, that is, a shared sense of what is right or wrong in terms of desirable defining characteristics of a community and its members. The Tyler and Jackson analysis demonstrated that these legitimacy influences were dis­ tinct from the influence of moral values upon voluntary compliance. Normative align­ ment, which reflects the view that the authorities and the public share normative values, had an influence that is distinct from that of morality. Morality—that is, the view that the law reflects the person’s personal values—similarly had an influence distinct (p. 416) from that of normative alignment. Hence, it was important both that the law reflected a person’s own sense of right and wrong and that they believed they shared values with le­ gal authorities.

19.9 Conclusion Two dynamic models of the connection between law and communities can be imagined. First, a focus on sanctions can predominate, which over time undermines values and rela­ tionships with authorities. This requires increasingly more resources to implement sanc­ tions, for example more resources to build prisons and to keep people in those prisons longer. The authorities become increasingly dependent upon punishment, and other link­ ages between people and the state diminish in significance. The alternative model involves an emphasis upon values. This makes values more salient, and the role of sanctions diminishes. People are increasingly self-regulatory, so society is increasingly free to use resources for education and job creation. This, in turn, leads to additional social motivations for being law-abiding. People increasingly have something to lose by breaking rules, so their personal motivation to be law-abiding becomes stronger. And, people’s behavior becomes more flexible and adaptive, responding well to local needs and conditions. Page 15 of 23

Value-Driven Behavior and the Law Overall, there are two arguments for value-based motivation. First, we gain the benefits of a value-based approach—that is, increasing voluntary cooperation. Second, we avoid the problems associated with instrumental approaches. To gain these advantages we need to move to a system in which value-based motivations are the primary motivation tapped, and instrumental motivations are the backup for a small group that have to be dealt with instrumentally because they are unable or unwilling to act on their values. Since values work for most people, we generally reap benefits from self-regulations, and using value-based approaches does nothing to undermine any subsequent reliance upon instrumental mechanisms. In other words, unlike instrumentality, which undermines val­ ues, values do not undermine instrumentality. And, as noted, this argument emerges from research on moral values just as it does from work on legitimacy. From both perspectives, it is crucial that people in positions of au­ thority focus on how to create and maintain values. It is only by doing so that they will be able to motivate the type of voluntary cooperation identified in discussions of the public behavior as most central to successful social functioning.

References Beetham, D. (1991). The Legitimation of Power. London: Macmillan. Blader, S. and Tyler, T. R. (2003). A four component model of procedural justice: Defining the meaning of a “fair” process. Personality and Social Psychology Bulletin, 29: 747–758. Blader, S. and Tyler, T. R. (2009). Testing and expanding the group engagement model. Journal of Applied Psychology, 94: 445–464. (p. 417)

Blumstein, Alfred, Jacqueline Cohen, and Daniel Nagin, eds. (1978). Deterrence and inca­ pacitation: Estimating the effects of criminal sanctions on crime rates. Washington, D.C.: National Academy of Sciences. Bottoms, Anthony E. (1999). “Interpersonal violence and social order in prisons,” in Michael Tonry and Joan Petersilia, eds., Crime and Justice: An Annual Review 26: 437– 513. Chicago: University of Chicago Press. Bottoms, A. and A. Von Hirsch (2010). “The crime preventive impact of penal sanctions,” in P. Cane and H. Kritzer, eds., The Oxford Handbook of Empirical Legal Research. Ox­ ford: Oxford University Press, pp. 96–124. Bradford, B. (2011). “Voice, neutrality and respect: Use of victim support services, proce­ dural justice, and confidence in the criminal justice system.” Criminology & Criminal Jus­ tice, 11: 345–366. Braithwaite, John and Toni Makkai (1991). “Testing an expected utility model of corporate offending.” Law and Society Review 25: 7–39.

Page 16 of 23

Value-Driven Behavior and the Law Breyer, S. (2010). Making our democracy work: A judge’s view. New York, NY: Random House. Chory-Assad, R. M. and M. L. Paulsel (2004). “Classroom justice: Student aggression and resistance as reactions to perceived unfairness.” Communication Education 53(3): 253– 273. Carlsmith, K. M., Darley, J. M., and Robinson, P. H. (2002). “Why do we punish?” Journal of Personality and Social Psychology, 83: 284–299. Darley, J. M., Carlsmith, K. M., and Robinson, P. H. (2000). “Incapacitation and just deserts as motives for punishment.” Law and Human Behavior 24: 659–683. Darley, J. M., Tyler, T. R., and Bilz, K. (2003). “Enacting justice: The interplay of individual and institutional perspectives.” In M.A. Hogg and J. Cooper (Eds.), The Sage Handbook of Social Psychology (pp. 458–476). London: Sage. Deci, Edward and Ryan, Richard M. (1985). Intrinsic motivation and Self-determination in human behavior. Deterrence and the death penalty (2012). Washington, D.C.: National Academy of Science. European Social Survey_Module 5 (2010). Data documentation report. European Social Survey. Fagan, J. and A. R. Piquero (2007). “Rational choice and developmental influences on re­ cidivism among adolescent felony offenders.” Journal of Empirical Legal Studies 4(4): 715–748. Fagan, J. and T. R. Tyler (2005). “Legal socialization of children and adolescents.” Social Justice Research 18(3): 217–242. Fischer, R., Harb, C., Al-Sarraf, S., and Nashabe, O. (2008). Support for resistance among Iraqi students. Basic and Applied Social Psychology 30: 167–175. Fuller, L. L. (1969). The morality of law. New Haven, CT: Yale University Press. Geller, W. and Belsky, L. (2009). “Building our way out of crime: The transformative pow­ er of police-community relationships.” Geller & Associates. Harcourt, Bernard E. 2001. Illusion of order: The false promise of broken windows polic­ ing. Cambridge: Harvard University Press. Harcourt, Bernard E. and Jens Ludwig. (2006). “Broken windows: new evidence from New York City and a five-city social experiment.” University of Chicago Law Review 73(1): 271–320. Haas, N. E., Keijser, J. W., and Bruinsma, G. J. N. (2014). “Public support for vigilantism, confidence in police and police responsiveness.” Policing & Society 24: 224–241. Page 17 of 23

Value-Driven Behavior and the Law Hohl, K., Stanko, B., and Newburn, T. (2012). “The effect of the 2011 London dis­ order on public opinion of the police and attitudes towards crime, disorder, and sentenc­ ing.” Policing 7: 12–20. (p. 418)

Hough, M., Jackson, J., and Bradford, B. (2013). “Legitimacy, trust and compliance.” In Tankebe, J. and Liebling, A. (Eds.), Legitimacy and criminal justice. New York, NY: Oxford University Press. Huselid, Mark A. (1995). “The impact of human resource management practices on turnover, productivity, and corporate financial performance.” Academy of Management Journal 38: 635–672. Jackson, J., B. Bradford, M. Hough, A. Myhill, P. Quinton, and T. R. Tyler (2012). “Why do people comply with the law? Legitimacy and the influence of legal institutions.” British Journal of Criminology 52: 1051–1071. Jackson, J., B. Bradford, B. Stanko, and K. Hold (2013). Just authority?: Trust in the police in England and Wales. Jackson, J., A. Z. Huq, B. Bradford, and T. R. Tyler (2013) . “Monopolizing force? Police le­ gitimacy and public attitudes toward the acceptability of violence.” Psychology, Public Pol­ icy and Law 19: 479–497. Kane, R. J. (2005). “Compromised police legitimacy as a predictor of violent crime in structurally disadvantages communities.” Criminology 43: 469–498. Katyal, Neil. (1997). Deterrence’s difficulty. Michigan Law Review 95: 2385–2476. Katz, D. and Gutek, B. (1976). Bureaucratic encounters. Ann Arbor, MI: Institute for So­ cial Research. Kelling, G. and M. H. Moore (1988). The evolving strategy of policing. Washington, D.C.: United States Department of Justice. Kelman, H. C. (1969). “Patterns of personal involvement in the national system: A socialpsychological analysis of political legitimacy.” In J. Rosenau (Ed.), International politics and foreign policy (pp. 276–288). New York: Free Press. Kelman, H. C. (2001) . “Reflections on social and psychological processes of legitimization and delegitimization,” in J. T. Jost and B. Major, eds., The psychology of legitimacy: Emerging perspective on ideology, justice, and intergroup relations. Cambridge: Cam­ bridge University Press. Kelman, H. C. and V. L. Hamilton (1989). Crimes of Obedience. New Haven, CT: Yale Uni­ versity Press. Kochel, T. R. (2012). Can police legitimacy promote collective efficacy? Justice Quarterly 29: 384–419. Page 18 of 23

Value-Driven Behavior and the Law LaFree, G. and Morris, N. A. (2012). “Does legitimacy matter? Attitudes toward antiAmerican violence in Egypt, Morocco, and Indonesia.” Crime & Delinquency, 58: 689– 719. Langevoort, Donald C. (2002). “Monitoring: The Behavioral Economics of Corporate Com­ pliance with Law.” Columbia Business Law Review, 2002: 71–118. Lind, E. A. Tyler, T. R. (1988). The social psychology of procedural justice. NY: Plenum. Levi, M., T. R. Tyler, and A. Sacks (2012). “The reasons for compliance with law,” in R. Goodman, D. Jinks, and A. Wood, eds., Understanding Social Action, Promoting Human Rights. Oxford: Oxford University Press. Lipsey, Mark W. and Francis T. Cullen (2007). “The effectiveness of correctional rehabili­ tation: A review of systematic reviews.” Annual Review of Law and Social Science 3: 297– 320. Lipsky, M. (1980). Street-level bureaucracy. New York, NY: Russell Sage Foundation. Loader, I. and Walker, N. (2007). Civilizing security. New York, NY: Cambridge University Press. (p. 419)

Lynch, James P. and William J. Sabol (1997). Did getting tough on crime pay? Washington, D.C.: The Urban Institute. McCord, J., C. S. Widom, and N. A. Crowell (2001). Juvenile Justice. National Research Council. MacCoun, R. (1993). “Drugs and the law: A psychological analysis of drug prohibition.” Psychological Bulletin 113: 497–512. Markell, David L. (2000). “The role of deterrence-based enforcement in a ‘reinvented’ state/Federal relationship.” Harvard Environmental Law Review, 24: 1–114. Mastrofski, S. D., Snipes, J. B., and Supina, A. E. (1996). “Compliance on demand: The public’s responses to specific police requests.” Journal of Crime and Delinquency, 33: 269–305. Martinson, B. C., M. S. Anderson, A. L. Crain, and R. De Vries (2006). “Scientists’ percep­ tions of organizational justice and self-reported misbehaviors.” Journal of Empirical Re­ search on Human Research Ethics 1(1): 51–66. Merelman, R. J. (1966). Learning and legitimacy. American Political Science Review 60: 548–561. Murphy, K. (2004). “The role of trust in nurturing compliance.” Law and Human Behavior 28(2): 187–209.

Page 19 of 23

Value-Driven Behavior and the Law Murphy, K. (2005). “Regulating more effectively: The relationship between procedural justice, legitimacy, and tax noncompliance.” Journal of Law and Society 32(4): 562–589. Murphy, K., T. R. Tyler, and A. Curtis (2009). “Nurturing regulatory compliance: Is proce­ dural justice effective when people question the legitimacy of the law?” Regulation and Governance 3(1): 1–26. Nagin, D. (1998). “Criminal deterrence research at the outset of the twenty-first century.” Crime and Justice 23: 1–42. Papachristos, A. V., T. Meares, and J. Fagan (2007). “Attention Felons: Evaluating Project Safe Neighborhoods in Chicago.” Journal of Empirical Legal Studies 4: 223–250. Papachristos, A. V., T. L. Meares, and J. Fagan (2012). “Why do criminals obey the law?” Journal of Criminal Law and Criminology 102: 397–440. Paternoster, Raymond (1987). “The deterrent effect of the perceived certainty and severi­ ty of punishment: A review of the evidence and issues.” Justice Quarterly 4(2): 173–217. Paternoster, Raymond (1989) . “Decisions to participate in and desist from four types of common delinquency.” Law and Society Review 23: 7–40. Paternoster, Raymond (2006). “How Much Do We Really Know About Criminal Deter­ rence?” Journal of Criminal Law and Criminology 100: 765–824. Paternoster, Raymond and Sally Simpson (1996). “Sanction threats and appeals to morali­ ty: Testing a rational choice model of corporate crime.” Law and Society Review 30: 549– 584. Piquero, A. R., Z. Gomez-Smith, and Lynn Langton (2004). “Discerning unfairness when others may not: Low self-control and unfair sanction perceptions.” Criminology 42(3): 699–733. Piquero, A. R., R. Paternoster, G. Pogarsky, and T. Loughran (2011). “Elaborating the Indi­ vidual Difference Component in Deterrence Theory.” Annual Review of Law and Social Science 7: 335–360. Podsakoff, Phillip M., Bommer, William. H., Podsakoff, Nathan P., and MacKenzie, Scott B. (2006). “Relationships between leader reward and punishment behavior and subordinate attitudes, perceptions, and behaviors.” Organizational Behavior and Human Decision Processes 99: 113–142. Pratt, Travis C., Francis T. Cullen, Kristie R. Blevens, Leah E. Daigle, and Tamara D. Madensen (2008). “The empirical status of deterrence theory: A meta-analysis,” in Fran­ cis T. Cullen,

(p. 420)

John Paul Wright, and Kristie R. Blevins, eds., Taking stock: The sta­

tus of criminological theory, 367–396. New Brunswick, NJ: Transaction Publishers.

Page 20 of 23

Value-Driven Behavior and the Law Reisig, M. D. (1998). “Rates of disorder in higher-custody state prisons: A comparative analysis of managerial practices.” Crime and Delinquency 44(2): 229–244. Reisig, M. D. and C. Lloyd (2009). “Procedural justice, police legitimacy, and helping the police fight crime.” Police Quarterly 12(1): 42–62. Reisig, M. D. and G. Mesko (2009). “Procedural justice, legitimacy and prisoner miscon­ duct.” Psychology, Crime and Law 15(1): 41–59. Reisig, M. D., J. Tankebe, and G. Mesko (2013). “Compliance with the law in Slovenia: The role of procedural justice and police legitimacy.” European Journal of Criminal Policy Re­ search, 20: 259–276. Robinson, P.H. and Darley, J. (1995). Justice, liability, and blame. Boulder, CO: Westview. Sampson, R. J., Raudenbush, S. W., and Earls, F. (1997). “Neighborhoods and violent crime: A multilevel study of collective efficacy.” Science 277: 918–924. Sargeant, E., Wickes, R., and Mazerolle, L. (2013). “Policing community problems.” Aus­ tralian and New Zealand Journal of Criminology 46: 70–87. Simpson, Sally (2002). Corporate crime, law, and social control. Cambridge: Cambridge University Press. Sklansky, D. A. (2011). The persistent pull of police professionalism. Cambridge, MA: Pro­ gram in Criminal Justice Policy and Management, Harvard Kennedy School. Sparks, R., A. Bottoms, and W. Hay (1996). Prisons and the Problem of Order. Oxford. Stuart, J., M. Fondarcaro, S. A. Miller, V. Brown, and E. M. Brank (2008). “Procedural jus­ tice in family conflict resolution and deviant peer group involvement among adolescents.” Journal of Youth and Adolescence 37: 674–684. Stewart, R. (February 10, 2003). Administrative law in the 21st century. Presentation at New York University Law School. Sunshine, J. and T. R. Tyler (2003). “The role of procedural justice and legitimacy in shap­ ing public support for policing.” Law and Society Review 37: 513–548. Sutinen, Jon G., and Kuperan, K. (1999). A socio-economic theory of regulatory compli­ ance. International journal of social economics, 26: 174–193. Tankebe, J. (2009). “Self-help, policing, and procedural justice: Ghanaian vigilantism and the rule of law.” Law & Society Review 43: 245–270. Tyler, T. R. (2005). “Promoting employee policy adherence and rule following in work set­ tings: The value of self-regulatory approaches.” Brooklyn Law Review 70: 1287–1312.

Page 21 of 23

Value-Driven Behavior and the Law Tyler, T. R. (2006a). “Legitimacy and legitimation.” Annual Review of Psychology 57: 375– 400. Tyler, T. R. (2006b). Why people obey the law: Procedural justice, legitimacy, and compli­ ance. Princeton: Princeton University Press. Tyler, T. R. (2011). Why people cooperate. Princeton: Princeton University Press. Tyler, T. R. and Blader, S. (2000). Cooperation in groups: Procedural justice, social identi­ ty, and behavioral engagement. Philadelphia, PA: Psychology Press. Tyler, T. R., Boeckmann, R., Smith, H. J. and Huo, Y. J. (1997). Social justice in a diverse society. Denver, CO: Westview. Tyler, T. R., Casper, J. D., and Fisher, B. (1989). “Maintaining allegiance toward political authorities: The role of prior attitudes and the use of fair procedures.” American Journal of Political Science 33: 629–652. Tyler, T. R. and Fagan, J. (2008). “Legitimacy and cooperation: Why do people help the po­ lice fight crime in their communities?” Ohio Journal of Criminal Justice. Tyler, T. R. Jackson, J. (2014). “Popular legitimacy and the exercise of legal author­ ity: Motivating compliance, cooperation and engagement.” Psychology, Public Policy, and Law 20: 78–95. (p. 421)

Tyler, T. R. and Y. J. Huo (2002). Trust in the law. New York: Russell-Sage. Tyler, T. R. and J. Jackson (2013). Popular legitimacy and the exercise of legal authority: Motivating compliance, cooperation and engagement. Unpublished manuscript, Yale Law School. Tyler, T. R., L. W. Sherman, H. Strang, G. C. Barnes, and D. J. Woods (2007). “Reintegra­ tive shaming, procedural justice, and recidivism: The engagement of offenders’ psycho­ logical mechanisms in the Canberra RISE drinking-and-driving experiment.” Law and So­ ciety Review 41(3): 553–586. Van Dijke, M. and P. Verboon (2010). “Trust in authorities as a boundary condition to pro­ cedural fairness effects on tax compliance.” Journal of Economic Psychology 31: 80–91. Walster, E., Walster, G. W., and Berscheid, E. (1978). Equity: Theory and research. Boston, MA: Allyn and Bacon. Weber, M. (1968). Economy and society. (G. Roth and C. Wittich, Eds.). Berkeley: Univer­ sity of California Press. Weisberg, Robert. (2005). “The death penalty meets social science: Deterrence and jury behavior under new scrutiny.” Annual Review of Law and Social Science 1: 151–170, p. 163. Page 22 of 23

Value-Driven Behavior and the Law Wenzel, M. (2006). “A letter from the tax office: Compliance effects of informational and interpersonal justice.” Social Justice Research 19(3): 345–364. Wright, Bradley, Caspi, Avshalom, and Moffitt, Terrie E. (2004). “Does the perceived risk of punishment deter criminally prone individuals?” Journal of Research in Crime and Delinquency 41: 180–213.

Tom R. Tyler

Tom R. Tyler Yale Law School Yale University New Haven, CT, USA

Page 23 of 23

Ex Ante vs. Ex Post

Ex Ante vs. Ex Post   Donald A. Wittman The Oxford Handbook of Law and Economics: Volume 1: Methodology and Concepts Edited by Francesco Parisi Print Publication Date: Apr 2017 Subject: Economics and Finance, Law and Economics, Econometrics, Experimental and Quanti­ tative Methods Online Publication Date: May 2017 DOI: 10.1093/oxfordhb/9780199684267.013.40

Abstract and Keywords Sometimes the sequence of events is important for establishing rights and obligations, and sometimes it is not. For example, if a nuisance was there before the neighbouring residences arrived, the nuisance is sometimes allowed to continue in the same location under the doctrine of coming to the nuisance. When and why should the doctrine be (or not be) upheld? While many concepts in law and economics do not explicitly have a time dimension, once we start thinking about ex ante versus ex post, a large number of seem­ ingly unrelated areas of the law involve similar issues of sequence. When new regulations are imposed, sometimes pre-existing businesses are exempt and sometimes not. In acci­ dent law, negligent behaviour by the first actor may require the second actor to take ac­ tion beyond the ordinarily efficient actions as can be seen in the doctrine of last clear chance. What is the underlying rational for the application of this rule? Rights typically go to the highest bidder, but at 4-way stop signs, rights are granted according to who was there first. In other areas involving traffic, being first accedes to other criteria such as majority rule. As a final example, priority in bankruptcy gives the right to the first credi­ tor of the same secured debt, but not to the first creditor of unsecured debt. Why? This article presents an efficiency-based framework for answering these questions. Keywords: ex ante, ex post, efficiency, rights, obligations, law

20.1 Introduction SOMETIMES the sequence of events is important for establishing rights and obligations and sometimes it is not. For example, if a nuisance was there before the neighboring resi­ dences arrived, the nuisance sometimes (but not always) is allowed to continue in the same location under the doctrine of coming to the nuisance. When and why should the doctrine be (or not be) upheld? While many concepts in law and economics do not explic­ itly have a time dimension (for example, supply and demand, the comparison of negli­ Page 1 of 18

Ex Ante vs. Ex Post gence rules to strict liability, and so forth), once we start thinking about ex ante versus ex post, a large number of seemingly unrelated areas of the law involve similar issues of se­ quence. When new regulations are imposed, sometimes pre-existing businesses are ex­ empt (grandfathered in) and sometimes not. In accident law, negligent behavior by the first actor may require the second actor to take action beyond the ordinarily efficient ac­ tions, as can be seen in the doctrine of last clear chance. What is the underlying rationale for the application of this rule? Rights typically go to the highest bidder, but at four-way stop signs, rights are granted according to who was there first. In other areas involving traffic, being first accedes to other criteria such as majority rule. As a final example, pri­ ority in bankruptcy gives the right to the first creditor of the same secured debt, but not to the first creditor of unsecured debt. Why? In this chapter, I present an efficiency-based framework for answering these questions. In the absence of market transaction costs, auctions are a simple device to allocate rights to the party who values the set of rights the most. We will often be dealing with situations where market transaction costs are high. In pollution cases, the existence of many pollu­ tees creates the free-rider problem when the pollutees are buying the right to clean air and the monopoly holdout problem when individual pollutees are selling the right. At a four-way stop sign, auctioning off the right to go first to the car that values the right the most would result in the cost of the allocation mechanism swamping the benefit of the correct assignment of rights. Nevertheless, we will use some idealized version (p. 423) of the market (one person owns everything, or market transaction costs are zero) as the standard for allocating rights correctly. We will then look at alternative allocation meth­ ods and choose that method which minimizes the sum of costs due to misallocation of rights and transaction costs. To provide an easy illustration of the methodology, let us go back to the four-way stop sign example. In the absence of further knowledge, giving the right to proceed first to the vehicle that came first is likely to approximate the outcome under an auction, as the cost to the second arrival under this system is likely to be less than the cost to the first arrival under the rule that the second vehicle has priority. Of course, in any particular case, this may not hold; there may have been more people in the second vehicle, which would likely result in the second vehicle outbidding the first if an auction had been held. But on average we would expect the same number of people in the first and second arriving vehicles. And so we rely on the low transaction cost method of first come, first served rather than the higher transaction cost method of auctioning off the rights. But first come, first served accedes to later arriving ambulances when their sirens are on because the benefit to the ambulance’s occupant from being first to go through the intersection very likely outweighs the cost of waiting to the occupants of the other vehicles.1

20.2 Determining the Optimal Sequence Not surprisingly, ex ante versus ex post suggests a two-period model (e.g. see Wittman, 1980; Shavell, 2008; Innes, 2008) and I will continue in this mode. To provide insight into the arguments, I will present a few topics in detail, but show how they might hold more Page 2 of 18

Ex Ante vs. Ex Post generally. I start with land use, where there is often a conflict between producers and res­ idences because of the smell, noise, or air pollution created by the former. Residences want the right to be free of these annoyances without having to move or undertake other costly methods that reduce their impact, while produces would like the right to produce such “annoyances” when they are a byproduct of their pursuit of profit. Producers, like homeowners, prefer to avoid costly methods of reducing pollution such as moving to iso­ lated areas or installing smoke-scrubbers. There are obvious costs to giving one side the right just because it was there first. This is because individuals may undertake premature investment just to gain the right. Building a residence, years before it will be used, is a costly method of preventing polluters from entering the area; and, similarly, building a factory years before it is otherwise economi­ cally viable just to gain the right to pollute is also economically costly. That is, (p. 424) giv­ ing rights to the side who was first may involve high transaction costs. Furthermore, the use that was there first may not have been the appropriate use for the area. As a result, either the more appropriate use buys out the less appropriate use (meaning still higher transaction costs) or the inappropriate use remains because transaction costs are so high that Pareto improving trades do take place.2 The law gets around some of the problems posed in the preceding paragraph by granting rights to the side that should have been first. For example, in Bove v. Donner-Hanna Coke Corporation, 258 N.Y.S. 229 (1932), the court ruled against Bove, the party that was there first. The court stated the following: “It is true that the appellant [Bove, the owner of a rooming house] was a resident of this locality for several years before defendant came on the scene … and that when the plaintiff built her house, the land on which these coke ovens now stand was a hickory grove. This region was never fitted for a residential dis­ trict [low land, river, seven railroads].” Swamps and railroad tracks are appropriate locations for heavy industry and inappropri­ ate for housing; hilltop locations in urban areas are appropriate for housing and inappro­ priate for gas stations, let alone heavy industry. But sometimes, large expanses of land are featureless, and the first use of the land establishes its character. This was the case in Mahlstadt v. City of Indianola, 100 N.W. 2d 189 (1959). Mahlstadt, a housing developer, had asked for a court injunction barring the operation of a city dump neighboring his property. The appellate court ruled in favor of the city. There was nothing peculiar to the location, making it particularly appropriate or inappropriate for housing or for a dump. Therefore, the first activity determined the character of the location. The court held that the dump’s “prior operation at that place should be given substantial weight in determin­ ing the character of the locality and the reasonableness or unreasonableness of operating it there.” City dumps need to be located near cities—the cost of transporting garbage is very high. The long-run stream of costs and benefits made it appropriate for the dump to be located there initially and to maintain operation as the area became more residential. However, if the residents were there first, it would have been inefficient to locate a dump next to them and the courts would have ruled in favor of the plaintiffs. In a nutshell, when

Page 3 of 18

Ex Ante vs. Ex Post the character of the place is determined by the first user, the doctrine of coming to the nuisance and its converse are invoked. Suppose that homeowners could collect damages for moving next to an appropriately lo­ cated dump.3 Would developers or homeowners be better off in the long run? The answer is no. Consider the case where homeowners receive $10,000 per house for being next to a pre-existing city dump. Then they would be willing to pay $10,000 more for their homes. Because the developer wants to maximize profits, the developer will charge $10,000 more per parcel in such cases. Because the developer gets $10,000 more (p. 425) for each par­ cel, the developer will be willing to pay $10,000 more for the land in the first place. And so the original owner of the undeveloped land, who is also interested in profit maximiz­ ing, will charge $10,000 more per parcel, as well. So, in this situation, a system of com­ pensation does not make homeowners or developers better off. The only person that is af­ fected by the rule is the original owner of the undeveloped property. And if this original owner owned the land on which the dump was built as well, such a transfer would go from the left pocket to the right. The main effect of collecting damages is that this in­ volves some transaction costs. Otherwise, the outcome does not change. The city would still build the dump where it did, the developer would still build the houses, and residents would still buy and live in the houses. There are many situations where the first party should have been there first, yet the sec­ ond party creates the dominating character of the place. Cities often expand into rural ar­ eas. Rural areas are appropriate for animal husbandry, but cities and cattle feedlots do not mix very well. Of the three possibilities—cities leapfrog around cattle feedlots, urban housing and businesses border cattle feedlots, and feedlots relocate—the last is clearly the most efficient And the law unambiguously reflects this economic logic. In such situations, the law is unlikely to compensate for the costs of relocation, even though the party was there first and should have been there first. There are several rea­ sons for such a policy. First, alternative land uses are likely to be more profitable; shop­ ping centers are more valuable than feedlots in suburban areas. The owner of the land is not being harmed by suburbanization and would naturally move his business to a more rural location. Second, over time, the owner of the property could have let the property depreciate rather than undertake repairs and upgrades. Third, a system of compensation would be costly because, in the absence of compensation, the outcome (shutting down the feedlot) is the same, and a court case is rarely needed. And fourth, a system of compen­ sating feedlots for moving would encourage owners to hold onto their businesses to col­ lect damages and more generally to not anticipate the future (the problems that arise in the absence of compensation, too few feedlots in period 1 or too much expansion of the urban area in period 2 are less likely except at the very margin). Often the rationale for compensation is to create the correct incentives for the person who is liable rather than to benefit the person who was harmed. In this case, compensation would, if anything, cre­ ate the wrong incentives. Much the same logic applies to quarries and other noxious ac­

Page 4 of 18

Ex Ante vs. Ex Post tivities. See, for example, Consolidated Rock Products Co. v. The City of Los Angeles, 370 P.2d 342 (1962), where a quarry was forced to shut down without compensation. The taking away of a property right without compensation might appear to violate some basic notions of the inviolability of property. But this paradox is resolved if we realize that all property rights are contingent in space and time. For example, I have the right to dig holes on my land but not to throw dirt at people passing by. Similarly, people have the right to have cattle feedlots on their property but not if the area is urban. That is, when the city grows around the cattle feedlot and the feedlot is forced to move without com­ pensation, the owner of the feedlot is not losing his property rights because (p. 426) the owner never had (or should never have had) the right to have a feedlot within an urban area. Even though, the cost of relocating feedlots is factored in, it is easy to see that the effi­ cient outcome is for the feedlots to move as the city expands. For other kinds of activity, where there are more substantial moving costs and smaller negative externalities, the stream of benefits may make it economically efficient to give rights to an activity when it should have been first, even if the activity would have been outlawed if it had not been first. Thus, there is “nonconforming land use” for those activities that should have been there first and should remain, while initiation of similar activities are prevented because the cost–benefit stream dictates that they should be undertaken elsewhere. There is a very marked distinction to be observed in reason and equity between the case of a business long established in a particular locality, which has become a nuisance from the growth of population and the erection of dwellings in proximity to it and that of a new erection or business threatened in such vicinity; and it re­ quires a much clearer case to justify a court of equity in interfering by injunction to compel a person to remove an establishment in which he has invested his capi­ tal and been carrying on business for a long period of time than would be required to prevent the establishment of an objectionable business by one who comes into the neighborhood proposing to establish such a business for the first time and is met at the threshold of his enterprise by a remonstrance of the inhabitants. (Barth v. Christian Psychopathic Hospital Association, 163 N.W. 62, 1917) The role of sequence in allowing for nonconforming land use is very evident when there is severe damage to the nonconforming structure due to fire, earthquake, and so on. Then the prior use is no longer prior and cannot be reinstated. “With the improvement substan­ tially destroyed, the land on which it is located will presumably have approximately as much value for use in conformity with the ordinance as otherwise and the public interest in conformity with the ordinance will be served if he is not permitted to continue the non­ conforming land use” (O’Mara v. Council of the City of Newark, 91 CA3d 156, 1965). Sometimes the nonconforming land use is allowed a certain amortization period. Consis­ tent with our economic analysis, such a scheme must consider all the costs and benefits. The relevant factors include remaining useful life, the harm to the public if the structure remains standing beyond the prescribed amortization period, and cost of moving. The lat­ Page 5 of 18

Ex Ante vs. Ex Post ter is relevant only if the nuisance is already there and should be there in the first place (see United Business Com. v. City of San Diego, 91 CA3d 156, 1979; City of Los Angeles v. Gage, 127 Cal. App. 2d 442, 1954). Note that the logic of the argument does not depend on whether it is a government regu­ lation or a private suit in a court of law. The optimal sequence is the optimal sequence whether it is enforced by private suit or government regulation. Much the same logic holds when new regulations are imposed. New houses and new addi­ tions to old houses may be required to have double-pane windows, while older houses are allowed to keep their single-pane windows. When installing a new window, (p. 427) the marginal cost of installing a double-pane window instead of a single-pane window is rela­ tively small in comparison to the cost of removing an old single-pane window and in­ stalling a new double-pane window. Hence new regulations are often grandfathered in (see Shaviro, 2000; Kaplow, 2003; Nash and Revesz, 2007; Shavell, 2008). In Spur Industries v. Del E. Webb Development Co., 494 P. 2d 700 (1972), Spur Industries had a cattle feedlot in an agricultural area well outside the boundaries of any city. Subse­ quently, Del Webb, a real estate developer, purchased land nearby and began to build a large retirement community. The court held that the developer was entitled to enjoin the cattle-feeding operation as a nuisance but was required to indemnify the cattle feedlot owner for the reasonable cost of moving or shutting down. The novel solution in this case reflects the fact that the cattle feedlot owner could not have reasonably foreseen the de­ velopment of a retirement community nearby (that is, the feedlot should have been first) and therefore should be compensated for all costs associated with moving his business. If the court had not questioned whether the developer should have been there in the first place, the court would have ruled for the feedlot to move without compensation, as it was cheaper to move the feedlot than to move the residents. But without compensation an in­ efficient precedent would have been established. Developers would choose areas inappro­ priate from a social cost–benefit analysis because the developers would not incur the costs to the adjacent nuisance of moving away. In future situations, even if it were cheap­ er for the builder to develop elsewhere than to make the feedlot move, the developer might move near the feedlot (if the developer did not have to compensate the feedlot owner). Thus courts are encouraging individuals to make efficient decisions by anticipat­ ing the future and considering the whole stream of costs and benefits and not just those costs and benefits that arise after both parties are located in the area.4 Before proceeding further, let us consider the previous cases altogether from a somewhat different perspective. In Spur Industries v. Del E. Webb Development Co., the victim moved next to the pre-existing and appropriately located injurer. Given this situation, the most efficient outcome was for the injurer to move. Del Webb had to pay for the moving costs, because either it was inefficient for Del Webb to move there in the first place or it was unclear whether it was inefficient, and the court wanted to make sure that the Del Webbs of the world would take into account the costs of making others move in their cal­ culations, thereby insuring efficient outcomes in the future. Clearly, if Spur Industries had Page 6 of 18

Ex Ante vs. Ex Post moved next to Del Webb, then Spur Industries would again have to move, but in this sce­ nario, Spur Industries would have to pay for the moving costs. In Bove v. Donner-Hanna Coke Corporation, Bove acted inefficiently when she built her rooming house in the area. It is clear from the facts of the case that it was efficient for Donner-Hanna to be built nearby even though Bove already had established her rooming house. That is, even considering the cost to Bove, it made sense to build the coke ovens. Although not discussed in the case, it appears that the optimal response by Donner-Han­ na was to not undertake any mitigating behavior. Given that the rooming (p. 428) house was there, the optimal response was to stay and possibly keep the rooming house win­ dows closed so as to reduce the unpleasantness of living there rather than tearing down the place and building elsewhere. However, it is likely that she would not have built there in the first place if Donner-Hanna had been there already. She was a “non-conforming vic­ tim.” In Mahlstadt v. City of Indianola, evidently the cost of moving the dump was greater than the cost to the residents from living next to the nuisance. Otherwise, we might have seen a solution similar to that of Spur Industries v. Del E. Webb Development Co. Because the dump was there first, the residents were not compensated by the city (the owner of the dump).

