The Oxford Handbook of Law and Economics 9780199684267, 9780199684205, 9780199684250, 9780198803737, 019968426X

Table of contents :
Cover
The Oxford Handbook of Law and Economics
Copyright
Contents
List of Figures
List of Tables
List of Contributors
Part I Private Law
1. Economics of Contract Law
2. (In)Efficient Breach of Contract
3. Economics of Tort Law
4. Estimating Pain-​and-​Suffering Damages
5. Medical Malpractice
6. Economics of Property Law
7. Commons and Anticommons
8. Economics of Intellectual Property Law
9. Trademarks and Unfair Competition
10. Law and Economics of Information
11. Open-​Access and Information Commons
12. Family and Household Economics
13. Economics of Remedies
Part II Corporate, Commercial, and Environmental Law
14. The Economic Nature of the Corporation
15. Market for Corporate Law Redux
16. Law and Economics of Agency and Partnership
17. Banking and Financial Regulation
18. Economics of Bankruptcy
19. Law and Economics of Insurance
20. Environmental Law and Economics
Author Index
Subject Index

Citation preview

 

T h e Ox f o r d H a n d b o o k o f

L AW A N D E C ON OM IC S

 

 

The Oxford Handbook of

LAW AND ECONOMICS Volume 2: Private and Commercial Law Edited by

FRANCESCO PARISI

1

 

3 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 Chapter 11 © Yochai Benkler The moral rights of the author‌have 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) 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.

 

Contents

List of Figures  List of Tables  List of Contributors 

vii ix xi

PA RT I   P R I VAT E  L AW 1. Economics of Contract Law  Douglas Baird

3

2. (In)Efficient Breach of Contract Daniel Markovits and Alan Schwartz

20

3. Economics of Tort Law  Jennifer Arlen

41

4. Estimating Pain-​and-​Suffering Damages  Ronen Avraham

96

5. Medical Malpractice  Ronen Avraham and Max M. Schanzenbach

120

6. Economics of Property Law  Henry E. Smith

148

7. Commons and Anticommons  Michael Heller

178

8. Economics of Intellectual Property Law  Robert P. Merges

200

9. Trademarks and Unfair Competition  Clarisa Long

220

10. Law and Economics of Information  Tim Wu

239

 

vi   Contents

11. Open-​Access and Information Commons  Yochai Benkler

256

12. Family and Household Economics  Amy Wax

280

13. Economics of Remedies  Ariel Porat

308

PA RT I I   C OR P OR AT E , C OM M E RC IA L , A N D E N V I RON M E N TA L  L AW 14. The Economic Nature of the Corporation  Lynn Stout

337

15. Market for Corporate Law Redux Roberta Romano

358

16. Law and Economics of Agency and Partnership  George M. Cohen

399

17. Banking and Financial Regulation  Steven L. Schwarcz

423

18. Economics of Bankruptcy  Michelle J. White

447

19. Law and Economics of Insurance  Daniel Schwarcz and Peter Siegelman

481

20. Environmental Law and Economics Michael A. Livermore and Richard L. Revesz

509

Author Index  Subject Index 

543 557

 

List of Figures



7.1 The standard solution to commons tragedy 

181



7.2 Revealing the hidden half of the ownership spectrum 

181



7.3 Ordinary use as the end point 

186



7.4 The trilogy of ownership 

186



7.5 The familiar split in ownership 

187



7.6 The new spectrum of use 

188



7.7 Goldilocks’ quest for the optimum 

188



7.8 An ownership puzzle 

189



7.9 The full spectrum of ownership 

189

7.10 The full spectrum of property, revealed 

190

7.11 Value symmetry in an anticommons and a commons 

191

7.12 Substitutes versus complements 

191

18.1 The insurance effect of bankruptcy 

464

 

 

List of Tables

11.1 Types of Open Access Commons by Governance and Provisioning Model 276 13.1 Allocating and Protecting Entitlements

310

19.1 Types of Insurance Policies with Identified Features 

499

 

 

List of Contributors

Jennifer Arlen, New York University School of Law Ronen Avraham, The University of Texas at Austin Douglas Baird, University of Chicago Law School Yochai Benkler, Harvard Law School George M. Cohen, University of Virginia Michael Heller, Columbia Law School Michael A. Livermore, University of Virginia School of Law Clarisa Long, Columbia Law School Daniel Markovits, Yale Law School Robert P. Merges, UC Berkeley School of Law Ariel Porat, Tel Aviv University Richard L. Revesz, New York University School of Law Roberta Romano, Yale Law School Max M. Schanzenbach, Northwestern Pritzker School of Law Daniel Schwarcz, University of Minnesota Law School Steven L. Schwarcz, Duke University School of Law Alan Schwartz, Yale Law School Peter Siegelman, University of Connecticut Law School Henry E. Smith, Harvard Law School Lynn Stout, Cornell Law School Amy Wax, University of Pennsylvania Law School Michelle P. White, University of California, San Diego Tim Wu, Columbia Law School

 

 

Pa rt I

P R I VAT E  L AW

 

 

Chapter 1

EC ONOM I C S OF C ONTRAC T  L AW Douglas Baird

Before law and economics entered the scene, contracts scholarship had not advanced much beyond Fuller and Perdue’s exploration of contract damages in an article in the Yale Law Journal published in 1936. By Fuller and Perdue’s account, contract law must in the first instance confront competing theories of relief—​reliance, expectation, and restitution. The choice between them required a normative assessment of the protection an innocent party deserved—​compensation equal to the full benefit of the promise, restoration to where she had been before the promise was made, or recovery of any benefits she had bestowed upon the party who had broken the promise. Contracts thinkers such as Oliver Wendell Holmes and Samuel Williston showed that common law courts had decisively favored the first approach—​expectation damages. By the middle of the twentieth century, however, the conventional wisdom had become that the middle course—​ reliance damages—​was the most sensible and most consistent with notions of corrective justice. The first generation of law-​and-​economic scholars entered this debate in the early 1970s (Posner 1972; Barton 1972). For them, notions of corrective justice and intuitions about fairness were off the mark. What mattered was that those who contemplated breaking a promise took account of the costs their conduct imposed on others. Far from being morally objectionable, keeping a promise or breaking it might be good or bad depending upon the circumstances. The role of contract law was to ensure that the incentives were aligned to ensure that breach took place only when it made economic sense. These accounts both vindicated the classical contract thinkers, such as Holmes, and showed that contract law itself possessed an inner logic. Expectation damages remain at the center of the law and economics of contract, but they are more contested. The contemporary view of expectation damages, and indeed of most other contract doctrines, starts with the idea that in a regime of free contracting, contract law provides a starting place. If parties are free to fashion their own terms, the debate should not be so much on whether expectation damages provide parties with the

 

4   DOUGLAS BAIRD proper incentives when the time comes to perform or breach, but on whether they provide a good starting place for parties to negotiate with each other and otherwise navigate the forces at work throughout the contractual relationship. This introductory essay sketches out the debate in its most general terms. The chapters that follow analyze these questions in more detail.

1.1.  The Expectation Damages Principle The common law, as a general matter, awarded disappointed litigants money damages. In the case of a breach of contract, those who broke promises were neither forced to perform nor punished. Instead, disappointed promisees were entitled to what Oliver Wendell Holmes called, “a compensatory sum” (1897). “Breachers” had to put their contracting opposites in the same position they would have been in had the promise been kept. They could not escape with less, but they did not have to pay more. Richard Posner (1972) showed that there was a very simple and straightforward way of making sense of such a rule. Making promises legally enforceable allows parties to count on the other’s performance. They can invest in the relationship, knowing that either the promise will be kept or they will be left in the same economic position as if it had. Because they will be compensated in the event of a breach, they will make investments that they would otherwise recoup only if the promise were performed. You can spend the resources needed to build the custom machine for me, confident that you will not be out of pocket if I breach and you are left with a machine that no one else can use. In a competitive market, expectation damages are an excellent way to capture all the ways, direct and indirect, in which someone relies on the existence of a contract. At the same time, the law should not give innocent parties more than expectation damages. Circumstances can change. My costs of keeping a promise may become larger than the benefits it brings to you. Keeping the promise is no longer in our joint interest. A law of contracts should focus on maximizing our joint welfare at the time of performance. It should induce parties to keep their promises when, but only when it is socially optimal that they do so. A rule that requires me to pay a compensatory sum when I breach forces me to take account of the harm my breach has on you. When I breach, you are paid an amount that makes you indifferent to my performance. I shall breach only if it makes me better off without leaving you worse off. The law of contract forces me to take into account the costs of breaking my promises just as the law of tort forces me to take account of the costs of socially undesirable conduct. In both instances, the common law rules work to induce individuals to internalize the harm that their conduct imposes on others. When Holmes asserted that the law of contracts gave the bad man an option to perform or pay a compensatory sum, he took no position on whether keeping a promise was good or bad. Holmes (1897) saw himself as a pragmatist giving an account of law as it was. Whether keeping a promise was required as a moral or ethical matter was not

 

ECONOMICS OF CONTRACT LAW    5 relevant to the task at hand. In providing his explicitly economic justification for the common law of contract, Richard Posner went a step further. In his view, the ethical notion that it is good to keep promises is wrong-​headed. The rule of expectation damages is desirable precisely because it encourages breach when it is efficient. Society as a whole is better off if people keep promises only in those situations in which it is socially efficient that they do. This effort to use economics to explain contract law has come under fierce criticism (Schiffrin 2007). The idea that breaching a promise was affirmatively desirable was provocative and perhaps deliberately so. The idea of “efficient breach” cuts strongly against the moral intuition that promises should be kept. But it is possible—​perhaps even desirable—​to talk about Posner’s contribution without engaging this debate. Quite apart from the morality of breaking promises, it was a significant advance over previous accounts of contract to show that expectation damages sensibly realign incentives, and it is a long step toward understanding the foundations of contract law. But Posner’s account of expectation damages, like all great revolutionary ideas, is too simple to be completely right. His account of expectation damages focused entirely on the incentives of the breaching party at the time of performance. It neglects the need to ensure that both parties’ incentives are properly aligned throughout the contractual relationship. Most obviously, at the same time one sets damages on the breaching party, one also needs to look at how these damages, in turn, affect the nonbreaching party. Just as the breaching party must have the right set of incentives, so must the beneficiary of the promise. We want that person to engage in the optimal amount of reliance. We want that person to invest, but not overinvest, in the contractual relationship (Shavell 1980). Imagine that a homeowner contemplates adding a driveway and a garage to her house. She values the driveway and garage at one amount if she starts building the garage early so they are both done at the same time, but at a somewhat smaller amount if the sole owner delays building the garage. Neither is worth anything without the other. Once construction of the driveway starts, however, the condition of the underlying soil will be revealed, and there is a chance that building the driveway will be unexpectedly high. Because of the chance that the driveway will prove unexpectedly costly, the homeowner might wait to build the garage. She would rather lose the benefit of having the garage and the driveway done at the same time than take the risk that she would incur most of the cost of building the garage only to discover that soil conditions were such that the project was not worth doing. If this homeowner were to enter a contract with a third party willing to build the driveway, however, she might build the garage sooner. She still captures the benefit if the other party keeps the promise, and the other party is obliged to cover the costs of the garage if she does not. Expectation damages require the third party to pay damages sufficient to pay all of the money the homeowner spends on the garage, plus whatever value she would have enjoyed from having the driveway and the garage. She enjoys the benefit of getting things sooner rather than later, but faces none of the cost if the project is abandoned. The homeowner is under no obligation to hold off the construction of the garage because of the possibility that the other party will not perform. The effect