20.3 No Optimal Sequence but the First Party Acted Suboptimally We are now ready to shift our focus of investigation from land use to other areas of the law. We expand on many of the ideas already put forward, but who should be first recedes into the background. If a motorcycle is parked illegally in the middle of a straight highway and a truck smashes into it, who should be liable for the damage to the motorcycle? When a farmer plants seed that is known to be defective, what is the extent of liability by the seller of the defective seed? Note that in both cases there is a clear sequence, and in both cases the first person in time has acted suboptimally. The motorcycle is parked in the middle of the road and then a truck comes—not a motorcycle is parked in the middle of the road just below the crest of the hill and no vehicle would have had enough time to undertake defensive ac­ tion. The farmer discovers that the seed is defective and then plants the seed—not the farmer plants the seed and then discovers that it is defective when the plants do not grow. When these inputs occur simultaneously (equivalently, when the second party in the se­ quence does not have sufficient warning ahead of time), a variety of liability rules create the proper incentives for an efficient outcome. For example, under both negligence with contributory negligence and comparative negligence, the owner would be liable for all of

Page 7 of 18

Ex Ante vs. Ex Post the damage to her motorcycle. This would discourage owners of motorcycles from ineffi­ ciently parking their motorcycles in the middle of the road in the first place. However, when the inputs occur sequentially, none of the damage-based liability rules create the optimal incentives both for long-run efficiency and for second-best outcomes when the long-run efficient choice is not made by the first party. For example, a compara­ tive negligence rule cannot discourage truck drivers from smashing into motorcycles and simultaneously discourage motorcyclists from leaving their motorcycles in the middle of the road. If comparative negligence provides the correct incentives for truck drivers to swerve and avoid hitting parked motorcycles, then motorcyclists will have (p. 429) insuffi­ cient incentive to park their motorcycles in appropriate areas in the first place. If compar­ ative negligence provides the correct incentives for motorcyclists to park appropriately, then they will rarely park in the middle of the road; but on those occasions when they do, there will be insufficient incentives for truck drivers to optimally avoid damaging motor­ cycles. In a nutshell, we need a system of rules that create the right incentives for effi­ cient behavior and at the same time create the right incentives for those second in time to respond efficiently if the first party in time acts inefficiently. The way out of this dilemma is to have a liability rule that is based on marginal cost im­ posed on the other party rather than on actual damage. The first party in the sequence is liable to the second party for any additional prevention cost that the second party should undertake in response to the first party’s suboptimal behavior plus any damage (to either party) that would still occur if the second party responds optimally to the situation creat­ ed by the first party. The second party is then liable for any damage and own prevention cost. Because the second party is being compensated by the first, if the second party re­ sponds optimally to the first party’s suboptimal behavior, then the net liability of the sec­ ond party for damage and additional prevention will be zero. Of course, if the second par­ ty acts suboptimally, the additional damage that occurs when the second party acts sub­ optimally instead of optimally will fall on the second party. In turn, this means that the second party’s additional costs will be greater than any savings that might be gained from not taking the appropriate additional prevention. Thus the second party will have the right incentives to undertake the additional precaution in the first place. The logic will now be elucidated within the context of several examples. In contract law, marginal cost liability is known as the doctrine of mitigation of damages. The plaintiff in a breach of contract case cannot collect for losses (damages) that the plaintiff could have avoided by reasonable effort without risk of substantial loss or injury. In economic terms, “reasonable effort” means best response. The plaintiff can collect, however, for the avoidance costs that he should undertake in response to the breach and any damage that would still occur if the optimal response were undertaken. His recovery against the defendant will be exactly the same regardless of whether he makes the effort to mitigate his loss. Consider Daley v. Irwin, 205 P. 76 (1922). In this case, the plaintiff purchased defective seed but found out that the seed was defective before planting it. The court ruled that a Page 8 of 18

Ex Ante vs. Ex Post buyer who plants defective seed, knowing of its defects and having an opportunity to avoid the loss of the crop, cannot recover damages for the losses on the doomed-to-fail crop. The buyer’s recovery is limited to the difference in market value between the seed as warranted and the defective seed delivered. The court decision makes sense in terms of our economic analysis. If the farmer knows that the seed is defective (that is, the farmer’s decision comes second in the sequence), we want the farmer to undertake costeffective strategies to mitigate the damage. In particular, the farmer should not go through the expense of planting seed that he knows to be defective; instead, he should purchase new seed that is not defective and plant it. So the court ruled that the farmer could only collect for the cost of purchasing good seed (whether he did so or not). In gen­ eral, the farmer should only collect for the additional cost of prevention and any (p. 430) subsequent damage that would occur (perhaps the new seed had to be sown a week later, reducing the output) if the farmer were to optimally mitigate. The farmer cannot collect for damage that could have been prevented by the farmer. Of course, if the farmer did not know that the seed was defective (perhaps it was irradiated and therefore the defect was not observable), then the farmer could also collect for the cost of planting the seed. Being second in the sequence only counts when the second party knows or has reason to know that the outcome of the first party was suboptimal. A good illustration of the concept is in the first Restatement of Contracts (1932) §336 at 538. A sells to B a quantity of pork packed in barrels of brine with a warranty that the barrels will not leak. B finds that some of the barrels are leaky but does not repack the pork in good barrels. B can get judgment for only the cost of repacking in good barrels and not the value of the pork spoiled after her discovery of the defects. The victim should be compensated for the additional prevention cost incurred by the victim and resulting damage when she mitigates optimally. The victim is compensated for this amount (here, the damage that occurred before the leaks were discovered and the cost of repacking af­ ter the leaks were discovered) whether or not the victim acted optimally. Of course, given this rule, the victim will have the appropriate incentive to act optimally because the cost of any inefficient response by the victim falls on the victim. So the breaching party is nev­ er liable for the damage that could have been mitigated under the doctrine of avoidable consequences. This rule is most important for its effect on the breaching party. It is true that we want the victim to mitigate damages optimally, but the victim would also mitigate damages if the victim received zero compensation or any other fixed amount independent of the victim’s behavior. What this mitigation of damage rule does is create the optimal incen­ tive for the breaching party. If the breaching party were not liable for damages, the breaching party would breach whenever the cost of preventing breach was greater than zero; if the breaching party were liable for all damages, even if not mitigated, the breach­ ing party would not breach, even if the cost of preventing the breach were greater than the cost imposed on the breached party. The only rule that creates the right incentives for both parties is the doctrine of avoidable consequences (or what I call marginal cost liabili­ ty). Page 9 of 18

Ex Ante vs. Ex Post In situations where a breach of contract does not give the plaintiff a last clear chance to avoid damages (i.e. the inputs do not take place sequentially but are simultaneous), the plaintiff can collect for the full amount of damages. In Parkersburg Rig and Reel Co. v. Freed Oil and Gas Co., 205 P. 1020 (1922), the court held that the plaintiff need not build dams in anticipation of breaks in oil tanks built by the defendant (a breach of contract) and therefore could recover for all damages. Another situation where the plaintiff does not have the last clear chance is when the defendant assures the plaintiff that the con­ tract is not repudiated. In these cases, too, the defendant is liable for all damages. Contracting involves relatively low transaction costs. The marginal cost liability rule re­ duces transaction costs still further because it is a reasonable starting point for any nego­ tiation. That is, the rule provides a good estimate of the likely agreement if the breach were anticipated beforehand. In all these cases, the breacher of the contract pays for the marginal cost that the breacher imposed upon the other party because the breaching party undertook less than the optimal amount of precaution. Less than optimal means that it would have been more cost-effective for the breaching party not to breach than for the party breached against to (p. 431)

mitigate the breach. Because the breaching party pays for the cost of mitigation by the breached-against party, the breaching party has the correct long-run incentives to act op­ timally. And because the breached-against party is liable for any damage beyond what would occur if the breached-against party acted optimally given the breach, the breached-against party will undertake the optimal amount of additional prevention. The doctrine of last clear chance makes the last person who could have reasonably avoid­ ed an accident liable. This doctrine originated in 1842 in the case of Davies v. Mann, 152 Eng. Rep, 588. The plaintiff negligently left his donkey tied up in the middle of the road. The defendant was driving his carriage too fast to avoid hitting the donkey. The donkey subsequently died. Had the defendant swerved in time, the donkey would not have been killed. The court found the carriage driver liable as he had the last clear chance to avoid the accident. Let us consider a modern version where a truck runs into a motorcycle parked in the mid­ dle of a road. The motorcyclist who left her motorcycle on the road was the least-cost avoider; but given that the motorcycle was already there, the truck driver should be en­ couraged not to smash into it. Making the motorcyclist liable for the damage will, in the future, discourage people from parking their motorcycles in the middle of the road, but not sufficiently discourage others from smashing into them. Making the truck driver li­ able for the damage will create the proper incentives for truck drivers to avoid smashing into motorcycles but will not sufficiently discourage people from parking their motorcy­ cles in the middle of road. Once again, the appropriate rule is to make the motorcyclist liable for the marginal costs that she imposes on drivers whether there is an accident or not. In this case, the motorcy­ clist who parks in the middle of the road would pay for the additional cost (swerving, braking, slowing down, etc.) that she imposes on other drivers, even in the absence of any Page 10 of 18

Ex Ante vs. Ex Post damage from an accident. This payment would internalize the cost that the motorcyclist is imposing on others. Therefore, most motorcyclists would not park in the middle of the road in the first place and those that did would pay. The next step is to make the second party, the truck driver, liable for the marginal costs that he imposes on the motorcyclist if he fails to respond properly to the dangerous situation. The truck driver is then liable for any costs that he imposes on the motorcyclist beyond those that would occur if he had acted optimally in response to the dangerous situation. Hence, the truck driver also has the incentive to swerve if this is the optimal thing to do (obviously, it would not be optimal for the truck driver to swerve into oncoming traffic). The role of sequence is clear. If the truck driver is otherwise driving appropriately and does not have enough time to avoid the accident (he does not have the last clear chance), the truck driver is not liable for the ensuing damage (regardless of which party is hurt). We do not want truck drivers to swerve when there is no reason to do so. The motorcyclist should pay (to the extent feasible) for the marginal costs she im­ poses on others. In this way, motorcyclists will tend to make efficient choices. Clearly, a system of liability to all drivers who swerve or drive on another road in order to avoid hit­ (p. 432)

ting the motorcycle is totally impractical. The cost to individual drivers is minimal, but the sum total to all the drivers may be quite significant; therefore, we want the driver of the parked motorcycle to pay for the cost she shifts onto others. The appropriate solution is to charge the motorcyclist a fine equal to the total marginal cost imposed on others (a Pigovian tax on the input). This fine will create the same incentives for the motorcyclist as marginal cost civil liability. Whether the motorcyclist pays to the state or to an individ­ ual, the incentive effect on the motorcyclist is the same. If the motorcyclist is not always caught parking illegally on the highway, the fine can be increased to compensate for the reduction in probability. For example, if the motorcyclist is only caught half the time, the fine would be twice as large as it would be if she were caught all the time. We now investigate the effect of this Pigovian tax solution on truck and car drivers. Un­ der a marginal cost liability scheme, these drivers are compensated for the cost of pre­ ventive behavior, regardless of their taking preventive measures. Compensation does not affect their short-run behavior because it is not based on any decision they make but rather on the decision the motorcyclist makes.5 Consequently, if the motorcyclist pays a fine to the state instead of being liable to drivers who swerved, her lack of payment to these other drivers will have no effect on their care level. In other words, compensation for marginal costs that should be incurred is not necessary for the system to work proper­ ly. We want drivers to take into account the marginal costs that they impose. The doctrine of last clear chance, by making drivers liable for the damage when they are truly the second in the sequence, encourages them to make efficient decisions. Thus, a ticket for illegally parking the motorcycle in the middle of the road and liability for the damage on the dri­ ver of the moving truck (because he had the last clear chance) is a marginal cost liability rule applied in sequence.

Page 11 of 18

Ex Ante vs. Ex Post This analysis also provides an important reason for fining people for inputs such as drunken or reckless driving instead of relying only on liability for the resulting damage. Other drivers may swerve out of the way to avoid an accident (and the legal system should encourage them to do so). To the extent that these other drivers are successful, reckless drivers will not pay for their behavior if liability depends only on actual damage. Reckless drivers are not sufficiently deterred if the costs of protection are shifted onto other drivers. In other words, if liability were only based on damages, reckless drivers would not pay for some of the negative externalities that they create. Therefore, we have fines to make these reckless drivers liable for the cost of damage prevention they shift on­ to others. As demonstrated earlier, when there is a sequence of events, marginal cost liability should be applied regardless of whether there is damage or who is actually damaged. (p. 433) The law, in fact, reflects the symmetry of this analysis. A person is fined for illegal parking regardless of whether an accident occurred. Depending on which party had the last clear chance, the doctrine can be used against the defendant in favor of the plaintiff or against the plaintiff in favor of the defendant. Let us consider this latter possibility in greater detail. Suppose that it was the truck that was parked illegally on the road and the motorcycle smashed into it, harming the motorcyclist but not the truck. The truck would get a ticket for illegal parking. If the motorcyclist had the last clear chance, then the motorcyclist would be liable for the damage to the motorcyclist; if the motorcyclist did not have the last clear chance, then the truck would also be liable for the damage to the motorcyclist. This reverse situation reflects an actual case, Topping v. Oshawa Street Railway, 66 Ont. L. R. 618 (App. Div. 1931) where a truck was inappropriately parked on streetcar tracks, but the streetcar failed to slow down and as a result its passengers were hurt. In accident law, marginal cost civil liability is rare, but in contract law, it is the rule. There are several reasons for these two areas of law having different methods of determining li­ ability: (1) In contract cases, the second party in the sequence (who can mitigate dam­ ages) always knows who created the initial wrong—the breacher of the contract. The same does not hold for accident cases. If a car swerved suddenly to avoid hitting another car driving in the other direction but on the wrong side of the road, it could be very diffi­ cult to discover afterward who the reckless driver was. (2) When a breach of contract oc­ curs, cost-effective preventive action by the person who has the last clear chance may be very costly; for example, repacking the leaky pickle barrels. Therefore, it pays for the plaintiffs in breach of contract cases to sue even when there is “no damage” because suf­ ficient marginal costs are involved in preventive action to make it worthwhile for the plaintiff to collect for these costs (which are also known as damages in the legal litera­ ture). In contrast, the amount of liability that any one driver could collect for swerving (and thereby avoiding an accident) would not be enough to compensate for the transac­ tion cost of litigation. (3) It is much easier to determine sequence (who was first and who was last) in breach of contract than in accident cases. For all these reasons, we would ex­ pect the sequence of events to be much more important for breach of contract—where Page 12 of 18

Ex Ante vs. Ex Post mitigation of damages is the rule—than for accidents where last clear chance is rarely ap­ plied. The differences in the two areas of civil law are thus due to the different costs of in­ formation and the benefits of litigation. In nuisance law, when “damages” are awarded, they are typically for the costs of reason­ able preventive behavior by the plaintiff and any damage occurring (or that would have occurred) after preventive action had been taken. In other words, there is marginal cost liability. For example, in United Verde Extension Mining v. Ralston, 296 P. 262 (1931), the court held that the plaintiff need not plant a hopeless crop that would be swept by sulfur fumes and should not recover for the damage to the crop if he did. In this case, the plain­ tiff had the last clear chance and could have mitigated damages and prevented wasted seed and crop damage by not planting. However, the plaintiff recovered for profits for­ gone in not planting the crop. Thus, we have marginal cost liability or, in legal terminolo­ gy, the doctrine of avoidable consequence. “This doctrine states that a party (p. 434) can­ not recover damages flowing from consequences which the party could reasonably have avoided…. The doctrine has a necessary corollary—the person who reasonably attempts to minimize his damages can recover expenses incurred.” All of this leads us back to Spur Industries v. Del E. Webb Development Corporation. The court held that the developer was entitled to enjoin the cattle-feeding operation as a nui­ sance, but it was required to indemnify the cattle feeder for the reasonable cost of mov­ ing or shutting down. This novel solution by the court reflects the fact that there are three stages to this sequential game: the feedlot was there first; the developer should have built elsewhere and thus had the “last clear chance”; but given that the plaintiff had built next to it, the feedlot had the last “last clear chance” to mitigate the damage. The lo­ cation of the feedlot far away from any city imposes little marginal cost on potential de­ velopers as they could readily avoid the nuisance by building elsewhere. However, given that the developer had located near the feedlot, it was best to have the feedlot move. Be­ cause the developer imposed the marginal costs, the developer should pay for them. As mentioned in the previous section, in Bove v. Donner-Hanna Coke Corporation, the rooming house was inefficiently built in the wrong place, but the optimal response by Donner-Hanna, the second party, was to do nothing to mitigate the damage. In Mahlstadt v. City of Indianola, it is unclear whether the developer acted inefficiently, but in any event, the optimal response was not for the dump to move. Moving a dump is more costly than moving a cattle feedlot. The same logic holds for the law of rescue. If a person can be rescued at low cost, it is clearly economically efficient to do so. The Continental version of the Good Samaritan rule encourages low-cost rescues in two ways. First, the rescuer is compensated for the small costs of rescue; second, if the potential rescuer’s costs are somewhat higher than the average so that the reward does not fully cover all of the rescuer’s costs, then the threat of being liable for non-rescue will motivate the person to rescue.

Page 13 of 18

Ex Ante vs. Ex Post The rule also provides the appropriate incentives for those who might need rescue. By charging for the average cost of the rescue, the rescuee takes the appropriate level of care. A higher price for rescue would result in the potential rescuee being overly cautious and needing too few rescues. For example, consider the possibility of a person deciding to swim at a beach. If the lifeguards charge $10,000 for a simple rescue, the person might not go swimming in the first place, or if the person does decide to go swimming, the per­ son might buy some very expensive equipment to haul along so that a self-rescue could be performed. However, if the price paid to rescuers is lower than the costs that the res­ cuers incur in rescue, then either there will be too many people needing rescue and being rescued (if the low price does not affect the supply of potential rescuers) or there will be too few rescuers and, as a consequence, too few people putting themselves at risk. To un­ derstand the logic behind the latter case, suppose that tow truck operators were paid less than their cost of towing stalled cars off the freeway to an auto repair facility. Then there would be fewer tow trucks and fewer than the optimal number of cars on the road that were likely to need towing. A less prosaic example is the rescue of boat people who were fleeing Vietnam. Ships are required to rescue people that are stranded at sea, but the res­ cuers are not compensated for their rescue efforts. Consequently, many (p. 435) ships un­ dertook circuitous routes to avoid areas where there were likely to be boat people in need of rescue. In turn, this meant that there were fewer boat people than optimal because the danger to them was unnecessarily high. In a nutshell, for there to be the correct economic incentives, both liability by the rescuee for being rescued and liability by the rescuer for non-rescue must be in place. The Conti­ nental rule employs both.6 In general, the Anglo-American rule employs neither; however, there are many exceptions based on special relationships. A number of statutes require a driver who is involved in an accident to offer assistance to the accident victims regardless of fault. In most American jurisdictions, the ship owner is liable if the captain fails to take reasonable measures to rescue passengers or crew who jump overboard. Passengers (crew) pay for these rescue services via higher ticket prices (lower wages). More impor­ tantly, “individuals in charge of a vessel shall render assistance to any individual found at sea in danger of being lost, so far as the … individual in charge can do so without serious danger to the … individual’s vessel or individuals on board (46 U.S.C. §2304; see also Canada Shipping Act, R.S.C. 1985 c. S-9, s. 451(1), and the U.K. Maritime Convention Act, 1911, 1 & 2 Geo. 5, c. 57, s. 6). Failure to do so may lead to a fine and/or imprisonment. As a final example, a physician who treats an unconscious person can collect her regular fee. The cost of rescue is not always trivial, and because of that, people are sometimes pre­ vented from undertaking certain risky actions in the first place. One cannot take off from the coast of Canada in an attempt to fly over the Atlantic Ocean in a hang glider with a snowblower engine because the likelihood of needing rescue is very high, with the cost of rescue being much higher than the would-be flyer is likely to have. This is an actual story (see the Santa Cruz Sentinel, July 23, 1980). More generally, mountain climbers are now required to post bond for their rescue.7 Page 14 of 18

Ex Ante vs. Ex Post

20.4 Sometimes Giving Rights to the First Is Ip­ so Facto Efficient In many cases, giving the right to the second party in time would completely undermine the incentives to invest that exist when the first party is given the right. Just think what it would mean if we gave the right to a patent to the second party in line or gave the right of salvage to the second discoverer of a sunken ship. The arguments for priority of the first (p. 436) over the second is so obvious in these cases that there is no need to dwell on the argument.8 So, we now turn to more subtle issues. It is generally the case that the first party in time knows less about the second party in time than the second party knows about the first party. And if this is the main distinguish­ ing characteristic, then the first party should have the right. We have already seen an ex­ ample in Mahlstadt v. City of Indianola. In the absence of zoning, the city that owned the dump knew very little about the dump’s future neighbors. They could be residences, chemical plants, or recycling businesses. But Mahlstadt knew about the dump. Rather than developing dumps in one place and then moving them, possibly more than once, it makes more sense for potential later arrivals to avoid building near the dump if the un­ pleasantness outweighs the lower cost of building nearby. Now dumps and houses are quite dissimilar, which may mask the essential argument. So lets us consider a situation where the actions are identical except that they take place se­ quentially in time. Sometimes secured debt is owned by two different entities. For exam­ ple, a homeowner may have borrowed $100,000 from lender A in 2003 and $100,000 from lender B in 2006. The homeowner may have used her $250,000 house as collateral. If the house were worth at least $200,000 at the time of bankruptcy in 2011, there would be no problem. But what if housing prices greatly deteriorated so that the house was worth only $150,000 when the homeowner declared bankruptcy? Three priority schemes are possible: (1) The first lender has priority; that is, lender A receives $100,000 and lender B receives $50,000. (2) The second lender has priority; lender A receives $50,000 and lender B receives $100,000. (3) Neither has priority; each receives $75,000. Bank­ ruptcy law has implemented the first scheme. This reduces the need for monitoring by the first creditor, although it increases it for the second creditor, who should have a com­ parative advantage in monitoring since she can see more recent housing prices and salary of the borrower and update the measure of risk involved. Most important, the sec­ ond lender knows about the existence of the first loan, but the first lender can only specu­ late about the existence and nature of a second loan. Giving the first creditor the rights to the asset first means that later creditors and the debtor cannot post-contract alter the value of the secured debt to the first creditor. If the first creditor shared instead of having priority (or was second in line), then the first creditor would probably insist on a new con­ tract whenever the asset had a second creditor. This would raise transaction costs. When the second creditor is second in line, she always knows what the bargain is when she steps into it (see Jackson and Kronman, 1979).

Page 15 of 18

Ex Ante vs. Ex Post Now it is conceivable that unsecured debt could also be prioritized (first to lend is first in line for being paid), but it seems highly impractical. To employ the language used in the introduction, the transaction costs would be very high. Would a person on night shift be in line after a person on day shift, and by how much? Would a seller of toilet paper be in line before the seller of paper clips, because the contract for the toilet paper was made before the paper clip contract, or would it be based on delivery, and how would one know the timing anyway or how much was owed? So it is not surprising that (p. 437) unsecured debt is not prioritized by time, but by other criteria (managers will know that the firm is going into bankruptcy before workers know and therefore the latter have priority) and that all members within a class are treated equally. It should be observed that priority does not depend on when a party demanded to be paid. So, rushing to be first is not a problem here. Turning our attention toward real estate sales, if A sells her house to B and then sells the same house to C, then B has priority over C. To do the reverse would clearly be problem­ atic because D might also come along after C, which would mean that no buyer would know whether he bought the house. Of course, no one has priority in time until the seller has agreed in writing to sell to the particular person. Until then, offers are just offers and no priority need be given to the first person to make an offer. Dividing lines are also found in hunting. In Pierson v. Post, 3 Cai R. 175 (N.Y.1805), Post was chasing a fox, but Pierson killed it. The court awarded the fox to Pierson because chasing (like offering to buy a house) is not sufficient to establish ownership. But once the fox was shot or captured, ownership is clear.

20.5 Areas for Future Research Much of the extant theory is based on the courts being fully informed. As in other areas of investigation, a natural intellectual step is to analyze situations where there is asymmet­ ric information. We have already seen how the likelihood of the second party in time knowing more than the first shifts rights to the first and responsibilities to the second. Further advances in our understanding of the law are likely to come about by considering several complications at once. Asymmetric information can lead to imperfect bargains by the parties involved in the conflict and/or mistaken decisions by judges. In turn, this may mean that the optimal set of rights and responsibilities of the first and second party change. Beginning forays have been undertaken by Pitchford and Snyder (2003) and Innes (2008). Both articles assume a more limited knowledge set by the judges than I have assumed here. More can be done along these lines in the future.

References Epstein, Richard (2002). “The Allocation of the Commons: Parking on Public Roads.” Jour­ nal of Legal Studies 31: 515–544.

Page 16 of 18

Ex Ante vs. Ex Post Gray, Kevin (2011). “The Legal Order of the Queue.” Cambridge University Working Pa­ per. Innes, Robert (2008). “Coming to the Nuisance: Revisiting Spur in a Model of Location Choice.” Journal of Law, Economics and Organization 25: 286–310. Jackson, Thomas H. and Anthony T. Kronman (1979). “Secured Financing and Priority among Creditors.” Yale Law Journal 88: 1143–1182. Kaplow, Louis (2003). “Transition Policy: A Conceptual Framework.” Journal of Contempo­ rary Legal Issues 13: 161–209. Lueck, Dean (1995). “The Rule of First Possession and the Design of the Law.” Journal of Law and Economics 38: 393–436. (p. 438)

Nash, Jonathan R. and Richard L. Revesz (2007). “Grandfathering and Environmental Regulation: The Law and Economics of New Source Review.” Northwestern University Law Review 102: 1677–1734. Perry, Ronen and Tal Z. Zarsky (2014). “Queues in Law.” Iowa Law Review 99: 1596–1658. Pitchford, Rohan and Christopher Snyder (2003). “Coming to the Nuisance: An Economic Analysis from an Incomplete Contracts Approach.” Journal of Law, Economics and Organi­ zation 19: 491–516. Shavell, Steven M. (1983). “Torts in Which Victim and Injurer Act Sequentially.” Journal of Law and Economics 26: 589–612. Shavell, Steven (2008). “On Optimal Legal Change: Past Behavior and Grandfathering.” Journal of Legal Studies 37: 37–85. Shaviro, Daniel (2000). When Rules Change: An Economic and Political Analysis of Transi­ tion Relief and Retroactivity. Chicago: University of Chicago Press. Wittman, Donald (1980). “First Come, First Served: An Economic Analysis of Coming to the Nuisance.” Journal of Legal Studies 9: 557–568. Wittman, Donald (1981). “Optimal Pricing of Sequential Inputs: Last Clear Chance, Miti­ gation of Damages and Related Doctrines in the Law.” Journal of Legal Studies 10: 65–91. Wittman, Donald (1982). “Efficient Rules of Thumb in Highway Safety and Sporting Activ­ ity.” American Economic Review 72: 78–90.

Notes: (1) See Wittman (1982) for a more extensive discussion of highway safety rules, Epstein (2002) for a discussion of parking rules, and Gray (2011) and Perry and Zarsky (2014) for a more general discussion of queues. Page 17 of 18

Ex Ante vs. Ex Post (2) Other things being equal, the side that has the most to gain is the most likely to under­ take the early investment. But other things need not be equal: the cost of premature in­ vestment to one polluter may be more or less than the cost of premature investment to one resident, and the first polluter (pollutee) may not capture the benefits accruing to lat­ er polluters (pollutees). (3) The option of moving the nuisance will be discussed later in this section when we dis­ cuss Spur Industries. (4) Spur is discussed at length in Wittman (1981), Pitchford and Snyder (2003), and Innes (2008). (5) Lack of compensation might affect their long-run behavior, in particular their decision whether to drive. (6) Note that it would be hard to determine whether boats that took circuitous routes were doing so in order to avoid having to rescue people. But if they were paid for the costs that they incurred for rescuing others, they would be unlikely to choose circuitous routes in the first place. Of course, the boat people were unlikely to have the wherewithal to pay for their rescue (a point that we will come back to). (7) For a more extensive discussion of the Good Samaritan Rule and related issues regard­ ing sequence, see Wittman (1981) and Shavell (1983). (8) However, why one uses sequence instead of the price system is not always so clear.

Donald A. Wittman

Donald A. Wittman is Professor of Economics at the University of California, Santa Cruz.

Page 18 of 18

Carrots vs. Sticks

Carrots vs. Sticks   Giuseppe Dari-Mattiacci and Gerrit DeGeest The Oxford Handbook of Law and Economics: Volume 1: Methodology and Concepts Edited by Francesco Parisi Print Publication Date: Apr 2017 Subject: Economics and Finance, Law and Economics, Econometrics, Experimental and Quanti­ tative Methods Online Publication Date: May 2017 DOI: 10.1093/oxfordhb/9780199684267.013.41

Abstract and Keywords This article draws a general picture of the differences between the metaphors of carrots and sticks. It discusses incentive effects (in principle, a $100 carrot creates the same in­ centives as a $100 stick, but there are exceptions); transaction costs (carrots are paid up­ on compliance, sticks upon violation, therefore sticks have lower transaction costs if the majority complies); risks (probabilistic carrots create risks for compliers, probabilistic sticks for violators); wealth and budget constraints (the maximum carrot depends on the principal's wealth, the maximum stick on the agent's wealth, but sticks can have a multi­ plication effect); distributive effects (carrots may overcompensate, sticks may undercom­ pensate; individualizing sanctions changes the distributive effect of carrots but not of sticks); activity level effects caused by these distributive effects; the principal's incentives to behave opportunistically; and the agent's incentives to self-report. The article also dis­ cusses special types (precompensated, annullable, combined, intra-group financed, re­ versible, strict liability carrots and sticks) and two extensions (political risks, behavioural effects). Keywords: incentive effects, transaction costs, risks, wealth constraints, budget constraints, distributive effects, activity level effects

21.1 Introduction INCENTIVES can be generated by either carrots—promises to reward, such as with prizes or bonuses—or sticks—threats to punish, such as with fines or damages.1 Carrots and sticks are prima facie equivalent, because any behavioral change induced by promis­ ing compliers a $100 reward can also be obtained by threatening violators with a $100 punishment. Yet, while carrots and sticks seem to produce the same effects, they are not chosen at random; some general patterns can be observed across legal systems. Incen­ tives for careful driving are generally created by holding negligent drivers liable under Page 1 of 29

Carrots vs. Sticks tort law, rather than by rewarding careful drivers. Likewise, theft is discouraged by pe­ nalizing thieves and not by rewarding those who do not steal. Incentives to invent are in­ stead created by rewarding successful inventors with patents or academic prizes rather than by punishing all others. The goal of this chapter is to draw a broad picture of the differences between carrots and sticks. Our goal is not only to identify the differences but also to examine their causes and the relationships among them. The chapter is organized as follows. In section 21.2, we characterize several versions of the incentive equivalence result. In section 21.3, we ana­ lyze the differences between carrots and sticks in terms of transaction costs and, in sec­ tion 21.4, of risk allocation. Next, we consider wealth and budget constraints (section 21.5); distributional effects (section 21.6); activity-level effects (section 21.7); and the principal’s incentive to behave opportunistically ex post and the agent’s incentives to selfreport in a biased way (section 21.8). Section 21.9 discusses pre-compensated, (p. 440) an­ nullable, combined, intra-group redistributed, reversible, and strict liability carrots and sticks. Section 21.10 considers two extensions (political risks and behavioral biases). We conclude in section 21.11 by offering guidelines for the optimal use of carrots and sticks.

21.2 The Incentive Equivalence Result In principle, a carrot has the same incentive effect as a stick of the same magnitude. If a principal wants an agent to do something with an effort cost of $99, the principal can of­ fer the agent a carrot of $100 or threaten the agent with a stick of $100. In both cases, the agent will comply. Nonetheless there can be cases in which a carrot and a stick of the same magnitude have different incentive effects. The most obvious case is when money has decreasing margin­ al utility for the agent, and a monetary carrot is used to induce a non-monetary effort.2 In this case, adding $100 to the agent’s wealth changes the total utility less than taking away $100 from the agent’s wealth. Since repeated carrots further increase the agent’s wealth, carrots may even have a saturation effect: there may be a wealth level at which a carrot can no longer incentivize an agent to make a significant effort.3 Aside from the decreasing marginal utility of money, the incentive equivalence result holds under a broad set of conditions. To identify these conditions, consider a simple mod­ el in which a benevolent ruler (the principal, “she”) enforces a rule of conduct over a pop­ ulation of wealth-maximizing risk-neutral individuals (the agents, “he”). Compliance with the rule yields a benefit for society but entails a private effort cost, which is different across individuals. The level of enforcement determines a cut-off level of effort so that all individuals with an effort cost less than or equal to the cut-off exert effort. We will there­ fore refer to the cut-off level of effort as level of enforcement. To be effective, the rule of conduct is supported either by a stick for violators or by a carrot for compliers.

Page 2 of 29

Carrots vs. Sticks We take the desired level of enforcement as given and ask the question whether this level can be best achieved through carrots or through sticks.4 As long as we assume risk neu­ trality of all parties involved, this question is equivalent to the question whether enforce­ ment is cheaper (or more effective) with carrots or with sticks.5 Let us first demonstrate three basic incentive equivalence results. (p. 441)

21.2.1 Simple Case: Certain Monitoring with No Errors

Let us introduce the following notation:6 e = the effort cost of an individual agent (e* is the level of enforcement); c = the reward (carrot); s = the punishment (stick); n = the number of agents in the population; F(e) = the distribution of the cost of effort. Assume that monitoring occurs with probability equal to one and that the ruler can per­ fectly and costlessly verify if an agent has exerted effort. Note that the ruler cannot verify the individual’s cost of effort and cannot use individualized carrots and sticks.7 If the ruler uses carrots, an individual earns c–e upon compliance and 0 upon violation; thus, an individual will comply if his effort cost is e ≤ c. Similarly, if the ruler uses sticks, an individual pays e upon compliance and s upon violation; thus, he will comply if e ≤ s. Carrots and sticks are equivalent in terms of incentives: a carrot and a stick of the same magnitude induce the same level of compliance. Therefore, if the ruler wants to induce all individuals with effort cost e ≤ e* to comply—so that, in expectation, nF(e*) individuals comply and violate—she can do so equivalently with a carrot or with a stick, so that we have the following incentive equivalence equation:8

(1)

21.2.2 Probabilistic Monitoring The ruler might monitor the behavior of individuals with a probability less than one, so that pc = the monitoring probability with carrots; ps = the monitoring probability with sticks.

Page 3 of 29

Carrots vs. Sticks With carrots, an individual will comply with the rule if e ≤ pc c. Likewise, with sticks, an individual will comply with the rule if e ≤ ps s. Again, carrots and sticks are equivalent (p. 442) with respect to incentives in that the same level of enforcement e* can be ob­ tained equivalently through carrots or through sticks:

(2)

Note that the combination of probability and sanction is not unique and that if c > s, then pc < ps and vice versa. If instead carrots and sticks are applied with the same probability pc = ps = p, then we have the following stronger version of the incentive equivalence equation with probabilistic monitoring:

(3)

21.2.3 Enforcement Errors While verifying whether an individual exerted effort, the ruler might make two types of errors. Even though an individual exerted effort, the ruler might erroneously believe that the individual exerted no effort (a type-I error). Similarly, even though an individual exert­ ed no effort, the ruler might erroneously believe that the individual exerted effort (a typeII error):9 εI = probability of type-I error; εII = probability of type-II error. With carrots, an individual earns

upon compliance and pc εII c upon viola­

tion, since there is a probability εI that a complier is not rewarded and a probability εII that a violator is rewarded. Hence, an individual will comply with the rule if e Likewise, with sticks, an individual pays violation and will comply with the rule if

upon compliance and

upon

.10 Again, the incentive ef­

fects of carrots and sticks are the same and are similarly diluted by errors. If the monitor­ ing probabilities differ, incentive equivalence takes the form

(4)

If instead carrots and sticks are applied with the same probability

, we have

the following version of the incentive equivalence equation with probabilistic monitoring and enforcement errors:

Page 4 of 29

.

Carrots vs. Sticks

(5)

(p. 443)

21.3 Transaction Costs

21.3.1 Carrots Are Applied Upon Compliance, Sticks Applied Upon Vi­ olation Assume now that applying carrots and sticks generates a transaction cost. Note that we need to consider only the cost of applying the sanction and not the cost of monitoring the agent; the latter cost depends on the probability of monitoring p (which is only indirectly affected by the choice of carrots or sticks through a possible change in the optimal proba­ bility of monitoring).11 Let: kc = the unit cost of applying a carrot; ks = the unit cost of applying a stick. This set-up is very general and it allows the unit costs to be zero, less than one or even greater than one (meaning that applying a sanction costs more to the ruler than the in­ centive it produces). With kc = ks = 0 we focus on the pure-transfer case, where a benevo­ lent ruler only considers the incentive effect of monetary sanctions, which simply move resources and hence yield no social costs. As the unit costs increase, the analysis covers situations in which, for instance, a ruler only bears the costs of paying carrots while not internalizing the costs of sticks (kc = 1, ks = 0), or situations in which both carrots and sticks imply some costs for the ruler . Note that this formulation also cov­ ers non-monetary carrots and sticks, which typically have different costs for the ruler and the agents. For instance, the cost ks can be interpreted as the ratio between the unit disu­ tility of a physical punishment (for instance whipping or incarceration) for an individual and the unit costs of applying that punishment for the ruler (for instance the number of hours that the individual cannot work due to the punishment). The total expected cost of applying a carrot is

(6)

which includes the probability of applying the carrot correctly to compliers incorrectly to violators pc εII. Similarly, with sticks, the total expected cost is

Page 5 of 29

and

Carrots vs. Sticks (7) (p. 444)

Given that the enforcement level is the same in both cases, we can set cpc = sps

(incentives equivalence). Thus, comparing the two costs, we have that the costs are less with carrots if

(8)

Otherwise, the costs are less with sticks. Note that the ratio on the right-hand side de­ pends on two things. First, it decreases if the level of enforcement e* increases, implying that, if more people comply, carrots become relatively more costly to apply. This is due to the fact that, although there are errors, carrots are mostly paid to compliers. Second, it depends on the balance between type-I and type-II errors. If εII grows relative to εI—that is, if there are more frequent errors with violators than with compliers—then applying carrots becomes relatively more costly than applying sticks. The intuition is that the costs of sanctions are realized when sanctions are applied. Thus, wrong applications of a sanc­ tion weigh more on costs than wrong denials of a sanction. With carrots, εII determines the probability of applying the carrot incorrectly and hence weighs more than εI, which determines the probability of wrongly denying a reward. The opposite is true with sticks. Note that this argument applies irrespective of the population n and of the probability of monitoring p and hence is valid also if there is only one individual or if the probability of monitoring is equal to one. Moreover, to abstract from the difference between type-I and type-II errors and focus in general on the quality of the information that the ruler re­ ceives, assume that εI = εII = ε. Thus, we have:

(9)

The right-hand side increases in ε if

½ and decreases in ε otherwise. This implies

that the effect of reduced enforcement errors on the costs of carrots and sticks depends on the level of enforcement. If more than half of the population complies, then higher en­ forcement errors translate into higher costs for sticks, because sticks are now frequently applied by mistake. From a different angle, an improved quality of the information that the ruler gathers makes carrots more desirable only if most of the population complies. The condition in (9) embeds two separate effects that need to be disentangled. First assume that there is full enforcement, F(e*) = 1. This might be the case if there is only one individual with a known effort cost, so that the expected sanction can be set such that the individual has incentives to comply. Then, carrots yield lower costs if the probability of errors is sufficiently high. That is, as the quality of the (p. 445) signal that Page 6 of 29

Carrots vs. Sticks the ruler receives deteriorates, carrots become more attractive than sticks (Dari-Mattiac­ ci, 2013).

(10)

21.3.2 Effect of Equilibrium Compliance (Majoritarian Criterion) Now assume that there are no errors. Here the optimal choice between carrots and sticks depends on the level of enforcement. Then, (9) becomes

(11)

If the costs of carrots and sticks are the same (kc = ks), we obtain a pure majoritarian cri­ terion: carrots should be applied if fewer than 50% of the agents comply, sticks otherwise (Wittman, 1984). With different costs (kc ≠ ks), the fractions of compliers and violators re­ ceive a weight proportional to these costs.

21.3.3 Effect of Information: Sticks May Punish Those Unable to Comply, Carrots May Reward Those Unable to Violate The previous result suggests that carrots are only superior when the majority of the agents will violate the rule in equilibrium. This raises the question why this would ever be the case in a rational choice framework. If an agent rationally violates a rule, this means that the carrot or stick is not high enough. But in that case, why does the principal not simply increase the carrot or stick until all agents comply? The reason is that the principal my not have enough information on the individual effort costs of all agents to determine which of them should follow the rule. Suppose the benefit of compliance is always $100, while the cost is $90 for agent A and $110 for agent B. If the principal is benevolent, she wants only agent A to comply with the rule. If she were fully informed, she would specify that the rule only applies to agent A and set the stick or carrot at $91 so that agent A would comply. But in the real world, individual effort costs may be difficult to verify. In this case, the principal should take into account that some agents will be unable to comply (for instance, if the rule is “compose good music,” most citizens might be unable to comply at reasonable cost). Similarly, there (p. 446) may be agents who are unable to violate (for instance, if the rule is “do not hack computers,” most citizens might be unable to violate the rule even if they wanted to). This suggests that carrots may be desirable when the principal has specification prob­ lems, that is, when the principal does not know what can be reasonably expected from Page 7 of 29

Carrots vs. Sticks the agent. This may help to explain why in an increasingly complex society, carrots are in­ creasingly used (De Geest and Dari-Mattiacci, 2013).12 An extreme case of enforcement with information problems is the volunteer’s dilemma: it is optimal for society that only one of the bystanders (ideally the one with the lowest ef­ fort costs) jumps in the water to save a drowning person. In this case, Leshem and Tab­ bach (2016) show that carrots are superior to sticks in that they minimize the transfers to be made.

21.4 Risk: Carrots Create Risk for Compliers, Sticks Create Risk for Violators Individuals might be risk-averse or risk-loving. Obviously, their reaction to probabilistic carrots and sticks changes. Yet, it is important to isolate the effect of risk-aversion from the confounding effects of increased wealth: carrots make individuals richer while sticks make them poorer, so that their behavior might change not only as a result of risk-bear­ ing but also because of endowment effects. In order to isolate the effect of risk-bearing, we focus on exponential utility functions, which have the unique property of exhibiting constant risk preferences, irrespective of wealth. Starting from risk-aversion, let: risk-averse individual’s utility of wealth, with risk-aversion coefficient a ≥ 0.

13

If there are no enforcement errors, with carrots, an individual complies if

(12)

where the left-hand side is what the individual earns upon compliance and the right-hand side is what he earns upon violation. Likewise, with sticks, an individual complies if

(13) (p. 447)

With pc = ps = 1, we find the standard equivalence result: the agent complies if c

= s ≥ e. However, with pc = ps = p < 1, risk-aversion makes individuals react more strong­ ly to sticks than to carrots. From (12) and (13) we have that a carrot and a stick applied with the same probability induce compliance with e* if

Page 8 of 29

Carrots vs. Sticks (14)

which implies

. Vice versa, if the carrot and the stick have the same magnitude,

the probability of monitoring needs to be greater under carrots, pc > ps. The left-hand side of (14) indicates that the probability of applying a carrot needs to balance the individual’s utility when exerting effort with his utility when receiving the carrot, as in the risk-neutral case. Instead, the right-hand side of (14) indicates that the probability of ap­ plying a stick needs to balance the individual’s utility when exerting effort with his utility when paying the stick, times an additional factor less than one, which is the ratio of the marginal disutilities of the stick and effort. This factor magnifies the effect of sticks (re­ quires lower probabilities of monitoring). Sticks have a stronger incentive effect than car­ rots because compliance in the case of sticks is risk-free, while it implies risk in the case of carrots. The case of risk-loving individuals yields the opposite result. Let Ewl – 1 = a risk-loving individual’s utility of wealth, with risk-loving coefficient l ≥ 0. The analysis is similar to the case of risk-aversion and yields s > c > e* or, conversely, ps > pc. This is because now risk is a valued feature and hence carrots, which imply risk for compliers, induce more compliance than sticks.

21.5 Wealth and Budget Constraints The existing wealth of the principal and agents matters for carrots and sticks in two re­ spects.14 First, a carrot is a transfer from the principal to the agent; a stick is a transfer from the agent to the principal. Therefore, the principal should be able to finance the car­ rot and agents should be able to pay the stick. The principal’s wealth (or the budget she is able to raise) is thus an upper constraint on the use of carrots; the agent’s wealth is an upper constraint on the use of sticks. Second, the application of the carrot or stick itself may be costly to the principal. This is especially the case for a non-monetary stick, such as imprisonment. Also in this case, the principal’s budget may become a binding con­ straint. We will now analyze how these constraints affect the use of carrots and sticks.

21.5.1 The Maximum Carrot Is Determined by the Wealth of the Principal, the Maximum Stick by the Wealth of the Agent (p. 448)

Carrots need to be financed and sticks need to be paid. If we assume that it is the princi­ pal who finances the carrot and consider a simple setting,15 with only one agent and one required effort, then the maximum carrot equals the principal’s wealth (or maximum bud­ get); similarly, the maximum stick equals the agent’s wealth.16 But when there are more agents or more required efforts and monitoring is probabilistic, the situation becomes more complex. Suppose that there are ten agents, all with a 10% Page 9 of 29

Carrots vs. Sticks chance to be monitored and to receive a $1 million bonus if found complying. Does the principal need a budget of $1 million or of $10 million? This depends on whether monitor­ ing is coordinated. If monitoring is organized in such a way that there is only exactly one agent monitored, the principal never has to pay more than $1 million. If monitoring is un­ coordinated, so that it may happen that all agents are monitored at the same time, the principal may have to pay $10 million. To analyze the wealth (and budget) constraints more formally, let: w = the wealth of an individual agent;17 b = the ruler’s budget. Carrots and sticks are bound by the ruler’s budget available to apply the carrot or the stick. At the individual level, the maximal stick that can be applied is also bound by the in­ dividual level of wealth smax = w. In general, one could envisage a boundary for carrots as well, which may be due to saturation, linked somehow to the individual’s wealth. In the following, we focus on the ruler’s constraints. From the ruler’s perspective, the constraints on the total maximum expenditures in equi­ librium can be derived from the total costs (6) and (7). Note, however, that these equa­ tions reflect expected total costs for the ruler. This implies that costs could be higher or lower, which is important for incentive purposes. For example, a monitoring probability equal to 50% and a carrot equal to $100 imply that the expected carrot is $50. However, if the ruler has a budget equal to $50, the individual will anticipate that the ruler will be able to pay only $50 (not $100), and hence his incentives will be diluted. To maintain the incentives unaltered, the ruler’s budget needs to be able to cover the worst-case sce­ nario, that is, her budget needs to be equal to the budget that would be necessary if pc = 1, which in this case is $100. Note that if the probability of monitoring (p. 449) were 10% and the carrot $500, the expected cost for the ruler would be again $50, but her budget (for incentive purposes) would need to be $500—ten times greater. The fraction of com­ plying individuals is also probabilistic if the ruler only knows the probability distribution of the costs of effort. A similar argument applies to errors. In essence, the ruler needs to be able to pay a carrot as if monitoring occurred with probability equal to one, the whole population complied, and there were no type-I errors. Likewise, the ruler needs to be able to apply a stick as if monitoring occurred with probability equal to one, the whole popula­ tion violated the rule, and there were no type-II errors. By setting all probabilistic variables in (6) and (7) at their worst-case-scenario level, we obtain the following constraint for carrots:

(15)

Similarly, with sticks, the ruler’s constraint is Page 10 of 29

Carrots vs. Sticks

(16)

Note that the maximal sanction seems to increase with n, but this effect is illusory, as the maximal sanction is calculated at the individual level. At the population level the total sanction is ncmax or nsmax, so that the effect of n cancels out from the ruler’s perspective. In the following section we examine how these constraints can be lessened.

21.5.2 Coordinated Monitoring To see how coordinated monitoring may increase the maximum sanction for both carrots and sticks, consider two monitoring strategies. In one case (uncoordinated monitoring), the ruler randomly decides whether to monitor a single individual by a lottery draw with probability p and then repeats the procedure for each individual. In another case (coordi­ nated monitoring), the ruler randomly draws pn individuals from the population and then monitors them.18 Clearly, the second strategy is constrained by the fact that pn must be a discrete number and hence only a discrete number of different probabilities are imple­ mentable. Yet, this constraint weakens as n grows. If n = 1, the only applicable coordinat­ ed-monitoring probability is p = 1; but if n = 10, coordinated monitoring admits ten differ­ ent probabilities: 1/10, 2/10, …, 9/10, 1. As n grows bigger this constraint becomes less binding. In both cases an individual faces a probability p of being monitored, hence, incentives are constant across the two treatments. However, from the perspective of the ruler (p. 450) these two strategies are markedly different, because coordinated monitoring lessens the budget constraints identified above. To see why, note that coordinated monitoring trans­ forms a probabilistic variable into a certain number for the ruler. Monitoring each individ­ ual with probability p yields a distribution of possible outcomes, which includes the worstcase scenario in which (albeit with a very small probability, pn)19 all individuals are moni­ tored. Instead, coordinated monitoring yields a certain number of monitored individuals, pn. Therefore, under coordinated monitoring the budget constraints in (15) and (16) reduce to:

(17)

and

(18)

Page 11 of 29

Carrots vs. Sticks Whenever p < 1, coordinated monitoring makes the constraints less binding (higher maxi­ mal sanctions). Moreover, the maximal sanction increases as the probability of monitoring decreases and does so proportionally. Coordinated monitoring increases the maximal sanction by a factor 1 / p both for carrots and for sticks. This effect implies that the minimal probability of monitoring is not bound­ ed by the ruler’s budget because the ruler, by reducing the probability of monitoring, in­ creases proportionally the maximal sanction applicable and hence can reduce the proba­ bility of monitoring at will.20

21.5.3 Population Effect Consider now two scenarios, which derive from two different interpretations of F(e). If F(e) is a probability distribution of the effort costs in the population, then nF(e*) is the ex­ pected number of compliers. If instead F(e) describes the real distribution of effort costs in the population—that is, if F(e) is the ratio of the number of individuals with effort costs equal to or less than e and the total number of individuals—then nF(e*) is the real number of compliers. The ruler knows more in the second scenario than in the first, although in both scenarios she does not know the effort costs of an individual. (The difference be­ tween these two scenarios is particularly clear if one considers a population composed of one individual.) If the ruler knows the real distribution, then she can predict with certainty how many individuals will be rewarded. If there are no type-II errors,21 this knowledge further lessens the budget constraint of carrots, which becomes: (p. 451)

(19)

(If we do not have coordinated monitoring pc = 1.) The same reasoning does not apply to sticks. The reason is that with sticks the ruler needs to be able to punish all individuals, not only the violators, or the threat of punishment would not be credible. To see this point in the sharpest way, consider a ruler who enforces a rule e* on an individual with effort cost e ≤ e*. The individual complies. If the constraint were determined only by violators, a ruler with no budget would be able to do the job. But this cannot be the case, since a ruler with no budget cannot credibly threaten the application of (costly) sanctions. The argument applies more broadly to populations of any size: the budget has to be large enough to punish all violators and to threaten (credibly) all compliers. The population effect can be summarized as follows: If the ruler knows the population, carrots have to be available to reward only actual compliers, while sticks have to be avail­ able to punish all potential violators (not only the actual violators).

Page 12 of 29

Carrots vs. Sticks

21.5.4 The Multiplication Effect of Sticks The fact that sticks need to be available for all potential violators needs to be reconsid­ ered in light of the fact that individuals might not be able to coordinate their actions. If they are able to coordinate, the analysis of the previous section applies. If they are not able to coordinate, then in fact one stick might be enough to incentivize all potential vio­ lators (Dari-Mattiacci and De Geest, 2010).22 To see why, consider the extreme case in which a dictator owns only one bullet, but uses this single bullet to make all seven billion world citizens live in slavery. To accomplish this, the dictator goes to the first citizen and presents him with the choice between living in slavery or being shot. The citizen opts for slavery and hence the bullet is not used; therefore, the dictator goes to the second citizen, who also opts for slavery. The dictator repeats this threat seven billion times. At the end of the day, the total effort cost of all seven billion citizens is much larger than the stick (which can destroy only one human life). For all agents taken as a group, it might be better to violate the rule and (p. 452) al­ low one of them to be punished, yet no agent has an individual incentive to sacrifice him­ self.23 A necessary condition for this multiplication effect to hold is that all agents know the or­ der in which the ruler will monitor and penalize them. If this is the case, the availability of one extra stick (besides those already used up on individuals who have higher costs of ef­ fort than the rule requires) is enough to incentivize the first individual who will be moni­ tored. But if the first individual complies, then the stick is not used up and will provide in­ centives to the second individual and so forth. At the limit, one stick is enough for any number of compliers. Therefore, (18) can be rewritten as follows:

(20)

Comparing the new ruler’s constraint under sticks with the constraint under carrots in (19), we have that carrots are preferred if

(21)

The three terms on the right-hand side of the inequality provide a balance of factors that favor carrots or sticks. Let us start with the simplest scenario in which the costs of car­ rots and sticks are the same (kc = ks) and they are applied with the same probability (pc = ps). Furthermore, let us disregard for simplicity the term 1 / n as it becomes very small for large n. Then, the condition in (21) says that the rulers’ constraint is less binding with carrots if Page 13 of 29

, which is the ratio of violators to compliers.24 This reveals two gen­

Carrots vs. Sticks eral principles. First, the multiplication effect of sticks is stronger (giving a stronger ad­ vantage to sticks over carrots) as n grows, since the condition becomes more difficult to satisfy. Second, this advantage also grows if the number of compliers increases. In nor­ mal circumstances, compliers outnumber violators and hence the condition becomes n < 1, so that carrots always yield a more stringent budget constraint for the ruler than sticks do.25 The reason why only sticks have a multiplication effect is as follows. Carrots are applied upon compliance, sticks upon violation. If the agent complies, the carrot is used up but the stick is not. Although a stick can be applied only once, the threat to apply the stick can be repeated several times. (p. 453)

Note that the multiplication effect is another qualification of the incentive equivalence re­ sult: a $100 stick can, under some conditions, have an incentive effect of more than $100; a $100 carrot, by contrast, can never have an incentive effect of more than $100. The upside of the multiplication effect is that it makes law enforcement more effective: it allows governing a country of 300 million citizens with relatively few prison cells—it is sufficient that there are some empty prison cells.26 The downside of the multiplication ef­ fect is that sticks carry an inherent risk of abuse, which is absent in carrots.

21.6 Distributional Effects The primary goal of carrots and sticks is to make an agent follow a certain rule. Nonethe­ less, carrots and sticks may have significant distributional side effects. Suppose a princi­ pal wants to make an agent exert an effort of $90. If the principal threatens the agent with a $100 stick, the agent will comply and become $90 poorer than before the rule. If the principal promises a $100 carrot instead, the agent will also comply but he will be­ come $10 wealthier than before the rule (by receiving $100 compensation for a $90 ef­ fort).

21.6.1 Sticks Undercompensate, Carrots May Overcompensate Sticks inherently undercompensate because the agent is always worse off compared to the status quo: either he incurs an effort cost or a penalty. (Note that indirectly the agent may receive compensation under a stick regime in the form of receiving the benefit of rule compliance by the other agents). Carrots, by contrast, have a built-in compensation mechanism in that they always allow the agent to opt for the status quo, in which the agent does not do the effort and simply does not receive the carrot. While carrots can never be undercompensatory (because then they would violate the in­ centive compatibility constraint), the question remains why they would ever be (p. 454) overcompensatory. Why cannot the principal set the carrot equal to the effort cost, so that there is no overcompensation?

Page 14 of 29

Carrots vs. Sticks A first reason is that the principal may have imperfect information on the individual effort cost. In this case, the carrot may be set higher than strictly required to induce perfor­ mance; the agent’s rent is then essentially an information rent. This is especially the case if agents have varying effort costs and the principal knows only the distribution F(e) of the cost (but not the individual costs). In this case, agents with a relatively low effort re­ ceive relatively high rents. A second reason why carrots may be overcompensatory is that the principal makes moni­ toring errors by mistakenly concluding that some violating agents complied. Since viola­ tion leads to an (even small) expected rent, compliance needs to include this expected rent too in order to give sufficient incentives to comply. Note that errors create fixed rents or impose fixed costs that are equal for all individuals in expected terms; therefore, the only distributional effects within the group of agents derive from differences in effort costs. With carrots, compliers and violators earn:

(22)

The pay-offs from complying and violating are the same only for individuals with e = e*. Given that the optimal carrot is such that and compliers have e ≤ e*, the pay-off of compliers is greater than zero and increases as e decreases, as indicated by the first inequality. Moreover, all individuals earn a rent equal to pc εII c. The intuition is that, due to type-II errors, some violators may earn a carrot. To restore incentives, the pay-off of compliers needs to be increased accordingly, thereby generating a rent for com­ pliers equal (at the margin) to the fraction of carrots paid to violators. Moreover, since the carrot is the same for all individuals, compliers below the margin earn an even greater rent, which increases as their effort cost decreases. With sticks, compliers and violators pay:

(23)

Again, the pay-offs from complying and violating are the same only for individuals with e = e*. Given that the optimal stick is such that and compliers have e ≤ e*, the cost for compliers is greater than zero and increases in e up to ps εI s + e*. The intuition is that, due to type-I errors, some compliers are punished. To restore incentives, the expected stick has to be raised above e*. These greater sticks also fall onto violators, generating a fixed cost for all individuals, which is greater than e* if there are type-I er­ rors. Page 15 of 29

Carrots vs. Sticks

21.6.2 General Carrots and Sticks Make Compliers with a High Effort Cost Relatively Poorer Compared to Compliers with a Low Ef­ fort Cost. This Effect Can Be Removed Through Individualized Car­ rots but Not Through Individualized Sticks (p. 455)

General carrots make compliers with lower effort cost richer. If the principal knows the individuals’ effort costs, carrots can be made individual-specific, so that each individual receives a carrot that is just enough to incentivize him. This removes the distributional ef­ fects that were caused by the general carrot, both with respect to non-agents—that is, in­ dividuals not subject to enforcement—and agents. Remarkably, individualizing sticks does not make their distributional effects disappear. To illustrate, suppose that agent A has an effort cost of $80 while agent B has an effort cost of $90. Instead of threatening all agents with a $100 stick, the principal now threatens agent A with a $81 stick and agent B with a $91 stick. Both agents comply. At the end of the day, compliance still made agent A $80 poorer and agent B $90 poorer. The reason why the distributional effects of individualized sticks are the same as of general sticks is that sticks are meant not to be applied. If they are not applied, their magnitude cannot change the distribution.

21.7 Activity-Level Distortions Caused by Dis­ tributional Effects We have seen that carrots and sticks have different distributional effects. These distribu­ tional effects may in turn cause activity-level effects. Stick regimes naturally undercom­ pensate and therefore also tend to violate the participation constraints of agents. Carrots, on the other hand, are either fully compensatory or overcompensatory. In the latter case, they generate a rent for the participating agent. Therefore, carrots and sticks also gener­ ate different incentives for entry into the regulated activity. Carrots attract entry of compliers, and incentives to enter increase as e decreases. If there are type-II errors, carrots generate a fixed rent from entry for both compliers and violators. Sticks discourage entry of compliers and violators. The complier’s incentives not to enter increase as e increases. If there are type-I errors, sticks generate a fixed cost of entry for both compliers and violators. Because of these activity-level effects, Wittman (1984) argued that, as a rule of thumb, carrots need to be used for positive externalities and sticks for negative externalities. In­ deed, the overcompensatory or undercompensatory effects can generate desirable (p. 456) activity-level effects when there are positive or negative externalities associated with the activity (Dari-Mattiacci, 2009).

Page 16 of 29

Carrots vs. Sticks

21.8 The Principal’s Opportunism and the Agent’s Distorted Incentives to Self-Report Caused by the Distributional Effects 21.8.1 The Principal’s Incentives to Behave Opportunistically So far we have assumed that after the agent has complied, the principal enforces the rule as announced, that is, she pays the promised carrot, or she does not apply the stick. In practice, however, a principal may behave opportunistically for two reasons. First, under a carrots regime, the principal may want to keep the bonus for herself. Second, under a sticks regime, the principal may falsely declare that the agent violated the rule, in order to receive the proceeds of the fine. More formally, assume the following timeline: at t0, the rule is announced, at t1 the agent complies with or violates the rule, at t2 the principal monitors and applies the carrot or stick. As always, opportunism is caused by the timing of the actions; the principal is that party who acts last and who therefore may benefit from not acting as announced. It is nonetheless important to see the relationship between this opportunism and the dis­ tributional effects of carrots and sticks. Carrots increase the agent’s wealth, sticks in­ crease the principal’s wealth (if the principal receives the proceeds of the fines). There­ fore, the principal may not want to apply the carrots and may want to apply the sticks. Because of this danger of opportunism, both carrots and sticks require commitment by the principal. If the principal cannot commit, the agent will violate the rule, either be­ cause he expects not to receive the carrot at t2 or because he expects to be punished at t2 even if he complies at t1. But opportunism is symmetric here: it exists irrespective of whether carrots or sticks are used. Yet the symmetry may disappear if special variants of carrots or sticks are used, as will be explained in section 21.9.

21.8.2 The Agent’s Incentives to Self-Report Agents have a natural incentive to self-report compliance under a carrots regime. Indeed, if they are found complying, they receive the carrot. By contrast, agents do not (p. 457) have natural incentives to self-report a violation under a sticks regime since, if they are found violating the rule, they receive a penalty.27 Another way of looking at this is that, in a probabilistic enforcement regime, agents have an incentive to manipulate the probability of monitoring p. Under a carrots regime, com­ plying agents try to increase p; under a sticks regime, violating agents try to decrease p.

Page 17 of 29

Carrots vs. Sticks Principals can use these features of carrots to increase p when p is naturally low, for in­ stance because the behavior is to a large extent non-verifiable without the cooperation of the agent.28

21.9 Special Types of Carrots and Sticks 21.9.1 Pre-Compensated Carrots and Sticks The distributional effect of sticks and carrots can be cancelled out through transfer pay­ ments. Stick regimes, which are naturally unattractive for agents, can be made attractive by offering agents an entry fee. Carrot regimes that generate rents for agents can be made less attractive to agents by requiring them to pay an entry fee; as long as the entry fee does not exceed the rents, the participation constraint of the agent is still met.29 When the distributional effects are removed in this way, the activity-level distortions caused by the distributional effects are removed as well. In practice, however, the legal system may forbid entry fees, especially when they are paid by employees. One concern is indeed that entry fees (paid by the agent) give the principal an additional incentive to be opportunistic. Similarly, entry bonuses paid to the agent (at t0) give the agent an incentive to be opportunistic.

21.9.2 Annullable Carrots and Sticks Annullable carrots are carrots that are paid unless the agent is monitored and found vio­ lating. They differ from normal carrots, which are paid if the agent is monitored and (p. 458) found complying. Similarly, annullable sticks are sticks that are applied unless the agent has been monitored and found complying. In essence, an annullable carrot is a threat to take back, an annullable stick a promise to give back (De Geest, Dari-Mattiacci, and Siegers, 2009). Annullable carrots are mathematically identical to normal carrots if the probability of monitoring is 100%. Indeed, the difference between the normal and the annullable vari­ ants consists in what happens in case of no monitoring. If all agents are monitored all the time (p = 1), the difference disappears. Consider the case in which an employee receives a bonus unless he is underperforming to the case in which the employee receives the bonus only if he is found performing. If monitoring takes place with certainty, the employ­ ee will receive a bonus if he complies and no bonus if he violates in both scenarios. Yet there is an important difference if the probability of monitoring is less than 100%. In those cases, violating employees who were lucky enough not to be monitored still receive a bonus under annullable carrots. An example of annullable carrots can be found in the efficiency wages literature (Shapiro and Stiglitz, 1984; Becker and Stigler, 1974). Here it has been argued that overpaying an employee can make sense because it gives the employee an extra incentive not to be fired. Indeed, the overpayment can be seen as a carrot that is paid unless the employee is Page 18 of 29

Carrots vs. Sticks monitored and found shirking. This extra incentive not to shirk may allow the employer to lower the monitoring levels and save on monitoring costs. This indicates that annullable carrots have a source of ineffectiveness that is absent in regular carrots. Regular carrots are only paid to agents who have been monitored (and found complying). Annullable carrots are paid not only to agents who have been moni­ tored (and found complying) but also to agents who have not been monitored (and who may have complied or shirked). Paying a bonus to an agent who has not been monitored obviously has no incentive effect, because incentives are generated by the wedge be­ tween what compliers and violators receive. Therefore, a carrot paid to non-monitored agents is, from the viewpoint of the employer, “wasted money.” This wasted money may indirectly lead to inefficiency. Indeed, a profit-maximizing em­ ployer faces a trade-off between paying rents and bearing monitoring costs: the higher the overpayment, the lower the monitoring costs can be. But if rents have to be paid also to the agents who are not monitored, the employer will adopt inefficiently high monitor­ ing levels in order to cut the frequency with which such rents are paid. This monitoringlevel distortion is not present in a normal carrots regime, in which the bonus is only paid to monitored, complying agents (De Geest, Dari-Mattiacci, and Siegers, 2009). If annullable carrots are intrinsically wasteful, why would rational principals ever use them? The answer is that the annullable variant changes the timing of the payments and therefore also the problem of opportunism. More specifically, annullable carrots give the principal an incentive to monitor as promised. In contrast, under normal carrots, the em­ ployer may ex post (after the agent has performed) decide not to monitor in order not to have to pay the agent a bonus. As a result, annullable carrots (such as efficiency (p. 459) wages) can be rational choices when the principal cannot credibly commit to paying car­ rots with a certain probability.30 Another way of looking at annullable carrots is to see them as fully pre-compensated sticks. Indeed, agents first virtually receive full pre-compensation c and virtually pay back c (that is, pay s) if found violating. Since the lump sum payment is sunk, this suggests that annullable carrots will have many of the characteristics of normal sticks. And indeed, they have the same risk properties (they create risk for violators), and treat non-monitored agents in the same way as monitored compliers—just like normal sticks do. Similarly, an­ nullable sticks have the risk properties of normal carrots. Yet the lump sum payment changes some of the distributional distortions. While normal sticks usually underpay (that is, they underpay unless there is sufficient in-kind compen­ sation in the form of the received benefit), annullable carrots do compensate the agents for their effort costs—as a matter of fact, they have to pay the same overcompensation to monitored compliers as normal carrots. But while normal carrots pay no compensation to non-monitored agents (and therefore correctly compensate in expected terms, at least if effort costs are identical), annullable carrots pay the same overcompensation to non-mon­ itored agents. Therefore, their distributional characteristics are different from those of Page 19 of 29

Carrots vs. Sticks normal sticks and carrots. Similarly, the distributional effects of annullable sticks are greater than those of normal sticks. The annullable variants also generate more transaction costs, though the analysis de­ pends on whether or not the original payment and the annihilation of that payment are set off against each other. Without set-off, the annullable variants clearly generate more transaction costs than their normal variants: they require a lump sum transaction in 100% of the cases and on top of that there will be an annihilation transaction for some of them. But even with set-off (where the lump sum pre-compensation is only virtually paid beforehand, but effectively paid ex post, in order to allow for transaction cost-saving setoffs) annullable carrots and sticks generate higher transaction costs because they also re­ quire transactions with non-monitored agents.31 Since annullable carrots can be seen as pre-compensated sticks, they will also make sense in most cases in which sticks make sense. For instance, when only shirking can be proven, (normal) sticks can work but (normal) carrots cannot. In this case, annullable (p. 460) carrots can work as well. An example is corruption, where typically the only par­ ties who have information on the corrupt act are the two parties involved (the bribed agent and the bribing third party). Therefore, it may be possible to prove corruption only when the third party betrays the agent or when the principal happens to acquire the rele­ vant information. Since the quality of the information is such that monitoring can only show that an agent is corrupt, but never reveal that the agent is honest, the fact that an agent has not been found to be corrupt during a monitoring session does not necessarily prove that he is honest.

21.9.3 Combined Carrots and Sticks The combined use of carrots and sticks increases the potential maximum sanction/reward and therefore may create stronger incentives than their single use. Indeed, the theoreti­ cal maximum of the sanction/reward under the combined use now becomes the wealth of the agent plus the total wealth of the principal. The combined use, however, also leads to higher transaction costs because a transfer is needed both in case of compliance and in case of violation. In addition, the combined use also changes the risk properties, by creat­ ing risks for both compliers and violators.

21.9.4 Intra-Group Financed Carrots and Intra-Group Redistributed Sticks So far we have assumed that it is the principal who both finances carrots and receives the proceeds of the sticks. Now consider the case in which it is the violating agents who fi­ nance the carrots for compliers, or the complying agents who receive the sticks paid by violators. An interesting question in such a regime is what happens when all agents comply. Since no agent is left to finance the carrots, all complying agents may have to finance the car­ rot, in which case no complier receives a net carrot. Similarly, if all agents violate, the Page 20 of 29

Carrots vs. Sticks proceeds of the fines may have to be redistributed to all agents, in which case no violator pays a net stick. But the fact that no agent pays or receives anything in this equilibrium does not mean that there are no incentives to comply. Indeed, what matters for incentives is that the first agent to comply or violate faces a change in pay-offs—which is the case here. This intra-group redistribution of carrots and sticks has three benefits. First, it removes the principal’s incentive to behave opportunistically by making the principal a neutral rather than a financially interested party.32 Second, for the reasons explained in the previ­ ous paragraphs, there are no transaction costs in the equilibrium in which all agents com­ ply or violate, which is a remarkable property for a regime that has an (explicit or implic­ it) carrot component. Third and foremost, intra-group redistribution (p. 461) strengthens incentives by creating combined sanctions—carrots for the compliers and sticks for the violators. Yet, the marginal incentives now become a function of the behavior of the others. Consid­ er first a stick system with intra-group redistribution of the proceeds of the stick (and in which individual sticks remain constant). The more agents violate, the more attractive it becomes to comply because the stick proceeds are both larger and have to be split among fewer compliers; if all others violate, the incentives to comply are at their strongest. This means that such a regime can be particularly effective for weakest-link offenses, such as cartels, in which it is sufficient that one individual cooperates with the competition au­ thority. Contrast this to a carrot system with intra-group redistribution (assuming again that the individual carrots remain constant). The more agents cooperate, the less attractive it now becomes to violate because the total carrots are both larger and need to be financed by fewer violators; if all others comply, the incentives to comply are the strongest. There­ fore, such a regime is most effective for rules that need to be followed by all agents.

21.9.5 Reversible Carrots Reversible rewards (Ben-Shahar and Bradford, 2012) are amounts of money that will ei­ ther be used to reward the agent (if he complies with the rule) or to finance his punish­ ment (if he violates the rule). The rewards are “reversible” because they are reversed in­ to stick enforcement funds if the agent violates the norm. Reversible rewards are a special type of combined sanctions. One special feature is that the reward is not combined with a stick of the same magnitude but with a stick enforce­ ment budget of the same magnitude. This budget may be lower or higher than the stick it enforces. Therefore, the incentive effect may be more (or less) than twice the incentive effect of the reward alone. Another feature is that the budget the principal is going to spend is independent of the choice of the agent. This fixed character of the amount makes it easier for the principal to pre-commit to spending it (for instance by transferring the amount to a fund controlled by a third party). This pre-commitment reduces the danger of Page 21 of 29

Carrots vs. Sticks ex post opportunism by the principal, which in turn makes the scheme more credible and effective. Reversible rewards can be useful when the enforcement costs associated with sticks are significant but cannot be financed through the proceeds of the stick (for instance because the stick is non-monetary). An example may be the economic boycott of another country, which is often as costly to the boycotter as to the boycotted.

21.9.6 Strict Liability Carrots and Sticks So far our analysis has focused on fault-based rules, in which the principal specifies good or bad behavior, and applies sanctions, possibly with errors, after observing good or bad behavior. Now consider strict liability rules, which measure only output but do not specify optimal input. Such rules can make sense when specification problems become too large because effort costs are insufficiently verifiable, while output is still sufficiently verifiable. (p. 462)

Under strict liability, sticks may also have to be paid upon “compliance.” Indeed, even if the agent reduced the output (for instance pollution) to the optimal level (by taking opti­ mal care), there is still a “sanction” applied as long as the output is not zero. As a result, sticks may lose a comparative advantage—that they do not have to be applied upon com­ pliance. Moreover, “compliers” (who can be defined here as those agents who do what is optimal from the perspective of the principal) still face a risk (though a lower one than “violators”) if monitoring of output is probabilistic. Another difficulty with strict liability is that defining carrots or sticks requires defining a baseline at which zero has to be paid. To illustrate, suppose that carrots would be used to reduce pollution. At what point should carrots start to be given? At the maximum possible level of pollution? But this may be infinite. This is why Lazear (1991) suggested (in the context of bonuses or penalties for workers) that sticks should be used if the maximum performance can more easily be defined, and carrots when the minimum performance can more easily be defined.

21.10 Extensions 21.10.1 Political Risks So far, we have assumed that the principal chooses the optimal type of enforcement mechanism. However, when the principal is a political institution, public choice mecha­ nisms may distort the decision-making process. Carrots (in the form of subsidies) may be more distortive in this respect, because well-organized groups may benefit more from ob­ taining a carrot for their own group than from obtaining a stick (Galle, 2012).

Page 22 of 29

Carrots vs. Sticks

21.10.2 Behavioral Biases Behavioral biases may further complicate the analysis. For instance, when agents have loss aversion (Kahneman and Tversky, 1979), negative incentives may be more effective than positive incentives. Or when agents are imperfectly aware of the rule, carrots may become relatively more effective because they tend to be more salient (Galle, 2014).33 Some findings in psychology seem to favor the use of carrots. For instance, when behav­ ior is followed by a positive consequence (“reinforcement”), it tends to be repeated; (p. 463) when it is followed by a negative consequence, it tends to be avoided (Thorndike, 1913; Ferster and Skinner, 1957). These results are in line with the activity-level distor­ tions caused by the different distributional effects of carrots and sticks. Similarly, the general tendency among psychologists to be more favorable toward carrots than toward sticks (see Kohn, 1993) is compatible with our observation that, in an increasingly com­ plex society, parents and employers increasingly face specification problems. There is a large literature in experimental economics studying carrots or sticks, but only a relatively small fraction of it deals with the choice between carrots and sticks.34 Most of the effects that we identify have not been tested empirically. Finally, a related literature studies the emergence of preferences for punishing or rewarding in an evolutionary mod­ el (Herold, 2012).