 

6   DOUGLAS BAIRD of expectation damages is that the beneficiary of a promise treats the third party’s performance as sure-​fire. She no longer has to take into account the possibility that building the driveway might be prohibitively expensive in deciding whether to get an early start on the garage. As a general matter, expectation damages give the innocent party an incentive to overrely on promises (Shavell 1980). In a competitive market, of course, the cost of this overreliance will not fall entirely on the driveway builder. Parties negotiate in the shadow of the legal rule. We should expect that promisor to command a price for the driveway contract that takes this possibility into account. But readjusting the price does not cure the underlying distortion. The homeowner will still have an incentive to build the garage too soon, and resources will be wasted. This is the problem of “overreliance.” Of course, if the parties could anticipate this risk, they would write a contract that obliged the homeowner to start building the garage only when doing so was efficient. But the builder of the driveway may not know enough to insist on such a clause. Not all forms of overreliance can be specified in detail, and, in any event, doing so is itself costly. More to the point, to the extent that a special contracts clause is necessary, expectation damages are no longer aligning the incentives of the parties. In the worst case, the expected costs of overreliance may keep parties from entering into a contract that is in their joint interest. The risk of overreliance, however, does not suggest that Fuller and Perdue’s alternatives to expectation damages were to be preferred. Merely capping damages at the homeowner’s reliance costs does not make the problem go away. Indeed, it makes it worse (Shavell 1980). The homeowner has an incentive to build the garage early even if doing so brings her no benefit at all. Assume that the homeowner receives no benefit from building the garage before the driveway. Under reliance damages, building the garage early is still worth doing because it provides a spur to the builder of the driveway. Building the garage increases the amount of the reliance damages that must be paid. The prospect of paying higher damages makes the other party more likely to perform. The builder may be better off completing the driveway after bad soil conditions are discovered, given the costs that have been sunk in building the garage. In a reliance-​based world, the innocent party is not indifferent between breach and performance. She prefers performance, and she is willing to spend resources to induce performance, even when it is socially wasteful. We could reshape the expectation-​damages rule and give the promisee only those reliance expenditures that were reasonable (Spier and Whinston 1995). If a sole owner would delay building the garage, then the promisee should not be able to recover those expenses if she built it early. But fine-​tuning expectation damages requires the court to solve the same kind of joint maximizing problem that makes Pigovian taxes virtually impossible to implement (Coase 1960). The larger point, however, goes beyond overreliance. The concept of expectation damages focuses on only one part of a complicated puzzle. As the rest of this chapter explores, the puzzle contains many other parts. Of these, one of the most important is the need to account for the possibility of renegotiation. The chapter next turns to this.

 

ECONOMICS OF CONTRACT LAW    7

1.2. Renegotiation When transaction costs are low enough, bargaining between the parties takes them to the Pareto frontier (Coase 1960). This now commonplace observation underscores an assumption embedded in the economic rationale for expectation damages. This rationale implicitly assumes that renegotiations between the parties after the fact are not possible. If they were, performance should go forward when and only when it is socially desirable regardless of the remedy for breach. Although renegotiations are far from costless, they are sometimes possible. A law-​and-​economics account of contract should consider them (Rogerson 1984). Return to the example of the garage and the driveway. Bad soil conditions are discovered, and it is no longer in the mutual interests of the parties for the driveway to be built, regardless of whether the homeowner has already built the garage. But as long as the costs of renegotiation are low enough, the driveway will not be built, even if the homeowner has a right to specific performance. Rather than building the driveway, the builder would reach some kind of bargain with the homeowner. In a world of expectation damages, the homeowner would insist on at least the amount that she values the completed garage and driveway (less costs saved as a result of the breach). But, given the unexpectedly high cost of the driveway, the builder would be willing to offer more than this. A bargain can be struck. As long as the costs are low enough, neither the builder nor the homeowner should walk away from the negotiating table. Rational parties do not leave money on the table. Again, there is not necessarily any advantage taking. The parties can take account of the bargaining game in the initial pricing of the contract. It might seem that, when renegotiation is possible, we should be indifferent between expectation damages and specific performance. But this is not so (Rogerson 1984). Renegotiation allows parties to reach the efficient outcome based on where things stand at the time of the renegotiation. The optimal rule needs to take into account the way the rule affects the behavior of the parties between the time of the contract and the time of the renegotiation. Rogerson (1984) shows the overreliance problem is lower in a specific performance regime than in one of expectation damages. With expectation damages, I build the garage whenever there is a benefit to me from having it early. You promise to build the driveway or put me in the same economic position I would have been in had the promise been kept. If you perform, I get the garage and the driveway. If you breach, I get an amount of money that puts me in the same position. Given that I can recapture the cost of building the garage if we fail to reach a deal as an element of my damages, I shall always receive it in any bargain that it is in my self-​interest to accept. By contrast, in a regime of specific performance, the cost of building the garage is not always something that I always recapture. To be sure, building the garage improves my bargaining position. Once the garage is built, my costs are sunk and the driveway may never be built. I value having the driveway more once I have already spent money

 

8   DOUGLAS BAIRD on the garage. But the amount by which my bargaining position improves is not necessarily enough to justify constructing the garage early. I would rather pocket the money you need to spend to buy your way out of the contract without spending any money on the garage. Building the garage early improves my bargaining position, but not necessarily enough to justify doing it. As renegotiation is possible, I shall tend to build the garage early more often in an expectation-​damages regime than in one of specific performance. Inducing each of the parties to behave optimally during the course of their performance is a complicated problem. Even if we focus simply on the problem of overreliance, expectation damages lose some of their luster. Moreover, overreliance is only part of a much larger problem. The problem of overreliance exists because the contract itself is not completely specified. If the parties were well informed and had time to negotiate, the contractor and the homeowner would bargain explicitly about the timing of the construction of the garage. The inability to specify everything with precision imposes costs that go well beyond the problem of overreliance. Just as the inability to spell out obligations completely may lead to overinvestment in the contractual relationship, it is equally possible that it can lead to underinvestment. Both buyer and seller might be better off if the seller invested more in the quality of the goods she was making, but if a judge has no way of telling whether the seller made the investment, she cannot impose a remedy. She cannot craft an order that obliges that party to perform if she does not know exactly what the person should do. Parties must write contracts taking account of the limits that judges face as best they can (Hart and Moore 1999). The challenge for contract law is not to design the optimal contract but rather to put in place a set of background rules that makes the parties’ task easier. The extent to which the law should allow renegotiation after performance begins and circumstances change is a question that exemplifies the difficulties contract law faces. Legal rules that facilitate renegotiation can make the parties better off, but, precisely because the renegotiations can be anticipated, they can also make the parties worse off. Parties can under-​or overinvest in the contractual relationship precisely because they know that the parties will later find themselves in a world in which the other will be better off renegotiating, rather than trying to compel performance. The value of imposing limits on their ability to renegotiate turns largely on the ability of the court to understand the dynamics of the original bargain. If the renegotiation arises as a result of changed circumstances, the law should facilitate it. On the other hand, if parties cannot spell out their obligations completely and a court is unable to tell whether parties have lived up to their obligations, parties are able to take advantage of the situation after the fact. A legal rule that limits the ability of the parties to renegotiate might make both better off before the fact. The canonical case is Alaska Packers (Threedy 2000). At the turn of the twentieth century, a group of hired laborers travels to a distant harbor in Alaska where they are to work for the salmon season for a packing company. There is a dispute,

 

ECONOMICS OF CONTRACT LAW    9 and the workers insist on a renegotiation of their initial contract. Once everyone returns to San Francisco, the workers try to enforce the new contract. The packer argues that contract law should refuse to allow renegotiated contracts under these circumstances. There were two possible stories. Under one, the nets turned out to be different from what the workers expected. As the workers were paid by the size of their catch, they were left worse off. The workers could assert that, even if the unusual nets were not the fault of the packer, the unexpected conditions justified the renegotiation. And if it were the packer’s fault, the breach would not only entitle the workers to refuse to perform without a new deal, but would indeed give them the right to sue the packer for damages once they returned. In either case, conditions have changed, and renegotiating the contract makes sense. Under the competing story, however, the packer provided the nets it was supposed to. The workers were simply taking advantage of the packer’s inability to bring up replacements owing to the shortness of the season. The parties themselves may know the true state of affairs, but as long as the court does not, each party has a chance to take advantage of the other. If the workers have no right to renegotiate, the packer can get away with providing subpar nets. But if there is a right to renegotiate, the workers can engage in advantage taking. Even if they were given the nets they were promised, the workers can force the packer to renegotiate. The court’s inability to tell whether the parties lived up to their promises keeps it from imposing a damage remedy that correctly aligns incentives. It is possible to write models in which the legal rule should be one that facilitates renegotiations, as well as ones in which parties are able to tie their hands in their original contract, and legal rules should limit the ability of parties to renegotiate. Many things, including the bargaining power of the parties, need to be taken into account (Choi and Triantis 2012). A general theory of the law governing reorganizations, if it exists at all, is yet undiscovered.

1.3.  The Limits of Expectation Damages Quite apart from the problem of overreliance and the possibility of renegotiation, there is considerable reason to doubt that the expectation-damages concept deserves the central place it has enjoyed in the law and economics of contract (Posner 2003). The canonical case in which expectation damages works arises when the promisor faces a change in circumstance. The performance of the promise was once in the joint interests of the parties but is no longer. A new opportunity has come along that allows me to put my human capital to a better use, and you can readily find someone else to take my place. Expectation damages, however, will ensure that I take advantage of this new opportunity only if a judge can calculate damages accurately, and there is considerable reason to doubt that she can.

 

10   DOUGLAS BAIRD The judge must assess accurately the costs you incurred as a result of my breach. These costs include the costs of delay and the costs of finding someone else to do the job, and they include any lowering of the quality of the performance when someone else does it. How well is a judge able to take all this into account? You may have bargained for my services for reasons that are important, but not readily transparent. If the court overestimates the harm you suffer from breach, I will perform even when everyone is better off if we parted company. If the court systematically underestimates the harm from my breach, I shall breach too often. A court could account for all these things in principle, but the case for expectation damages rests on the idea that courts are good at making this sort of calculation. Expectation damages are easiest to measure in the common case in which the market price of a fungible commodity changes after the contract is entered. But exactly here where expectation damages are easiest to measure is where the efficient-​breach argument has little traction. Disputes arising from these contracts rarely are litigated in common law courts. Contracts involving purely fungible commodities tend to arise in self-​regulated markets, such as the Chicago Board of Trade. More important, contract breaches involving corn or wheat are not likely to be ones where there is likely to be much social benefit from breach. If it is a fungible commodity, why do I need to break my contract to provide it to some new buyer? Presumably, this new buyer can find someone else willing to sell at the market price. Even if the entire supply were locked up, new buyers, if they are indeed the highest-​valued users, can go to those who already have goods under contract and acquire the rights from them. Such contracts are typically assignable. Even if not, when the old buyer acquires the goods, she should be willing to resell them to those who value them more. To be sure, there are the costs of two sales, but these may be less than the costs of the breach. Fungible goods are often stored with third parties. They are bought and sold without any change in the physical location. Breach may arise in these cases not because breach will put the goods to a higher-​ valued use, but because the breaching party believes that expectation damages, once the dust settles, will turn out to be undercompensatory. There are many common law doctrines—​such as limiting individuals to damages that are reasonably foreseeable and nonspeculative—​that make expectation damages less than completely compensatory. But even if courts turned out to provide, on average, damages that equaled expectation damages, expectation damages would still not provide the correct incentives if those who are considering breach could assess before the fact whether the judge would underestimate damages in their particular case. It does no good to have an unbiased judge if the parties can predict in advance when she will be too high and when too low. This opportunity for strategic behavior exists whenever the promisor knows more about the costs of breach than the judge does. The possibility that judges are not going to do a good job of assessing expectation damages—​particularly the possibility that they might systematically underestimate them—​naturally leads to the question of whether the common law’s ban on supercompensatory damages is at all sensible (Baird 2013).