21.11 Conclusion: The Optimal Choice Between Carrots and Sticks Overall, should carrots or sticks be used? Sticks have an intrinsic advantage over carrots: they do not have to be applied if the agent complies. Moreover, probabilistic sticks create risks for violators rather than for compliers, which further increases their effectiveness when agents are risk-averse. Because of these advantages, sticks will be superior in sim­ ple settings, in which the principal has sufficient information about the agent’s effort cost to make sure that the agent should and will comply. Yet, carrots may be superior in two cases (De Geest and Dari-Mattiacci, 2013). The first is when the principal faces specification problems, that is, when she does not know what to expect from each individual citizen because she does not know their effort costs. In those cases, sticks may punish agents who are unable to comply with the norm at reasonable costs, which leads to higher transaction costs, risk costs, and undesirable wealth effects. Specification problems may explain why carrots (in the form of copyrights) are used to in­ centivize composing music (the legal system does not know which individuals are able to do this) or rescuing at sea (the legal system does not know which part of the cargo of a sinking ship should be rescued in emergency situations with time constraints). The second case in which carrots may be superior is when a significantly higher effort is required from some individual agents than from others (for instance, when only some families need to send a family member to the army, or only some families need to sacrifice Page 23 of 29

Carrots vs. Sticks land for a highway project). In such cases of singling-out, sticks would cause significant unintended distributional distortions (impoverishing those from whom much is (p. 464) re­ quired). This may help to explain why carrots are increasingly used in complex societies. When agents specialize, specification and singling-out problems increase.

Acknowledgments We thank Brian Galle, Theo Offerman, Alexander Stremitzer, and Donald Wittman for comments. We also thank Megan Lasswell for research assistance.

References Abbink, Klaus, Bernd Irlenbusch, and Elke Renner (2000). “The Moonlighting Game: An Experimental Study on Reciprocity and Retribution.” Journal of Economic Behavior & Or­ ganization 42: 265–277. Andreoni, James, William Harbaugh, and Lise Westerlund (2003). “The Carrot or the Stick: Rewards, Punishment, and Cooperation.” American Economic Review 93: 893–902. Ayres, Ian (2010). Carrots and Sticks: Unlock the Power of Incentives to Get Things Done. New York: Random House. Bar-Gill, Oren and Omri Ben-Shahar (2009). “The Prisoners’ (Plea Bargaining) Dilemma.” Journal of Legal Analysis 1: 737–773. Becker, Gary S. (1968). “Crime and Punishment: An Economic Approach.” Journal of Polit­ ical Economy 76: 169–217. Becker, Gary S. and George Stigler (1974). “Law Enforcement, Malfeasance, and Com­ pensation of Enforcers.” Journal of Legal Studies 3: 1–18. Ben-Shahar, Omri and Anu Bradford (2012). “Reversible Rewards.” American Law and Economics Review, 15(1): 156–186. Brooks, Richard R. W., Alexander Stremitzer, and Stephan Tontrup (2012). “Framing Con­ tracts: Why Loss Framing Increases Effort.” Journal of Institutional and Theoretical Eco­ nomics 168: 62–82. Dari-Mattiacci, Giuseppe (2009). “Negative Liability.” Journal of Legal Studies 38: 21–60. Dari-Mattiacci, Giuseppe (2013). “Slavery and Information.” Journal of Economic History 73: 79–116. Dari-Mattiacci, Giuseppe and Gerrit De Geest (2010). “Carrots, Sticks, and the Multiplica­ tion Effect.” Journal of Law, Economics, and Organization 26: 365–384. De Geest, Gerrit and Giuseppe Dari-Mattiacci (2013). “The Rise of Carrots and the De­ cline of Sticks.” University of Chicago Law Review 80: 341–392. Page 24 of 29

Carrots vs. Sticks De Geest, Gerrit, Giuseppe Dari-Mattiacci, and Jacques J. Siegers (2009). “Annullable Bonuses and Penalties.” International Review of Law and Economics 29: 349–359. Ferster, Charles B. and Burrhus Frederic Skinner (1957). Schedules of Reinforcement. Englewood Cliffs, NJ: Prentice-Hall. Galle, Brian D. (2012). “The Tragedy of the Carrots: Economics and Politics in the Choice of Price Instruments.” Stanford Law Review 64: 797–850. Galle, Brian D. (2014). “Carrots, Sticks, and Salience.” Tax Law Review 66: 53–110. Garoupa, Nuno and Matteo Rizzolli (2012). “Wrongful Convictions Do Lower Deterrence.” Journal of Institutional and Theoretical Economics 168: 224–231. Herold, Florian (2012). “Carrot or Stick? The Evolution of Reciprocal Preferences in a Haystack Model.” American Economic Review 102: 914–940. (p. 465)

Kahneman, Daniel and Amos Tversky (1979). “Prospect Theory: An Analysis of Decision under Risk.” Econometrica 47: 263–291. Kohn, Alfie (1993). Punished by Rewards. Boston, MA: Houghton Mifflin. Lando, H. (2006). “Does Wrongful Conviction Lower Deterrence?” Journal of Legal Stud­ ies 35: 327–338. Lazear, Edward P. (1991). “Labor Economics and the Psychology of Organizations.” Jour­ nal of Economics Perspectives 5: 89–110. Leshem, Shmuel and Avraham Tabbach (2016). “Solving the Volunteer’s Dilemma: The Superiority of Rewards over Punishments.” American Law and Economics Review 18: 1– 32. Miceli, Thomas and Kathleen Segerson (2007). “Punishing the Innocent along with the Guilty: The Economics of Individual versus Group Punishment.” Journal of Legal Studies 36: 81–106. Nosenzo, Daniele, Theo Offerman, Martin Sefton, and Ailko van der Veen (2014). “En­ couraging Compliance: Bonuses Versus Fines in Inspection Games.” Journal of Law, Eco­ nomics, & Organization 30: 623–648. Nosenzo, Daniele, Theo Offerman, Martin Sefton, and Ailko van der Veen (2016). “Discre­ tionary Sanctions and Rewards in the Repeated Inspection Game.” Management Science 62: 512–517. Offerman, Theo (2002). “Hurting Hurts More Than Helping Helps.” European Economic Review 46: 1423–1437. Png, I. P. L. (1986). “Optimal Subsidies and Damages in the Presence of Judicial Error.” International Review of Law and Economics 6(1): 101–105. Page 25 of 29

Carrots vs. Sticks Polinsky, A. Mitch and Steven Shavell (2000). “The Economic Theory of Public Enforce­ ment of Law.” Journal of Economic Literature 38: 45–76. Rizzolli, M. and M. Saraceno (2013). “Better that Ten Guilty Persons Escape: Punishment Costs Explain the Standard of Evidence.” Public Choice 155: 395–411. Shapiro, Carl and Joseph E. Stiglitz (1984). “Equilibrium Unemployment as a Worker Dis­ cipline Device.” American Economic Review 74: 433–444. Tabbach, Avraham (2010). “The Social Desirability of Punishment Avoidance.” Journal of Law, Economics and Organization 26: 265–289. Thorndike, Edward Lee (1913). Educational Psychology. New York: Teachers College Press. Wittman, Donald A. (1984). “Liability for Harm or Restitution of Benefit?” Journal of Legal Studies 13: 57–80.

Notes: (1) A carrot can be defined as a payment from the principal to the agent upon compliance of the agent. A stick is then a payment from the agent to the principal upon violation by the agent. While a carrot can sometimes be rewritten as a mathematically identical stick through the use of entry fees, this is not the case when enforcement is probabilistic. See De Geest and Dari-Mattiacci (2013). (2) Risk aversion might also be invoked to justify a difference. (3) Note that saturation may also occur when the effort is monetary but the carrot is over­ compensatory, so that the agent receives a rent and becomes wealthier in each round. (4) The socially optimal level of enforcement is a separate question that is the object of the (extensive) economic literature on the public enforcement of law (see Polinsky and Shavell, 2000). Note that this literature focuses on sticks (because criminal law is a stickbased regime) and that there may be unexplored subtle differences between carrots and sticks in this respect. (5) Risk adds complications that will be discussed in section 21.4. (6) Unless otherwise noted, all variables are positive. All probability variables are be­ tween 0 and 1. (7) We discuss individualized carrots and sticks in section 21.6.2. (8) The level of enforcement e* is taken as given; its optimal setting is the focus of a large literature on public law enforcement (see note 4). (9) It is natural to assume that εI and εII are less than ½. Page 26 of 29

Carrots vs. Sticks (10) Png (1986) was the first to show that type-I and type-II errors dilute deterrence in the same way with respect to sticks. In the literature on public enforcement of law there are several papers identifying differential effects of type-I and type-II errors (for instance Lando, 2006; Garoupa and Rizzolli, 2012; Rizzolli and Saraceno, 2013). (11) The choice of the optimal level of monitoring (and of monitoring costs) and the opti­ mal mix between probability of monitoring and magnitude of the sanction is treated in the literature on public enforcement of law and we abstract from it here; see note 4. (12) Specification problems and errors in the application of the sanction may even explain why Roman slave-masters used sticks for simple, physical tasks and carrots for complex tasks (such as running a business) (Dari-Mattiacci, 2013). (13) Not to confuse it with effort, we denote the exponential function as E(·). (14) Wealth constraints may be binding when the principal wants to minimize the proba­ bility of monitoring, as in Becker (1968). (15) In section 21.9.4 we consider intra-group financing of carrots, that is, financing by the violating agents. (16) For simplicity, we assume away the part of the parties’ wealth that needs to be re­ served for paying the transaction costs associated with monitoring and with the applica­ tion of the carrot or stick. (17) We assume for simplicity that the cost of effort does not reduce an individual’s wealth. (18) Collective responsibility comes close to coordinated monitoring in that it spreads re­ sponsibility across the population of agents (Miceli and Segerson, 2007). (19) To illustrate, if there are three individuals and the monitoring probability is p = 1/10, the probability that all three are monitored is p3 = (1/10)3 = 1/1000. (20) This finding is implicit in the observation made in the literature on public enforce­ ment of law that the maximal sanction is optimal even if the punishment is imprisonment (Polinsky and Shavell, 2000). (21) If there are type-II errors, some non-compliers will be rewarded and hence knowing the exact number of compliers does not help. (22) Dari-Mattiacci (2009) employs the multiplication effect to show that tort liability (a stick) provides stronger incentives than restitution (a carrot). (23) Some readers, and even we, might disagree with the value of life implied by the ex­ ample, but this is not the point here. (24) Note that this condition can never be satisfied if F(e) is the real distribution of effort in the population. Page 27 of 29

Carrots vs. Sticks (25) The same results are obtained if one does not drop the term 1 / n from (21). Moreover, the result can be easily generalized to situations in which coordination costs are not fixed (either zero or prohibitive as in our example) but rise with the number of individuals in­ volved in the coordination. In this case, the multiplication effect works as long as there are enough sticks available to punish the group of coordinators. As coordination costs rise, the number of individuals who can be feasibly involved in coordination decreases and so does the number of sticks that need to be available. If coordination costs rise very sharply, the analysis collapses to the basic case described in the text. (26) Bar-Gill and Ben-Shahar (2009) explain that prosecutors are able to strike plea-bar­ gaining deals with most defendants despite their limited resources because they can threaten each individual defendant to go to trial. (27) Incentives for self-reporting in a stick regime can be created by leniency programs that annul the punishment for those who report themselves to the authorities. Such an­ nullable sticks, however, can be seen as precompensated carrots. See section 21.9.2. (28) Tabbach (2010) shows that when manipulating p is costly for the agent (avoidance) it might be desirable that agents try to avoid punishment. (29) An example can be found in Becker and Stigler (1974), who suggest to pay efficiency wages (which are technically annullable bonuses) to police officers in order to give them an incentive not to be corrupt, and to let them pay a fee for the right to become a police officer in order to cancel out the rents. (30) The reason why the principal may offer a device that protects the agent against op­ portunism by the principal is that the principal must meet the participation constraint of the agent. Note that efficiency wages do not prevent opportunism consisting of falsely de­ claring that the employee shirked or flat-out refusing to pay the bonus to monitored nonshirkers. But the legal system may be able to observe such forms of opportunism, while opportunism consisting in lowering the probabilities of monitoring may be non-verifiable. (31) This can be illustrated with a numerical example. Suppose 10% monitoring probabili­ ty and 20% violators. In this case a carrot is paid to 8% of the agents, a stick is paid by 2% of the agents, an annullable carrot without set-off requires payment to or from 100% + 2% of the agents (100% receive the carrot, and 2% have to give it back), an annullable carrot with set-off requires payment to 98% of the agents, an annullable stick without setoff requires payment to or from 100% + 8% of the agents, and an annullable stick with set-off requires payment to 92% of the agents. (32) It often occurs in firms, organizations, and commitment contracts that penalties are paid to third parties (Ayres, 2010). (33) More specifically, Galle (2014) considers a trade-off between moral hazard and riskspreading and argues that if government can control the salience of carrots, it may be

Page 28 of 29

Carrots vs. Sticks able to mitigate some of the moral hazard carrots usually present. This would shift the op­ timal mix of instruments farther towards carrots. (34) A non-exhaustive list includes Abbink, Irlenbusch, and Renner (2000); Offerman (2002); Andreoni, Harbaugh, and Westerlund (2003); Brooks, Stremitzer, and Tontrup (2012); Nosenzo et al. (2016).

Giuseppe Dari-Mattiacci

Giuseppe Dari-Mattiacci, Professor of Law and Economics, University of Amsterdam. Gerrit DeGeest

Gerrit De Geest, Professor of Law and Director of the Center for Law, Innovation & Economic Growth, Washington University in St. Louis Law.

Page 29 of 29

Law and Social Norms

Law and Social Norms   Emanuela Carbonara The Oxford Handbook of Law and Economics: Volume 1: Methodology and Concepts Edited by Francesco Parisi Print Publication Date: Apr 2017 Subject: Economics and Finance, Law and Economics, Econometrics, Experimental and Quanti­ tative Methods Online Publication Date: May 2017 DOI: 10.1093/oxfordhb/9780199684267.013.032

Abstract and Keywords Legal norms are often seen as a means to regulate behaviour when neither self-interest nor social norms produce the desired behaviour in individuals. This suggests, on the one hand, that the law should regulate those areas in which social norms do not exist and pro­ vide support and extra enforcement in those areas where social norms exist. It also sug­ gests on the other hand that there seems to be no questioning of the intrinsic efficiency and fairness of existing social norms. This article first looks at the genesis of social norms and the mechanism of their enforcement. This allows a closer inspection of the efficiency and fairness concepts. It then considers the impact that introducing legal norms has in contexts in which social norms already exist and in those that social interaction left un­ regulated. The main issue here is that the social norms prevailing at some historical mo­ ment may be just an equilibrium among multiple equilibriums. Given many possible equi­ libriums, we need to explain why and how one equilibrium is selected and others are re­ jected. The scholarship on social norms emphasizes that expressive acts in law can select the equilibrium. Legal norms seemingly reinforce existing social norms, bending them to­ wards the law when discrepancy exists and favouring their creation where social norms do not exist. However, legal regulation can also destroy existing social norms (crowding out) or it can be defeated by them (legal backlash and countervailing effects). Keywords: law, self-interest, social norms, efficiency, fairness, legal norms

22.1 Introduction LEGAL norms are often seen as a means to regulate individuals when self-interest does not produce the desired behavior as measured by efficiency and fairness.1 As Oliver Wen­ dell Holmes, Jr. (1897) posited, laws are for the “bad man,” a man “who cares nothing for

Page 1 of 19

Law and Social Norms an ethical rule which is believed and practised by his neighbours.” Then, laws should mat­ ter when neither self-interest nor social norms provide the right incentives to individuals. This latter statement paves the way for a twofold argument. On the one hand, it seems to suggest that the law should regulate those areas in which social norms do not exist and provide support and extra enforcement in those areas where social norms exist. On the other hand, there seems to be no questioning of the intrinsic efficiency and fairness of ex­ isting social norms. In this chapter we discuss both issues. First, we look at the genesis of social norms and the mechanism of their enforcement. This allows a closer inspection of the efficiency and fairness concepts. In the study of social norms, “efficiency” has several standard mean­ ings, notably Pareto efficiency, cost–benefit efficiency, and welfare maximization. “Unfair­ ness” also has several possible meanings, but the most frequently discussed today is dis­ crimination based on race, ethnicity, gender, or sexual orientation. In some circum­ stances, social norms are efficient and fair, requiring no regulation so long as private and criminal law operate in the background. In other circumstances, unregulated social norms waste resources or discriminate against individuals or groups. In principle, regula­ tion can correct these normative failures. In practice, regulations correct failures in so­ cial norms or worsen them depending on the politics of regulation—who has power, and who benefits from efficiency and fairness. This leads us to consider the (p. 467) impact that introducing legal norms has in contexts in which social norms already exist and in those that social interaction has left unregulated. The main issue here is that the social norms prevailing at some historical moment may be just an equilibrium among multiple equilibriums. Given many possible equilibriums, we need to explain why and how one equilibrium is selected and others are rejected. A com­ plete model of law and social norms encompasses the equilibriums and the selection mechanism. As we will show, the scholarship on social norms emphasizes that expressive acts in law can select the equilibrium. Legal norms seemingly reinforce existing social norms, bending them towards the law when discrepancy exists and favoring their cre­ ation where social norms do not exist. However, legal regulation can also destroy existing social norms (crowding out) or it can be defeated by them (legal backlash and counter­ vailing effects). The law and economics analysis of social norms deals with individual preferences and the way in which these are formed. Generally, in other areas of applied economics, individual preferences are given and stable. Social norms instead depend on internalized values, not just external threats. Internalizing a social norm modifies a person’s commitment to val­ ues. Commitments to values affect personal identity, and personal identity affects a person’s understanding of his self-interest. Thus, in order to explain how social norms work, a good theory should look at how people develop “preferences,” analyzing socializa­ tion, internalization, and identity.

Page 2 of 19

Law and Social Norms

22.2 Conforming with Social Norms: Coordina­ tion, Sanctions, Internalization Why do social norms arise and why do people conform? In the law and economics litera­ ture, “social norm” sometimes refers to any regularity in the behavior of people, regard­ less of its cause.2 However, the phrase is often used more restrictively. Most scholars, in fact, restrict the phrase “social norms” to behavioral regularities with particular causes.3 We are going to identify three causes of conformity: a) coordination in a multiple-equilib­ riums environment; b) fear of non-legal sanctions; c) internalization. First, some people conform to social norms because they gain from doing the same thing as others. Conforming to the norm serves their interests so long as other people also con­ form to the norm, which implies that conforming is a Nash equilibrium. For example, most people will drive on the left-hand side of the road if they believe that others will also drive on the left-hand side of the road. In these circumstances, game theorists often call the behavior a “convention.” Thus Young (1993) defines a “convention” as (p. 468) a “pat­ tern of behaviour that is customary, expected and self-enforcing. Everyone conforms, everyone expects others to conform, and everyone wants to conform given that everyone else conforms.” The second reason for conforming to a social norm is fear of a social sanction. Social norms typically require some individuals to bear costs or forgo benefits. Absent sanctions for non-conformity, the dominant strategies favor violating the social norm (e.g. prisoner’s dilemma-like situations). Many social norms are obligations backed by non-legal sanc­ tions. When violating a social norm serves immediate self-interest, fear of non-legal sanc­ tions can tip the balance of self-interest in favor of conforming. The behavioral regularity described by the social norm may depend on the effective threat of social sanctions. By analogy with the “imperative theory of law,” according to which a legal rule is an obligation backed by a state sanction, Cooter (2000) defines a social norm as an “obliga­ tion backed by a non-legal sanction.” The word “non-legal” in this context refers to the nature of the sanction (like shaming, stigma, shunning, criticizing, refusal to deal) and the kind of people enforcing it (private citizens). Sanctions are necessary to sustain social norms that are not pure conventions that bene­ fit everyone. Pure conventions help people coordinating and are self-enforcing. The fact that not all social norms are pure conventions explains why people publicly uphold some norms and privately disobey them when unobserved.4 In his study of the Ik traditions and habits, Turnbull (1972) reports that members of this ethnic group, living in the northeast­ ern mountains of Uganda and suffering from extensive famine, often tried to elude situa­ tions where compliance with social norms of reciprocity was expected from them. Coordination and fear of non-legal sanctions are two reasons why people conform to so­ cial norms. In “The Grammar of Society: The Nature and Dynamics of Social Norms,” the Page 3 of 19

Law and Social Norms philosopher and psychologist Cristina Bicchieri combines these reasons into a definition of social norms. Social norms are “conditional rules” such that individuals prefer to con­ form when they expect a sufficiently large proportion of the population to conform (coor­ dination or “empirical expectation”), and they believe that a sufficiently large proportion of the population might sanction them if they don’t (fear of sanction or “normative expec­ tation”). The empirical expectation of conformity is the belief that a behavioral regularity exists, and the normative expectation is the belief that social sanctions enforce it. Coordination and the fear of sanctions are not the only reasons why people conform to so­ cial norms. The third cause of conforming to a social norm is internalization of the obliga­ tion. Even without fear of legal sanctions, some people do their duty from conviction. They will sacrifice self-interest to do their duty as they perceive it. In economic language, people who internalize an obligation acquire a preference to conform to it. They (p. 469) are willing to pay to conform to a social obligation, just as they are willing to pay for a boat ride on a lake. Performing an internalized duty is like consumption in that a person will pay to do it. Conversely, violating an internalized duty is like work in that a person must be paid to do it. Coordination, sanction, and internalization are three causes of behavioral regularities in social norms. A definition of “social norms” must comprehend all three causes in order to encompass the models in law and economics. Thus we could define a “perfect social norm” as a behavioral regularity caused by coordination, non-legal sanctions, and inter­ nalization. However, we will not labor over the definition. Economics is more concerned with causes than meanings, and so are we. Having explained three causes of individual behavior that sustain social norms, we now turn to how social norms arise and to their ef­ ficiency.

22.3 The Efficiency and Fairness of Social Norms We have just seen that a convention can be a social norm, whereas a social norm is not necessarily a convention, as it may not be self-enforcing. This raises two different issues. First, how are social norms created and what legitimizes them? Second, are social norms always efficient? These two issues can hardly be separat­ ed. Many social norms develop from conventions or religious imperatives. As Sugden (1989) effectively explains, conventions typically spread because of past experience, com­ mon background, and analogy. There is no guarantee that the actions that allowed coordi­ nation in the past are the most efficient (whichever definition of efficiency we deploy). Most decisions are made in a backward-looking, myopic approach. One example should clarify. Sugden (1989) reports that in a village on the Yorkshire coast, people shared drift­ wood scattered after a storm following a first-on rule. There is, however, no historical ac­ count of the reasons that led to its adoption and, even if the first-on rule satisfies some ef­ ficiency criteria (it is indeed a very cheap way to establish property rights over valuable Page 4 of 19

Law and Social Norms objects), it is just one of many possible rules, some of which may be preferable in some respects. In theory homo oeconomicus is supposed to follow a forward-looking, rational expectation approach. Evidence seems to indicate that this does not happen in practice. On the contrary, backward-looking decisions typically imply systematic errors and may lead to the adoption of a rule, which would then be enforced and kept thereafter. The costs of changing it would be perceived to be too high compared to the benefits, especial­ ly in coordination games. This means that, in a multiple equilibriums setting, there is not only the problem of equilibrium selection but also a potential “path-dependency trap,” due to the costs of moving from an equilibrium to another.5 Similar considerations can be made regarding the fairness of social norms. Many social conventions have turned into strict social norms that limit the rights and the digni­ ty of political and ethnic minorities in conflicts.6 The interplay of honor, stigma, and the law is often responsible for the perpetuation of such rules. Benabou and Tirole (2011) show how such forces, together with the expressive power of the law, may explain why people resist legal changes that would enhance efficiency and lead to more “effective” laws. Societies tend to renounce cruel and unusual punishments, even if potentially very (p. 470)

effective, to express the value they attribute to being civilized. Similar forces may explain the resilience to change discriminatory norms, as this might undermine the current de­ sign of institutions or reduce job availability for people belonging to dominant social groups. For instance, an ill-placed attention to “family issues” may explain the resistance to changing social norms that limit freedom and career possibilities for women. Given path-dependency and resilience, how do social norms change and what drives their evolution? Often, societies seem to lack a mechanism to correct for “bad” social norms. Changes follow great, unexpected, and possibly traumatic historical events. There is, however, the possibility for direct intervention, conceivably paternalistic in nature, by the lawmaker, and we are going to analyze it.

22.4 Changing Social Norms The change in social norms operates through coordination, sanction, and internalization. First we discuss change through coordination, as with a pure convention. The essence of a coordination problem is selecting among multiple equilibriums. When a coordination problem is solved for the first time, a social norm emerges. When people change the coor­ dination equilibrium, a social norm changes. As explained, individuals will conform to a coordination equilibrium if they believe that others will conform. Thus creating or chang­ ing a coordination equilibrium requires making enough people believe that others will conform to the new social norm. An entrepreneur, according to Kirzner, is someone who knows future prices, so he can make more profitable investments than others.7 Similarly, a “norm entrepreneur” (Ellick­ son, 2001) knows the future coordination equilibrium. Norm entrepreneurs gain by re­ ceiving social esteem and empowerment from others. Since a norm entrepreneur fore­ Page 5 of 19

Law and Social Norms sees the future, he can induce others to believe that they will gain from complying with a new norm. For coordination norms, the entrepreneur induces people to select a new equilibrium. This possibility implies the existence of multiple equilibriums. The ease of selecting a dif­ ferent equilibrium depends on disequilibrium dynamics. Change from one norm to (p. 471) another is particularly fast when the old norm is unstable and the new norm is stable. Consequently, a disturbance that moves people away from the unstable norm will cause them to converge towards the stable norm. Sunstein (1996) considers convergence to one stable norm as a “bandwagon” or “cas­ cade” effect. People hide their true preferences for fear of getting a social sanction if their belief is different from the social norm. Carbonara et al. (2008) consider instead a theory encompassing more possible outcomes, and explaining a wider variety of possible reactions to legal innovation. Instead of converging to a single social norm, another possi­ bility is that different groups of people converge to different social norms. Here there are different stable equilibriums, and individuals in one group proceed towards one of them, while individuals in another group proceed towards another one. Individuals cluster around different beliefs and multiple social norms coexist in the community, one for each cluster. At the limit, society may end up very polarized, with people clustering around op­ posite social norms. We have been discussing cases where change to a new social norm is easy because the initial social norm is unstable. Conversely, change from an old norm to a new one is diffi­ cult when the old norm is stable. Given a stable equilibrium, small disturbances have no lasting effect—the system departs temporarily from the old equilibrium and subsequently returns to it. Thus social norms are often described as “sticky” or “conservative.” Chang­ ing a sticky norm requires a large disturbance that takes the system far from the stable region. The intervention of the norm entrepreneur would be particularly useful when either many competing social norms regulate behavior in a given society (for example, because soci­ ety is fragmented into several groups, often in conflict) or when a unique, inefficient, and stable social norm exists. A social norm can be changed also by changing a social sanction. In many economic mod­ els, the threat of sanctions determines the level of conformity to the norm. To illustrate, assume that individuals must decide whether to do A or B, where A and B are substitutes. “A” might refer to conforming to a social norm, and “B” might refer to violating it. Purely self-interested individuals will choose the act with the higher pay-off. In these circum­ stances, the willingness of more people to sanction wrongdoers will result in more confor­ mity with the social norm. These facts have important consequences for norm leadership by entrepreneurs and charismatic figures. A norm leader may change behavior also by convincing more people to sanction norm breakers, besides convincing more people to conform to the norm. Will­ Page 6 of 19

Law and Social Norms ingness to punish non-conformity affects the equilibrium as much as willingness to con­ form. These facts have important consequences for norm leadership by entrepreneurs and charismatic figures. A norm leader may change behavior primarily by convincing more people to sanction norm breakers, rather than by convincing more people to conform to the norm. So far we have been discussing the change in social norms that can be produced by norm entrepreneurs, without specifying whether they have a public or political role, (p. 472) or they are private citizens engaging in social activities and opinion leadership. Social norms can be changed also through law. The law, in fact, possesses the power to “express” the values that a society has or should have and can be used to reinforce or change existing social norms. Caution is, however, needed when the law operates in the presence of con­ trary social norms, since the two types of norms may interact in an unexpected fashion, leading to undesired outcomes. We now turn to the analysis of the complex interaction of law and social norms.

22.5 The Connection of Social Norms and State Laws Expressive law theories stress the ability of formal rules to reinforce, bend, and modify social norms and consequent behavior. The expressive effect of the law operates indepen­ dently of sanctions, meaning that even sanctionless rules will be internalized and obeyed by individuals. If the rule is also accompanied by a sanction, the expressive effect will be stronger and the presence of the external incentive will simply speed up internalization. However, legal regulation can also destroy existing social norms (crowding out) or it can be defeated by them (legal backlash and countervailing effects).

22.5.1 The Expressive Function of the Law As mentioned, the law carries the power to express social values. This is what scholars define as “the expressive power of the law,” which has the ability to signal what society expects from its members and to mold individual preferences. In contexts in which a social norm does not exist or where multiple norms coexist, law­ makers may be willing to “manage” social norms, creating new ones or bending existing ones towards what they reckon to be more acceptable behavior. This is likely when exist­ ing norms are considered either inefficient or unfair towards parts of the society. In their “management” activity, lawmakers can use the “expressive” function of the law (Sunstein, 1996; McAdams, 1997; Cooter, 1998, 2000): by enacting a law, the government makes a “statement” that strengthens the desired norms embodied in the law while weakening un­ desired ones.

Page 7 of 19

Law and Social Norms The expressive power of the law exerts its effect through different channels. 1) Legal innovation can change the social meaning of given actions and behavior; 2) New laws help create “focal points,” attracting attention towards certain actions and changing individuals’ expectations about other people’s behavior, favoring coor­ dination; (p. 473) 3) Legal norms can change preferences, prompting individuals to “internalize” the values embodied in the law. We are now going to discuss these channels in turn. It is, however, interesting to notice that, through each of these channels, the law can influence behavior independently of the sanction it imposes (McAdams, 2000a). Let us consider first the link between the law and social meaning. To understand this link, some anecdotal evidence can be useful. In the United States, the ban on public smoking has reduced the number of young black Americans who smoke by changing the social meaning of smoking from rebellion to dirtiness and foolishness. Legal innovation can therefore be used to change the social meaning of given actions and behavior. Social meaning is the semiotic content, or symbolism, attached to a particular action or gesture. Specifically, it is the way individuals belonging to a particular society or commu­ nity understand and interpret a given action (or inaction).8 Once individuals are aware of social meaning, they can use it, as they can be “tools—means to a chosen end.”9 There are many examples of such selective use of actions: clothing, the selection of certain words in special contexts, insults, and so forth. What is most interesting for our argument is that social meanings can be used by everyone, either groups or single individuals, and can be used by governments, too. For instance, if the majority of the population follows traditional religious beliefs, the government can use “family values” to advance state goals; if the majority has health concerns, the government can leverage them to impose a ban on smoking in public places. Laws expressing existing social meanings have greater chances to generate compliance. Social meaning can be built and changed through law. An enlightening example is provid­ ed by Lessig (1996) and is related to possible techniques used to eliminate dueling. While prohibiting duels in itself is neutral for social meaning, the sanction provided to punish vi­ olators does affect it and might improve substantially the efficacy of the prohibition. Con­ sider two possible legal strategies: imprisoning duelers and inflicting a sanction that, in addition to imprisonment, prevents duelers from holding public office. Both techniques have the effect of increasing the cost of dueling but have very different impacts on social meaning. The first type of sanction simply increases the cost of dueling, raising its ex­ pected harm. The meaning of the act, however, remains unaltered: duelers are gentlemen defending their honor and that of their community. Disqualifying duelers from public of­ fice not only increases the cost for duelers but it also changes the meaning of the act and, more precisely, the meaning of disobeying the law.

Page 8 of 19

Law and Social Norms Assume in fact that the regulation imposes imprisonment: complying with the law means that the dueler is not willing to defend his community’s honor because the cost of doing so is too high. He may well be seen as a coward, someone willing to escape his duty to serve his community. Consider the alternative case, in which the sanction is disqualifica­ tion from public office. Now, if someone refuses to duel, it might either be (p. 474) be­ cause she refuses the challenge and is therefore a coward, or because she privileges an­ other type of public duty that the law has put in competition with the duty of defending the community’s honor. The second type of sanction “ambiguates” the meaning of refusing to duel, changing the possible meaning of the act and therefore reducing its cost, rendering compliance with the law more attractive. The idea that a social meaning is attributed to actions and behavior is particularly impor­ tant in case of the so-called “top-down” lawmaking. As the previous example well illus­ trates, legal innovation that openly challenges current social meaning might reinforce a given behavior rather than reduce its prevalence. Manipulation of social meaning by the legislator has therefore to be rather subtle, creating new incentives that counteract the ones provided by the original social meaning. In this sense, the lawmaker has to act as a clever “norm entrepreneur.” Changing a social meaning through lawmaking is a paternalistic act that legislators might sometimes want to perform to produce merit goods. More often, a legislator will choose to embed the existing social meaning into a law, which therefore becomes an imperative expression of social meaning. If social meaning already exists and is shared in a commu­ nity, why do we need to express it also by law? In other words, why do we need the ex­ pressive power of the law? This leads us straight to the second channel of the expressive power of the law, namely the creation of focal points. The expressive power of the law plays a major role in situa­ tions characterized by coordination problems (McAdams, 2000a). For instance, by stating that drivers should keep to the right, the law creates a “focal point,” solving a coordina­ tion problem. Moreover, laws legitimized by a democratic voting process are usually posi­ tively correlated with “popular attitudes” (McAdams, 2000b) and thus provide a signal of those attitudes, helping individuals to form beliefs about what others will think of their behavior. Given that people normally care about being approved or disapproved by oth­ ers, the law can influence behavior even without a legal sanction. Another example of situations requiring coordination due to multiple equilibriums are po­ tentially conflicting situations in which individuals have to share resources or, more gen­ erally, yield to others.10 A possible equilibrium would be to yield when the other does not yield or vice versa. The worst possible scenario occurs when no one yields. A law might help in this case, because it might allocate entitlements, specifying who has to yield and under which circumstances.11

Page 9 of 19

Law and Social Norms Possibly the strongest effect of expression is the change in preference that the law might produce in individuals, which is the third channel of expressive power. Particularly, the law might induce individuals to internalize the values it embodies (Cooter, 1998, 2000). To see how that happens, consider that obeying a norm is a costly act (money, opportunities, etc.) and individuals are willing to obey if and only if they (p. 475) derive some kind of ben­ efit from it. There might be an extrinsic value from obeying (say, I keep my promise be­ cause I want to be perceived a trustworthy person, which would enhance my business op­ portunities). More importantly, there might be an intrinsic value: obeying benefits me inasmuch as I have internalized the norm and I feel a warm glow when conforming to my moral imperative. Intrinsic values are the psychological equivalent of “tastes” or “prefer­ ences.” People who have a taste for obeying a norm are prepared to bear the costs of compliance independently of any resulting advantage or disadvantage. What is the role of the law in such a scenario? If people believe it a moral duty to comply with legal norms (Tyler, 1990), their willingness to pay for compliance increases. The taste for obeying the law can also explain compliance with legal rules that are not aligned with current morali­ ty and social norms, since it increases the cost of disobedience and enhances internaliza­ tion. More precisely, people tend to align their concept of morality with the law (Tyler, 1990 describes such a process with respect to law; and Tyler and Huo, 2002 extend the argument to the decisions of authorities). Understandably, the closer the content of the law to existing social norms, the greater its legitimacy. Tyler (1990) and Sunshine and Tyler (2003) argue that the public’s percep­ tions of legitimacy influence people’s compliance with the law. As we will see below, laws can have unexpected effects on behavior when they interact with social norms, no matter whether the law and norms are aligned. Such unexpected ef­ fects counteract the expressive power of the law and might require heavy sanctioning to induce the desired behavior in people.