 

ECONOMICS OF CONTRACT LAW    11 Let us assume I want my portrait painted and I know that artists are, as a group, unreliable. Timely performance is very important for me, but the judge is likely to underestimate how important it is. In such a case, it would be useful for me to be able to bargain for supercompensatory damages. I shall have to pay more in order to persuade someone to enter into such a contract, but it allows me to screen the painters. Only those most likely to perform on time will be willing to accept my contract. Even if renegotiation is costly and there is a possibility that the painter will perform even though her efforts are better spent elsewhere, there are offsetting benefits as the willingness to pay a penalty signals private information. There is, of course, the possibility that the penalty clause reflects advantage taking against an unsophisticated consumer, but it is not obvious that one needs a special rule governing damages as opposed to a general mechanism against unconscionability. Using supercompensatory damages as a screening mechanism is one of many reasons why parties might not opt for expectation damages. The ability to find such rationales for other damage measures leads one to doubt that the traditional law-​and-​economics explanation for expectation damages is compelling. Although expectation damages may induce parties to take account of the costs of breach—​and this is an eminently sensible role for a contract damages rule to play—​it is not the only possible role. To justify expectation damages, one needs to explain why, in the mine-​run of cases, it is more compelling compared with the others. If one accepts the idea that two sophisticated parties should be able to bargain for any measure of damages they find in their mutual interest, the case for expectation damages rests upon the idea that it is a sensible baseline from which parties can depart, rather than that it induces optimal decisions about breach and that this concern trumps everything else. To see if expectation damages are even a logical starting point, engage in the following thought experiment. At the outset, I tell you explicitly that there is a possibility that it may not be in our mutual self-​interest that I perform. But I also tell you that I do not want to be a bad person, and I have no intention of breaching any contract. Instead, what I want is an option built into the contract that gives me an option not to perform. There is a clause in the contract in which I have the right to cancel the contract in exchange for giving you a certain amount of money. If we negotiated this option, what is the option price upon which we would agree? (Scott and Triantis 2004). Once the remedy for breach of contract is money damages, it would seem that it should equal the price parties would agree upon for the right not to perform. The challenge of contract law is setting the presumptive price of that option. If we characterize the problem in that way, it is not at all obvious that the right way to price the option is to use expectation damages (Scott and Triantis 2004). As in the case of the painter, it could be much more, but it many other cases it could be much less. Placing the problem of contract damages inside a framework that promotes free contracting replicates a more general problem in the law and economics of contract. Contract law provides a backdrop. Identifying exactly what purpose this backdrop should serve has been a long-​ standing question in the law and economics of contract, and it is the focus of the next section.

 

12   DOUGLAS BAIRD

1.4. Default Terms At the same time it was wrestling with the question of expectation damages and whether it was in the mutual interests of the parties, economics was used to shed light on contract’s substantive rules. Contract law simply provided the parties with terms that advanced their mutual interest. In a world in which we are committed to free contracting, these terms, like expectation damages, should be the rule only until parties negotiating at arms’ length bargain for something else. Contracts are incomplete, and the law is needed to fill in the gaps. When parties enter into a deal with each other, there are many things left unspoken. It could be a warranty term, a delivery date, or questions about who bears the risk of loss when the goods are destroyed in transit. Merchants ordinarily do not have the time or the money to spell out every particular detail. In the 1970s, the literature talked about contract law as providing parties with a set of “off-​the-​rack terms” (Goetz and Scott 1977). Contract law is not a set of abstract principles but rather a set of premade contract terms. Contract law saves parties the costs of drafting a long contract every time they deal with one another. Contract law provides a set of off-​the-​rack terms. Like off-​the-​rack clothes, they do not necessarily fit perfectly, but they fit most people, most of the time. Tailoring contracts, like tailoring clothes, takes time and money. Like bespoke clothes, they are more costly. For the vast majority of people, however, ordinary terms will do. Most are content with the terms contract law gives them. You may be a special person. You may be especially large, extra tall, extra short, or extra whatever. The clothes you find on the rack do not fit you. You may want to get your clothes custom-​made. Similarly, if you are a contracting party, when your needs are special, you can get your own customized terms. The imperfections in expectation damages seem tolerable from this perspective. One need not make the argument that they serve everyone, only that they work for most. It is as good a baseline as anything else is. Even if the inability to opt out of expectation damages under current law does not make sense, it may still be a sensible off-​the-​rack rule. Expectation damages are not necessarily God-​given, but they do align incentives at the time of the breach decision, and this is important. They are what parties would bargain for most of the time. The account of the law and economics of contract as it emerged at the end of the 1970s put no special magic into much of contract law. The rules governing warranties, the risk of loss, and the like were simply sensible terms. For example, at common law and under the law merchant, manufacturers sell their goods with an implied warranty of merchantability. Unless the manufacturer disclaims, they come with the promise that they pass without objection in the trade under the contract description. If I am the shoemaker and you buy shoes from me, an implicit term of the bargain is that the shoes I give you pass without objection in the trade as “shoes,” and they are suitable for the purposes

 

ECONOMICS OF CONTRACT LAW    13 for which shoes are typically used. This feature of the law has been in place since the twelfth or thirteenth century. It is not controversial and it never has been controversial. Appropriately limited, the implied warranty of merchantability is a sensible off-​the-​ rack rule. If I make something and do not want to promise that it passes without objection in the trade under the contract description, I am the one who is doing something unusual. Hence, I should be the one who has to bargain for special terms. It is often said that the common law was a regime of caveat emptor. This is not true (Scheppele 1988). Caveat emptor was a doctrine that put sensible limits on the warranty of merchantability. It applied most often in contexts in which the seller was not a manufacturer. The seller had no more expertise than the buyer did. I have a ship with lots of different kinds of cargo, including exotic wood. I am an intermediary and I am selling you goods I have acquired from others. It would be odd if I were making warranties about qualities that you can judge much better than I can. You have expertise with respect to wood, and I do not. The other cases to which caveat emptor applied arose when the buyer’s obligation to inspect was more important than seller’s promises with respect to quality were. If I sell you a horse, you have a duty to inspect it. You are responsible for any defects that an inspection should have revealed. The cases worth litigating typically involved horses and other assets that were both valuable and subject to change. A seller is obliged to reveal latent defects, but not those a reasonable inspection would reveal. A different rule would inevitably lead to litigation over whether the horse became lame before or after the sale. Quite apart from providing specific terms, contract law also provides general rules of engagement (Baird 2013). There is a general duty of good faith, but there is no affirmative duty to disclose information. The most sensible trade-​off here is again not obvious. On the one hand, it does not make sense to force the buyer to engage in an endless game of twenty questions to find out what the seller knows, but neither does it make sense to force the parties to disclose information that may be costly to acquire. But even this background rule needs to be seen in light of the general prohibition on fraud and its corollary—​you must disclose information that once you have learned it makes previous statements misleading. Again, it is hard to argue that these particular background rules are inevitable, but it is quite easy to argue that they provide sensible rules of engagement. The value of contract law is not that it provides any particular set of terms, but that it provides some set of terms and that these, in the main, be sensible. When you enter into a contract, you should be confident that if you do not bargain for special terms, you will get a set of terms that make sense for most people most of the time. The common law of contracts provides a set of terms that gives the vast majority the right set of incentives. The iconic (and now rather dated) example deals with the person who wants to climb Mt. Everest and buys an ordinary camera and an ordinary roll of film to take her picture when she reaches the top. The camera and film prove defective. The seller is liable for damages, but how should these be calculated? The sensible allocation of damages should take into account not only the need to give the seller incentives to make sure the film is not defective but should also give the buyer an incentive to take precautions against the possibility of defects. It makes sense that a

 

14   DOUGLAS BAIRD photographer who will incur large damages in the event that the film proves defective take precautions, such as buying an especially reliable camera or bringing a back-​up, rather than imposing the entire risk on the seller, who is unable to take such precautions, or forcing the seller to bargain with each buyer to shift the risk back. Such a rule makes sense in a world in which we want sellers of cameras and film to be able to sell them to people cheaply. This rationale provides an explanation for the rule of Hadley v. Baxendale, the rule that limits consequential damages for those risks that are reasonably foreseeable. As the use of photographic film suggests, this account of contracts belongs to a different era (Posner 1972). Much has changed. Among other things, a new vocabulary has emerged. Law and economics switched from using the idea of “off-​the-​rack” terms to “default” terms in the 1980s. This new vocabulary came from the introduction of personal computers. Software could be configured in many different ways, but when first installed, the software came configured in a standard way. These were known as “default settings.” You could change them if you wanted, but they provided a decent starting place. Contract rules could be seen in the same way. Contract rules were like factory defaults in your word processing program. The literature replaced “off-​the-​rack terms” with “default terms.” The change in vocabulary brought with it a subtle change in the way people thought about the problem of contract law. The idea of “off-​the-​rack” clothes reinforced the idea that contract law was simply the standard contract that most people wanted because it sensibly allocated risk. The metaphor of “default” rules underscores the idea that the contract terms the law provides are simply a starting place. What most want might be a good starting place, but somewhere else might be as good or better. The starting place, for example, might be set to induce those with important information to reveal it during the course of contract negotiations. Parties might be better off if, rather than allocate risks optimally, the default rules put parties in a position that forced the disclosure of information (Ayres and Gertner 1989). Consider a variation on the facts of Peevyhouse (Baird, Gertner, and Picker 1994). A strip-​mining company approaches a couple who own a farm. It wants to acquire the right to strip mine the land. The couple requires that the company promise to restore the land once they have finished. Their subjective value is high. They put a much higher value on having the land restored than other farmers would. When the couple entered into the contract, they cared intensely not simply about damages in the event of breach or the incentives of the coal mining company, but also about the likelihood of breach, a question on which the company was much better informed than they were. A rule the holds the strip-​mining company liable for the entire cost of restoring the land—​regardless of whether the couple even thinks it worth that much—​may make sense. The prospect of facing that kind of damage award will force the coal mining company to disclose information, either explicitly or by opting out of the default rule and adding a clause limiting damages in the event of a breach. Rather than terms that fit parties most of the time, it might make more sense to choose rules that provide the best starting place for negotiations. As this example illustrates, a

 

ECONOMICS OF CONTRACT LAW    15 good starting place may focus on providing an incentive to those with information to reveal it. Alternatively, it might be one that provides terms that are suitable for the unsophisticated. Because sophisticated parties are better able to reshape contract terms, it makes sense to impose upon them the obligation to do it. One can take the analogy to default terms in computer programs a step further (Ayres 2012). Lawmakers should pay attention not only to the content of default terms but also to the rules that govern opting out of them. Just as it ordinarily takes multiple keystrokes to delete a file permanently, it may make sense to require parties to jump through special hoops when opting out of a particularly important default. Opting out of the right to litigate the dispute in court and instead agreeing to arbitration may require more steps and deliberation than a change in delivery date.

1.5.  Contract Terms as a Product Attribute It has become a commonplace that contract law in the first instance provides a set of terms that parties can use or not as they please. From this, it is only a short step to treat a contract term as just another product feature. A contract term that accompanies the sale of a computer is no different from its hard disk or its battery. It may be good or bad. Some buy a product after learning all its features; others buy knowing comparatively few. Whether a buyer knows the content of various contract terms may be no different from whether the buyer knows about other features of the product. This perspective sheds some light on how we might think about standardized contracts (Baird 2013). The typical contract we have today is, by and large, not negotiated. If I buy a piece of software, I will typically click and agree to certain terms. But it is most unlikely that I am going to read the terms. As a practical matter, I either take Microsoft’s or Apple’s terms sight unseen or do not buy their product. Courts still put forward legal tests that make the enforceability of standardized terms turn on whether the typical buyer is likely to have read them. It is hard to believe that the judges who write these opinions actually read fine print in the manner they ascribe to their hypothetically reasonable person. The rest of us certainly do not (Bakos, Marotta-​ Wurgler, and Trossen 2014). A coherent account of the law and economics of contract must confront the dynamics of the marketplace. In this respect, an account of contracts that focuses upon default terms that are merely starting places for further negotiation does not capture much of what is going on in many contracting environments. One suspects that if the legal terms were in fact important, at least some consumers would find out about them by reading product reviews or advertising rather than reading the contract. They would not negotiate, but neither would they buy products that contained terms they did not like. What matters is not whether every consumer knows about the hidden legal terms, but rather whether a sufficient number of sophisticated

 

16   DOUGLAS BAIRD people know about them to ensure that sellers have to offer good terms in order to gain sufficient business (Schwartz and Wilde 1979). Apple sells to Fortune 500 companies, as well as to ordinary consumers. To the extent Apple sells the same product with the same terms, it may have no ability to exploit consumers even if it were inclined to do so. Its need to attract both types of customers may allow the ignorant to free ride on the expertise of the sophisticated. This perspective reminds us that we should take seriously the idea that contract law can only do a limited amount of work. Those of an economic bent are more inclined than others are to think that markets will tend to be self-​policing. But regardless of how much people trust markets, they should be able to agree on where trouble, if it exists, is most likely to lie. Advantage taking is more probable in specialized markets that focus on the unsophisticated—​such as the market for payday loans or rent-​to-​own furniture—​than it is in mass markets in which large firms sell on the same terms to everyone (Baird 2013). For some, the consuming problem with respect to standardized contracts is the nightmare scenario in which you buy a product, take it out of the box, use it, and only several weeks later learn that the seventieth page of the fine-​print contract contains language to the effect saying that you owe the seller an extra $1000. Because the common law of contract is based on the arms’-​length bargain, there is no neat doctrinal response to this problem. The buyer became bound by all the terms of the contract when she accepted the goods. It was her responsibility to read the contract and to object to any terms not to her liking. It is worth noting, however, that this problem appears much more in the law school classroom than it does in mass consumer markets. Reputational forces matter. Moreover, rules against fraud offer more protection that contract law does. Any representations made in advertising or any places that represent the product as having particular features that the seller knows the product does not have is a form of common law fraud. It is punishable criminally. The egregious conduct that contracts professors fear may not be properly the province of contract law at all. When it arises, contract law is neither necessary nor appropriate. To the extent that we worry about hidden terms, it may make sense not to worry about the process of negotiation itself and instead mandate forms of protection—​such as the implied warranty of merchantability—​that ensure that unknown terms cannot work much of a surprise. If I sell you a product, it has to pass without objection to the trade under the contract description. One can mandate such a warranty in all contracts. Alternatively, one can keep it as a default rule, but require special hoops to jump through in the event that a seller wants to opt out (such as requiring such terms to be highlighted). But lawmakers need to be careful. Making many terms prominent ensures that none of them is prominent (Ben-​Shahar and Schneider 2011). Solving this problem requires going beyond thinking of contracts as a discrete document that normally emerges from arms’ length bargaining. The law must regulate the market as a whole. Economics has much to say about how markets as a whole should be regulated. There is no need to create an acoustical separation between the economic

 

ECONOMICS OF CONTRACT LAW    17 theory of such regulation and the law and economics of contract, but the more one focuses on this problem, the farther one is stepping away from Richard Posner’s original enterprise, which was to make sense of the core doctrines of the common law of contracts.

1.6.  Contract Law and Contract Theory Much of the challenge in exploring contract law is identifying what is distinctive about contract law. The law and economics of contract law needs to be located between two poles. As discussed in the last section, there is the challenge of regulating the market as a whole. Economics has much to say about regulation, but one can argue that this is distinct from the domain of the law and economics of contract. Regulation arguably stands apart from contract law proper. Contract law may be best conceived as an enterprise that focuses on the terms and conditions on which two individuals enter into a bargain with each other. At the other extreme are the bargains themselves. If two people want to contract with each other, they face many problems. It is very hard to spell out everything in detail. When things are left unsaid, there is the possibility of shirking by either side. All contracts are incomplete. It may be that you know what you should be doing, and I know what you should be doing, but it is going to be hard for any judge or indeed any third party arbiter to figure it out. It may also be that I simply do not have enough information to know how you should carry out the job. You have promised to mow my lawn and you have expertise about how to do the job that I do not have. For this and other reasons, I do not know enough to specify in the contract exactly how you should go about performing your job. You may cut corners and keep the promises you made to me without promising me everything that I would have demanded had I sufficient knowledge. Moral hazard and adverse selection problems are everywhere. Economics has much to say about how parties overcome agency and other problems in designing their contracts. But the problems of contract design are the province of contract theory. They are not the same as contract law. Contract theory has spawned its own robust literature (Bolton and Dewatripont 2004), but it is a literature that, as far as the law is concerned, has little to say about the content of contract law beyond a general mandate for courts to enforce contracts as written. Contract law is an enterprise distinct from contract theory. Rather than trying to understand contracts themselves, it focuses on the rules that tell a court that comes how to interpret them, it is the world in which parties find themselves when they start to write their contracts. Contract law is a set of ground rules. Largely, a sensible contract law merely provides parties that want to engage in trade with a starting place from which they can fashion their own bargain. But what makes some starting places better

 

18   DOUGLAS BAIRD than others? What are the stakes? What turns on this? For many years, advances in the law and economics of contracts have started with such questions. We may anticipate that this will continue.

References Ayres, Ian. 2012. “Regulating Opt-​Out: An Economic Theory of Altering Rules.” Yale Law Journal 121, pp. 2032–​2116. Ayres, Ian and Gertner, Robert H. 1989. “Filling in Gaps in Incomplete Contracts: An Economic Theory of Default Rules.” Yale Law Journal 99, pp. 87–​130. Baird, Douglas G. 2013. Reconstructing Contracts. Cambridge, MA: Harvard University Press. Baird, Douglas G., Gertner, Robert H., and Picker, Randal C. 1994. Game Theory and the Law. Cambridge, MA: Harvard University Press. Bakos, Yannis, Marotta-​Wergler, Florencia, and Trossen, David R. 2014. “Does Anyone Read the Fine Print? Testing a Law and Economics Approach to Standard Form Contracts.” Journal of Legal Studies 43, pp. 1–​35. Barton, John H. 1972. “The Economic Basis of Damages for Breach of Contract.” Journal of Legal Studies 1, pp. 277–​304. Ben-​ Shahar, Omri and Schneider, Carl E. 2011. “The Failure of Mandated Disclosure.” University of Pennsylvania Law Review 159, pp. 647–​749. Bolton, Patrick and Dewatripont, Mathias. 2004. Contract Theory. Cambridge, MA: MIT Press. Choi, Albert and Triantis, George. 2012. “The Effect of Bargaining Power on Contract Design.” Virginia Law Review 98, pp. 1665–​1743. Coase, R. H. 1960. “The Problem of Social Cost.” Journal of Law and Economics 3, pp. 1–​44. Fuller, L. L. and Perdue, William R. 1936. “The Reliance Interest in Contract Damages (Part I). Yale Law Journal 46, pp. 52–​96. Goetz, Charles J., and Scott, Robert E. 1977. “Liquidated Damages, Penalties and the Just Compensation Principle: Some Notes on an Enforcement Model and a Theory of Efficient Breach.” Columbia Law Review 77, pp. 554–​594. Hart, Oliver and Moore, John. 1999. “Foundations of Incomplete Contracts.” Review of Economic Studies 66, pp. 115–​138. Holmes, Oliver Wendell. 1897. “The Path of the Law.” Harvard Law Review 10, pp. 457–​478. Posner, Eric A. 2003. “Economic Analysis of Contract Law after Three Decades: Success or Failure.” Yale Law Journal 112, pp. 829–​880. Posner, Richard A. 1972. Economic Analysis of Law, Boston, MA: Little, Brown. Rogerson, William P. 1984. “Efficient Reliance and Damage Measures for Breach of Contract.” RAND Journal of Economics 15, pp. 39–​53. Scheppele, Kim L. 1988. Legal Secrets: Equality and Efficiency in the Common Law Chicago, IL: University of Chicago Press. Schiffrin, Seana Valentine. 2007. “The Divergence of Contract and Promise.” Harvard Law Review 120, pp. 708–​753. Schwartz, Alan and Wilde, Louis L. 1979. “Intervening in Markets on the Basis of Imperfect Information: A Legal and Economic Analysis.” University of Pennsylvania Law Review 127, pp. 630–​682. Scott, Robert E. and Triantis, George C. 2004. “Embedded Options and the Case against Compensation in Contract Law.” Columbia Law Review 104, pp. 1428–​1491.

 

ECONOMICS OF CONTRACT LAW    19 Shavell, Steven. 1980. “Damage Measures for Breach of Contract.” Bell Journal of Economics 11, pp. 466–​490. Spier, Kathryn E. and Whinston, Michael D. 1995. “On the Efficiency of Privately Stipulated Damages for Breach of Contract:  Entry Barriers, Reliance, and Renegotiation.” RAND Journal of Economics 26, pp. 180–​202. Threedy, Deborah L. 2000. “A Fish Story: Alaska Packers’ Association v. Domenico.” Utah Law Review 2000, pp. 185–​221.