22.5.2 Crowding Out and Crowding In Legal norms seemingly reinforce existing social norms, bending them towards the law when discrepancy exists and favoring their creation where social norms do not exist (ex­ pressive law theories). However, legal regulation can also destroy existing social norms (crowding out). More generally, legal norms and sanctions especially are often proven to curb cooperative behavior and destroy social norms of cooperation. Frey and Jegan (2001) and Frey and Oberholzer-Gee (1997) argue this might occur because the law crowds out intrinsic moti­ vation, by hitting on individuals’ self-determination and self-esteem. Somehow, the legal rule is perceived as a lack of acknowledgment of individuals’ intrinsic motivation and as a lack of trust. It may also happen that individuals adopt cooperative behaviors and comply with social norms of cooperation when they want to signal their trustworthiness and in­ trinsic motivations to others. If a law prescribes cooperation, it becomes impossible to Page 10 of 19

Law and Social Norms distinguish whether somebody is being cooperative for fear of sanction or for intrinsic motivation. This might discourage their cooperation. In his famous example of blood donors, Titmuss (1970) argues that providing incentives to blood donors may reduce blood supply, as purely altruistic donors are demotivated by the reward. It may also reduce the quality of the donated blood. Hepatitis rates from blood transfusions significantly decreased when the blood was donated rather (p. 476) than pur­ chased. When monetary incentives are not involved, people supplying blood are donors who have no reason to hide an illness. Interestingly, Costa-Font, Jofre-Bonet, and Yen (2011) find that the nature of the rewards matters. They collected data on blood dona­ tions in fifteen European countries in 2002, showing that monetary rewards may indeed crowd out blood donations, whereas non-monetary rewards do not. Similar effects can be found in case of norms of cooperation, in particular with norms of cooperation favoring an entire community. Frey and Oberholzer-Gee (1997) present a field experiment on the implementation of a nuclear waste facility in a Swiss town. Citi­ zens were asked whether they would vote in favor of the establishment of the facility in their community. More than half of them agreed. However, they were later offered mone­ tary compensation to allow the waste facility and were asked to vote again. The level of support for the project dropped by more than 50%. Several explanations could be offered for this phenomenon. On the one hand, by refusing to agree on the public project when offered compensation, citizens might be simply trying to raise their stakes, like in an anti­ commons scenario.12 They expect that their refusal might trigger an increase in the com­ pensation offered by the government. Alternatively, an offer for compensation might be perceived as a signal that the facility implies serious risks for public health, risks that citi­ zens were not able to consider at the time of their first vote. Whatever the explanation, this is considered clear evidence that the positive feeling that citizens develop when they support public projects benefiting an entire community (in this case, the entire Switzer­ land) is crowded out by the offer of monetary compensation. The explanation is, however, important if the legislator is willing to enact a law avoiding such crowding effects. If the first explanation prevails, crowding out can be avoided by making it clear that compensa­ tion cannot be renegotiated and that there will be no increase following a refusal. If the second prevails, an information campaign stressing how safe waste facilities are might be the preferable option.

22.5.3 Legal Backlash and Countervailing Effects We have seen that legal norms can shape social norms and the behavior driven by them via the expressive function and by crowding in. However, legal norms can have a detri­ mental effect on social norms (crowding them out), defeating a system of private enforce­ ment and totally shifting its costs onto the public sector. Legal norms can have even worse effects, not only defeating possibly virtuous social norms but polarizing societies, exacerbating possible social conflicts, and even worsening individual behavior, increasing the prevalence of actions that the law intends to limit.

Page 11 of 19

Law and Social Norms Before looking at these undesired effects of laws, it is interesting to mention that, in the presence of very strong social norms, legal rules may be irrelevant. That is to say that “norms control individual behaviour to the exclusion of law” (see McAdams, 1997, p. 347). Consider, for instance, Ellickson (1991), who argues that the way neighbors (p. 477) resolved disputes in Shasta County was independent of the property regime adopted, since a unique social norm governed, irrespective of the legal norm. On a different key, Kahan (2000) points to the existence of social norms so strongly embedded in common behavior that enforcers fail to enforce laws contrary to such norms (the “sticky-norm” problem). The sticky-norm problem affects laws that depart too much from current social norms and has the consequence of reinforcing current social norms. Not sanctioning the behavior prohibited by the legal norm but admitted by the social norm reasserts the be­ havior the law was trying to change, encouraging and possibly perpetuating it. When a new law is enacted, it has a coordination and internalization effect, which at­ tracts individual opinions towards the value embodied in the law. Such effect is stronger the closer the law is to existing social norms. Otherwise, there is the possibility that coun­ tervailing effects arise, where legal intervention contrary to existing social opinions re­ pels those who hold strong contrary beliefs. The final impact depends on the relative strength of such two forces, with the countervailing effects more likely to dominate the farther away the law is from the prevailing social norm (Carbonara et al., 2008a, 2008b). What happens when a new law is enacted where either segmentation or polarization of social norms exists? The easiest case to consider is a new law with a strong internaliza­ tion and coordination effect. This reduces the attractiveness of social norms that lie far away from the new law. Such norms will be abandoned in favor of social norms closer to the new law and, if the attraction exerted by the latter norms is strong enough, we might witness a convergence to a unique social norm very close, if not identical, to the law. This is what happened after smoking was banned from public spaces in many countries around the world. Even smokers complied with the law because they felt it was embodying a widespread belief about the health hazards of smoking. The social norm changed through internalization of the values expressed by the law. The opposite would occur in a case where the opinions expressed in disagreement with the new law acted as strong attractors, thus offsetting the coordination and internaliza­ tion power of the law. Social norms far off the new law are reinforced, and this might re­ sult in different groups within a society following different norms: some of them might vi­ olate the legal prohibition, some might adopt stricter rules than the law, and finally, some might comply with the law. In the worst-case scenario, society might end up very polar­ ized and social conflicts might possibly arise. Even if polarization does not occur, and vio­ lation of the law is the prevalent behavior, the law would anyway be ineffective in forging a pattern of behavior consistent with the law: society would converge to a unique social norm contrary to the law. So far we have considered how the law impacts on social norms, possibly changing them in unexpected and undesirable directions. Legal rules in the presence of countervailing Page 12 of 19

Law and Social Norms social norms may not only impact existing social norms but also drive behavior in direc­ tions opposite to that intended by the lawmaker. Carbonara, Parisi, and von Wangenheim (2012) present a model explaining in what circumstances this might happen. According to this model, the incentives to comply with a law depend on the (p. 478) magnitude of the sanction, the expressive power of the law, and the degree of perceived legitimacy. While the first two elements push towards compliance, when perceived legitimacy is low people may be dragged into disobedience. Perceived legitimacy may be low when the legal rule has little expressive power and the values embodied in the law clash with society’s shared values. A new law that is contrary to current social values or more restrictive than people deem fair triggers opposition, where such opposition can take the form of either open protest or civil disobedience. An enlightening example is the reaction by the Internet community to the severe punish­ ments that were adopted against copyright infringers in the past few years. Internet users often reacted to the huge sanctions raised against private citizens who downloaded music from the Internet illegally. They protested fiercely, boycotted entertainment majors, and even engaged in fundraising to fully cover fines.13 The result of such protests was a steady increase in the number of users of peer-to-peer technologies, notwithstanding in­ creasingly severe sanctions.14 Moreover, there is experimental evidence that sanctions re­ inforce the norms prevailing in the file-sharing community, producing polarization.15 Protest indicates social disapproval of the legal rule and induces individuals to expect that society will approve or tolerate infringement, thus “subtracting” from the legal sanc­ tion (as in the case of fundraising to cover the fines of music pirates). If such effect is strong enough to compensate for the expressive power and for the incentive effect of the sanction, outcomes opposite to those intended by the lawmaker might ensue. Similarly, countervailing effects may be observed when society perceives a legal norm as too le­ nient relative to the pre-existing social norm. In that case, individuals may be inclined to “add” social sanctions to the behavior (disapprobation, further punishment, etc.). Countervailing effects may be particularly strong if protest reacts faster than actual be­ havior and if the reaction to legal innovation depends also on the relative change in the strictness of the law. Protest has the opportunity to react faster (or, at least, earlier) than behavior whenever a new law is announced or debated before it is enacted. This affects opinions and protest before actual incentive effects on behavior are possible. Countervail­ ing effects are thus exacerbated, since protest may result in a reinforced social norm. When the law is finally enacted, it is counterbalanced by a much stronger contrary social norm and countervailing effects are more likely to occur. Similarly, protest tends to react both to the absolute strictness of a law and to the change relative to a previously existing law (or to “a zero sanction”, if no law regulating a certain matter existed before). Then, if the sanction for music piracy is raised from $100 to $10,000, much more protest is likely to arise than in the case where the sanction increas­ es from $100 to $500. The implication of this is a possibly huge reaction when a new law is passed that departs very much from current social values, which can make contrary so­ cial norms extremely strong. After some time, people are likely to get used to (p. 479) the Page 13 of 19

Law and Social Norms new law and protest may subside. However, the initial reaction may reinforce contrary so­ cial norms so much that countervailing effects may endure in the long run, even after the initial effect subsides. In general, the existence of countervailing effects indicates that laws intending to induce substantial shifts from current norms may have to be introduced gradually, in small, con­ secutive steps allowing for adaptation of individual values to the content of the law. More interestingly, countervailing effects of social reaction could be used to induce legal deter­ rence by means of a reduction in the strictness of the law. When facing a low sanction, people might not object to its application, thus magnifying its deterrence effect.

22.6 Conclusions: Substitutes, Complements, Antagonists? The analysis in the previous sections shows that the impact of legal innovation on social values and behavior depends on several, often contrasting factors. Particularly, they should balance extrinsic incentives (sanctions reducing the net benefit from action), the expressive power, and possible crowding-out and countervailing effects. First of all, ab­ sent a strong expressive power (which implies a reduced ability of individuals to internal­ ize the values embodied in the law), laws differing substantially from prevailing social norms should be introduced gradually, to allow individuals to adapt their values to the content of the law. Notice that phenomena like crowding-out and countervailing effects in particular reduce the power of extrinsic incentives, to the point that it may be counter­ productive to impose a very high sanction. Internalization then becomes the crucial vari­ able for successful legal innovation. This leads to another interesting and apparently paradoxical insight. Legal deterrence could increase if the strictness of the law is re­ duced. What, then, is the relationship between legal and social norms? Are they substitutes or complements? Expressive law theories point to the possible complementarity of legal and social norms. The law can reinforce existing social norms by helping internalization, rein­ forcing compliance within a particular social group, and spreading them among individu­ als that do not belong to that social group but interact with it regularly. When law and so­ cial norms are aligned, their coexistence magnifies their respective power to change be­ havior. Things change quite substantially if legal and social norms are not aligned and the law tries to change behavior sustained by social norms. Furthermore, the ability of legal and social norms to reinforce each other is questioned every time legal innovation crowds out existing virtuous social norms. Interestingly, Zasu (2007) argues that social norms and the law tend to be complements when societies are not well “connected,” so that social sanctions are hardly (p. 480) ap­ plied. Conversely, if social norms are consistent with welfare maximization, social and le­ Page 14 of 19

Law and Social Norms gal norms are substitutes and costly legal norms are not required, as they risk crowding out “virtuous” social norms. This process of erosion of social values is described also by Posner (1998, 2000). Legal and social norms tend to be antagonists whenever they pursue conflicting objec­ tives (e.g. when legal norms tend to maximize overall social welfare and social norms per­ petuate a predatory behavior by one group against another) and when they are heavily misaligned. In such cases, not only do they not reinforce each other, but also the introduc­ tion of the law may well be totally counterproductive, in particular when a legal sanction is applied. When law and social norms are antagonists, social norms tend to prevail and costly legal intervention is wasted. That is why other instruments, different from sanctions, may be preferable. For instance, taxes might be a better option, as they are perceived as a “price” that individuals are free to pay if they want to indulge in a given, socially costly, behavior (like polluting). Also, positive incentives, like rewards and prizes, may be more effective. For instance, reward­ ing compliance may be a good idea in case of tax evasion.16 The literature on social norms and the law is vast, the effects of their interaction diverse and difficult to predict. In this work, we have tried to provide a systematic view of the subject, to help interested scholars and lawmakers to disentangle the forces involved in this process and understand the circumstances in which each of them could prevail.

Acknowledgment This chapter greatly benefited from comments and suggestions by Robert D. Cooter, who generously shared with me his profound knowledge on the interaction between the law and social norms.

References Banabou, R. and J. Tirole (2011). “Laws and Norms.” NBER Working Paper 17579. Bicchieri, Cristina (2006). The Grammar of Society. The Nature and Dynamics of Social Norms. New York: Cambridge University Press. Carbonara, Emanuela, Francesco Parisi, and Georg von Wangenheim (2008a). “Lawmak­ ers as Norm Entrepreneurs.” Review of Law and Economics 4: 779–799. Carbonara, E., F. Parisi, and G. von Wangenheim (2008b). “Legal Innovation and the Com­ pliance Paradox.” Minnesota Journal of Law, Science and Technology 9: 837–860. Carbonara, E., F. Parisi, and G. von Wangenheim (2012). “Unjust Laws and Illegal Norms.” International Review of Law and Economics 32: 285–299. Carbonara, E. and P. Pasotti (2010). “Social Dynamics and Minority Protection.” Interna­ tional Review of Law and Economics 40: 317–328. Page 15 of 19

Law and Social Norms (p. 481)

Cooter, Robert D. (1984). “Prices and Sanctions.” Columbia Law Review 84: 1523–

1560. Cooter, Robert D. (1996). “Decentralized Law for a Complex Economy: The Structural Ap­ proach to Adjudicating The New Law Merchant.” University of Pennsylvania Law Review 144: 1643–1696. Cooter, Robert D. (1998). “Expressive Law and Economics.” Journal of Legal Studies 27: 585–608. Cooter, Robert D. (2000). “Do Good Laws Make Good Citizens? An Economic Analysis of Internalized Norms.” Virginia Law Review 86: 1577–1601. Costa-Font, J., M. Jofre-Bonet, and S. Yen (2011). “Not all incentives wash out the warm glow: The case of blood donation revisited.” CESifo working paper: Public Finance, No. 3527. Depoorter, B., F. Parisi, and S. Svanneste (2005). “Problems with the Enforcement of Copyright Law: Is There a Social Norm Backlash?” International Journal of the Economics of Business 12: 361–369. Depoorter, B. and S. Vanneste (2005). “Norms and Enforcement: The Case Against Copy­ right Litigation.” Oregon Law Review 84: 1127–1180. Ellickson, Robert C. (1991). Order Without Law: How Neighbors Settle Disputes. Cam­ bridge, MA: Harvard University Press. Ellickson, Robert C. (2001). “The Market for Social Norms.” American Law and Econom­ ics Review 3: 1–49. Fabbri, Marco (2015). “Shaping Tax Norms Through Lotteries.” International Review of Law and Economics 44: 8–15. Frey, B. S. and R. Jegan (2001). “Motivation Crowding Theory.” Journal of Economic Sur­ veys 15: 589–611. Frey, B. S. and F. Oberholzer-Gee (1997). “The Costs of Price Incentives: An Empirical Analysis of Motivation Crowding Out.” American Economic Review 87: 746–755. Heller, M. A. (1998). “The Tragedy of the Anticommons: Property in the Transition from Marx to Markets.” Harvard Law Review 111: 621–688. Holmes, Oliver W. (1897). “The Path of the Law.” Harvard Law Review 10: 457–478. Kahan, Dan M. (2000). “Gentle Nudges v. Hard Shoves: Solving the Sticky Norms Prob­ lem.” University of Chicago Law Review 67: 607–646.

Page 16 of 19

Law and Social Norms Kirzner, Israel M. (1995). “The Nature of Profits: Some Economic Insights and their Ethi­ cal Implications,” in R. Cowan and Mario J. Rizzo, eds., Profits and Morality. Chicago and London: University of Chicago Press: 22–47. Kuran, Timor (1997). Private Truths, Public Lies. The Social Consequences of Preference Falsification. Cambridge, MA: Harvard University Press. Lessig, Lawrence (1995). “The Regulation of Social Meaning.” University of Chicago Law Review 62: 943–1045. Lessig, Lawrence (1996). “Social Meaning and Social Norms.” University of Pennsylvania Law Review 144: 2181–2189. McAdams, Richard H. (1995). “Cooperation and Conflict: The Economics of Group Status Production and Race Discrimination.” Harvard Law Review 108: 1003–1084. McAdams, Richard H. (1997). “The Origin, Development, and Regulation of Norms.” Michigan Law Review 96: 338–433. McAdams, Richard H. (2000a). “A Focal Point Theory of Expressive Law.” Virginia Law Re­ view 86: 1649–1729. McAdams, Richard H. (2000b). “An Attitudinal Theory of Expressive Law.” Virginia Law Review 79: 339–390. McAdams, R. H. and E. B. Rasmusen (2007). “Norms in Law and Economics,” in A. Mitchell Polinsky and S. Shavell, eds., Handbook of Law and Economics, Volume II. Am­ sterdam: Elsevier: 1573–1618. (p. 482)

Mahoney, P. G. and C. W. Sanchirico (2000). “Competing Norms and Social Evolution: Is the Fittest Norm Efficient?” University of Pennsylvania Law Review 149: 2027–2062. Oksanen, V. and M. Välimäki (2007). “Theory of Deterrence and Individual Behavior. Can Lawsuits Control File Sharing on the Internet?” Review of Law and Economics 3: 693– 714. Posner, Eric A. (1996). “Law, Economics, and Inefficient Norms.” University of Pennsylva­ nia Law Review 144: 1697–1744. Posner, Eric A. (1998). “Symbols, Signals, and the Social Norms in Politics and the Law.” Journal of Legal Studies 27: 765–798. Posner, Eric A. (2000). Law and Social Norms. Cambridge, MA: Harvard University Press. Sugden, Robert (1989). “Spontaneous Order.” Journal of Economic Perspectives 3: 85–97. Sugden, Robert (2004). The Economics of Rights, Co-operation and Welfare. London: Pal­ grave Macmillan.

Page 17 of 19

Law and Social Norms Sunshine, J. and T. R. Tyler (2003). “The role of procedural justice and legitimacy in shap­ ing public support for policing.” Law and Society Review 37: 555–589. Sunstein, Cass R. (1996). “On the Expressive Function of Law.” University of Pennsylva­ nia Law Review 144: 2021–2053. Titmuss, Richard M. (1970). The Gift Relationship. London: Allen and Unwin. Turnbull, Colin M. (1972). The Mountain People. New York: Simon & Schuster. Tyler, Tom R. (1990). Why People Obey the Law. New Haven, CT: Yale University Press. Tyler, Tom R. (1994). “Psychological Models of the Justice Motive.” Journal of Personality and Social Psychology 67: 850–863. Tyler, T. R. and Y. J. Huo (2002). Trust in the Law. New York: Russell-Sage. Young, H. Peyton (1993). “The Evolution of Conventions.” Econometrica 61: 57–84. Zasu, Yoshinobu (2007). “Sanctions by Social Norms and the Law: Substitutes or Comple­ ments?” Journal of Legal Studies 36: 379–396.

Notes: (1) See McAdams and Rasmusen (2007), p. 1575. (2) See Ellickson (1991). (3) See, among others, Cooter (1996), Ellickson (1991), McAdams (1997, 2000a, 2000b). (4) In such a case, public lies serve the purpose to avoid sanctioning, and private behavior follows beliefs about the real “rule” followed by the majority of the people in the society. An example of this could be adherence to religious prescriptions or adultery. See also Ku­ ran (1997). (5) On the efficiency of social norms, see Posner (1996) and Mahoney and Sanchirico (2000). (6) See McAdams (1995) and Carbonara and Pasotti (2010). (7) Kirzner (1995). (8) See Lessig (1995). (9) Lessig (1995), p. 956. (10) The typical coordination game we refer to in this case is the hawk–dove (also known as the “chicken”) game. See Sugden (2004). (11) See Carbonara and Pasotti (2010). Page 18 of 19

Law and Social Norms (12) On anticommons, see Heller (1998). (13) See Oksanen and Välimäki (2007). (14) See Depoorter and Vanneste (2005). (15) Depoorter, Parisi, and Vanneste (2005). (16) See Fabbri (2015).

Emanuela Carbonara

Emanuela Carbonara, Associate Professor, Department of Economics, University of Bologna

Page 19 of 19

Mechanism Design and the Law

Mechanism Design and the Law   Werner Güth The Oxford Handbook of Law and Economics: Volume 1: Methodology and Concepts Edited by Francesco Parisi Print Publication Date: Apr 2017 Subject: Economics and Finance, Law and Economics, Econometrics, Experimental and Quanti­ tative Methods Online Publication Date: May 2017 DOI: 10.1093/oxfordhb/9780199684267.013.033

Abstract and Keywords Mechanism design is the game theoretic jargon for institutional design and the even old­ er tradition (in German) of ‘Ordnungspolitik’ (institutional design policy). When imple­ menting institutions or mechanisms (or simply rules of conduct) such regulation should usually be codified by complementing the law appropriately. This article first derives and discusses legal rules as traditionally justified and implemented legally. This is then con­ fronted with game theoretic mechanism design, relying on Dominance Solvability or the Revelation Principle. It is argued that the Revelation Principle is very useful for welfaris­ tic or, more generally, consequentialistic explorations of what is attainable but offers no practical basis for legal mechanism design due to its unrealistic common knowledge re­ strictions. Keywords: law, legal mechanism design, institutional design policy, Dominance Solvability, Revelation Principle

23.1 Introduction MECHANISM design is the game-theoretic jargon for institutional design and the even older tradition (in German) of Ordnungspolitik (institutional design policy). When imple­ menting institutions or mechanisms (or simply rules of conduct) one should usually codify such regulation by complementing the law appropriately. We first derive and discuss legal rules as traditionally justified and implemented legally (section 23.2). This is then con­ fronted with game-theoretic mechanism design (section 23.3), relying on Dominance Solv­ ability or the Revelation Principle, before concluding.

Page 1 of 13

Mechanism Design and the Law

23.2 Examples of Legal Mechanism Design Economists are naturally is interested in legal rules regulating public provision, that is, in the productive aspect of state intervention. As an example, imagine a public university or a local community trying to get some construction work done, for example a lecture hall or a town hall. Usually there will be several private enterprises i, for example construc­ tion firms i in private ownership, able to provide such construction work. Specifically, they can compete for the delivery by submitting a sealed bid . Let denote the vector of all bids that are submitted. A legal mechanism is usually not an ad hoc rule for a specific demand, for example for construction work, but rather for all similar cases. In practice, the rules applied rarely change (see Gandenberger, 1961, who shows that the German tradition dates back to the fifteenth century). What do such rules regulate? Since, when a law is passed, it is not (p. 484) obvious when it will be applied and in which situation, the rules have to deter­ mine for all possible bid vectors b 1) which of the bidders i will be chosen as the provider: let us refer to this bidder i as the winner , and 2) which price the winner receives for delivering the construction work. What one should note here is that in the legal tradition the rules determining and are thought to apply to possibly many future public provision problems about which little or nothing is known when passing the law. Nevertheless the legal tradition is far from arbitrary and quite systematically based on obvious ethical requirements. So postulating (EF) “according to his bid no bidder i should prefer the net exchange of another bidder to his own one” requires for­ mally:

respectively

for all bidders . Let us look at the upper inequalities: the left one guarantees that the winner w (b) who receives the price p (b) and, according to his bid , claims cost of

Page 2 of 13

Mechanism Design and the Law does not prefer to be a non-winner who receives and provides nothing. Similarly, the right inequality makes sure also that the non-winners prefer to be a non-winner over providing at price p (b) according to their cost claims . Altogether, (EF) implies for all possible bid vectors that the winner w (b) has to be a lowest bidder and that the price lies in the interval from lowest to second-lowest bid. Additionally, asking for (EQ) “according to their bids all net exchanges of bidders are evaluated equally” determines via , the so-called “first price rule” due to . This lowest bid price rule has prevailed in Germany for centuries (see Gandenberger, 1961). Since non-winners are left empty-handed, the winner , according to his bid , should earn zero, too. But why did one pay little or no attention to efficiency? That this would be possible can easily be seen when imposing (IC) “for all potential providers i bidding where denotes their true provision costs should be the only weakly dominant bidding be­ havior”, instead of (EQ) in addition to requirement (EF). The two requirements (EF) and (IC) imply , that is, the second rather than the first price (p. 485) rule, in our procurement set-up the second-lowest bid price rule (see Vickrey, 1961; Güth, 1986). More importantly, when all potential providers bid truthfully, i.e. for all i, the lowest-cost provider will always win, that is, public provision will be efficient. Rather than complaining about legal design missing such an efficiency-enhancing mecha­ nism, one should ask why legal scholars may have consciously regulated otherwise. Whether by conscious mechanism design or by legal evolution,1 there are good reasons to believe that legal scholars were more influenced by antitrust concerns, here by fear of ring formation, than by efficiency considerations. To demonstrate this (see Fehl and Güth, 1987 and Güth and Peleg, 1996 for details) assume the incentive compatible second-low­ est bid price rule. If most bidders i agree that some bidder i* should win and earn a high profit P, they can arrange this by letting i* bid so low that no other bidder i will ever want to underbid whereas those bidders i wanting to earn P choose bids . Clearly, if all bid in this way, the designated winner Page 3 of 13

Mechanism Design and the Law wins and earns at least P due to the second-lowest bid price rule. Furthermore, no bidder i can profitably underbid when for all . We conclude that the incentive compatible and efficient mechanism invites ring forma­ tion. Thus it is understandable that legal scholars abstain from using it. Compared with the incentive compatible pricing rule, ring formation for the lowest bid price rule is much more difficult since obviously the designated winner would have to bid when trying to earn . This allows other bidders (ring members as well as outsiders) to profitably interfere by underbidding . In fact, the same requirements can be applied to legal mechanisms for terminating joint ventures like joint venture firms and, in family law, divorce and inheritance conflict reso­ lution. Here, however, incentive compatible pricing rules are impossible due to the addi­ tional restrictions prevailing in such situations (see Güth, 2011). To illustrate this, assume that bidders collectively own a firm but want to terminate their joint venture by selling it to one of them who has to financially compensate all former partners. Again we assume that all submit a monetary bid and that the rules must determine for all possible bid vectors the winner and the monetary compensations of for all . Clearly, (EF) requires that all former partners get the same monetary compensation, that is, for all . Additionally, (EF) implies

for all or

Thus, the firm has to be sold to a highest bidder at a price between the secondhighest and highest bid which is shared equally among all n collective owners. (p. 486)

Now, however, there exists no incentive compatible pricing rule satisfying (EF) and (IC). If, for instance, is determined by the second-lowest bid, the non-winners will want to strategically overbid their true evaluation Page 4 of 13

Mechanism Design and the Law of the firm in order to raise the price and thus their compensation payment . For the highest bid price rule, in turn, the winner will want to underbid in order to decrease and thus the compensation payments for the other collective owners. Whereas guarantee­ ing both, (EF) and (IC), is impossible, maintaining (EF) and (EQ) has the straightforward implication or , that is, of the first price rule and that, according to bids, all collective owners receive the same net gain, namely . This illustrates how asking for (IC), in addition to (EF), is often impossible whereas im­ posing (EQ) and (EF) can be maintained. Let us further consider the productive aspect of public authorities and the legal rules ap­ plying to this. Consistent with the requirements (EF) and (EQ) one can also design mech­ anisms allowing people to actively influence whether and what should be provided pub­ licly. Rather than via referendum, as most prominently used in Switzerland, allowing peo­ ple to accept or reject a certain public project, people submit monetary bids and deter­ mine via their bids endogenously what is collectively provided in a way granting them ve­ to power. Let denote the public projects p that can have positive and negative consequences for differ­ ent people i.2 The monetary bids by persons i can thus be negative as well as non-negative numbers. Since each bidder i submits a bid for all projects p, the bid vectors have individual elements which are lists of bids. Let for all p denote the “cost” of public project p, which can be negative.3 If for all p the “cost” of the public project p is given and known,4 one could require (CB) “if some is provided, its cost should be covered by what all people i have to pay” and (E) “only if for some the for which the surplus is maximal will be provided.” (p. 487)

Page 5 of 13

The equality requirement (EQ) implies, for the case at hand,

Mechanism Design and the Law (EQ) “if is provided the payments of persons have to satisfy for all individual pairs j and k. These conditions together imply that when some is provided, the payments are

for all persons j. Due to (E) the surplus is non-negative and no person i has to pay more than his bid . Thus the mechanism satisfies an important voluntariness condition (V), similar to veto power, to be exerted by bidding sufficiently low (see Güth and Kliemt, 2013 for a discus­ sion). Due to (EQ) also requirement (EF) is granted, what guarantees institutional and thus legal consistency for all situations considered so far. At least as far as public provi­ sion is concerned, it seems possible to let people participate more actively in determining what is provided publicly and how the “costs” are shared—in our view, an adequate legal system of a liberal society.

23.3 Welfaristic Mechanism Design When considering incentive compatibility as a desirable requirement of mechanism de­ sign, impossibility or general non-applicability is often due to asking not only that truthful bidding is optimal but that it is optimal for all possible choice constellations of the other bidders—a requirement known as dominance solvability. One can render incentive com­ patibility applicable across the board by weakening it to (EIC) “when all bid truthfully, e.g. in the sense of for all bidders , then for all choosing is optimal.” The general applicability of (EIC) follows from the so-called Revelation Principle (e.g. My­ erson, 1979). This is a construction device by which one determines for each equilibrium of each game a so-called revelation game for which general truthful bidding is an equilib­ rium implying the same allocation outcome as the equilibrium of the game one has start­ ed with. We will illustrate the Revelation Principle for the procurement auction of the pre­ vious section with just two bidders and the lowest bid price auction which is not incentive compatible but underbidding proof (all bids are weakly dominated). Page 6 of 13

Mechanism Design and the Law (Welfare) Optimization is, of course, constrained by incentive compatibility condi­ tions, captured by restricting the (welfare) potential to what can be guaranteed by the al­ location outcomes of truth-telling equilibria of revelation games. To illustrate this, assume that two bidders and compete in the lowest bid price procurement auction and that and have commonly known true costs and , and that the public authority would only accept deals with prices not exceeding R, where R with is also commonly known. Note that we thereby have defined a game that is commonly known, an aspect that we did not have to bother with at all before. (p. 488)

Denoting by a monetary bid marginally exceeding allows us to describe the equilibria of this auction with—in game-theoretic jargon—com­ plete information by with . Of course, requires that bidder uses a weakly dominated strategy illustrating that the auction has many equilibria in spite of its unique equilibrium solution in weakly undominated strategies. The revelation games for each with specify for all cost claims the pay-offs

and

so that is a truth-telling equilibrium with pay-offs and for all . The sum is obviously maximal for when restricting the potential to truth-telling equilibria of the revelation games, captured by the constraint . This constraint is binding since only non-equilibrium bid vectors in the range of the lowest bid price procurement auction would imply pay-off sums exceeding one. Pay-off sums

Page 7 of 13

Mechanism Design and the Law exceeding one are feasible but not as equilibrium outcomes and therefore are excluded by the incentive compatibility constraint of the Revelation Principle. This illustrates that if a game has multiple equilibria, each of them—each with in the exercise above—implies its specific revelation game and that (welfare) optimization is restricted to equilibrium outcomes. The Revelation Principle is usually not applied to complete information games but to games with incomplete or private information, which will be illustrated by adjusting the example above. Assume that and are both independently and randomly determined according to the uniform density on the interval from zero to one and that both, and , are risk-neutral, that is, maximizing their own expected profit. Private information means that only is informed about the randomly selected and that bids are now functions and , respectively, of what is privately known by and . (p. 489) It is important that all these assumptions have to be commonly known, that is, again we need a well-defined, especially informationally closed, game (Harsanyi, 1967– 68), an aspect that is not required for the procedural approach outlined in section 23.2 and based more on the legal tradition. Due to their a priori symmetry let for all denote the symmetric equilibrium bid function in the sense of for all that we want to derive. For bidder ’s expected profit for is thus

due to being generated according to the uniform density. The first-order condition for optimality

is

or, after substituting ,

Page 8 of 13

Mechanism Design and the Law This differential equation is solved by for all and also satisfies the second-order condition for optimality. Thus, for all is an equilibrium of the lowest bid price procurement auction with incomplete informa­ tion. The Revelation Principle assumes as strategies cost declarations and with and constructs the revelation game such that are truth-telling equilibria for all . This is achieved by substituting the equilibrium bids into the pay-off functions of the original lowest bid price auction so that for the expected profit

is a function of . The first-order condition for optimality, namely (p. 490)

implies for all . Since also the second-order condition for optimality holds, general truth telling (here for all and ) is an equilibrium of the revelation game with the same allocation outcome as the sym­ metric -equilibrium of the original procurement auction via construction. Doing the same for oth­ er than the lowest bid price rule would obviously allow us to check for the (welfare) opti­ mal pricing rule in the narrow class of revelation games.5 Altogether this demonstrates 1) the weakness of the Revelation Principle, since it requires an ad hoc mechanism for each equilibrium of each game, an impossible endeavor for legal scholars; 2) its strength, since when analyzing what is (welfare) optimal, given the incentive compatibility constraints, one can focus on revelation games and their truth-telling equilibria (see Güth and Hellwig, 1986, 1987); 3) its generic non-applicability, since it is based on equilibria of well-defined games whose common knowledge requirements are outrageously unrealistic. In sum, the Revelation Principle is very useful for welfaristic or, more generally, conse­ quentialistic explorations of what is attainable but offers no practical basis for legal mechanism design due to its unrealistic common knowledge restrictions. Page 9 of 13

Mechanism Design and the Law

23.4 On the Literature of Mechanism Design An economist is not well qualified to report on the history of law, which is why the focus here is on the history of mechanism design. Sticking to our guiding example of procure­ ment auctions, there is a long tradition of using auctions (see Milgrom, 1989), partly very delicate applications like auctions to “trade” slaves. For the German-speaking areas, we have already mentioned Gandenberger (1961) who documents an impressive constancy in relying on the lowest bid price procurement auction. Whether as procurement auctions by which a buyer wants to contract with one of the potential suppliers or as sales auc­ tions where a seller wants to find a buyer, the literature now offers a lot of auction theory (Wilson, 1985), auction experiments (e.g. Kagel, 2016), and related field studies (Mil­ grom, 1989). The voluntary supply of public projects, most typically in the form of pure public goods (see, e.g., Ledyard, 1995) or common pool resources (Ostrom, Walker, and Gard­ ner, 1992), also has a long tradition of applications. Rather than as a problem of mecha­ nism design, it is usually studied by social dilemma games where community members (p. 491)

nevertheless manage to cooperate to some extent, for example, due to repeated interac­ tion, or some other aspects of their social life like monitoring and punishing (those who “free-ride”). Especially, the group in Bloomington has done an immense amount of field research and has also run experiments to study voluntary provision of public projects (see Ostrom, 2015). Mechanism design, however, has a more recent, already very successful history (the No­ bel laureates in 2007, L. Hurwicz, E. S. Mascin, and R. B. Myerson, were awarded for their achievements in the theory of mechanism design). Hurwicz (1960, 1973) introduced the notion of incentive compatible mechanisms which in the form of dominance solvabili­ ty were applied by Groves and Ledyard (1977) to voluntary public good provision. The Revelation Principle had been propagated by Myerson (1979, 1981, 1982); Baron and My­ erson (1982); Mascin (1999); Dasgupta, Hammond, and Mascin (1979); Meyerson and Sattertwaithe (1983); and Wilson (1985), where we exclude the exercises for “large economies” (e.g. Bierbrauer and Hellwig, 2011). The Revelation Principle is now a stan­ dard tool of mechanism design theory (see the more recent contributions like Bergemann and Morris, 2005 and Jehiel and Moldovanu, 2001).

23.5 Conclusions The law, with its long tradition, has evolved to be very functional where, of course, func­ tionality or, in evolutionary terminology, the notion of fitness may be debatable. Some le­ gal scholars, especially of law and economics, would claim that economic efficiency is what determines institutional and thus legal change. We have challenged this proposition in section 23.2 by a procedural approach to legal mechanism design postulating that nonarbitrariness and fairness are crucial. It is illustrated by examples that this alternative ap­

Page 10 of 13

Mechanism Design and the Law proach stands in the tradition of what has been legally regulated and can allow for people’s active involvement. For Dominance Solvability we have conceded its efficiency but discussed also its disad­ vantages. For the Revelation Principle it has been illustrated how it allows to explore the welfare potential of each and every specific situation, but based on very unrealistic com­ mon knowledge assumptions. Even when these are granted, it suggests ad hoc designs of legal institutions. In our view, legal scholars are therefore justified in propagating proce­ durally fair law rather than rules, justified by efficiency claims. The law existed long before game-theoretic mechanism design. There is no need to quick­ ly adopt this new method of exploring what is possible under incentive compatibility con­ straints, simply because law is and should be guided more by procedural than by conse­ quentialistic requirements. There may be an increasing and fruitful exchange (p. 492) in the future. But legal scholars have good reasons to maintain their more procedural ap­ proach. That this does not mean abstaining from more rigorous characterizations of legal institutions is hopefully demonstrated by deriving bidding mechanism from basic require­ ments.