 

Chapter 2

(IN) EFFICIE NT BRE AC H OF C ONTRAC T Daniel Markovits and Alan Schwartz

2.1. Introduction 2.1.1. The Law The common law of contracts vindicates promisees’ rights principally through the expectation remedy. Although the Restatement recognizes three distinct contractual interests—​expectation, reliance, and restitution—​it expressly privileges the expectation interest over the other two. The law, the Restatement observes, “[o]‌rdinarily … enforces the broken promise by protecting the expectation that the injured party had when he made the contract” (Restatement 1981, § 344 cmt. a). Courts thus respond to breach “by attempting to put [the promisee] in as good a position as he would have been in had the contract been performed”; that is, by “giv[ing] the injured party the ‘benefit of the bargain’ ” (Restatement 1981, § 344 cmt. a) The Uniform Commercial Code similarly recites that contract remedies are designed to put the aggrieved party “in as good a position as if the other party had fully performed” (UCC § 1-​305 [a]). Conventional contract law thus appears to vindicate contractual promises not directly but only through substitutionary means, by requiring the promisor to pay money damages that equal the benefit of the promisee’s lost bargain. As a consequence, promisees become indifferent, ex post, between performance and breach-​plus-​damages, transaction costs aside. An implication of contract law’s “benefit of the bargain damages” is that a promisor, after paying them, is free to realize gains made possible by diverting the promised performance to a higher-​valuing third party, or by not performing at all. In the current lexicon, contractual expectations conventionally thus receive liability rather than property rule protection (Calabresi and Melamed 1972). The rationale behind the law’s decision to accord limited protection to promisees has a name—​the “theory

 

(In)Efficient Breach of Contract    21 of efficient breach.” The theory has become sufficiently familiar to appear in textbooks intended for students (Ferriell 2009, 715–​17). It also is unpopular outside the lawyer-​ economist community.1

2.1.2. The Theory Restated Efficient breach theory should vanish because no breach is efficient. To see why, assume that contracting agents can contract in their self-​interest and that it is in the agents’ self-​ interest to choose efficient contract terms: that is, terms that maximize the expected surplus that the agents’ deal makes possible. Then define a breach of contract as an agent’s failure to take an action that the contract requires or her taking an action that the contract forbids.2 Under this definition, no breach is efficient because to breach would be not to take an action that the contract efficiently requires or to take an action that the contract efficiently forbids. It follows that the analyst’s or the court’s first-​order question is interpretative: what does the contract at issue require, permit, or forbid? We illustrate with the canonical case. Consider two contracts. Contract A lays out a single path to performance: The promisor must trade a specified good or service for a price. Contract B, which we call a dual-​performance contract, provides alternative paths to performance: The promisor must trade the good or service item or transfer a sum to the promisee that equals the gain the promisee would have realized from the trade. Let the promisor not trade the item, but transfer the sum instead. This would be a breach if the parties wrote contract A, but not if the parties wrote contract B. The promisor would breach even Contract B, however, if she failed both to trade and to transfer the sum. Neither the failure to trade nor the failure to transfer could be efficient breaches if the applicable contract terms were 1  An article by Roy Ryden Anderson (2015) asserts, “Unsurprisingly, most dispassionate contracts scholars have rejected the theory out of hand.” Professor Anderson summarizes the “now rather infamous theory of ‘efficient breach’—​a self-​indulgent proclamation that is noxious to even neoclassical contract law. The heretical supposition of an efficient breach is that the compensated promisee will be indifferent to the breach, the breaching promisor will be most pleased with herself, and the rest of us will somehow happily benefit … by having the performance of the breached contract wind up with someone who … values the performance more than does the promisee” (2015, 957). According to Professor Anderson, and others, efficient breach theory “since it first slithered into view … has been almost exclusively a creature of the academic literature and the law school classroom … because the idea is based on so many unrealistic and unprovable assumptions that run counter to the workday world, to life experience, and to the natural intuition of fair-​minded people” (2015, 957–​58 [footnotes within quotes omitted]). 2  The UCC does not define a breach, but §2-​106(2) defines conforming as actions that are “in accordance with the obligations under the contract.” It follows that a breach would be to take an action that is not in accordance with those obligations. Similarly, Restatement (Second) of Contracts §2(1) defines an enforceable promise as “a manifestation of intention to act or refrain from acting in a specified way, so made as to justify a promisee as understanding that a commitment has been made.” Hence, the failure to act, or not to refrain from acting, in the “specified way” would be a breach. In accordance with this inference, Restatement §235(2) provides that “When performance of a duty under a contract is due any non-​performance is a breach.”

 

22    DANIEL MARKOVITS AND ALAN SCHWARTZ efficient. Hence, to ask whether a breach is efficient is to ask the wrong question. Moreover, it is a question that courts generally know not to ask, as the law’s foundational commitment to freedom of contract entails that courts do not evaluate the efficiency of terms in business contracts. Finally, the right question is what did the contract require the promisor to do? Once again, this question does not distinctively concern remedies at all; rather, it concerns interpretation. Richard Posner (1972), the most prominent and effective early proponent of efficient breach theory, made the initial mistake. He assumed that parties always wrote Contract A, but he recognized that, in his examples, it would be ex post efficient for the promisor to pay off the promisee and sell the specified property item to another. That the parties wrote Contract A implied that nondelivery was a breach; that the promisor should not deliver thus was an efficient breach. But if it would be efficient, again on his examples, for the promisor to transfer rather than trade, the parties would have written Contract B. Of course, Contract A might actually be efficient, for example, when the promisee’s expectation is not monetizable. In almost every example of efficient breach the literature provides, this condition is not satisfied. The correct conclusion, then, is that the parties wrote Contract B, so that nondelivery alone would not be a breach. As should be apparent, Contract B tracks current law, under which an agent is free not to trade, but must then transfer the promisee’s expectation. The contract is attractive when parties can inform the court of the promisee’s gain: the value a buyer would have realized3 or the cost the seller would have incurred.4 When values and costs are verifiable, the Contract makes each party the residual claimant with respect to the breach decision.5 As an example, let the seller discover an opportunity to sell to a third party or to redeploy her resources to some entirely distinct purpose, rather than perform under the initial agreement. Contract B gives the seller efficient incentives to decide whether to take the opportunity because she would realize the gain but bear the cost—​the buyer’s expectation. Similarly, the Contract requires the buyer to internalize the gains and costs were he to perform elsewhere, or not perform. Therefore, under the contract (which reflects the law today), a promisor will choose the permitted action—​to transfer the specified property or to pay—​that produces the greater overall gain.

2.1.3. The Expectation Interest as a Default There is a well-​grounded view that parties have limited contractual control over remedies. For example, a court will not ask whether, all economic things considered, it was “reasonable” of the parties to contract to trade 200 widgets rather than 150 widgets, but a court will ask whether, all economic things considered, a party estimate of the profit the 3 

The law awards the buyer his value less the price. The law awards the seller the price less her cost. 5  An early elaboration of this rationale is Birmingham (1970, 273–​92). The first formal explanation is Shavell (1980). 4 

 

(In)Efficient Breach of Contract    23 promisee would realize from a deal was “reasonable,” if that estimate was instantiated in a liquidated damage clause. Furthermore, parties cannot require a court to enforce their promises specifically. On the other hand, much scope remains for parties to contract away from the standard expectation interest remedy. Parties thus have substantial discretion to liquidate damages, to under-​liquidate damages (Schwartz 2010; Stole 1992), to limit remedies (UCC §2-​7 19), to exclude consequential damages, and often successfully to request specific performance by setting out their need for the remedy in a contract’s whereas clauses.6 That this much contracting scope exists supports our claim, assumed in Section 1.2 and set out in more detail below, that parties actually choose expectation interest contracts such as Contract B when their contracts are silent regarding remedies. The remainder of this chapter expands the claims in Section 1 in more detail, considers moral and economic objections to Contract B, and analyzes contract doctrine to show that the law actually presupposes that parties write this Contract.

2.2.  Critiques of Efficient Breach of Contract Certain moral theorists, for some time and with increasing force, have adopted a strong line of attack against orthodox contract law. This attack rests on the premise that a promisor who refuses to trade commits a breach. The moralists, that is, make the same mistake that Posner made: they assume that parties write Contract A when parties actually write Contract B. On the moralists’ assumption, orthodox contract remedies price breach, and they set prices so low (at levels that enable breaching promisors to profit from their wrongs) as to encourage breaches of the very obligations that contract law purports to establish. This feature of the orthodox doctrine, these critics say, undermines the immanent normativity of contract obligation and causes contract law to diverge from the morality of promise in unattractive ways.7 Moral critics of orthodox contract also, and relatedly, attack other features of established law, for example the mitigation doctrine. This doctrine supports the expectation remedy by requiring promisees to respond to breach by taking steps to minimize their costs, and thus the promisor’s damages, from the promisor’s failure to trade. Critics of orthodox contract law charge that the doctrine authorizes a breaching promisor to draft her promisee involuntarily into the promisor’s service, specifically by requiring the promisee to exercise initiative in order to reduce the damages that breaching promisors owe (Shiffrin 2012).8 Supracompensatory remedies, 6  Parties cannot contract for penalties but, in symmetric information environments, they do not want penalties (see Schwartz 1990). When information is asymmetric and parties contract partly to induce efficient investment, penalties can be efficient, however. (See Edlin and Schwartz 2003.) 7  A variety of these claims appears in, for example, Friedman (1989); Shiffrin (2007; 2009). 8  Shiffrin’s (2012) concern is incorrect. Mitigation protects the promisee’s expectation at the lowest cost, an efficiency that is reflected in the price. Promisees do not need to be coerced to mitigate because

 

24    DANIEL MARKOVITS AND ALAN SCHWARTZ moral critics of orthodox contract say, would avoid all of these wrongs. A legal regime that responded to breach of contract by ordering specific performance, restitutionary disgorgement, or even punitive damages would truly sanction rather than merely price breach. Such a regime would thus support the internal norms of contract obligation and bring the law of contract into conformity with the morality of promise. A legal rule whose theoretical foundation appears shaky to many commentators sometimes is replaced, so it is unsurprising that the expectation interest’s critics have made real-​world progress; courts and other legal authorities have begun to adjust the law in response to their critiques. The Uniform Commercial Code liberalized the right to specific performance. Specific performance, however, sometimes is impractical: it takes too long to get. Specific performance also is impossible when the promisor has sold the property to a good faith purchaser. The disgorgement remedy ameliorates these difficulties, because it requires the promisor to transfer to the promisee the gain the promisor realized from redeploying her resources. The recently adopted Restatement of Restitution permits a court to replace the expectation remedy with disgorgement when the remedy is inadequate,9 and “gain based damages” are beginning to be awarded in England and Israel.10 American courts also increasingly exhibit distaste for restricting contract damages to the promisee’s loss when the promisor benefited from not performing the contract’s substantive terms (Restatement [Third] of Restitution and Unjust Enrichment, § 39[1]‌[2011]).11 Critics even argue that the promisee’s right to they realize mitigation gains. And they are not in fact coerced because they have assumed the duty to mitigate and memorialized this duty in the contract through the lower price. 9  A thoughtful analysis by the drafter appears in Kull (2001). A more recent thorough treatment, which summarizes newer cases and commentary, is Anderson (2015). 10  For England, see Cunnington (2008). For Israel, see Adras Building Material Ltd v Harlow & Jones Gmbh, (1995) 3 Restitution L Rev 235, 240, 242, 246, 249, 261, 268. Adras is criticized in Botterell (2010, 135–​60). 11  Traditionally, disgorgement remedies were awarded only in narrow classes of cases. For example, disgorgement might be awarded in cases in which the breaching party violated a fiduciary or quasi-​ fiduciary duty owed to the victim of the breach. See, e.g., Snepp v United States, 444 US 507, 515–​16 (1980) (awarding the CIA disgorgement of profits earned by a former agent in connection with a tell-​all book published without being vetted as required by his contract of employment with the CIA); X-​It Prods, LLC v Walter Kidde Portable Equip, Inc, 155 F Supp 2d 577, 658 (ED Va 2001) (observing that “equitable remedies” including disgorgement, may be awarded in cases of “wrongfully obtained profits in a variety of contexts, including breach of fiduciary obligation or breach of contract” but suggesting that the proper award of these remedies depends on a factual finding of “unclean hands”); Ajaxo Inc v E*Trade Grp, Inc, 37 Cal Rptr 3d 221, 248 (Cal Ct App 2005) (awarding a disgorgement remedy in connection with breach of a nondisclosure agreement governing trade secrets or proprietary confidential information). Another narrow class of cases, in which disgorgement remedies have historically been awarded, concerns breaches of contract to sell land, where courts, when they are unable to adopt the traditional specific performance remedy for technical reasons, have imposed constructive trusts on breaching sellers and required them to disgorge their gains from breach. See, e.g., Gassner v Lockett, 101 So 2d 33, 34 (Fla 1958) (imposing a constructive trust against a breaching seller where a new buyer’s recording of title prevented the court from ordering specific performance of the sale to the buyer against whom the seller breached). This use of disgorgement comports with our views. Insofar as the common law treats land as unique and makes contracts to sell land specifically enforceable, such contracts do not, properly interpreted, permit sellers to transfer rather than trade. Refusals to trade are, consequently, breaches, full stop.

 

(In)Efficient Breach of Contract    25 trade sometimes should be protected with the strongest remedy: punitive damages (see Shiffrin 2007).

2.3.  A Poorly Framed Dispute Sections 1 and 2 imply that the dispute between proponents and critics of efficient breach theory is poorly framed. This misframing gives considerable force to criticisms of a law that would permit breaches on payment of a price. The critics’ concerns are misplaced, however, because they suppose breaches to exist where there are no breaches. Critics and proponents of the theory incorrectly assume that some contract breaches are efficient but disagree over the morality of particular breaches. If no breach is efficient, however, there is no dispute: a truly breaching promisor is behaving both inefficiently and immorally. Using the example above, when parties write Contract A, the promisor’s failure to trade is an immoral and inefficient breach. When parties write Contract B, the failure to trade is neither. Hence, the only thing to argue about is which contract the parties wrote. Section 3 thus analyzes the two assumptions that underlie our claim that typical parties would and do write Contract B. The assumptions are (a) that renegotiation is infinitely costly; and (b) that contract creation costs are zero. Assumption (a) implies that parties cannot renegotiate ex post, so their choice of a contract is consequential; assumption (b) implies that parties are free to write whatever contract they like. To see why these assumptions matter, first let renegotiation costs be zero. The Coase Theorem teaches that if agents are informed, as in common contracting situations, agents bargain to the efficient allocation of property rights whenever the status quo allocation is suboptimal. To apply the Theorem here, let the law protect the expectation with a property rule—​specific performance—​and assume that trade would be inefficient: the promisor’s loss from trade would exceed the promisee’s gain. In this circumstance, parties would renegotiate to excuse the promisor from trading: that is, the parties would renegotiate to the result that a liability rule would have yielded on its own. Now let there be a liability rule and assume that trade would be efficient: the promisee’s gain from trade exceeds the promisor’s costs. In this circumstance, the promisor simply trades, which is the result that a property rule would have directed. And to summarize, in the ideal Coase Theorem environment that efficient breach theory assumes, efficient performance (and efficient nonperformance) occurs under every rule type. Therefore, though Recently, there has been a trend toward looking more favorably on disgorgement remedies in connection with ordinary breaches of contract. See EarthInfo, Inc v Hydrosphere Res Consultants, Inc, 900 P2d 113, 119–​20 (Colo 1995) (approving restitution for breach of contract where a defendant’s wrongdoing leads directly to her profits); see also Univ of Colo Found, Inc v Am Cyanamid Co, 342 F3d 1298, 1311 (Fed Cir 2003) (recognizing EarthInfo as stating Colorado law); Dastgheib v Genentech, Inc, 438 F Supp 2d 546, 552 (ED Pa 2006)(approving of disgorgement in EarthInfo type situations); and Daily v Gusto Records, Inc, No 3:94-​1090, 2000 US Dist LEXIS 22537, at *6 (MD Tenn Mar 31,( 2000) (endorsing the EarthInfo logic).

 

26    DANIEL MARKOVITS AND ALAN SCHWARTZ parties could write Contract B, they also could write Contract A, which would yield the same result. The theory, that is, cannot explain why parties would contract for any remedy in the feasible set rather than any other remedy. A similar analysis shows that the moral issue concerning efficient breach is not properly joined. Imagine that agents make Contract B. Under it, recall, the parties agree to alternative performances: either they will trade goods or services in exchange for a price or the promisor will make a monetary transfer to the promisee, equal to the gain the promisee would have realized from receiving the goods or services. If contractual promises are understood in this way, as involving a promise either to trade or to transfer, then, as suggested above, a seller who voluntarily transfers rather than trades does not in fact breach the contract. So called efficient breaches are not in fact breaches at all, and a court that requires the promisor to pay expectation damages provides not substitutionary but rather direct relief, specifically enforcing the transfer prong of the dual performance promise. Critics of efficient breach theory may object to this argument because, in their view, Contract B (or the expectation interest remedy that replicates it) is unfair. Under this Contract, critics worry, the promisor captures all of the gains from reallocating her resources to another use, but promisees are entitled to those gains. To see why this fairness objection is misplaced, realize first that parties contract because they anticipate that trading will produce net gains. They anticipate, that is, that buyers’ values of trade will exceed sellers’ costs, so that trade will produce joint surplus.12 But between the time of contract and the time of trade, a buyer’s valuation may fall, or a seller’s costs may rise. Accordingly, even as the parties anticipate gains from trade, it always remains possible that, as events transpire, the parties’ joint surplus will be maximized by not trading. This logic is of course available to the parties when they fix the terms of their contracts. Accordingly, a seller who retains the right to capture nontrading gains at performance time will pay for this right up front, at formation—​her payment will be memorialized in a lower price. The lower price, moreover, increases the difference between the buyer’s value and the contract price and hence the buyer’s expectation damages in the event that the seller opts to transfer rather than to trade. Thus, although the expectation remedy appears to allow the seller to capture the entire gains from “efficient breach,” this appearance deceives. The price mechanism gives buyers a share of all the gains from contracting, including of the nontrading gains associated with “efficient breach.” Put concisely, the buyer’s gain is invariant to the legal remedy.13 12 

For expositional purposes only, and without loss of generality, we refer to promisors as sellers and promisees as buyers. 13  It may help to state this result formally. Let the buyer have property rule protection. The price of the contracted for item, pPR, equals the seller’s costs, c. There are two cost components. The first, denoted (i), is production cost. The second is the bribe, b, the seller expects to pay to the buyer for permission to exit if transfer of the item would be inefficient. Let the probability of inefficient transfer be 0 < θ < 1. The bribe’s expected cost, denoted (ii), thus is θb, and pPR = (i) + (ii). The buyer would realize v from transfer plus the expected value of the bribe. His expectation under property rule protection thus is v + (ii) – p ​ PR = v + (ii) – ((i) + (ii)) = v – (​ i). Now let the buyer have liability rule protection. The seller need not pay a bribe so her costs fall by the bribe’s amount. Hence, pLR = (i). The buyer’s expectation

 

(In)Efficient Breach of Contract    27 On this view, the morality of promise therefore does not favor mandating trade or requiring sellers to disgorge their gains from diverting trade to higher-​valuing third parties. Indeed, a buyer who pays the lower price associated with the expectation remedy but subsequently insists on specific performance (or disgorgement of the nontrading gain produced by an “efficient breach”) claims a benefit that she has not paid for, and that she forswore in her contract (through its price term). One might even say that such a buyer acts in bad faith. The conventional disputes about the theory of efficient breach are therefore poorly framed, and the central issues the expectation remedy raises—​both economic and moral—​are not truly joined. Whereas these issues are framed (by all sides) as concerning the law’s choice of contract remedies, they in fact concern the parties’ choice of contract terms. And whereas the assessment of the expectation remedy is supposed (again, by all sides) to turn on substantive economic or moral arguments, it in fact turns on how best to interpret party intent.

2.4.  Efficient Breach Reconsidered: Would Parties Write Contract B? Parties sometimes explicitly adopt one or the other contract type, as when they contract specifically for trade,14 or when they incorporate a conventionally explicit dual performance “take-​or-​pay” clause. But many contracts give courts less direction on the interpretive question than one might hope; hence the confusion about the theory of efficient breach. How, then, should a court construe a contractual promise whose character is not obvious? Continuing with our illustration, when should a court decide which of Contracts A or B particular parties wrote? becomes v – pLR = v – (i). Therefore, the buyer’s expected gain is invariant to the legal rule. This result is proved for the general case in Markovits and Schwartz (2011, 1959–​1970). 14 

We defend the expectation remedy as a default, not as a mandatory rule. In most jurisdictions, however, current law is mandatory: courts do not enforce contracts for specific performance. See Yorio (1989), supplemented in Thel and Siegelman (2011, 233–​241). The proposed new UCC Article 2, in § 2-​7 16(1), recommends that courts should enforce specific performance terms in commercial contracts, but no states have adopted this Article, and it is about to be withdrawn. Thel and Seigelman (2011) make two claims regarding the disgorgement remedy: (a) it is awarded frequently, and (b) it is an efficient remedy when courts cannot fully protect the promisee’s expectation. The first claim restates the expectation remedy. Thus, market damages—​the difference between the contract price and the ex post market price—​are thought to protect the expectation. A promisor who sells to a third party at the ex post market price must pay market damages to the promisee. Because these damages equal her gain, Thel and Siegelman (2011) classify market damages as a disgorgement remedy. We use the more common classification of these damages here. These authors recognize elsewhere that parties may bargain for liability rule protection or property rule protection when free to do so. See Thel and Seigelman (2009).