References Baron, D. P. and R. B. Myerson (1982). “Regulating a monopolist with unknown costs.” Econometrica: Journal of the Econometric Society 4: 911–930. Bergemann, D. and S. Morris (2005). “Robust mechanism design.” Econometrica 73(6): 1771–1813. Bierbrauer, F. and M. Hellwig (2011). Mechanism Design and Voting for Public-Good Pro­ vision. Preprints of the Max Planck Institute for Research on Collective Goods Bonn 2011/31. Dasgupta, P. S., P. J. Hammond, and E. S. Mascin (1979). “The implementation of social choice rules: some general results on incentive compatibility.” Review of Economic Stud­ ies 46: 185–216. Fehl, U. and W. Güth (1987). “Internal and external stability of bidder cartels in auctions and public tenders.” International Journal of Industrial Organization 5: 303–313. Gandenberger, O. (1961). Die Ausschreibung. Heidelberg: Quelle and Meyer. Groves, T. and J. Ledyard (1977). “Optimal Allocation of Public Goods: A Solution to the ‘Free Rider’ Problem.” Econometrica 45(4): 783–809. Güth, W. (1986). “Auctions, public tenders, and fair division games: An axiomatic ap­ proach.” Mathematical Social Sciences 11: 283–294. Güth, W. (2011). “Rules (of Bidding) to Generate Equal Stated Profits: An Axiomatic Ap­ proach.” Journal of Institutional and Theoretical Economics 167(4): 608–612. Page 11 of 13

Mechanism Design and the Law Güth, W. and M. Hellwig (1986). “The private supply of a public good.” Zeitschrift für Na­ tionalökonomie, Supplement 5, Welfare Economics of the Second Best: 121–159. Güth, W. and M. Hellwig (1987). “Competition versus monopoly in the supply of public goods.” In Efficiency, Institutions, and Economic Policy,183–217. Berlin and Heidelberg: Springer. Güth, W., R. Ivanova-Stenzel, and S. Kröger (2006). “Procurement Experiments with Un­ known Costs of Quality.” Pacific Economic Review, 11(2): 133–148. Güth, W. and H. Kliemt (2013). “Consumer Sovereignty Goes Collective. Ethical basis, ax­ iomatic characterization and experimental evidence,” in Martin Held, Gisela Kubon-Gilke, and Richard Sturn, eds., Normative und institutionelle Grundfragen der Ökonomik. Gren­ zen der Konsumentensouveränität. Jahrbuch 12, 83–98. Marburg: Metropolis. Güth, W. and B. Peleg (1996). “On Ring formation in auctions.” Mathematical Social Sciences 32: 1–37. Güth and E. Van Damme (1986). “A comparison of pricing rules for auctions and fair divi­ sion games.” Social Choice and Welfare 3: 177–198. Harsanyi, J. C. (1967–68). “Games with incomplete information played by ‘Bayesian’ play­ ers.” Parts I-III. Management Science 14:159–182, 320–334, 486–502. Hurwicz, L. (1960). “Optimality and informational efficiency in resource allocation processes,” in K. J. Arrow, S. Karlin, and P. Suppes, eds., Mathematical Methods in Social Sciences, 27–46. Stanford, CA: Stanford University Press. Hurwicz, L. (1973). “The design of mechanisms for resource allocation.” American Eco­ nomic Review 63(2): 1–30. Jehiel, P. and B. Moldovanu (2001). “Efficient design with interdependent valua­ tions.” Econometrica 69(5): 1237–1259. (p. 493)

Kagel, J. H. and Roth, A. E. (2016). The Handbook of Experimental Economics, Volume 2. Princeton, NJ: Princeton University Press. Maskin, E. S. (1999). “Nash equilibrium and welfare optimality.” Review of Economic Studies 66(1): 23–38. Milgrom, P. (1989). “Auctions and Bidding: A Primer.” Journal of Economic Perspectives 3(3): 3–22. Myerson, R. B. (1979). “Incentive compatibility and the bargaining problem.” Econometri­ ca 47(1): 61–73. Myerson, R. B. (1981). “Optimal Auction Design.” Mathematics of Operations Research 6(1): 58–73. Page 12 of 13

Mechanism Design and the Law Myerson, R. B. (1982). “Optimal coordination mechanisms in generalized principal–agent problems.” Journal of Mathematical Economics 10(1): 67–81. Myerson, R. B. and M. A. Satterthwaite (1983). “Efficient mechanisms for bilateral trad­ ing.” Journal of Economic Theory 29(2): 265–281. Ostrom, E. (2015). Governing the Commons. Cambridge: Cambridge University Press. Ostrom, E., J. Walker, and R. Gardner (1992). “Covenants with and without a Sword: SelfGovernance is Possible.” American Political Science Review 86(2): 404–417. Vickrey, W. (1961). “Counterspeculation, Auctions, and Competitive Sealed Tenders.” Journal of Finance 16(1): 8–37. Wilson, R. (1985). “Incentive efficiency of double auctions.” Econometrica 53(5): 1101– 1115.

Notes: (1) What is meant by legal evolution is that legal mechanisms are adapted in view of past experiences, for example, of past success of different rules like lowest bid or second-low­ est bid price rule. (2) For example, if stands for a new highway, some will suffer from its noise and pollution whereas others will enjoy it when traveling. (3) A public project for which people pay fees when using it can obviously generate more revenue than it costs. (4) A public authority could run a vector procurement auction to learn about the costs of the different projects (see Güth, Ivanova-Stenzel, and Kröger, 2006). (5) Due to a well-known equivalence result (see, e.g., Güth and Van Damme, 1986), such an analysis would only differentiate between pricing rules in case of a priori asymmetric auctions with incomplete information.

Werner Güth

Werner Güth is Director of the Strategic Interaction Group, Max Planck Institute of Economics. He has made pivotal contributions to game theory, experimental econom­ ics, microeconomics, psychology, philosophy, evolutionary biology, and the political sciences.

Page 13 of 13

Collective Decision-Making and Jury Theorems

Collective Decision-Making and Jury Theorems   Shmuel Nitzan and Jacob Paroush The Oxford Handbook of Law and Economics: Volume 1: Methodology and Concepts Edited by Francesco Parisi Print Publication Date: Apr 2017 Subject: Economics and Finance, Law and Economics, Econometrics, Experimental and Quanti­ tative Methods Online Publication Date: May 2017 DOI: 10.1093/oxfordhb/9780199684267.013.035

Abstract and Keywords Issues related to collective decision making and to Condorcet jury theorems have been studied and publicly discussed for over two hundred years. Recently, there is a burgeon­ ing interest in the topic by academicians as well as practitioners in the fields of Law, Eco­ nomics, Political Science, and Psychology. Typical questions are: What is the optimal size of a panel of decision makers such as a jury, a political committee, or a board of direc­ tors? Which decision rule to utilize? Who should be the members of the team, representa­ tives or professionals? What is the effect of strategic behaviour, group dynamics, conflict of interests, free riding, social interactions, and personal interdependencies on the final collective decision? This article presents current thinking in the field, offers suggestions for further research, and alludes to possible future developments regarding public choice and collective decision making. Keywords: collective decisions, Condorcet jury theorems, decision rules, strategic behavior, free-riding, personal interdependencies

24.1 Introduction THE Marquis de Condorcet (1743–1794) is considered one of the pioneers of the social sciences. In the English literature, Baker (1976) and Black (1958) were among the first to turn the attention of the scientific community to the importance of Condorcet’s writings (see Young, 1995). There is no doubt that his seminal work from 1785 has formed the ba­ sis for the development of social choice and collective decision-making as modern re­ search fields. In the last forty years, hundreds of articles have been written in these fields, and at least five new journals mainly devoted to these subjects have been pub­ lished. This chapter focuses on Condorcet jury theorems and related collective decision

Page 1 of 26

Collective Decision-Making and Jury Theorems issues which might be of interest to scholars in the areas of law, economics, and political science. In 1785 no jury existed in France. Condorcet applied probability theory to judicial ques­ tions and argued that the English demand for unanimity among jurors was unreasonable, suggesting instead a jury of twelve members that can convict with a majority of at least ten. In 1815 the first French juries used Condorcet’s rule but later adopted the simple majority rule. At that time the mathematician Laplace argued that simple majority is a dangerous decision rule for juries. Since 1837 juries had been established on several dif­ ferent plans, but the French law has never believed that one could count on twelve peo­ ple agreeing (see Hacking, 1990, ch. 11). Since the 1970s, several works have analyzed the jury system by applying probability theory as well as statistical data. Gelfand and Solomon (1973, 1975), Gerardi (2000), Klevorick and Rothschild (1979), and Lee, Broed­ ersz, and Bialek (2013) are a few such studies. The common English jury system is an ex­ treme form of a qualified majority rule, while the French version is less so. The debate about the size and the composition of the jury and the kind of decision rule it should use has been going on for more than two centuries and will probably contin­ (p. 495)

ue. This chapter presents the state of the art in collective decision-making and jury theo­ rems as well as some expectations about further developments in the field. Condorcet (1785) makes the following three-part statement: 1) The probability that a team of decision-makers would collectively make the cor­ rect decision is higher than the probability that any single member of the team makes such a decision. 2) This advantage of the team over the individual performance monotonically in­ creases with the size of the team. 3) There is a complete certainty that the team’s decision is right if the size of the team tends to infinity, that is, the probability of this event tends to one with the team’s size. A “Condorcet Jury Theorem” (henceforth, CJT) is a formulation of sufficient conditions that validate the above statements. There are many CJTs, but Laplace (1815) was the first to propose such a theorem. Some of the following conditions are explicitly stated by Laplace as sufficient conditions and the others are only implicitly assumed. A major con­ cern of the modern collective decision-making literature is whether or not each of these conditions is also necessary for the validity of Condorcet’s statements. These conditions, which are listed below, will serve as a road map for the rest of the chapter. Specification of the problem faced by the jury: 1) There are two alternatives. 2) The alternatives are symmetric.

Page 2 of 26

Collective Decision-Making and Jury Theorems Specification of the decision rule applied by the team: 3) The team applies the simple majority rule. Specification of the properties of the jury members: 4) All members possess identical competencies. 5) The decisional competencies are fixed. 6) All members share identical preferences. Specification of the behavior of the jury members: 7) Voting is independent. 8) Voting is sincere. The core of Laplace’s proof is the calculation of the probability of making the right collective decision (to be denoted by Π), where Π is calculated by using Bernoulli’s theorem (1713). Laplace shows, first, that Π is larger than the probability of making the right decision P by any single member of the team; second, that Π is a monotone increas­ (p. 496)

ing function of the size of the team; and third, that Π tends to unity with the size of the team. Of course, besides the above conditions there is an additional trivial condition that the decisional capabilities of decision-makers are not worse than that of tossing a fair coin. Namely, the probability P of making the correct decision is not less than one-half. Other properties of Π are that it is monotone increasing and concave in the size of the team, n, and in the competence of the individuals, P. Thus, given the costs and benefits of the team members’ identical competence, one can find the optimal size of a team and the optimal individual’s competence by comparing marginal costs to marginal benefits. The above eight conditions are quite strong and there is much doubt whether they exist in reality. The present chapter studies the consequences of the relaxation of these condi­ tions, which has been and apparently will continue to be a major topic in the literature of collective decision-making related to CJT. Given the limited space, it is possible to cover only some of the important subjects. Natu­ rally, to some extent this chapter will be biased towards the interest of the authors.

24.2 Heterogeneous Competencies Let us start by studying the consequences of relaxing the assumption that the members of the team possess identical competencies. Suppose that there are n members in a team facing two symmetric alternatives with decisional competence that can be represented by their fixed probabilities of making the right decision. Denote these probabilities by p = (p1, …, pn), where the pis are not necessarily identical. By simple numerical examples it is shown that the first two parts of Condorcet’s statement are no longer necessarily valid.

Page 3 of 26

Collective Decision-Making and Jury Theorems Suppose p = (0.95,0.80,0.80). Assuming that the team utilizes a simple majority rule, it is possible to find that the team’s competence in this case is Π = 0.944. One can see that the competence of the first member of the team is larger. Thus, the first part of the state­ ment is violated. Add now two more members to the team both with a competence of 0.6 to obtain the enlarged team where p = (0.95,0.80,0.80,0.60,0.60). Utilizing simple majori­ ty rule, the team’s competence is now reduced to Π = 0.910. The second part of the state­ ment is therefore also violated (the accuracy of majority decisions in groups with added members is discussed in Feld and Grofman, 1984). The numerical examples that are presented here are far from being extreme. For in­ stance, Nitzan and Paroush (1985, p. 68) find that if one draws a random sample of individual’s competencies from a uniform distribution over the range [1∕2, 1], then within five-member teams, in more than 20% of the cases the first part of Condorcet’s (p. 497) statement is violated. To wit, the competence of a random five-member team that utilizes a simple majority rule is smaller in at least 20% of the cases than the probability that the most qualified person of the team would decide correctly. In contrast to the first two parts of Condorcet’s statement, the survival of the third part is somehow surprising. Many attempts have been made to prove the validity of the third part in case of heterogeneous teams (see Boland, 1989; Fey, 2003; Kanazawa, 1998; and Owen, Grofman, and Feld, 1989). In fact, the following is a well-known version of CJT: “If a team of decision-makers utilizes a simple majority rule, the decision would be perfectly correct in the limit given that the size of the team tends to infinity, even if the individual competencies, the pis, are different, provided that pi ≥ 1∕2+ε, where the value of ε is a positive constant regardless of how small it is.” The proof of the theorem relies on the proof of Laplace where P = 1∕2+ε combined with the fact that Π is an increasing function of the team members’ competencies. One comment is in order. Paroush (1998) shows by means of an example that Π = 1∕2 in the limit even if all pi > 1∕2. This counterintuitive example occurs when the sequence of the pis decreases rapidly from above towards 1∕2. Thus, it seems that regardless of how small it is, the presence of a fixed ε is not only a sufficient but also a necessary condition. But it turns out that existence of a fixed ε is inessential for CJT. Relying on the law of large numbers, Berend and Paroush (1998) found a necessary and sufficient condition for the validity of the third part of the statement, even in the case that competencies are not identical. The condition is: (An-1∕2) √n →∞, where An = ∑pi ∕ n. The meaning of this condition is that if all pi > 1∕2, the team’s competence Π is still equal to one in the limit, even if the sequence of the pis decreases towards 1∕2 but only at a “moderate pace.” The next section shows that the first part of the statement can be retained if the team uti­ lizes not a simple but a weighted majority rule, where the weights are adjusted to the in­ dividuals’ competencies given that they are known.

Page 4 of 26

Collective Decision-Making and Jury Theorems

24.3 Optimal Decision Rules Condorcet, who was a great fan of the wisdom of the crowd, was among the first to lay down the philosophical foundations of democracy. In particular, he strongly believed in the superiority of simple majority over other decision rules (see Grofman, 1975). But, it is known today that if the team members’ competencies are “public knowledge,” the simple majority rule may lose its superiority. Many studies have suggested alternative criteria for the optimality of a decision rule (see, e.g., Rae, 1969; Straffin, 1977; Fishburn and Gehrlein, 1977). However, in what follows, the assumption that the team’s competence, Π, is the criterion for the optimality of the decision rule is preserved. Note that given that the team members share homogenous preferences, this criterion is also consistent with (equivalent to) expected utility. Allowing heterogeneous decisional capabilities, Shapley and Grofman (1981, 1984) and Nitzan and Paroush (1982, 1984b) find that the optimal decision rule is a weighted majority rule (WMR) rather than a simple majority rule (SMR). By maximization of the likelihood that the team makes the better of the two choices it confronts, they also (p. 498)

establish that the optimal weights are proportional to the log of the odds of the individu­ als’ competencies, that is, the weights, wi, are proportional to log[pi ∕(1-pi)]. Let us apply this result to the case of a team of three members with known competencies of p1 > p2 > p3, where pi is the probability that member i decides correctly. It is obvious that the simple majority is the superior decision rule if and only if . Thus, the simple majority rule loses its superiority to the ex­ pert rule if either p1 is close enough to 1 or p3 is close enough to 1∕2. For the general case of a team of n members, one can make the following statement: A sufficient condition for the violation of the first part of Condorcet’s statement is the in­ equality: p1 ⁄ (1-p1) > Π [pi ⁄ (1-pi)] where p1 is the competence of the most competent person in the team and Π indicates here the multiplication of the odds of the rest of the members. In the case of a team with more than three members, besides the SMR and the expert rule (ER), there are other efficient rules that are potentially optimal. For instance, in a five-member team there exist seven different efficient potentially optimal rules. These rules include the “almost expert rule,” the “almost majority rule,” and the “tie-breaking chairman rule.” The number of efficient rules increases very rapidly with the team size. For instance, in a team of nine members the number of efficient rules is already 172,958 (see Isbell, 1959). Now the following question is raised: Is there a mathematical relation between the size of the team and the exact number of efficient rules? This simple ques­ tion is still an open one. Note that a choice of the optimal size of a team may be considered as a special case of a choice of the optimal decision rule. For instance, the optimal rule within a team of seven Page 5 of 26

Collective Decision-Making and Jury Theorems members could be a restricted majority rule where simple majority rule is utilized only by the three or by the five most qualified members, where the rest of the members are inessential (their decisions are disregarded, that is, the weight assigned to an inessential member is zero). Paroush and Karotkin (1989) investigate the robustness of optimal restricted majority rules over teams with changing size. One of the useful findings is that optimal restricted majority rules are robust over reduction of the size of the team. For instance, suppose that a SMR is the optimal rule in a team of seven members, then a SMR will stay optimal even in a team of five where the two least qualified members of the original team are dis­ carded. The clarification of the conditions for the robustness of optimal decision rules when the team members’ competencies are allowed to vary is an important topic for fu­ ture research.

(p. 499)

24.4 Order of Decision Rules

The existence of order relations among decision rules is first noted in Nitzan and Paroush (1985, p. 35). The existence of such an order means, first, that the number of rankings of m efficient rules is significantly smaller than the theoretical number m! of all possible rankings of these rules and, second, that its existence is independent of the team’s com­ petence. Beyond the theoretical interest in studying order among efficient decision rules, the infor­ mation about the order has useful applications. Since the order relations are independent of the specific competencies of the decision-makers, the knowledge about the order of the rules is important in cases where the competencies are unknown or only partially known. For instance, if for some reason (e.g. excessive costs) the optimal rule cannot be used, then even in the absence of knowledge about the decisional competencies, the team can identify by the known order of the decision rules the second-best rule, the third-best, and so on. Karotkin, Nitzan, and Paroush (1988) show that six out of the seven efficient rules avail­ able to a team of five members generate a complete scale or an essential ranking. Thus, only fewer than 200 rankings out of 7! = 5040 possibilities are feasible. Paroush (1997) finds that the SMR and the ER are always polar rules in the sense that if one of them is optimal and thus the most efficient, the other one is the least efficient. He also identified second-best rules and penultimate rules in cases that majority rules (sim­ ple or restricted) are optimal or the most inferior. Karotkin (1998) exposes an interesting analogy between the network yielding the ranking of all weighted majority rules by efficiency and the weighted majority games in terms of winning and losing coalitions.

Page 6 of 26

Collective Decision-Making and Jury Theorems Paroush (1990) shows how to apply the knowledge of essential ranking of decision rules when the team of decision-makers faces multi-choice problems. Karotkin and Paroush (1994) apply the essential order in cases where the team members’ competencies may be subject to some fluctuations. For instance, if the optimal rule loses its optimality status due to fluctuations, then one of its neighbors on the ranking will become the second best. However, the complete order relations among efficient decision rules as well as the coiled network among them are still to be discovered; the task of studying these issues is far from being exhausted.

24.5 Competencies Are Not Fixed Assume that the individual’s decisional competence pi is not constant anymore but monot­ onically increasing (in a decreasing rate) with some invested resources, ci, which (p. 500) are intended to improve the individual’s competence pi. Such an investment in human capital might be a direct cost in the form of pecuniary outlay, or an indirect cost in the form of money equivalent of the time and efforts necessary to collect data or evidence, to elaborate, process, and transfer the relevant information into a decision. It also includes the costs necessary to shorten the delay of reaching a decision. Note that in most cases such an investment is very specific and depreciates completely immediately after the vot­ ing. The question of optimal investment in human capital is important and still open for future research. Only a few results are available and they have been obtained under restrictive assumptions. Assume that all the team members possess an identical learning function. To wit, for each one of the members pi(ci) = p(ci). It is worth noting that this assumption is consistent with a liberal viewpoint that the differences among individuals’ competencies are due to rea­ sons such as unequal opportunities and are not due to genetic, gender, race, or origin dif­ ferences. Based on this model, Nitzan and Paroush (1980), Karotkin and Paroush (1995), and BenYashar and Paroush (2003) obtain the following results: 1) Given a standard learning function such as the logistic function, which is common­ ly used in psychology and education, the optimal social policy is first to invest in the least competent member of the team and only then in the more competent ones and only finally in the most competent member. 2) Denote by c* = (c*1,c*2, … ,c*n) the socially optimal investment in each member where the investment in each person is decided in a centralized system. Denote by c** = (c**1,c**2, … , c**n) the optimal investment that is determined by each individ­ ual in a decentralized system. Assume that individual’s reward is independent of her own vote but depends only on the success of the whole team. Under this assumption, the motivation to free-ride emerges and the necessary result is that c* > c**. Page 7 of 26

Collective Decision-Making and Jury Theorems 3) An incentive scheme that compensates not only for the team’s success but also for the individual’s success is a possible remedy against free-riding. However, such a remedy has two deficiencies as by-products. First, it encourages individuals to vote insincerely (see section 24.9). Second, it encourages individuals to violate the as­ sumption of independence (see section 24.8). For instance, even if the simple majori­ ty is the optimal decision rule and even if individuals share identical preferences, each of them will try in this case to copy the vote of the most competent member in order to increase the likelihood of her personal reward. 4) Modern technology offers ways to overcome the above deficiency. Secret voting together with automatic recording of votes will guarantee both the independence as well as the possibility to compensate individuals for correct voting. It is worth noting that free-riding behavior in information acquisition affects not only the optimal investment in human capital, but also the optimal size of the team. (p. 501) Grad­ stein and Nitzan (1987) and Mukhopadhaya (2003) find, for instance, that in contrast to Condorcet’s second statement, a larger jury may reach worse decisions. In an extended setting of endogenous competencies, Ben-Yashar and Nitzan (2001c) demonstrate that it is no longer the case that the SMR is superior to the ER, even if indi­ viduals share identical skills. Since now the objective of the team takes into account both the probability of making a correct collective decision as well as the aggregate costs asso­ ciated with investment in human capital of the team members, it may very well be the case that it is better to invest in a single member and apply the ER rather than in all the members. Thus, Condorcet’s statement may not be valid. Further discussion on variability of decisional skills and of information acquisition within teams appears in Gerling et al. (2005), Gradstein and Nitzan (1988), and Nitzan and Paroush (1984c). In concluding this section, we wish to point out that the area of the economics of deci­ sion-making that takes into account various elements of cost and benefit is only in its ini­ tial stage and further developments are expected.

24.6 Diverse Preferences Consider a group of individuals with different preferences (note that such a group can no longer be called a “team”). A typical such group is a political committee, a parliament, a randomly selected jury, or any representative decision panel whose members have differ­ ent utilities or social values. In a binary setting, it is clear that the alternative that is considered “correct” by one of the members might be considered “incorrect” by anoth­ er. There is no agreed upon “truth” to seek; each member seeks her own subjective truth. Social choice theory struggles to find a decision rule that aggregates the diverse individ­ ual preferences and aims to come up with some “desirable” social preference relation or some common social objective. The classical normative approach establishes such objec­ tives or social welfare functions on the basis of philosophical considerations or principles of justice (Rawls, 1971). But the more modern approach in the social choice literature is Page 8 of 26

Collective Decision-Making and Jury Theorems axiomatic. The axiomatic approach assumes a set of “reasonable” axioms or “desirable” properties that are deemed necessary and sufficient conditions for the derivation of an aggregation rule that enables society to take a collective action even in situations where conflicts of interest prevail among its members. Among the studies that use the axiomatic approach in a binary setting, May (1952) derives SMR, Fishburn (1973) and Nitzan and Paroush (1981) characterize WMR, and Houy (2007) and Quesada (2010) come up with qualified majority rules. The following dilemma still remains: Which of the two juristic systems is socially prefer­ able, a randomly selected group of jurors who possess diverse preferences as well as a variety of social norms or a team of professional judges who share a common goal of seeking the truth? Obviously, this dilemma has important applications, but as far as we know, there is neither an answer to this question nor even a theoretical framework within which the question can be analyzed. In the context of Condorcet’s setting, given the individuals’ common objective and diverse information which yields their decisions, the optimal collective decision rule can be identified, as we have seen in section 24.3. However, in a binary setting and diverse (p. 502)

preferences, one can reach these same optimal collective decision rules by their unique axiomatic characterization. In a more general multi-alternative setting, however, the po­ tential success of the axiomatic approach is clouded by the classical Impossibility Theo­ rems of Arrow (1951) and his followers. As is well known, if few reasonable axioms have to be satisfied by the aggregation rule, a social welfare function does not exist. This type of finding, which has become one of the cornerstones of social choice theory, has generat­ ed a wave of works that have attempted (mostly in vain) to disperse the pessimistic at­ mosphere implied by Arrow’s “Impossibility Theorem.”

24.7 Asymmetric Alternatives Suppose now that asymmetry exists between the two alternatives. Asymmetry may stem from three different sources. First, the priors of the two states of nature (the defendant being innocent and the defendant being guilty) may be different. For example, assume that most people are not criminals, so in order to convict a defendant the state “innocent” is considered as a status quo that has to be refuted and the state “guilty” is deemed as the alternative that has to be proved. Thus, a collective decision to convict is expected to be “beyond any doubt,” whereas collective acquittal may remain doubtful. Second, the net benefits of a correct decision under the two states of nature can be different. The two types of errors, acquittal of a guilty defendant and conviction of an innocent defendant, may have different costs. Third, an individual’s decisional competency may depend on the state of nature. In particular, the probability to decide correctly if the state is “innocent” can be different from the probability to decide correctly if the state is “guilty.” In any case, the decisional competency of an individual is not parameterized by a single proba­ bility of making a correct choice, but by two probabilities of deciding correctly in the two states of nature. Page 9 of 26

Collective Decision-Making and Jury Theorems Under asymmetry it is required that one of the two alternatives, say conviction, will be the collective choice only when it receives the support of a special majority with a quota larger than one-half. Thus, under asymmetry the decision rule should be a qualified ma­ jority rule (QMR). QMRs are discussed in several works in the political context of constitutions and funda­ mental laws as well as in relation to juries (see, e.g., Buchanan and Tullock, 1962; Rae, 1969). Nitzan and Paroush (1984d) were the first to derive the exact quota necessary for the optimal QMR. However, their quota is derived under the restrictive conditions of iden­ tical competencies that are invariant to the state of nature. The special case of identical competencies that depend on the state of nature was extensively analyzed by Sah and Stiglitz (1988) and Sah (1990, 1991). Allowing heterogeneous and state-dependent competencies, Ben-Yashar and Nitzan (1997) specify the expression for both the weight that has to be assigned to each member of the team under the optimal rule as well as the desirable quota of votes neces­ sary for the rejection of the status quo. The optimal rule in this more general case there­ fore becomes a weighted qualified majority rule, WQMR. The optimal weight is now pro­ (p. 503)

portional to the average of the logs of the odds of the two probabilities of making a cor­ rect choice and the optimal quota is a function of four parameters: the log of the two probabilities of making a correct decision, the log of the odd of the prior probability, and the log of the ratio of the two net benefits. The QMR may take two extreme forms. In the first one, the quota is maximal and equal to 1 and in the second form the quota is minimal and equal to 1/n where n is the number of members in the team. In the former case the choice of the alternative with the unfavor­ able bias requires unanimity (as in the case of the English and American jury) and in the latter case it requires the minimal support of a single member. Ben-Yashar and Nitzan (1997) present necessary and sufficient conditions for the optimality of an extreme quota, and Sah and Stiglitz (1985, 1986, 1988) compare the two extreme QMRs in an organiza­ tional context. Further discussion of these extreme QMRs appear in Ben-Yashar and Nitzan (2001b) and Koh (2005). Recently, Ben-Yashar and Nitzan (2014) present sufficient conditions for the superiority of a QMR that is solely based on the prior and entirely disregards the members’ decisions. This degenerate “prior based rule” emerges as the most efficient rule in cases where the prior probability is large enough in comparison to the individuals’ ability to decide cor­ rectly.

24.8 On the Violation of Independence Before taking a vote there is, in most cases, a preliminary process of deliberation and elaboration. Such an early interaction among the members of the panel has two major im­ portant contrasting effects on the likelihood to reach collectively a correct decision. One is positive (it increases Π) and the other is negative (it decreases Π). The positive effect is Page 10 of 26

Collective Decision-Making and Jury Theorems the learning factor which improves the competencies, exactly as investment in human capital does (see section 24.5). For instance, in many cases, the process includes witness­ ing evidences, hearing testimony, listening to other opinions, collecting bits of informa­ tion, observing signals about states of nature, and being exposed to different viewpoints, where all these factors elevate the decisional competencies to a higher level. The negative effect is entirely due to factors of social psychology that distort the indepen­ dence. For instance, in many cases, social pressures, false persuasive arguments, threats, influential power, or leadership charisma enhance conformity, affecting the weights of the optimal WMR and thus reducing the collective competency. Some enlightening examples of negative impacts of interdependencies on the correct decision (p. 504) are presented in the following two bestsellers: Surowiecki (2004, ch. 3) and Kahneman (2011, ch. 7). Within the dichotomous choice framework, Nitzan and Paroush (1984e) establish that in­ dependent voting is always superior to almost any pattern of dependent voting, given that the team utilizes the optimal decision rule. In case of violation of the independency, their study also shows how it is possible to adjust the individuals’ weights in order to derive a new optimal WMR. To illustrate the argument, consider a team with the competencies p = (0.70, 0.65, 0.65, 0.65, 0.65). The optimal rule in this case is SMR, which yields under independence Π = 0.78. Suppose now that the most qualified member exercises her leadership or personal charisma on the rest of the team so that the other members become followers who try to mimic her vote with the mistaken idea that they can increase Π by increasing their own ability from 0.65 to 0.70. Obviously, under such dependency the value of Π is reduced to 0.70. Expecting this result, the weights can be adjusted by turning the leader together with another member to be inessentials (with zero weights). After this adjustment the SMR yields Π = 0.7182. Let us add that the negative effects of social psychology have stronger impact when the members of the panel possess diverse rather than identical preferences. The reason is that, in the absence of conflicts of interests, the members’ self consciousness helps to overcome the potential negative psychological effects. This result magnifies the necessity and the importance of secret voting. Secret voting may eliminate, at least in part, the negative effects of dependencies. Also note that it is very difficult to impose secret voting on a representative group of decision-makers, like a political com­ mittee, because the conspicuous vote serves as a report to the public about the members’ actual opinion or about their loyalty to their declared attitude. However, it is evident that governments allocate a lot of resources to carry out secret elections. A discussion of the general case where independence of decisional skills does not exist appears in several studies, for example Berg (1993) and Ladha (1992, 1993), who prove a CJT under correlated votes by imposing some restrictions on the correlation coefficients (see also Peleg and Zamir, 2012). Their results display again the robustness of Condorcet’s third statement.

Page 11 of 26

Collective Decision-Making and Jury Theorems Besides the available results in the psychological literature, it is expected that further studies of the impact of group dynamics or social behavior in general on collective deci­ sion-making will yield more results. Such studies will certainly contribute to the debate about the necessity of secret voting.

24.9 Strategic Behavior It seems that no individual has an incentive to manipulate her vote and act insincerely. It turns out, however, that this may not be a valid assumption; that is, it is possible that indi­ viduals have an incentive to vote insincerely. Using a game-theoretic analysis, Austen-Smith and Banks (1996) show that nonstrategic voting may be inconsistent with Nash equilibrium, even when all members have identical preferences and beliefs. More precisely, if the number of voters is sufficiently large, then voting based on private information only (informative voting) is generically not a Nash equilibrium of a Bayesian game that formally represents a CJT. The general idea is that an individual’s vote affects the collective decision only when it is pivotal. But (p. 505)

if all the others vote informatively, the fact that they are tied may be very informative in the sense that this reveals more precise information than that privately held by the indi­ vidual. Following this newly revealed information, it may be rational not to vote according to one’s own private information. The following example illustrates the argument. Consider three individuals with identical preferences over two alternatives, A and B. There is uncertainty about the true state of the world, which may be either state A or state B. In each state, individuals receive a payoff of one if the alternative of the specific state is chosen and a payoff of zero otherwise. There is a common prior probability that the true state is A. Individuals have private in­ formation about the true state of the world. Majority rule is used to select an alternative. There are two additional assumptions on individuals’ beliefs: (1) sincere voting is informa­ tive; and (2) the common prior belief that the true state is A is sufficiently strong, such that if any individual were to observe all the three individual signals, then that individual believes B is the true state only if all the available evidence supports the true state being B. In this example sincere voting is not rational. Suppose that you are playing this game and the two other individuals vote sincerely. You must then be in one of the three following situations: (1) the others have observed that the state is A and accordingly vote for A; (2) the others have both observed B and vote for B; or (3) the others have observed different signals and one votes for A and the other for B. In the first two scenarios the aggregated outcome is independent of your own vote, and in the third scenario your vote is decisive. However, if you are in the third scenario your best decision is to vote for A. Therefore, voting sincerely is not a Nash equilibrium.

Page 12 of 26

Collective Decision-Making and Jury Theorems In response to the above finding, McLennan (1998) and Wit (1998) found conditions un­ der which Nash equilibrium behavior, although it may be inconsistent with non-strategic voting, still predicts that groups are more likely to make correct decisions than individu­ als. Feddersen and Pesendorfer (1997, 1998) adapt the general framework of AustenSmith and Banks (1996) to the specific case of jury procedures in criminal trials. In their model it is never a Nash equilibrium for all jurors to vote non-strategically under the una­ nimity rule. Moreover, Nash equilibrium behavior leads to higher probabilities of both convicting the innocent and acquitting the guilty under the unanimity rule than under al­ ternatives rules, including the simple majority rule. Ben-Yashar and Milchtaich (2007) examine the question of strategic voting when voters are solely concerned with the common collective interest. They find that under the opti­ mal WMR, individuals do not have an incentive to vote strategically and non-informative­ ly. Such strategy-proofness does not hold under second-best anonymous (p. 506) voting rules. Thus, assigning the proper weight to each voter achieves the optimal team perfor­ mance and guarantees sincere voting.

24.10 Latent Competencies In section 24.3 we presented the optimal WMR where the weight assigned to each team’s member is proportional to the log of the odd of her competency. In reality, individuals’ competencies are mostly latent and far from being public knowledge. But, in some cases there are observed signals. It is evident that, in order to demonstrate their expertise and display their excellence, professional consultants such as lawyers and medical doctors decorate the walls of their offices with the records of their education, experience, seniori­ ty, and so forth. With the help of “word of mouth” such signals might even become public knowledge. The public should also be aware of false signals and thus, in general, at­ tempts at subjective evaluation of decisional skills might be unreliable so that the optimal WMR cannot be applied. However, in certain cases it is possible to use objective statistical data for estimating per­ sonal competencies. For instance, Nitzan and Paroush (1985, p. 107) use 362 medical records to estimate competencies of cardiologists, and Karotkin (1994) uses data of suc­ cessful appeals for estimation of the competencies (in fact, lack of competencies) of pro­ fessional judges. Grofman and Feld (1983) and Nurmi (2002), among others, also suggest the estimation of voters’ decisional capability by the extent that their observed past decisions align with those of the majority. In other words, the majority decision is used as a plausible endoge­ nous proxy for the true alternative. Recently, Baharad et al. (2011, 2012) make use of this idea along with the algorithm suggested by Dempser, Laird, and Rubin (1977) to general­ ize and optimize the method by constructing a specific algorithm that iteratively updates the weights assigned to each of the team members. The procedure converges to consis­

Page 13 of 26

Collective Decision-Making and Jury Theorems tent and stable evaluation of the voters’ skill and, in turn, it enables the identification of the optimal aggregation rule. Voter equality as implicit in SMR is justified in situations where decisional skills are all of sufficient quality and homogeneity. In this case, a sufficiently large number of voters re­ sults in convergence to the maximal collective performance. In contrast, optimal decisionmaking that is based on the voting procedure proposed in Baharad et al. (2011, 2012) is always warranted, deriving its superiority from its ability to identify skills through learn­ ing from experience. The merit of this procedure and its advantage over SMR are espe­ cially high when skills are sufficiently spread out and the track record of individual deci­ sions is sufficiently abundant. Many studies attempt to draw conclusions regarding the desirable collective decision rule when competencies are only partially known. For instance, consider the case where the statistical distribution from which competencies are randomly drawn can (p. 507) be spec­ ified (see, e.g., Nitzan and Paroush, 1985, ch. 5; Nitzan and Paroush, 1984a; Ben-Yashar and Paroush, 2000; Berend and Harmse, 1993). Let us conclude with the following comments. Even in the absence of any information on individuals’ decisional ability, if a team utilizes SMR the first and the third Condorcet’s statements are valid as long as the voters stay away from being fair coins. 1) The value of the unknown probability of correct decision of a team that utilizes SMR exceeds the expected value of the competency of a random representative of the team, that is, Π >∑pi ∕n (see section 24.2). Thus, the first statement is valid. Nitzan and Paroush (1985, p. 69) offer a simple proof of this result for the case of a three-member team. 2) The third statement is also valid and apparently several democracies use this fact to run public opinion polls, especially where binary critical questions are at stake. Some restrictions, like age limitation, are imposed to guarantee that voters’ compe­ tencies are larger than one half. Is it a coincidence that the custom of using public opinion polls is more popular in the French-speaking democracies?