 

28    DANIEL MARKOVITS AND ALAN SCHWARTZ Section 1.3 argued that courts should assume parties wrote Contract B because they could conveniently have written Contract A. Economic, moral, and doctrinal considerations reinforce this argument. Contracts that are silent about remedies should be read to make dual performance promises. Beginning with the economic argument, recall our assumptions of infinite renegotiation costs and zero contract-​writing costs. Both sets of costs actually are positive, and when this is so, the common-​law expectation remedy promotes efficiency. To see why, consider the renegotiation case, in which the seller could realize a nontrading gain, and so must decide whether to trade under the contract or not. The expectation remedy allows the seller unilateral discretion to choose, subject to an incentive regime that requires her to internalize the full social costs of her choice. (To observe this simply recapitulates the conventional theory of efficient breach.) Now suppose that the buyer is entitled to specific performance or, analogously, to disgorgement of any gains that the seller captures by declining to trade. It is one thing for the buyer to possess a legal right to these gains, however, and quite another for the buyer in fact to obtain them. Even under a specific performance regime, the seller can deprive her buyer of the nontrading gain, simply by trading. Accordingly, a seller and buyer contemplating nontrading gains will enter into an ex post renegotiation of their contract, in which the seller will purchase the right to exit by offering the buyer his expectation plus a portion of the nontrading gain.15 Renegotiations are costly in general, and there is every reason to suspect that property rule-​triggered renegotiations will be especially costly. When a promisor/​seller rejects trade, a specific performance remedy forces the parties into a bilateral monopoly. Bargaining in this game is zero sum: every dollar the seller pays to the buyer is a dollar that the seller must lose, and the seller’s incentive to resist payment is increasing in the size of the payment the buyer seeks. Moreover, no outside offers from third parties exist to narrow the bargaining space or discipline the parties’ conduct. Because the promisee/​ buyer has a higher ex post payoff under a property rule than under a liability rule, the seller bargains harder, and perhaps less ethically, to avoid paying him under the property rule. Hence, Contract A is more costly to enforce than Contract B is. Commonly used contracts, that is, are insufficiently state contingent and so produce ex post inefficient outcomes with positive probability. Parties ex ante thus have a choice between writing Contract A, which will have to be renegotiated if an inefficient state materializes, and Contract B, which permits unilateral promisor exit and so needs not be renegotiated if an inefficient state materializes. When the parties’ gross return is the same under either contract and renegotiation costs are positive, parties therefore prefer Contract B, which tracks current law. 15 

A restitutionary remedy that permits the promisee to choose between trade and promisor breach plus disgorgement of breach gains would also make the promisee the residual claimant. Moreover, such a remedy might avoid renegotiation costs by empowering the promisee to compel the promisor to “breach” and pay restitution. See Brooks (2006). We focus on the traditional expectation interest remedy because it is much the more common. In addition, as we have argued elsewhere, a promise with the power to compel “breach” and disgorgement is more naturally said to own the promisor, so that contract, as a distinctive form of legal ordering, falls from view. See Markovits and Schwartz (2011, 1992–​1993).

 

(In)Efficient Breach of Contract    29 This preference is strengthened when the zero contracting costs assumption is relaxed, because Contract B is less costly to write than Contract A is. Contract B conditions only on two states of the world—​when trade is efficient and when it is not—​and need not anticipate the parties’ ex post bargaining returns. Contract A must condition on the trade and no-​trade states but also on the state when disgorgement is needed because the promisor has sold the subject of sale, and Contract A also requires the parties to anticipate and so price the renegotiation result.16 Again, because the parties’ gross gain is the same under either contract and contracting costs are lower under Contract B, parties prefer Contract B.

2.5.  Revisting the Moral Issues When parties write Contract B, the promisee trades his right to property rule protection in return for a larger expected return.17 As a consequence, nothing in the nature of dual performance promises renders them any less solemn, and nothing in the nature of the expectation remedy renders the law’s enforcement of dual performance promises any less complete than the conventional morality of promise and contract recommends. Dual performance contracts, thus, are not formally distinctive:  they announce real obligations, which promisors really recognize, even (indeed just as intently) when they honor their promises’ transfer prongs as when they trade. The same holds for the law. When a court awards expectation damages in connection with Contract B—​the dual performance contract—​the court acknowledges rather than denies the contract’s promissory obligation; and, the expectation remedy provides direct rather than substitutionary relief, constituting specific performance of one of the dual performance promise’s two prongs. Indeed, nothing in the nature of dual performance promises precludes the law’s sanctioning true breaches, in which a promisor declines both to trade and to transfer, including even through punitive damages. Moreover, dual performance promises deploy the formal morality of promise in the service of distinctively appealing substantive arrangements. Parties to Contract B recognize that they are assuming genuinely promissory moral and legal obligations. Dual performance contracts prescribe conduct for both the trade and the no-​trade states. But as long as parties respect the constraints that these prescriptions create, they are as free to pursue their respective self-​interests within these promissory relations as they were without them, including when they initially negotiated their promissory engagements. Dual performance contracts reflect negotiated settlements, which simultaneously bind 16 

The contracting cost concern is considered in detail in Markovits and Schwartz (2011, 1973–​1974). Because the parties’ gross gain is the same under either contract, but total transaction costs are lower under Contract B, this contract maximizes net expected surplus. And because the transaction cost savings are reflected in a lower price, the buyer realizes his share of those savings. Hence, the buyer does better under Contract B than under Contract A. 17 

 

30    DANIEL MARKOVITS AND ALAN SCHWARTZ the parties to a real agreement and free the parties from any further legal need or duty to renegotiate, going forward. Unlike fiduciary engagements, which, by their nature, impose open-​ended obligations of general benevolence toward the other party and thus require continuing renegotiations, contracts, properly understood, limit the obligations they impose according to the terms of the initial agreements that the parties struck. Dual performance contracts, and the expectation remedy with which they are associated, enact this central feature of contract. The promisor may remain as self-​interested within her contract as she was without it, and may consult her private advantage in deciding whether to trade or to transfer, as long only as she also vindicates her promisee’s contractual expectations. By contrast, supracompensatory remedies impress contracting parties into obligations of fiduciary benevolence and ongoing readjustment in the administration of contracts, under which promisors’ acquire open-​ended duties to consult their promisees’ interests as they administer their contracts. Supracompensatory contractual remedies thus carry with them implicit instructions concerning how contracting parties must carry out the renegotiations that such remedies inevitably engender. A promisor who applied ruthless self-​interest to these negotiations would surely fall foul of the ideals in whose name the supracompensatory remedies are proposed: a breaching promisor who uses her bargaining power to retain any share of the surplus generated by the breach continues to profit from her own “wrong” just as surely as one who simply keeps the entire ex post surplus under the traditional expectation remedy. Proponents of supracompensatory remedies must therefore condemn such hard bargaining ex post, for the very same reason they (mistakenly) condemn the “opportunistic” efficient breach that they associate with current law. Honoring the ideals that underwrite the case for supracompensatory remedies thus requires parties to renegotiate in a manner that departs radically from the norms governing their initial negotiations. And this is the essence of a fiduciary relationship.18 18  This formulation suggests that the dispute between the conventional understanding of contract and critics who seek to remake contract on a fiduciary model might be characterized in another way, emphasizing doctrines concerning performance rather than remedies for breach. Arguments that favor supracompensatory remedies by invoking fiduciary-​like principles of other regard within the contract relation propose, in effect, to revise the duty of good faith in contract performance. The duty of good faith in performance—​imposed through one of the rare mandatory rules of contract law—​has conventionally been understood to require only that promisors refrain from using vulnerabilities that promisees incurred as a result of contracting to deprive the promisees of benefits that the contracts were intended to secure. See UCC (2011, § 1-​201) (limiting good faith to “honesty in fact and the observance of reasonable commercial standards of fair dealing”). The substantive content of the duty of good faith in performance is thus conventionally fixed according to the terms of the parties’ agreement. This renders it entirely consistent with good faith for a promisor single-​mindedly to pursue her self-​interest within the contract relation, as long as she accepts the contract as a side-​constraint. Therefore, it is not bad faith for a promisor to comply with Contract B by honoring its transfer rather than its trade prong. Critics of the expectation remedy thus misapply the duty of good faith, on its conventional understanding, because it cannot be bad faith to comply with a contract. Alternatively, the critics are implicitly rejecting the conventional understanding in favor of an alternative account, on which good faith requires not just respecting the promise but an open-​ended, fiduciary-​like regard for the other party. See Markovits (2014).

 

(In)Efficient Breach of Contract    31 Contract B thus not only is more efficient than Contract A—​the contract associated with supracompensatory remedies; Contract B also is more solicitous of contract law’s core moral values. Because the dual performance contract leaves the contracting parties in control of the substance of their obligations, and, in particular, because the Contract imposes only obligations the parties assumed ex ante, the Contract better respects freedom of contract. Dual performance contracts also work better than fiduciary contracts at fashioning contractual trust in a manner appropriate to the relations among traders who approach one another at arm’s length in open, cosmopolitan economies (as even the critics of the expectation remedy accept that traders typically do). Trading partners in such economies approach their exchanges not just with divergent interests; they may also have divergent conceptions of fairness and the general interest. A large part of contract law’s moral and political appeal lies in its capacity to support respectful exchange among persons who agree on little (or perhaps nothing) besides the terms of their exchange. In order to support exchange under such circumstances of pluralism, contract law must allow trading partners to cabin their obligations as narrowly as they wish, and thus to remain, in spite of their contracts, at arm’s length in respect of all matters that their agreements do not cover. By allowing such arms’ length exchange, moreover, contract law supports a valuable and distinctive form of interpersonal respect. Exchange partners appreciate each other’s universal humanity, recognizing each other in respect of the deliberative and intentional capacities that all persons share (whatever their interests) rather than in respect of more particular, and hence more contingent, substantive interests and concerns. Precisely because it cabins contractual enforcement according to the terms of the parties’ initial agreement, the expectation remedy supports broad-​based contracting that embodies this model of interpersonal respect. A contract law that imposed supracompensatory remedies and embraced a fiduciary ideal would cast its net more narrowly and replace this conception of respect with another.19

2.6.  Contract Doctrine Instantiates Dual Performance Obligations Recall once again that though parties contract in order to realize gains from trade, sophisticated parties know that trading may not generate gains in every future state and that sometimes surplus can be increased by not trading. The parties (who wish to maximize total surplus ex ante) thus have reason to plan for both the case in which trade is efficient ex post and the case in which not trading is efficient. In a transactions costly world, for reasons rehearsed earlier, the expected contractual surplus across both states

19 

See, generally, Markovits (2011); Markovits and Schwartz (2011).

 

32    DANIEL MARKOVITS AND ALAN SCHWARTZ is maximized by agreeing that when trade turns out to be inefficient ex post, the seller may divert performance as efficiency requires, as long as she makes a transfer equal to the buyer’s valuation of trade. Observe, further, that allowing the seller to retain the entire gains received ex post from refusing to trade does not entail that she acquires the entire expected value of the nontrading surplus ex ante. The contracting parties, after all, bargain ex ante over the whole expected contractual surplus, and a contract term that increases nontrading surplus ex post must also increase contractual surplus ex ante and thus (because the term cannot by itself alter the parties’ relative bargaining powers) also increase the buyer’s surplus ex ante. This result is achieved through the mechanism of the price; the contract price falls where a contract allows the seller to avoid inefficient trade by making a transfer equal to the buyer’s expectation. The lower price allows the buyer to share in both the trading and the nontrading gains from contracting. The lower price has another consequence. Because the buyer’s gains from contracting equal her value minus the price, the lower price just contemplated increases the buyer’s expectation interest. In this way, the lower price also increases the transfer that the seller must make in lieu of trading. The contract’s price term and transfer term are thus necessarily intertwined. Put differently, contracting parties (who can anticipate not just trade but also the possibility of not trading) necessarily choose the remedies associated with the prices that they set. The transfer prong of a dual performance contract is therefore just as real as the trade prong; it is memorialized in the contract through the price term (which, once again, fixes the gain to buyers not just from trade but from transfer, because the required transfer amount equals the value to them of trade, minus the price). A contract’s price term is thus an implied-​in-​fact liquidated damages clause that establishes the liquidated amount as an alternative performance. Ordinary contracts, rightly understood, are dual performance contracts. The positive law acknowledges and incorporates this insight. To be sure, the Restatement appears superficially to take another view. Section 361 thus recites, “Specific performance or an injunction may be granted to enforce a duty even though there is a provision for liquidated damages for breach of that duty” (Restatement 1981, § 361). The first comment to this section adds that “[m]‌erely by providing for liquidated damages, the parties are not taken to have fixed a price to be paid for the privilege not to perform,” so that “[s]uch a provision does not, therefore, preclude the granting of specific performance or an injunction if that relief would otherwise be granted” (Restatement 1981, § 361cmt a). If liquidated damages are substitutionary, as the Restatement apparently suggests, then the expectation remedy should be substitutionary as well. A better interpretation of the Restatement, however, recognizes that these remarks do not reject the dual performance hypothesis but rather underestimate how frequently parties contract in the alternative. Although section 361 instructs courts not to infer a party intention to price the refusal to perform the contract’s trade terms from the “mere” presence of a liquidated damage clause, parties plainly remain permitted to price this refusal. The comment goes on to say, “there is no reason why parties may not fix such a