24.11 Miscellaneous 24.11.1 The Hung-Jury Problem In many states juries must reach a unanimous decision. While a larger jury may be more likely to make the correct decision (Condorcet’s second statement), it may not reach any decision at all. Therefore, the balance between making the correct decision and reaching a decision whatsoever must be found. Several papers have researched jury decision-mak­ ing under the unanimity method.

Page 14 of 26

Collective Decision-Making and Jury Theorems The intuition that larger juries may be more likely to end in hung juries is demonstrated by Grofman (1976). He shows that if the unanimity rule is used, then juries of twelve are more likely to be hung juries compared to juries of six. Alternatively, Luppi and Parisi (2012) show that when information cascades exist, the ef­ fect of jury size on hung juries can be reduced or even reversed. Information cascades oc­ cur when people make a decision based on the observation of others while ignoring or discounting their private information. The intuition here is that with an increase in the size of the jury, the potential for different opinions to result in hung juries increases but, at the same time, if the opinions are expressed in succession, the jurors may adjust their views in light of what others have expressed and an information cascade can be initiated. Therefore, the probability of a unanimous outcome may increase with the size of (p. 508) the jury. Consequently, consensus will be more likely to occur in larger groups if enough jurors are open to opinions of other jurors. In a different strategic setting, Feddersen and Pesendorfer (1998) study the intuition that the unanimous decision requirement reduces the probability of convicting an innocent de­ fendant, while increasing the probability for acquitting the guilty defendant. They show that under strategic voting the unanimity rule can lead to a high probability of both types of errors, and the probability of convicting an innocent defendant might actually increase with the number of jurors.

24.11.2 Multi-Criteria Decisions The problem of aggregating individual judgments may take the form of a multi-issue (cri­ teria) problem where individual judgments need to be the basis of the collective outcome. In this generalized setting the simple majority rule can be applied sequentially but in dif­ ferent forms. Consider a three-member jury that has to come to a conclusion on whether a defendant is guilty or not. Let the available evidence consist of fingerprints found at the scene of the crime, an eyewitness, and the motive for the crime. Each juror decides to convict the de­ fendant if she believes that at least two of the evidence criteria justify this decision, and conviction requires the support of a majority of the jury. For example, let us look at Table 24.1. In this situation the defendant is found guilty because, according to the judgments of Ju­ ror 1 and Juror 2, she is guilty. Notice, however, that the collective decision could have been based not on the majority of the jury members’ judgments, but on the majority of the criteria where the signal of each criterion is based on the majority of the jurors’ judg­ ments regarding that criterion. In the above example a majority of the jury believes that, according to the fingerprints criterion and according to the eyewitness criterion, the ac­ cused is innocent, so the defendant is declared innocent according to a majority of the criteria. The different collective outcomes obtained under the two alternative majoritybased decision rules raise the so-called doctrinal paradox. Page 15 of 26

Collective Decision-Making and Jury Theorems Table 24.1 An Example of the Doctrinal Paradox Fingerprints

Witness

Motive

Majority

Juror 1

Guilty

Innocent

Guilty

Guilty

Juror 2

Innocent

Guilty

Guilty

Guilty

Juror 3

Innocent

Innocent

Innocent

Innocent

Majority

Innocent

Innocent

Guilty

Innocent/Guilty

Page 16 of 26

Collective Decision-Making and Jury Theorems Kornhauser and Sager (1986) illustrate the doctrinal paradox using a three- mem­ ber court where a defendant is liable according to the legal doctrine if and only if she committed some action which, by contract, she was obliged not to commit. The paradox arises in this context if a majority of the judges find that the defendant is not liable, whereas a majority may still find that she committed some action that, by contract, she was obliged not to perform. (p. 509)

List (2006) discusses possible procedures to escape the doctrinal paradox and obtain a reasonable final decision. One procedure is the minimally liberal one, where each of the judges reaches her own decision and then a majority rule is used. An alternative proce­ dure is the comprehensive deliberate procedure, where each judge gives her judgment on each of the specific issues and these are aggregated by majority rule. More recently, Hartmann, Pigozzi, and Sprenger (2010) shift the focus to the search for the procedure that, as in our setting, maximizes the probability of selecting the correct result. In pursuit of the most desirable judgment aggregation procedure, they use proba­ bilistic methods, looking at the reliability of the judges involved focusing on two ap­ proaches: distance-based methods and Bayesian analysis. Let us conclude by noting that the existence of the doctrinal paradox fortunately implies that, for a particular profile of judgments, at least one of the sequential majority-based decision rules presented above yields the appropriate outcome. A more troubling possibil­ ity is the existence of a “truth-tracking paradox” whereby, for a particular profile of judg­ ments, the doctrinal paradox does not exist, but nevertheless the appropriate collective decision is missed. In other words, the two majority-based rules may result in the same collective decision, however, the optimal majority rule surprisingly supports the alterna­ tive possible outcome (see Baharad, Ben-Yashar, and Nitzan, 2013).

24.11.3 Applications The study of uncertain dichotomous choices has numerous applications in a variety of fields where collective decision-making is of major significance. This is true in particular in law, medicine, economics, business administration, and political science. The relevance to law is clear in court and jury decision-making (Gelfand and Solomon, 1973, 1975; Ger­ ardi, 2000; Karotkin, 1994; Romeijn and Atkinson, 2011; Klevorick, Rothschild, and Win­ ship, 1984). The same is true in the area of medical diagnostics (Nitzan and Paroush, 1985). Within economics (public economics, industrial organization) and business admin­ istration, the insights we have discussed can shed light on strategies for an organization’s investment in human capital under conditions of uncertainty (Ben-Yashar and Nitzan, 1998, 2001a); on the size of decision-making bodies such as boards of directors (Kahana and Paroush, 1995; Gradstein, Nitzan, and Paroush, 1990); on the decision rules applied by firms (Nitzan and Procaccia, 1986); on the organizational form of industries, for exam­ ple partnership (Paroush, 1985); and on methods of personnel (p. 510) selection in organi­ zations (Ben-Yashar, Nitzan, and Vos, 2006). In political science it is central, for instance, to debates on the “epistemic” conception of democracy (List and Goodin, 2001). And it Page 17 of 26

Collective Decision-Making and Jury Theorems can even shed light on decision-making by editors of academic journals and, in particular, on whether referees are sufficiently informed about a journal’s editorial practices (BenYashar and Nitzan, 2001d).

Acknowledgment The authors are indebted to Assaf Basevich for his most useful assistance.

References Arrow, K. J. (1951). Social Choice and Individual Values (New York: John Wiley & Sons). Austen-Smith, D. and J. Banks (1996). “Information aggregation, rationality and the Con­ dorcet jury theorem.” American Political Science Review 90(1): 34–45. Baharad, E., R. Ben-Yashar, and S. Nitzan (2013). “Do consistent majority judgments guar­ antee the truth?” Mimeo. Baharad, E., M. Koppel, J. Goldberger, and S. Nitzan (2011). “Distilling the Wisdom of Crowds: Weighted Aggregation of Decisions on Multiple Issues.” Autonomous Agents and Multi-Agent Systems 22(1): 31–42. Baharad, E., M. Koppel, J. Goldberger, and S. Nitzan (2012). “Beyond Condorcet: Optimal Judgment Aggregation Using Voting Records.” Theory and Decision 72(1): 113–130. Baker, M. K., ed. (1976). Condorcet: selected writings. Indianapolis, IN: Bobbs-Merrill. Ben-Yashar, R. and I. Milchtaich (2007). “First and second best voting rules in commit­ tees.” Social Choice and Welfare 29(3): 453–486. Ben-Yashar, R. and S. Nitzan (1997). “The optimal decision rule for fixed size committees in dichotomous choice situations—The general result.” International Economic Review 38(1): 175–187. Ben-Yashar, R. and S. Nitzan (1998). “Quality and structure of organizational decisionmaking.” Journal of Economic Behavior and Organization 36(4): 521–534. Ben-Yashar, R. and S. Nitzan (2001a). “Investment criteria in single and multi-member or­ ganizations.” Public Choice 109(1-2): 1–13. Ben-Yashar, R. and S. Nitzan (2001b). “The robustness of optimal organizational architec­ tures: A note on hierarchies and polyarchies.” Social Choice and Welfare 18(1): 155–163. Ben-Yashar, R. and S. Nitzan (2001c). “The invalidity of Condorcet jury theorem under en­ dogenous decisional skills.” Economics of Governance 2(3): 243–249. Ben-Yashar, R. and S. Nitzan (2001d). “Are referees sufficiently informed about the editor’s practice?” Theory and Decision 51(1): 1–11. Page 18 of 26

Collective Decision-Making and Jury Theorems Ben-Yashar, R. and S. Nitzan (2014). “On the significance of the prior of a correct decision in committees.” Theory and Decision 76(3): 317–327. Ben-Yashar, R., S. Nitzan, and H. Vos (2006). “Comparison of optimal cutoff points for sin­ gle and multiple tests in personnel selection.” Psicothema 18(3): 652–660. Ben-Yashar, R. and J. Paroush (2000). “A non-asymptotic Condorcet jury theorem.” Social Choice and Welfare 17: 189–199. Ben-Yashar, R. and J. Paroush (2003). “Investment in human capital in team mem­ bers who are involved in collective decision making.” Journal of Public Economic Theory 5(3): 449–548. (p. 511)

Berend, D. and J. E. Harmse (1993). “Expert rule versus majority rule under partial infor­ mation.” Theory and Decision 35(2): 179–197. Berend, D. and J. Paroush (1998). “When is Condorcet’s jury theorem valid.” Social Choice and Welfare 15(4): 481–488. Berg, S. (1993). “Condorcet’s jury theorem, dependency among voters.” Social Choice and Welfare 10(1): 87–95. Black, D. (1958). The Theory of Committees and Elections. Cambridge: Cambridge Uni­ versity Press. Boland, P. J. (1989). “Majority systems and the Condorcet jury theorem.” The Statistician 38(3): 181–189. Buchanan, J. M. and G. Tullock (1962). The Calculus of Consent: Logical Foundations of Constitutional Democracy. Ann Arbor, MI: University of Michigan Press. De Condorcet, N. C. (1785). Essai sur l’application de l’analyse a la probabilite des deci­ sions rendues a la pluralite des voix. Paris: De l’imprimerie royale. Dempster, A. P., N. M. Laird, and D. B. Rubin (1977). “Maximum likelihood from incom­ plete data via the EM algorithm.” Journal of the Royal Statistical Society. Series B (Methodological) 39(1): 1–38. Feddersen, T. J. and W. Pesendorfer (1997). “Voting behavior and information aggregation in elections with private information.” Econometrica 65(5): 1029–1058. Feddersen, T. J. and W. Pesendorfer (1998). “Convicting the innocent: The inferiority of unanimity jury verdicts under strategic voting.” American Political Science Review 92(1): 23–35. Feld, S. and B. Grofman (1984). “The accuracy of group majority decisions in groups with added numbers.” Public Choice 42(3): 273–285.

Page 19 of 26

Collective Decision-Making and Jury Theorems Fey, M. (2003). “A note on the Condorcet jury theorem with supermajority rules.” Social Choice and Welfare 20(1): 27–32. Fishburn, P. C. (1973). The theory of social choice. Princeton: Princeton University Press. Fishburn, P. C. and W. V. Gehrlein (1977). “Collective rationality versus distribution of power of binary social choice functions.” Journal of Economic Theory 15(1): 72–91. Gelfand, A. and H. Solomon (1973). “A study of Poisson’s model for jury verdicts in crimi­ nal and civil courts.” Journal of the American Statistical Association 68(342): 271–278. Gelfand, A. and H. Solomon (1975). “Analyzing the decision-making process of the Ameri­ can jury.” Journal of the American Statistical Association 70(350): 305–310. Gerardi, D. (2000). “Jury verdicts and preference diversity.” American Political Science Review 94(2): 395–406. Gerling, K., H. P. Grüner, A. Kiel, and E. Schulte (2005). “Information acquisition and deci­ sion making in committees: A survey.” European Journal of Political Economy 21(3): 563– 597. Gradstein, M. and S. Nitzan (1987). “Organizational decision making quality and the severity of the free-riding problem.” Economics Letters 23(4): 335–339. Gradstein, M. and S. Nitzan (1988). “Participation decision aggregation and internal in­ formation gathering in organizational decision making.” Journal of Economic Behavior and Organization 10(4): 415–431. Gradstein, M., S. Nitzan, and J. Paroush (1990). “Collective decision making and the limits on the organization’s size.” Public Choice 66(3): 279–291. Grofman, B. (1975). “A comment on ‘Democratic Theory: A Preliminary Mathematical Model.” Public Choice 21(1): 100–103. Grofman, B. (1976). “Not necessarily twelve and not necessarily unanimous,” in G. Bermant, C. Nemeth, and N. Vidmar, eds., Psychology and the Law, 149–168. Lexington, MA: Lexington Books. (p. 512)

Grofman, B. and S. L. Feld (1983). “Determining optimal weights for expert judgment,” in B. Grofman and G. Owen, eds., Information Pooling and Group Decision Making: Proceed­ ings of the Second University of California, Irvine, Conference on Political Economy. Greenwich, CT: JAI Press Inc. Hacking, I. (1990). The Taming of Chance. Cambridge: Cambridge University Press. Hartmann, S., G. Pigozzi, and J. Sprenger (2010). “Reliable Methods of Judgment Aggre­ gation.” Journal of Computational Logic 20(2): 603–617.

Page 20 of 26

Collective Decision-Making and Jury Theorems Houy, N. (2007). “A characterization for qualified majority voting rules.” Mathematical So­ cial Sciences 54(1): 17–23. Isbell, J. R. (1959). “On the enumeration of majority games.” Mathematical Tables and Other Aids of Computation 13(65): 21–28. Kahana, N. and J. Paroush (1995). “The size of the executive board of labor-managed firms.” Advances in the Economic Analysis of Participatory and Labor-Managed Firms 5: 175–183. Kahneman, D. (2011). Thinking, Fast and Slow. New York: Farrar, Straus and Giroux. Kanazawa, S. (1998). “A brief note on a further refinement of Condorcet Jury Theorem for heterogenous groups.” Mathematical Social Sciences 35(1): 69–73. Karotkin, D. (1994). “Effect of the size of the bench on the correctness of court judg­ ments: The case of Israel.” International Review of Law and Economics.14(3): 371–375. Karotkin, D. (1998). “The network of weighted majority rules and weighted majority games.” Games and Economic Behavior 22(2): 299–315. Karotkin, D., S. Nitzan, and J. Paroush (1988). “The essential ranking of decision rules in small panels of experts.” Theory and Decision 24(3): 253–268. Karotkin, D. and J. Paroush (1994). “Variability of decisional ability and the essential or­ der of decision rules.” Journal of Economic Behavior and Organization 23(3): 343–354. Karotkin, D. and J. Paroush (1995). “Incentive schemes for investment in human capital by members of a team of decision makers.” Labor Economics 2(1): 41–51. Klevorick, A. K. and M. Rothschild (1979). “A model of the jury decision process.” Journal of Legal Studies 8(1): 141–164. Klevorick, A. K., M. Rothschild, and C. Winship (1984). “Information processing and jury decision making.” Journal of Public Economics 23(3): 245–278. Koh, W. T. H. (2005). “The optimal design of fallible organizations; Invariance of optimal decision criterion and uniqueness of hierarchy and polyarchy structures.” Social Choice and Welfare 25(1): 207–220. Kornhauser, L. A. and L. G. Sager (1986). “Unpacking the court.” The Yale Law Review 96(1): 82–117. Ladha, K. K. (1992). “The Condorcet jury theorem, free speech and correlated votes.” American Journal of Political Science 36(3): 617–634. Ladha, K. K. (1993). “Condorcet’s jury theorem in light of de Finetti’s theorem, Majority rule with correlated votes.” Social Choice and Welfare 10(1): 69–86.

Page 21 of 26

Collective Decision-Making and Jury Theorems Laplace, P. S. de (1815). Theorie analytique des probabilities. Paris: n.p. Lee, E. D., C. P. Broedersz, and W. Bialek (2013). “Statistical mechanics of the US Supreme Court.” arXiv preprint arXiv:1306.5004. List, C. (2006). “The Discursive Dilemma and Public Reason.” Ethics 116(2): 362–402. List, C. and R. E. Goodin (2001). “Epistemic democracy: Generalizing the Condorcet Jury Theorem.” Journal of Political Philosophy 9(3): 277–301. Luppi, B. and F. Parisi (2012). “Jury Size and the Hung-Jury Paradox.” University of St. Thomas Legal Studies Research Paper 12–31. (p. 513)

McLenan, A. (1998). “Consequences of the Condorcet jury theorem for beneficial informa­ tion aggregation by rational agents.” American Political Science Review 92(2): 413–418. May, K. O. (1952). “A set of independent necessary and sufficient conditions for simple majority decision.” Econometrica: Journal of the Econometric Society 20(4): 680–684. Mukhopadhaya, K. (2003). “Jury size and the free rider problem.” Journal of Law, Eco­ nomics and Organization 19(1): 24–44. Nitzan, S. and J. Paroush (1980). “Investment in human capital and social self protection under uncertainty.” International Economic Review 21(3): 547–557. Nitzan, S. and J. Paroush (1981). “The characterization of decisive weighted majority rules.” Economics Letters 7(2): 119–123. Nitzan, S. and J. Paroush (1982). “Optimal decision rules in uncertain dichotomous choice situations.” International Economic Review 23(2): 289–297. Nitzan, S. and J. Paroush (1984a). “Partial information on decisional skills and the desir­ ability of the expert rule in uncertain dichotomous choice situations.” Theory and Deci­ sion 17(3): 275–286. Nitzan, S. and J. Paroush (1984b). “A general theorem and eight corollaries in search of a correct decision.” Theory and Decision 17(3): 211–220. Nitzan, S. and J. Paroush (1984c). “Potential variability of decisional skills in uncertain di­ chotomous choice situations.” Mathematical Social Sciences 8(3): 217–227. Nitzan, S. and J. Paroush (1984d). “Are qualified majority rules special?” Public Choice 42(3): 257–272. Nitzan, S. and J. Paroush (1984e). “The significance of independent decisions in uncertain dichotomous choice situations.” Theory and Decision 17(1): 47–60. Nitzan, S. and J. Paroush (1985). Collective Decision Making: An Economic Outlook. Cam­ bridge: Cambridge University Press. Page 22 of 26

Collective Decision-Making and Jury Theorems Nitzan, S., and U. Procaccia (1986). “Optimal voting procedures for profit maximizing firms.” Public Choice 51(2): 191–208. Nurmi, H. (2002). Voting Procedures under Uncertainty. Berlin, Heidelberg, and New York: Springer-Verlag. Owen, G., B. Grofman, and S. Feld (1989). “Proving a distribution free generalization of the Condorcet jury theorem.” Mathematical Social Sciences 17(1): 1–16. Paroush, J. (1985). “Notes on partnerships in the services sector.” Journal of Economic Be­ havior and Organization 6(1): 79–87. Paroush, J. (1990). “Multi-choice problems and the essential order among decision rules.” Economics Letters 32(2): 121–125. Paroush, J. (1997). “Order relations among efficient decision rules.” Theory and Decision 43(3): 209–218. Paroush, J. (1998). “Stay away from fair coins: A Condorcet jury theorem.” Social Choice and Welfare 15(1): 15–20. Paroush, J. and D. Karotkin (1989). “Robustness of optimal majority rules over teams with changing size.” Social Choice and Welfare 6(2): 127–138. Peleg, B. and S. Zamir (2012). “Extending the Condorcet Jury Theorem to a general de­ pendent jury.” Social Choice and Welfare 39(1): 92–125. Quesada, A. (2010). “Two axioms for the majority rule.” Economics Bulletin 30(4): 3033– 3037. Rae, D. W. (1969). “Decision-rules and individual values in constitutional choice.” Ameri­ can Political Science Review 63(1): 40–56. (p. 514)

Rawls, J. (1971). A Theory of Justice. Cambridge, MA: Harvard University Press.

Romeijn, J. W. and D. Atkinson (2011). “Learning jury competence: a generalized Con­ dorcet Jury Theorem.” Politics, Philosophy and Economics 10(3): 237–262. Sah, R. K. (1990). “An explicit closed-form formula for profit-maximizing k-out-of-n sys­ tems subject to two kinds of failures.” Microelectronics and Reliability 30(6): 1123–1130. Sah, R. K. (1991). “Fallibility in human organizations and political systems.” Journal of Economic Perspectives 5(2): 67–88. Sah, R. K. and J. Stiglitz (1985). “Human fallibility in economic organizations.” American Economic Review 75(2): 292–297. Sah, R. K. and J. Stiglitz (1986). “The architecture of economic systems: Hierarchies and polyarchies.” American Economic Review 76(4): 716–727. Page 23 of 26

Collective Decision-Making and Jury Theorems Sah, R. K. and J. Stiglitz (1988). “Committees, hierarchies and polyarchies.” Economic Journal 98(391): 451–470. Shapley, L. and B. Grofman (1984). “Optimizing group judgmental accuracy in the pres­ ence of interdependence.” Public Choice 43(3): 329–343. Straffin, Jr., P. D. (1977). “Majority rule and general decision rules.” Theory and Decision 8(4): 351–360. Surowiecki, J. (2004). The Wisdom of Crowds. New York: Anchor Books. Wit, J. (1998). “Rational choice and the Condorcet jury theorem.” Games and Economic Behavior 22(2): 364–376. Young, P. (1995). “Optimal Voting Rules.” Journal of Economic Perspectives 9(1): 51–64.

Further Reading Batchelder, W. H. and A. K. Romney (1986). “The statistical analysis of a general Con­ dorcet model for dichotomous choice situations,” in B. Grofman and G. Owen, eds., Infor­ mation Pooling and Group Decision Making, 103–112. Greenwich, CT: JAI Press. Ben-Yashar, R. and S. Kraus (2002). “Optimal collective dichotomous choice under quota constraints.” Economic Theory 19(4): 839–852. Ben-Yashar, R., S. Kraus, and S. Khuller (2001). “Optimal collective dichotomous choice under partial order constraints.” Mathematical Social Science 41(3): 349–364. Berend, D. and Y. Chernyavsky (2008). “Effectiveness of weighted majority rules with ran­ dom decision power distribution.” Journal of Public Economic Theory 10(3): 423–439. Berend, D. and L. Sapir (2005). “Monotonicity in Condorcet Jury Theorems.” Social Choice and Welfare 24(1): 83–92. Berg, S. (1994). “Evaluation of some weighted majority decision rules under a dependen­ cy model.” Mathematical Social Sciences 28(2): 71–83. Berg, S. and J. Paroush (1998). “Collective decision making in hierarchies.” Mathematical Social Sciences 35(3): 233–244. Boland, P. J., F. Proschan, and Y. L. Tong (1989). “Modelling dependence in simple and in­ direct majority systems.” Journal of Applied Probability 26(1): 81–88. Catalani, M. S. and G. F. Clerica (1995). Decision Making Structures. Heidelberg: PhysicaVerlag. Dari-Mattiacci, G., E. S. Hendriks, and M. Havinga (2011). “A generalized jury theorem.” Amsterdam Center for Law and Economics Working Paper: 2011–2012.

Page 24 of 26

Collective Decision-Making and Jury Theorems Gradstein, M. (1986). “Necessary and sufficient conditions for the optimality of the sim­ ple majority and restricted majority decision rules.” Theory and Decision 21(2): 181–187. Gradstein, M. and S. Nitzan (1986). “Performance evaluation of some special classes of weighted majority rules.” Mathematical Social Sciences 12(1): 31–46. (p. 515)

Grofman, B. (1978). “Judgmental competence of individuals and groups in a dichotomous choice situation: Is a majority of heads better than one?” Journal of Mathematical Sociolo­ gy 6(1): 47–60. Heiner, R. A. (1988). “Imperfect decisions in organizations.” Journal of Economic Behav­ ior and Organization 9(1): 25–44. Kaniovski, S. and A. Zaigraev (2011). “Optimal jury design for homogeneous juries with correlated votes.” Theory and Decision 71(4): 439–459. Karotkin, D. (1992). “Optimality among restricted majority decision rules.” Social Choice and Welfare 9(2): 159–165. Karotkin, D. (1993). “Inferiority of restricted majority decision rules.” Public Choice 77(2): 249–258. Karotkin, D. (1996). “Justification of the simple majority and chairman rules.” Social Choice and Welfare 13(4): 479–486. Karotkin, D. and S. Nitzan (1995a). “Two remarks on the effect of increased equalitarian­ ism in decisional skills on the number of individuals that maximizes group judgmental competence.” Public Choice 85(3–4): 307–311. Karotkin, D. and S. Nitzan (1995b). “The effect of expansions and substitutions on group decision making.” Mathematical Social Sciences 30(3): 263–271. Karotkin, D. and S. Nitzan (1996a). “A note on restricted majority rules: Invariance to rule selection and outcome distinctiveness.” Social Choice and Welfare 13(3): 269–274. Karotkin, D. and S. Nitzan (1996b). “On two properties of the marginal contribution of in­ dividual decisional skills.” Mathematical Social Sciences 34(1): 29–36. Koh, W. T. H. (1992a). “Variable evaluation costs and the design of fallible hierarchies and polyarchies.” Economics Letters 38(3): 313–318. Koh, W. T. H. (1992b). “Human fallibility and sequential decision making: Hierarchy ver­ sus polyarchy.” Journal of Economic Behavior and Organization 18(3): 317–345. Koh, W. T. H.. (1993). “First mover advantage and organizational structure.” Economics Letters 43(1): 47–52. Koh, W. T. H. (1994a). “Making decision in committees. A human fallibility approach.” Journal of Economic Behavior and Organization 23(2): 195–214. Page 25 of 26

Collective Decision-Making and Jury Theorems Koh, W. T. H. (1994b). “Fallibility and sequential decision making.” Journal of Institutional and Theoretical Economics 150(2): 362–374. Koh, W. T. H. (2005). “Optimal sequential decision architectures and the robustness of hi­ erarchies and polyarchies.” Social Choice and Welfare 24(3): 397–411. Lam, L. and C. Y. Suen (1996). “Majority vote of even and odd experts in a polychotomous choice situation.” Theory and Decision 41(1): 13–36. Lindner, I. (2008). “A generalization of Condorcet’s Jury Theorem to weighted voting games with many small voters.” Economic Theory 35(3): 607–611. Meyerson, R. (1998). “Extended Poisson games and the Condorcet jury theorem.” Games and Economic Behavior 25(1): 111–131. Miller, N. (1986). “Information, electorates, and democracy: Some extensions and inter­ pretation of the Condorcet jury theorem,” in B. Grofman and G. Owen, eds., Information Pooling and Group Decision Making, 173–194. Greenwich, CT: JAI Press. Nitzan, S. and J. Paroush (1983). “Small panels of experts in uncertain dichotomous choice situations.” Decision Sciences 14(3): 314–325. Paroush, J. (1993). “A simultaneous determination of size and quality of teams of decision makers.” Economic Notes 22: 123–127. (p. 516)

Pete, A., K. R. Pattipati, and D. L. Kleinman (1993). “Optimal team and individual rules in uncertain dichotomous situations.” Public Choice 75(3): 205–230. Tang, Z. B., K. R. Pattipati, and D. L. Kleinman (1991). “An algorithm for determining the decision thresholds in a distributed detection problem.” IEEE Transactions on Systems, Man and Cybernetics 21(1): 231–237. Young, P. (1988). “Condorcet’s Theory of Voting.” American Political Science Review 82(4): 1231–1244. Young, P. (1996). “Group choice and individual judgment,” in Dennis Mueller, ed., Per­ spectives on Public Choice, 181–200. New York: Cambridge University Press.

Shmuel Nitzan

Shmuel Nitzan is the Sir Isaac Wolfson Chair in Economics at Bar Ilan University Jacob Paroush

Jacob Paroush is Full Professor of Economics at Bar Ilan University

Page 26 of 26

Name Index

Name Index   The Oxford Handbook of Law and Economics: Volume 1: Methodology and Concepts Edited by Francesco Parisi Print Publication Date: Apr 2017 Subject: Economics and Finance Online Publication Date: May 2017

(p. 517)

Name Index

Abrams, D. 38 Acemoglu, D. 5 Alchian, A. 163 Alt, J. 214 Anderson, T. 163 Angelova, V. 93 Arrow, K. 184–6, 255, 257, 373–4 Atanasov, V. 52 Austen-Smith, D. 503 Babcock, L. 108 Baddeley, M. 88 Baharad, E. 504 Baker, E. 370 Baker, M. 492 Baldus, D. 132 Balganesh, S. 117 Banabou, R. 468 Banks, J. 503 Bar-Gill, O. 173, 451 Barnett, R. 275–6 Baron, J. 108, 112 Bartels, D. 111 Baxter, W. 1 Beard, C. 215 Bebchuk, L. 3 Beccaria, C. 11 Becker, G. 1–5, 11, 13, 17, 78, 296, 455 Ben-Shahar, O. 451 Ben-Yashar, R. 498, 499, 501, 503 Bentham, J. 1, 11 Berend, D. 495 Page 1 of 10

Name Index Berg, S. 502 Bibas, S. 62–4 Bicchieri, C. 466 Bilz, K. 142 Black, B. 52 Black, D. 492 Blankart, C. 213 Blomberg, S. 214 Blount, J. 216 Blume, L. 207, 209, 212, 215 Boadway, R. 390 Boardman, A. 377 Bork, R. 304 Bornstein, B. 110 Braithwaite, J. 403 Braman, D. 118 Braucher, J. 120 Bregant, J. 119 Brown, J. 15 Buccafusco, C. 116–17 Buchanan, J. 181, 203–5, 248, 250, 251, 252, 253, 254, 255, 256, 257, 258, 259, 263, 264 Burgess, D. 373 Burns, Z. 111, 117 Calabresi, G. 1, 11, 20, 22, 78 Camerer, C. 94 Carbonara, E. 168, 169, 173, 469, 475 Caruso, E. 110, 111 Cecil, J. 54 Chadee, D. 132 Chalfin, A. 49–50 Charness, G. 88 Cheibub, J. 210 Clay, K. 163, 167 Coase, R. 1, 2, 19, 20, 21, 31, 78, 268, 295–6, 362 Cohen, D. 120 Cole, D. 373, 377 Coleman, J. 368–70 Congleton, R. 216 Cooter, R. 15, 16, 17, 165, 167, 168, 170, 466 Cosgel, M. 166 Costa-Font, J. 474 Croley, S. 185 (p. 518) Dalton, R. 214–15 Dari-Mattiacci, G. 449 Darley, J. 142–3, 144, 410, 411 Davis, J. 128 De Condorcet,N. 492–5, 499, 505 De Geest, G. 346 Page 2 of 10

Name Index Deaton, A. 32, 38 Dekel, E. 167 Dempster, A. 504 Demsetz, H. 1, 162–4, 271–2 Djik, F. van 87–8 Dolan, P. 396–7 Duverger, M. 206 Earley, P. 116 Earls 413 Eberhardt, J. 132 Eckardt, M. 171 Ehrlich, I. 166 Eichenberger, R. 172 Eigen, Z. 114 Elkins, Z. 210, 216 Ellickson, R. 163, 259, 474–5 Epstein, R. 274 Evans, W. 49 Feddersen, T. 503, 506 Feld, L. 210, 212, 214, 215 Feld, S. 504 Feldman, Y. 114 Ferejohn, J. 216 Fernadez, P. 166 Finnis, J. 294 Fischhoff, B. 105 Fletcher, D. 96 Fon, V. 165 Frederick, S. 372 Frey, B. 172, 473, 474 Friedman, B. 55 Friedman, D. 65–6 Funk, P. 214 Galle, B. 460 Gamage, D. 335 Garoupa, N. 166 Gathmann, C. 214 Gelbach, J. 53–4, 55 Gennaioli, N. 166 Georgakopoulos, N. 166, 171 Gerring, J. 209 Gigerenzer, G. 96 Gilbert, D. 110 Ginsburg, T. 210, 216 Glaeser, E. 71 Goeree, J. 88 Gollier, C. 373 Goodman, J. 142 Page 3 of 10

Name Index Gradstein, M. 499 Grady, M. 15 Graham, J. 377 Greenstone, M. 52 Grofman, B. 496, 504, 505 Grunwald, B. 49 Guarnaschelli, S. 88 Gutmann, J. 210 Haddock, D. 271 Haidt, J. 117 Hamilton, V. 407 Harper, D. 283 Harrod, R. 361 Hartmann, S. 507 Hastie, R. 110 Hatzis, A. 166 Hayek, F. 168, 173, 211, 250, 264, 269, 272–3, 282 Hayo, B. 216 Heckman, J. 38, 44 Hess, G. 214 Hicks, J. 289, 358, 361 Hill, P. 163 Hirschman, A. 172 Hoffman, D. 112, 115, 118 Hoffman, E. 90, 91 Hoffrage, U. 96 Holmes, O. 10, 294, 464 Holsinger, S. 68–9 Holt, C. 89 Hubbard, W. 54 Hume, D. 248–9, 264 Huo, Y. 473 Innes, R. 435 Iverson, T. 207–8 (p. 519) Jackson, J. 415 Jegan, R. 473 Jofre-Bonet, M. 474 Jolls, C. 72 Just, R. 369 Kahan, D. 118, 475 Kahneman, D. 63, 382–3 Kaldor, N. 289, 360–1 Kalven, H. 127 Kanfer, R. 116 Kant, I. 291 Kaplow, L. 12, 293–4, 346, 366 Karotkin, D. 496, 497, 498, 504 Kelling, G. 402 Page 4 of 10

Name Index Kelman, H. 407, 408 Kennedy, R. 394 Kevelson, R. 305–6 Khalil, E. 166–7, 173 Khan, B. 171 Kiesling, L. 271 Kirchgässner, G. 212, 214 Kirzner, I. 273, 468 Klein, B. 85 Klick, J. 49, 51, 52 Koh, W. 501 Kornhauser, L. 92–3, 165, 170, 507 Kraus, J. 168 Krecké, E. 282–3 Kronman, A. 29 Lachmann, L. 273, 275, 277 Ladha, K. 502 Laird, N. 504 Landes, W. 1, 29 Laplace, P. 493–4 Lassen, D. 214 Lawless, R. 120 Layard, R. 396 Lazear, E. 460 Leamer, E. 32 Lederman, D. 209 Lee, L. 170 Leeson, P. 271 Leshem, S. 444 Lessig, L. 471 Levitt, S. 48 Lichtenstein, S. 105 Ligüerre, C. 166 Lind, E. 116 List, C. 507 Lizzeri, A. 207 Loayza, N. 209 Loewenstein, G. 94, 108 Loftus, E. 142 Luppi, B. 505 Lutz, D. 216 Maass, A. 356–7 McCloskey, D. 367 MacCoun, R. 116 McCrary, J. 48, 49–50 MacDonald, J. 49 McGuire, R. 215 McKelvey, R. 88 Page 5 of 10

Name Index McLennan, A. 503 Makkai, T. 403 Malloy, R. 306 Mandel, G.N. 94 Mankiw, N. 345 Manne, H. 1, 52 Manta, I. 117 Marciano, A. 166–7, 173 Martin, A. 283, 284 Martin, N. 271 Matsusaka, J. 214 Mattei, U. 173 Melamed, A. 20, 22 Melton, J. 216 Menger, C. 269, 282 Metcalfe, R. 396–7 Miceli, T. 16, 17, 165, 166 Milchtaich, I. 503 Milgram, S. 105–6 Miller, M. 110 Mirrlees, J. 322, 347 Mises, L.v. 269–70 Moore, M. 402 Morriss, A. 166 Mukhopadhaya, K. 499 Müller, J. 207 Nitzan, S. 494, 496, 497, 498, 499, 501, 502, 504 Nurmi, H. 504 (p. 520) Oates, W. 172 Oberholzer-Gee, F. 473, 474 Ohsfeldt, R. 215 Olson, M. 248 Ordeshook, P. 203 Osborne, E. 166 Oswald, A. 389 Owens, E. 49 Oyer, P. 52 Palfrey, T. 88 Palmon, O. 373 Parisi, F. 165, 169, 173, 475, 505 Parkinson, S. 88 Paroush, J. 494, 495, 496, 497, 498, 499, 502, 504 Pasotti, P. 168 Payne, J. 110 Peirce, C. 305 Persico, N. 207 Persson, T. 206, 207, 208 Pesendorfer, W. 503, 506 Page 6 of 10

Name Index Peters, E. 119 Pigozzi, G. 507 Piquero, A. 403 Pistor, K. 171 Pitchford, R. 435 Polinsky, A. 17 Ponzetto, G. 166 Porter, T. 377 Posner, E. 478 Posner, R. 1–5, 12, 13, 17, 166, 227, 246, 259–63, 268, 281, 291–2, 300, 304, 305, 364, 366 Powdthavee, N. 389 Priest, G. 85, 165, 283 Pulaski, C. 132 Rajagopalan, S. 283–4 Ramsey, F. 372 Rasch, B. 216 Raudenbush 413 Rawls, J. 205 Ritov, I. 108 Rizzo, M. 64–5, 274, 280–1 Robbennolt, J. 109 Robbins, L. 361 Robinson 410, 411 Robinson, J. 216 Robinson, P. 142–3 Rodden, J. 211, 212 Roland, G. 208 Rowell, A. 119 Rubin, D. 504 Rubin, P. 165, 166, 283 Rung, L. 110 Saez, E. 345 Sager, L. 507 Sah, R. 500, 501 Sampson 413 Sanchirico, C. 345, 346, 350 Savioz, M. 215 Schaltegger, C. 212, 214 Schargrodsky, E. 49 Schkade, D. 110 Schmitz, A. 369 Schnellenbach, J. 214 Schotter, A. 92–3 Schroeder, C. 367 Schwab, R. 172 Schwab, S. 91 Schwartz, A. 67, 70 Scott 365 Page 7 of 10

Name Index Segerson, K. 16 Severance, L. 142 Shapiro, S. 367 Shapley, L. 496 Shavell, S. 2, 12, 17, 293–4, 346, 366–7 Shen, F. 143 Shleifer, A. 5, 166 Simon, H. 60, 62 Simpson 403 Sitkoff, R. 51, 52 Slain, A. 111 Slovic, P. 105 Smith, A. 89, 258, 270, 310 Smith, J. 38 Smith, V. 78 Snyder, C. 435 Soares, R. 209 Solow, J. 96 (p. 521) Sonnemans, J. 87–8 Sood, A. 144 Soskice, D. 207–8 Spiller, P. 229 Spitzer, M. 90, 91, 229 Sprenger, J. 507 Sprigman, C. 116–17 Stegarescu, D. 212 Stigler, G. 52, 97, 455 Stiglitz, J. 395–6, 500, 501 Stolle, D. 111 Storr, V. 271 Stringham, E. 282 Sugden, R. 467 Sunshine, J. 409, 473 Sunstein, C. 72, 74, 109, 469 Sutter, M. 88 Sykes, A. 16 Tabarrok, A. 49 Tabbach, A. 444, 455 Tabellini, G. 206, 207, 208 Tanzi, V. 211 Teichman, D. 114 Tella, R. Di 49 Tetlock, P. 117 Thacker, S. 209 Thaler, R. 72 Thomas, D. 284 Ticchi, D. 216 Tiebout, C. 172, 190, 211 Page 8 of 10

Name Index Tirole, J. 468 Titmuss, R. 473 Torvik, R. 216 Treisman, D. 209 Tullock, G. 181, 204–5, 252, 254, 259 Turnbull, C. 466 Tversky, A. 63 Tyler, T. 116, 403, 404, 409, 410, 414, 415, 473 Ulen, T. 17 Umbeck, J. 163 van den Berg, B. 389 Van den Bos, K. 116 Varian, H. 23 Vaubel, R. 213 Vaughn, K. 272 Veenhoven, R. 394–7 Verchik, R. 367 Vindigni, A. 216 Vissing-Jorgensen, A. 52 Voigt, S. 205, 207, 210, 212, 216 Wagner, R. 283 Wangenheim, G. v. 169, 170, 475 Weber, M. 94, 407, 409 Weerapana, A. 214 Weinstein, N. 68 Weinzierl, M. 345 Weisbach, D. 392–3 Weitzman, M. 373, 374, 375 Whitman, D. 64–5, 166, 170, 283 Wicksell, K. 204 Wilde, L. 67, 70 Wilkinson-Ryan, T. 112, 113, 114, 115, 117 Wilson, T. 110 Wit, J. 503 Wittman, D. 93, 453–4 Woeckener, B. 170 Wolf, C. 207 Wright, G. 163 Wu, D. 94 Yariv, L. 88 Yen, S. 474 Yoon, A. 38 Young, P. 465–6 Zasu, Y. 477–8 Zeckhauser, R. 373 Zeisel, H. 127 Zerbe, R. 365, 367, 369, 373 Zipursky, B. 93 Page 9 of 10

Name Index Zywicki, T. 272, 274, 275, 282, 283

Page 10 of 10

(p. 522)

Subject Index

Subject Index   The Oxford Handbook of Law and Economics: Volume 1: Methodology and Concepts Edited by Francesco Parisi Print Publication Date: Apr 2017 Subject: Economics and Finance Online Publication Date: May 2017

(p. 523)

Subject Index

accident avoidance 13–14 accident model 24 act utilitarianism 291 ad coelum rule 278 affective forecasting errors 386 agencies, delegation to 194 agency behaviour 194–5 allocative efficiency 300 almost expert rule 496 almost majority rule 496 American Pragmatism 305 anchoring effects, in adjudication 87–8 apologies, and culpability 108–10 arbitrary/capricious decision-making challenges 226 Army Corps of Engineers 363 Arrow’s Theorem 183–4, 186, 500 artificial commodity 78 Ashcroft v. Iqbal 53–5 asymmetric information hypothesis 86–7 Austrian law 268–85 analysis method 269 coherence in law 280–2 common law incentives 281 coordination in society 276–9 decentralized knowledge 272–6 dispersed knowledge 273 distinctiveness of 284 emergence of legal rules 269 emergence of money 269 endogenous legal rules 280 entrepreneurship 282–5 Page 1 of 23

Subject Index knowledge problems 273, 279 legal institutions 269–72, 275 and legal order 279–82, 285 processes 269 rules 273–6 spontaneous processes 271–2 time and ignorance, economics of 272–6 and trade development 270–1 automobile safety/fatalities 71–4 availability heuristic debiasing 68 in risk estimation 66–8 avoidable consequences doctrine 428 bargaining experiments 83 Batson v. Kentucky 131 Bayes’ Rule 96–7 BCA (benefit-cost analysis) see CBA (cost-benefit analysis) behavioural economics 11 and law 60–74 and nonomniscience 61–74 belief elicitation 89 Bell Atlantic Corp. v. Twombly 53–5 benefit versus expected loss liability 93 bicameral legislature 229 bicameralism 223 bilateral care model 14, 15, 16 bounded rationality and law 60–74 and moral philosophy 295 and nonomniscience see nonomniscience Bove v. Donner-Hanna Coke Corporation 422, 425–6, 432 BPS (Becker-Polinsky-Shavell) enforcement model 17–18 breach of contract 112–14 Bush, President George W. 376, 377 calculus of choice 311–12 Calculus of Consent 204, 248, 254, 258, 264 carrots and sticks 437–63 activity-level effects 437, 453–4 agents’ compliance/violation 458 (p. 524) agents’ self-reporting 437, 454–5 annullable 455–8 behavior reinforcement 460 behavioral biases 438 behavioural economics 460–1 certain monitoring with no errors 439 combined 458 compliance and carrots 441–5 coordinated monitoring 447–8 Page 2 of 23

Subject Index and corruption 458 distributional effects 437, 451–5 enforcement errors 440 enforcement with information problems 443–4 enforcement level 438 equilibrium compliance 443 incentive equivalence result 437, 438–40 information effects 443–5 intra-group financed carrots 458–9 intra-group redistributed sticks 458–9 liability 438 majoritarian criterion 443 multiplication effect of sticks 449–51 optimal choice between 438, 461–2 overcompensating carrots 451–2 political risks 438, 460 population effect 448–9 pre-compensated 455 as prima facie equivalent 437 principals’ opportunistic behaviour 437, 454 probabilistic monitoring 439–40 reversible rewards 459 risk allocation/aversion 437, 444–5 saturation effect 438 strict liability 459–60 transactions costs 437, 441–4 undercompensating sticks 451–2 violation and sticks 441–5 wealth of agent, and stick 446–7 wealth of principal, and carrot 446–7 wealth/budget constraints (agents and principals) 437, 445–51 causal inference, in experimental economics 84 causality, of self-serving bias 86 causation 19 CBA (cost-benefit analysis) Baker’s critique 370 comparable units 360 constant rates 371–3 consumer and producer surpluses 358 critiques of law and economics 304 CV (compensating variation) 358 definition 357–60 discount factor 373–4 discount rate 370–5 distributionally weighted 390–3 early use of 355–7 economic origins of 360–2 EKH (Expanded Kaldor-Hicks) 364–5, 366–7, 373 Page 3 of 23

Subject Index general equilibrium analysis 357–8 history of use 363–4 issues with 366–75 KH (Kaldor-Hicks) 358–9, 360–2, 364–5, 368, 373 in legal decision-making 355–77 legal landscape 355–7 limitations of 304–5 measurement in gains and losses 359 mentions in legal cases 356 and moral sentiments 366–8 Nash equilibrium behavior 503 partial equilibrium analysis 357–8 PCT (potential compensation test) 358, 362, 365, 368, 370 policing 50 PPF (project production possibility frontier) 368 present values 360 principles and standards 375–7 project rule 365 RIA (Regulatory Impact Analyses) 363 Scitovsky reversals 365, 368–70 SOC (Social Opportunity Cost of Capital) 371–3 SPC (Shadow Price of Capital) 371 STP (Social rate of Time Preference) 371–2 and SWB 381–2, 384 time declining rates 373–5 (p. 525) TP (time preference) 371 WTA (willingness to accept) 358–9, 365, 367 WTP (willingness to pay) 358–9, 362, 365 centralization, in federal states 213 certainty of the law 275 Chevron doctrine 224, 226, 236 Chicago school 12, 305 civil procedure, empirical law 53–5 Clinton, President Bill 363–4, 385 clubs theory 172 Coase Theorem 19–20, 90–2, 173 Coleman’s preference reversal example 369–70 collective decision-making almost expert rule 496 almost majority rule 496 applications 507–8 asymmetric alternatives 500–1 diverse preferences 499–500 doctrinal paradox example 506–7 endogenous competencies 497–9 ER (expert rule) 496, 497 heterogenous competencies 494–5 hung-jury problem 505–6 Page 4 of 23

Subject Index information cascades 505 and jury theorems 492–508 latent competencies 504–5 multi-criteria decisions 506–7 optimal decision rules 495–6 optimal human capital investment 498 optimal restricted majority rules 496 order of decision rules 497 QMR (qualified majority rule) 500–1 SMR (simple majority rule) 496, 497, 499, 502, 504, 505 strategic behavior 502–4 strategic voting 503–4 tie-breaking chairman rule 496 violation of independence 501–2 WMR (weighted majority rule) 496, 499, 501–2, 503 WQMR (weighted qualified majority rule) 501 see also juries common law development 283 common law incentives 281 Community Oriented Policing Services program 49 compensation 15, 23 concrete information 68 Condorcet Jury Theorem (CJT) 492–4, 495, 499–500, 502, 503, 505 Condorcet’s paradox 185 Conley v.Gibson 53 consequentialism 290–1 consideration instrument 226 Consolidated Rock Products Co. v. The City of Los Angeles 423 constitutional economics amendment culture 217 constitutional assemblies 216 constitutional interpretation by courts 218 corruption 211–12, 214 datasets/indicators 217 decision-making costs 204 democracy and productivity 215 efficiency 205 electoral rules 206–8 electoral systems 206–8 endogenizing constitutional rules 215–17 external costs 204 federalism 211–13 fiscal policy 214 forms of government 208–10 future development 217–18 governance quality 214–15 horizontal separation of powers 208–10 interdepence costs 204 Page 5 of 23

Subject Index and judiciary 210 and law 202–18 methodological foundations 203–5 MR (majority rule) systems 206–8, 217 normative constitutional economics 203–6 omitted variables bias 218 PCE (positive constitutional economics) 206–17 PR (proportional representation) systems 206–8, 217 presidential vs. parliamentary systems 208–10 and principal-agent framework 214 representative vs. direct democracy 213–15 vertical separation of powers 211–13 (p. 526) constitutional rules 202–18 consumer nonomniscience 67 contract law 16 and social psychology 147–8 contract performance/breach instrument 226 contract transaction costs 428–9 contractarian perspectives choice individualism 256, 261 choice of rules 251 consent vs. legitimization by consent 262–4 and constitutional economics 265 constitutional political economy 250–2 constitutional vs. welfare economics 254–6 constraints choice 251 contractarian conjectural history 248 contractarian theory 247 contractarianism issues 247–50 contractual relations and market rules 251 CPE (constitutional political economy) research 246, 248, 250–60, 262, 264 economics of rules 250 efficiency 256 exogenous standards 252 idealized consent 264 in law and economics 246–65 level of actions 250 methodological individualism 247 new institutional economics 250 normative individualism 247 normative standards 252 order of actions 250 order of rules 250 original/ongoing agreement distinction 249–50 outcome evaluation vs. constitutional reform 257–9 politics as exchange 252–4 preference aggregation 257 public/self interest 258 Page 6 of 23

Subject Index reconciliation 262–4 social welfare function 255, 256 sub-constitutional/constitutional level distinction 251 unanimity principle 252–4 utility individualism 255, 256, 260, 261 wealth maximization and consent 259–62 contracts breach of 112–14 exculpatory clauses 111–12 moral psychology of 111–15 and promissory obligation 111–15 and rational wealth maximization 147 willingness to breach 112 contractual promise 16 contractual uncertainty 114 convergence to equilibrium 93 convergent evolution 17 coordination, in society 276–9 corporate law and economics 50–3 cost-benefit analysis see CBA courts and asymmetric information 435 constitutional interpretation 218 in democracies 195 CPE (constitutional political economy) see contractarian perspectives credibility revolution 30, 47–50 critiques of law and economics 300–13 calculus of choice 311–12 CBA (cost-benefit analysis) 304 efficiency in economic analysis 303 fact-value distinction 312 interpretative critique of economic analysis of law 305–12 invariance principle 310–11 methodological critiques 303–5 methodological individualism 310 normative critiques 300–3 normative value of wealth maximization 300–2, 312 point of viewlessness 309–10 systems boundaries in economic analysis 304 utility and income distribution 304 wealth/welfare maximization and efficiency 302–3 culpability 142 cultural cognition 117–19 curse of knowledge 93 CV (compensating variation) 358 Daley v. Irwin 427–8 damages assesment 111 (p. 527) damages instrument 226 Page 7 of 23

Subject Index damages liability 16, 109–11 data restriction 44 Daubert v. Merrell Dow Pharmaceuticals Inc. 95 Davies v. Mann 429 debiasing and availability heuristic 68 educational loans 68–9 through experiments 96–7 through law 61, 69–72 decentralized knowledge 272–6 decision utility 382 decomposability of rules system 274 deontology 290–1 deregulation 305 deterrence, and compensation 15 DHS (Department of Homeland Security) 49 dice lottery 89 difficulty-of-ratification index 216 discrepancy-reducing sentencing guidelines 64 dispersed knowledge 273 dispute resolution, experimental psychology 115–16 doctrinal paradox 506–7 driver’s licenses 71–4 DRM (day reconstruction method) 383, 397 due care standard 14 early remorse 110 Economic Analysis of Law 281 economic models of law 9–25 assumptions 10 behavioural economics 11 Coase Theorem 19–20 criminal law 17–18, 22–3 economic analysis of law 12, 13–18 economics of crime 17–18, 47–50 and empiricism 11 general transaction structure 21–2 goals/objectives of law 12 and incentives 9, 24 law and economics 18–23 law and markets 12, 13–18 methodology 9 model building 23–5 model of precaution 16–17 nature of models 10–11 property rules vs. liability rules 20–1 shared characteristics 11 social welfare 12 tests of 10–11 Page 8 of 23

Subject Index tort law 13–15 economic terms, in law 308 economics of rules 250 Edmonson v. Leesville Concrete Company 131 efficiency, in economic analysis 303, 308 EKH (Expanded Kaldor-Hicks) test 364–5, 366–7, 373 ELE (evolutionary law and economics) Coase Theorem 173, 174 common law 165–6 competing jurisdictions 172–5 continuity hypothesis 162 definition of evolution 161–2 discrimination 168, 170 dynamic competition 174 and economics 161–75 and efficiency 165–7 interdependent jurisdictions 171, 172–5 judge-made law 165–6 legal evolution in stable environment 164–7 microeconomic models of legal evolution 164–71 neo-institutional approaches 162–4, 175 normative perspective 168 property rights 162–4 social norms and law 167–70 technology and law 170–1 third-degree path dependance 167 transplanted law 168 Universal Darwinism 161–2 empirical law challenges and credibility 33–47 civil procedure 53–5 comparison groups 41 control function 39–42 control variables 30–1 corporate law and economics 50–3 credibility revolution 30, 47–50 difference estimators 41 and economics 29–56 external validity 32 (p. 528) function approach 39 history of 30–3 instrumental variable estimation 32, 42–4, 48 internal validity 32 linear-in-controls assumption 39 omitted variables bias 30–1, 34, 39, 48 policy changes 31 random assignment 37–8 real/natural experiments 41 Page 9 of 23

Subject Index reduced form modeling 33–7 regression discontinuity approaches 44–7 scaling up approach 45 structural modeling 33 employment discrimination 133 endowment effect 92 in intellectual property 116–17 entrepreneurship 282–4 ER (expert rule) 496, 497 ESM (experience sampling method) 382–3, 397 event lottery 89 evolutionary law and economics see ELE ex ante probability 94, 110 ex ante vs. ex post 420–35 allocating and rights/obligations 420 asymmetric information and courts 435 auctions and rights 420 avoidable consequences doctrine 428 bankruptcy 434–5 contract transaction costs 428–9 damage mitigation 427–8 damage prevention costs 430–1 first party rights 433–5 hunting 435 land use 421–6 last clear chance doctrine 429, 432 marginal cost civil liability 431–2 marginal cost liability 427, 428, 430 mitigation of damages 427 negligence and liability rules 426–7 optimal sequence determination 421–6 real estate sales 435 rescue law 432–3 secured/unsecured debt 434–5 suboptimal behavior 426–33 exculpatory clauses, contracts 111–12 excuse of performance for mistake instrument 226 Executive Order (EO) 12291 375, 376 Executive Order (EO) 12498 376 Executive Order (EO) 12866 363–4, 376 Executive Order (EO) 13258 376 Executive Order (EO) 13497 376 Executive Order (EO) 13514 376 experimental economics anchoring effects in adjudication 87–8 applications 95 assumption-based 80 at trial 94–7 Page 10 of 23

Subject Index bargaining experiments 83 causal inference 84 Coase Theorem 90–2 controlled experimental design 79–81 as evidence 94–6 and evidence admissability 95–6 experiment sessions 84 in-court institutions 87–90 induced-value theory 81–2 judges and juries 87–90 and law 78–98 and legal institutions 85–90 in legal practice 94–7 in litigation strategy 96–7 methodology of 79–84 negligence/tort liability 92–4 neuroeconomic experiments 83 out-of-court institutions 85–7 repetition and replication 84 scaled observation 82–3 settlement bargaining 85–7 settlement failure model 85 statistical inference on data 84 understanding legal doctrine 90–4 experimental psychology apologies 108–9 bankruptcy filers 120–1 cognitive styles 117 contracts and promissory obligation 111–15 cultural cognition 117–19 deception 105–6 as descriptive enterprise 105 dispute resolution 115–16 (p. 529) future of 119–21 incentive compatibility 106 individual differences 117–19 intellectual property 116–17 and law 104–21 main effects 117 methodological considerations 105–7 moral psychology in contracts 111–15 numeracy and decision-making 119 race differences 119–20 settlement biases 108–11 statistical significance testing 107 subjects 106–7 tort law biases 108–11 fact-based legal instrument 235 Page 11 of 23

Subject Index fact-value distinction 312 fair process effect 115–16 false convictions 88 false positive 97 federalism 211–13 FJC (Federal Judicial Center) 53 FOCJ (functional overlapping competing jurisdictions) 172 Frye v. United States 95 game theory, and negligence law 15 GDP (Gross Domestic Product), happiness-based alternatives to 394–7 general equilibrium effects 38 GTS (General Transaction Structure) 21–3 Hadley vs. Baxendale 16 HDI (Human Development Index) 381–2, 394 hedonic psychology 382 hindsight bias 93, 94, 110, 124 HLE (Happy Life Expectancy) 394, 397 human happiness 302–3, 386 hypothetical consent 292 IAH (Inequality-Adjusted Happiness) 396, 397 IAT (Implict Association Test) 130 idealized consent 264 imperative theory of law 466 imperfect compliance 45 inalienability rules 23 incentives see carrots and sticks induced-value theory 81–2 institutional design see mechanism design institutional self-dealing 216 intellectual property 116–17 interest group theory 184–5 interpretation theory 311 interpreting texts 306–7 interpretive critique 305–12 interpretive turn 307 interrogation see social psychology invariance principle 310–11 Iqbal (Ashcroft v. Iqbal) 53–5 judge-made law, and efficiency 165 judicial decision-making 87–90, 210, 222–42 background 224–8 balancing test 227 basic elements 224 challenges to regulation 226 and constitutional economics 210 decision costs 227–8 decision instruments 225–6 doctrinal form 226–7 Page 12 of 23

Subject Index doctrine choice 232–5 and efficiency 227 horizontal competition 225, 228–31, 242 instruments 226, 232, 235–7 internal theater of competition 237–41 law as strategy 228–41 legal doctrine/precedent 225–6 legal rules 230 panel effects 225, 238–41 rules vs. standards 226–7 theaters of competition 225, 228–37 vertical competition 225, 232–7, 242 judiciary independence 195–6, 210 motivation of 196–7 political ideologies 196–7 and precedent 197, 238 juries 87–90 and apologies/remorse 138–9 and confessions 136–7 decision-making models 127–8 doctrinal paradox example 506–7 hung juries 505–6 (p. 530) intentional harms 145 motivated reasoning 144–5 multi-criteria decisions 506–7 pre-deliberation preference 127 race issues 132 see also collective decision-making juror decision-making 126 jury awards, and time 110 jury size, and damage awards 128 jury theorems, and collective decision-making see collective decision-making Kaldor-Hicks 205, 261, 277–9, 289–90, 301, 358–9, 360–2, 364–5, 367, 368, 373 Kelo v. City of New London 146 knowledge problems 273 labor earnings tax system/taxation 323 last clear chance doctrine 429, 432 LATE (Local Average Treatment Effect) parameter 44 law certainty of 275 coherence in 280–2 common law incentives 281 economic models of see economic models of law economic terms in 308 and future of economics 1–5 and PPT (positive political theory) 181, 222–42 and precedents 282 Page 13 of 23

Subject Index and social norms 167–70 as strategy 223–4 and technology 170–1 Law as a System of Signs 306 law-based legal instrument 235 legal craft 233 legal doctrine 225–6 legal efficiency criterion 307 legal entrepreneurship 282–4 legal institutions, and public choice theory see public choice theory legal intervention, Coase Theorem 19–20 legal order, and economic activity 279–82 legal rules 346 lex mercatoria rules 167–8 liability rule 20–3, 90 life satisfaction survey 383 litigation choice, and efficiency 166 litigation strategy, experiments in 96–7 logical coherence of law 281 logical-possibility test 53 loss aversion 62–4 Lotka-Volterra model 170 LSAT scores 44–7 lump sum tax types 326 Mahlstadt v. City of Indianola 422, 426, 432, 434 mandatory disclosure 52 marginal cost civil liability 431 marginal cost liability 427, 428, 430 market-for-corporate-control hypothesis 51, 52 market/legal exchange, and property rules 20 Mechanical Turk 107 mechanism design auction theory 488 Dominance Solvability 481, 485, 489 incentive compatibility 483, 484, 485 and the law 481–90 legal evolution 483 legal mechanism examples 481–5 literature on 488–9 procurement auctions 488 Revelation Principle 481, 485–9 ring formation 483 Welfare Optimization 486 welfaristic 485–8 mens rea 142 methodological individualism 310 mitigation of damages 427 model building 23–5 Page 14 of 23

Subject Index model of precaution 16–17 moral character, and punishment 139–41 moral philosophy consequentialism 290–1 and criticism of law and economics 293–7 and defense of law and economics 291–3 deontology 290–1 and economic analysis 288–9 efficiency and justice 293–4 (p. 531) efficiency of law and economics 288–90 efficiency and morality 288–91 human behaviour modeling 294–5 hypothetical consent 292 in law and economics 288–97 and non-commercial behaviour 296 and public choice theory 296–7 secular 290 and tort law 294–5 types of 290–1 understanding/misunderstanding law 295–6 virtue ethics 290–1 and wealth maximization 290, 291–2, 293 welfare economics 293–4 moral psychology, in contracts 111–15 motivated reasoning 144–5 Mountain State College v. Holsinger 68–9 MPC (Model Penal Code) 142–3 mutual assent instrument 226 Nash equilibrium 15 negligence law 13–15, 92–4, 124, 260 negotiation 128–30 neuroeconomic experiments 83 new institutional economics 250 no liability rule 15 non-commercial behaviour, and moral philosophy 296 non-experimental randomization 38 non-infringement defense instrument 226 nonconforming land use 424 nonomniscience automobile fatalities 71–4 and availability heuristic 66–8 in behavioural law and economics 61–6 and criminal justice system 62–4 and debiasing through law 66–74 driver’s licenses 71–4 mandated disclosure in educational loans 66–71 and obesity 65–6 optimism-based 63 Page 15 of 23

Subject Index organ donation 71–4 and self-regulation 64–6 nonoptimization and criminal justice system 62–4 and obesity 61, 65–6 and self-regulation 64–5 norm entrepreneurs 468–9 Obama, President Barack 376, 385 OIRA (Office of Information and Regulatory Affairs) 375–7 OLS (ordinary least squares) estimator 35–7, 44, 51 OMB (Office of Management and Budget) 372, 375–7 omitted variables bias 30–1, 34, 39, 48, 218 opinion formation model 169 optimal criminal punishment 18 optimal human capital investment 498 optimal labor earnings taxation 338 optimal redistributional instruments arbitrage approach 341–4 arbitrage-like conditions for eclecticism 343–4 artificially separated legal system 346 behaviourally responsive attributes 345 distributing through labor earnings tax 348–9 distributional policy 349–50 distributionally motivated adjustments 341 in law and economics 321–50 literature on 322, 350 optimal taxation 322, 341–6 policy certainty in conventional framework 347 policy eclecticism 322, 333, 341–6, 347 policy uncertainty 322, 346–50 tax substitution argument see tax substitution argument taxing immutable tags 344–5 optimal restricted majority rules 496 optimal tax and policy framework 322, 347 optimism bias 62–4, 67, 70–2 organ donation 71–4 ou-of-court settlements 129 panel effects 225, 238–41 paradox of compensation 15 Pareto analysis 289–90 Parkersburg Rig and Reel Co. v. Freed Oil and Gas Co. 428 (p. 532) patent invalidity defense instrument 226 The Path of the Law 10 PCE (positive constitutional economics) 206–17 PCT (potential compensation test) 358, 362, 365 Pierson v. Post 435 Pigovian taxation 13, 327, 329 pirates, and governance 271 Page 16 of 23

Subject Index plausibility standard 53 plea bargaining 63–4 point of viewlessness 309–10 police interrogation see social psychology policing, and crime 48–50, 55 political entrepreneurship 284 Popitz’s law 213 PPF (project production possibility frontier) 368 PPT (positive political theory) 181, 222–42 see also judicial decision-making prediction theory 10 preference aggregation 257 preference induction 82 principal-agent theory of panel effects 238–41 The Problem of Social Cost 295 procedural justice 129 products liability 12 promissory obligation, and contracts 111–15 property law, and social psychology 145–7 property rights, evolutionary law 162–4 property rule 20–1, 90 public choice theory adminstrative agencies 194–5 agency behavior 194–5 constitutional design 188–9 constitutional structure 188–90, 189–90 courts 195–7 creation/implementation of legislation 191–8 decentralization 189–90 delegation to agencies 194 federalism 189–90 institutional structures 181–8 interbranch interactions 197–8 interest groups 184–5, 198 judicial independence 195–6 judicial motivation 196–7 and legal institutions 181–99 legislative structure/process 192–4 legislator motivation 191–2 legislatures 191–4 and moral philosophy 296–7 option symmetry 186–7 overview of 182–8 and political structures/processes 199 procedural rules 193 public choice and public law 198–9 re-election theory 191–2 statutory interpretation 198 Page 17 of 23

Subject Index voting and agendas 185–8 public policy, and well-being see well-being punishment and moral character 139–41 theories/goals 141–4 punitive damages 109–10 QALYs (quality-adjusted life years) 386, 388–9 QMR (qualified majority rule) 500–1 race, and criminal justice 131–2 racial bias, in legal system 130–3 random assignments 38 rapport, and negotiation 129 rational actor model 295 rational reconstruction 295 rational wealth maximization 147 RCTs (randomized controlled trials) 38 RD (regression discontinuity) 44–7, 49 Reagan, President Ronald 363, 375, 376, 384 reasonable foresight doctrine 16 reasonably prudent person liability 93 reciprocity of causation 296 reduced form econometric studies 33 regulatory takings law 16 related hindsight bias 110 rescue law 432–3 Restatement of Contracts 428 restorative justice goals 141 RIA (Regulatory Impact Analyses) 363 risk aversion 349 risk estimation, availability heuristic in 67–8 Rule 12(b)(6) motions 53–5 Rule of Four 223, 224 (p. 533) rule utilitarianism 291 rules and constraints on individuals 275 decomposability of 274–5 givenness of 280 and wealth maximization 281 Rules Enabling Act 53 Sarkozy, President Nicolas 395 satisficing option 62 scaled observation 82–3 scaling up approach 45 Scitovsky Paradox 290 Scitovsky reversals 365, 368–70 SDS (Social Decision Schemes) model 127 self-regulation 64–5 self-serving bias 86, 108 Page 18 of 23

Subject Index semiotics 305–6 sentencing guidelines 63–4 sequential accidents 15 settlement bargaining asymmetric information hypothesis 86–7 in experimental economics 85–7 settlement biases 108–11 settlement failure model 85 settlement inefficiencies 86 Skidmore doctrine 226, 236 SMR (simple majority rule) 496, 497, 499, 502, 504, 505 SOC (Social Opportunity Cost of Capital) 371–3 Social Judgement Scheme 128 social norms changing 468–70 conforming with 465–7 convention 465–6 and cooperation 473–4 coordination equilibrium 465–7, 468–9 countervailing effects 474–7 crowding out/in 473–4, 477 efficiency of 467–8 empirical expectation 466 expressive function of the law 470–3, 477 fairness of 467–8 imperative theory of law 466 internalization 465–7 and law 464–78 legal backlash 474–7 and legitimacy 473, 476 non-legal sanctions 465–7 norm entrepreneurs 468–9 path-dependency 467–8 preference change 472–3 sanctions 471–2, 477 social meaning 471–2 and state laws 470–7 sticky norm problem 475 and welfare maximization 478 zero sanction 476 social optimum 255 social psychology apologies and remorse 138–9 assumptions 124–5 blame 137–45 coherence-based reasoning 126 confession and judges/juries 136–7 contract law 147–8 Page 19 of 23

Subject Index employment discrimination 133 interrogation 133–6 interrogation, and lie detection 133–4 interrogations, and confession 134–6 juror decision-making 126 jury decision-making 127–8 and law 124–48 mental state and punishment 141–4 moral character and punishment 139–41 moral reasoning and motivation 144–5 moral reasoning and punishment 141–4 morality 137–45 negotiation 128–30 pre-deliberation preference 127 property law 145–7 punishment 137–45 race and criminal justice 131–2 substantive law 145–8 soda law (New York) 61, 65–6 SPC (Shadow Price Capital) 371 spontaneous processes 272 Spur Industries v. Del E. Webb Development Co. 425, 432 stare decisis doctrine 306 statistical inference 52 statistical significance testing 107 status-quo bias 64 (p. 534) statutary interpretation 226 statutory interpretation 198 The Stern Review on the Economics of Climate Change 372 Stiglitz Commission 395, 396, 397 STP (Social rate of Time Preference) 371 structural econometric studies 33 substantial similarity, in copyright 117 SWB (subjective well-being) 381–98 affective forecasting errors 386 agency and benefits 385 CBA (cost-benefit analysis) 384–93 cost-benefit-weighting principle 386 decision utility 382 DRM (day reconstruction method) 383, 397 ESM (experience sampling method) 382–3, 397 happiness data as prices 388–90 happiness-based alternatives to GDP 394–7 HLE (Happy Life Expectancy) 394, 397 IAH (Inequality-Adjusted Happiness) 396, 397 life satisfaction survey 383 national well-being 394 QALYs (quality-adjusted life years) 386, 388–9 Page 20 of 23

Subject Index SWB data 382–4 SWF (social welfare function) 391 U-Index 395–6 WBA (well-being analysis) 382, 384–93, 397, 398 WBU (well-being units) 386 SWF (social welfare function) 391 tags 344–5 tax substitution argument 321, 323–40, 341 behavioural response in labor earnings 330–1 differential commodity taxation 325–6 distortion counting 335 double distortion argument 334–6 empirical agnosticism 332 externality correction inclusive of legal rules 326–7 full tax substitution assumption 330 indicator goods 340 information role 332–3 informational problem 329–30 informational program 328 labor productivity 337–8 legal rules 346 lump sum tax types 326 and mechanical equivalence 333–4 missing third margin 334–6 optimal labor earnings taxation 338 Pigovian taxation 327, 329 policy X adjustments and labor earnings 334, 344 policy X, inefficient distributionally motivated adjustment to 324–5 policy X in literature 325–7 policy X revenue price of redistribution 342–3 potential misperceptions 331–40 private law rules 327 problematic descriptions of 334–6 prohibitory regulations 326–7 public goods and government programs 327 single-individual illustrations 336 tax substitution assumption 323–4, 324–31, 336–40 theoretical versatility 332 “theory of the second best” problem 328, 330 utility-equivalent lump sum scheme 325, 327, 329, 330, 331 weak separability of leisure 337–40 within income group differences 338 “theory of the second best” problem 328, 330 threshold of behaviour 14 tie-breaking chairman rule 496 time, and jury awards 110 Topping v. Oshawa Street Railway 431 tort damages 109–11, 334 Page 21 of 23

Subject Index tort law, and moral philosophy 294–5 tort law biases 108–11 tort liability 13–15, 22, 92–4 TP (time preference) 371 treatment effect 46 Truth in Lending Act 67 Twombly (Bell Atlantic Corp. v. Twombly) 53–5 (p. 535) U-Index 395–6 unanimity principle 252–4 uncertainty aversion 349 Unfunded Mandates Reform Act (1995) 36466 363–4 unilateral care model 14 United States Constitution First Amendment 229 Commerce Clause 229 and interest groups 215 and minorities 188–9 and Philadelphia Convention 215 Presentment Clause 223 Separation of Powers Clause 224 United Verde Extension Mining v. Ralston 431 unity of law 17 Universal Darwinism 161–2 U.S. v. Carroll Towing Co. 13 The Use of Knowledge in Society 273 utilitarianism 291 utility individualism 255 value-driven behavior deterrent effects 403–4 engagement goal 413–16 and law 402–16 legal authority 402–5 legal system goals 412–13 and legitimacy 406–11, 413–14 and moral values 409–11 normative alignment 415–16 organizational incentives 404 and punishment 403–4, 412 and self-regulation 410–11 target behaviors 415 value-based motivation 405–6, 411–12, 416 values 405–6 veilonomics 205 Violent Crime Control and Law Enforcement Act (1994) 49 virtue ethics 290–1 warrantless arrests 146 WBA (well-being analysis) 382, 384–93, 397, 398 WBU (well-being units) 386 Page 22 of 23

Subject Index wealth effects 292 wealth maximization 147, 259–62, 281, 290, 291–2, 293, 300–3, 308, 366 wealth of Nations 258 welfare economics 293–4 well-being, and public policy 381–98 Why People Obey the Law 412 within income group differences 338 WMR (weighted majority rule) 496, 499, 501–2, 503 WQMR (weighted qualified majority rule) 501 WTA (willingness to accept) 358–9, 365, 367 WTP (willingness to pay) 358–9, 362, 365, 381, 391

Page 23 of 23