 

(In)Efficient Breach of Contract    33 price [at which a promisor may replace trade with transfer] if they so choose.” Further, the comment instructs courts to deny an injunction to enforce a contract’s trade terms whenever “a contract contains a provision for the payment of such a price as a true alternative performance” (Restatement (Second) of Contracts 1981, § 361 cmt b [emphasis added]). Moreover, courts applying the Restatement are receptive to the possibility that a contract might incorporate an alternative promise. Although courts will specifically enforce trade despite an express transfer term where the transfer is “intended to be mere security for the performance of the principle [sic] obligation,” they also insist that “where the sum specified may be substituted for the performance of the act at the election of the person by whom the money is to be paid, or the act done, equity will deny specific performance” (Duckwall v Rees, 86 NE2d 460, 462 [Ind App 1949] [en banc]). Courts following this reasoning deny specific performance of the trade terms to a promisee whose promisor complied with a liquidated damages clause. See, e.g., Davis v Isenstein, 100 NE 940 (Ill 1913); O’Shield v Lakeside Bank, 781 NE2d 1114 (Ill App Ct 2002); Moss & Raley v Wren, 120 SW 847 (Tex 1909). Pomeroy’s (1897) treatise on Specific Performance of Contracts thus observes: Where the parties to an agreement, whatever may be the subject-​matter or the terms, have added a provision for the payment, in case of a breach, of a certain sum which is truly liquidated damages and not a penalty—​in other words, where the contract stipulates for one of two things in the alternative, the performance of certain acts, or the payment of a certain amount of money in lieu thereof—​equity will not interfere to decree a specific performance of the first alternative, but will leave the injured party to his legal remedy of recovering the money specified in the second.20 (emphasis added)

Courts apply this general principle even when they hold that in a particular case a promisee’s right to specific performance of trade survives a liquidated damages clause. For example, a Florida case holding that “[a]‌provision for liquidated damages in a contract does not necessarily bar injunctive relief against its breach” adds, in the next sentence, “If, however, it appears that the liquidated damages clause was intended to furnish a party the alternative of performance or payment or was to be an exclusive remedy, an injunction will not be issued” (Beery v Plastridge Agency, Inc, 142 So 2d 332, 334 [Fla Dist Ct App 1962]).21 The law thus adopts the dual performance hypothesis’s understanding of the expectation remedy as a conceptual matter, even where the law concludes, with respect to a particular contract, that a promisor is not entitled to transfer in lieu of trade. Of course, our earlier economic analysis suggests that sophisticated parties typically do write dual performance contracts. Finally, a case may even be made that the law generally or at least commonly interprets contracts that do not contain liquidated damages clauses as making an appropriate

20 

See also Duckwall, 86 NE2d at 462. The court cites, among other authorities, to Restatement (First) of Contracts § 378, which is the predecessor to Restatement (Second) § 361. Id. 21 

 

34    DANIEL MARKOVITS AND ALAN SCHWARTZ transfer into an alternative performance. Consider, for example, the treatment of cases in which a contract fails explicitly to make a stipulated remedy exclusive, the promisor volunteers transfer under the remedy, and the promisee seeks more. The UCC addresses these cases in two provisions: Section 2-​7 18 directs courts to enforce liquidated damage clauses and Section 2-​7 19 directs courts to enforce clauses that limit remedies, though it adds, in Section 2-​7 19(1)(b), that “resort to a remedy as provided [by the contracting parties] is optional unless the remedy is expressly agreed to be exclusive” (UCC 2004, §2-​7 19(1)(b)). This language raises the question whether Section 2-​7 19(1)(b)’s requirement of an express exclusivity agreement for remedy limitations also applies to the liquidated damage clauses that Section 2-​7 18 regulates. The answer is that it does not: liquidated damages clauses adopted under Section 2-​7 18 are interpreted to exclude other remedies even without an express statement. To see why, begin with the language of the statute, which recites that liquidated damages are not subject to the express exclusivity requirement that governs remedy limitations. Section 2-​7 19(1) makes its exclusivity requirement “subject to … the preceding section on liquidation and limitation of damages” (§ 2-​7 19(1)). Courts have taken this direction seriously; a prominent opinion observes that a liquidated damage clause is “not a limitation on a remedy” (N Ill Gas Co v Energy Coop, Inc, 461 NE2d 1049, 1056 [Ill App Ct 1984]).22 Rather, “the concepts are separate and distinct,” so that liquidated damage clauses are governed not by Section 2-​7 19(1)(b) but rather by Section 2-​7 18, which does not “impose the additional restraints” attached to limitations of remedy (N Ill Gas Co v Energy Coop, Inc, 461 NE2d 1049, 1056 [Ill App Ct 1984]). A liquidated damage clause, the court concluded, is the promisee’s sole remedy against a promisor who rejects trade even when the clause is not made expressly exclusive. Indeed, the court reached this broad conclusion although, in the case at hand, the liquidated damages transfer constituted less than 10 percent of the value the promisee would have realized from trade. Hence, the clause had the same practical effect as a limitation of remedy (N Ill Gas Co v Energy Coop, Inc, 461 NE2d 1049, 1056 [Ill App Ct 1984]). As another court observed, “a liquidated damages clause, without evidence to the contrary, is so inconsistent with any other [d]‌amage remedy as to require a conclusion that it contemplates exclusiveness” (Ray Farmers Union Elevator Co v Weyrauch, 238 NW2d 47, 50 [ND 1975]).23 This approach—​and in particular, its insistence on the formal distinction between remedy limitations and liquidated damages clauses even in the face of the two provisions’ 22 

The court’s view is correct. A remedy limitation attempts to allocate efficiently between sellers and buyers the duty to repair and preserve the goods. The remedy aspect requires the seller to repair or replace when she can do this most efficiently; the limitation aspect puts risks on the buyer that he can best reduce by careful use. A liquidated damages clause has a different function. 23  The inclusion of the phrase “without evidence to the contrary” signals that the parties may, of course, create a contract that identifies a sum of money as merely substitutionary relief if they so choose. And some courts, latching on to the specifics of contractual language in particular cases, have given particular provisions this effect. See, also, Beck v Mason, 580 NE2d 290, 293 (Ind Ct App 1991). This possibility is entirely consistent with the dual performance hypothesis, which does not make the rule that transfer can count as performance mandatory.

 

(In)Efficient Breach of Contract    35 substantially identical practical effect—​would be nonsensical if contract damages provided only substitutionary relief. The comment to UCC Section 2-​719 observes: “Subsection (1) (b) creates a presumption that clauses prescribing remedies are cumulative rather than exclusive” (UCC § 2-​719 cmt 2 [2004]). Accordingly, if a liquidated damages clause were understood merely to identify a substitutionary remedy for the promisor’s refusal to trade, then this remedy would naturally fall under the UCC’s presumption in favor of cumulation and against exclusivity. That the remedy does not fall under this presumption, but instead is held to be exclusive even if the contract does not so state could reflect rational drafting only if the UCC understands liquidated damages clauses to be different things from remedy limitations. In the Code’s view, parties do not write liquidated damage clauses merely to identify substitutes for a promisor’s failure to perform the contract’s trade terms; rather, as the dual-​performance hypothesis proposes, parties intend compliance with these clauses to be an alternative means of performance. Consequently, a court that orders a transfer pursuant to a liquidated damage clause is directing specific relief of an obligation in the alternative. This is the basic insight achieved by courts when they observe that UCC Section 2-​ 718, which governs liquidated damages provisions, takes precedence over Section 2-​7 19’s rules concerning limitations of remedies. As another court explained, it is a “valid argument generally … that if a liquidated damage clause [is] created under Section 2.718, it … logically self operate[s]‌” so that the other remedies generally contemplated by the law are no longer available (McFadden v Fuentes, 790 SW2d 736, 738 ([Tex Ct App 1990]).24 The turn to the idea of “self-​operating” is instructive; it suggests, as under the dual-​ performance hypothesis, that the liquidated damages clause in itself identifies the content of the promisor’s contractual obligation. There is no need to feed the clause through the conventional substitutionary logic of the law of remedies in order to fix its legal effect. Indeed, there are opinions that adopt the language in which the dual performance hypothesis is cast, saying that an appropriate “provision for liquidated damages limits the non-​breaching party to a suit for specific performance of the liquidated-​damages provision” Mkt Place P’ship v Hollywood Hangar, LLC, No CA 06-​749, 2007 WL 1086573 (Ark Ct App Apr 11, 2007) (emphasis added).25 The UCC’s treatment of liquidated damage provisions is neither a narrow technicality nor an outlier (that is, a departure from the common law). Instead, it tracks the interpretive presumptions associated with the expectation remedy quite generally.26 The UCC is striking merely because it recapitulates these presumptions in a somewhat surprising arena—​surprising because of the contrast between the treatment of liquidated damage 24  The edits in the quotation indicate that the court’s statement was counterfactual because the court found that the liquidated damages clause at issue in that case was invalid for independent reasons. 25  It is commonplace for courts to hold that a liquidated damages clause forecloses claims not just for the money value of trade as substitutionary relief but also for specific performance of the promise to trade. See Gilman Yacht Sales, Inc v FMB Invs, Inc, 766 So 2d 294, 296 (Fla Dist Ct App 2000). 26  Other sections of the UCC also support this chapter’s view, at least in some measure. The chapter confines discussion of these sections to the margin because they do not go directly to the dual performance hypothesis, in its core application, so much as reveal that law decides peripheral matters in ways that are consistent with the dual performance approach. Section 2-​508 (which governs cure by

 

36    DANIEL MARKOVITS AND ALAN SCHWARTZ provisions under Section 2-​7 18 and remedy limitations under Section 2-​7 19(1)(b), a contrast that lays bare what would otherwise remain hidden. These observations naturally carry over to contracts without express liquidated damages provisions, supporting the view that the expectation remedy is not a form of substitutionary relief provided by the law but is adopted by the parties as a liquidated damage clause, whose textual expression appears through the price term and that term’s implications for surplus sharing.27 The law’s preference for the expectation remedy over specific performance, conventionally understood, is now revealed to be a direct analog to the UCC’s treatment of more ordinary liquidated damages clauses: just as the UCC treats conventional liquidated damages clauses as presumptively exclusive, so the common law treats the expectation remedy as presumptively exclusive, that is, as precluding specific performance of trade terms. And just as with respect to the UCC’s rules, the common law’s approach would be inexplicable if the expectation remedy merely gave promisees access to a form of substitutionary relief. The same intuition that underlies UCC Section 2-​7 19(2)—​that contract remedies should be cumulative—​would apply to the common law also, to reject the interpretive presumption that the expectation remedy is a promisee’s exclusive recourse when a promisor refuses to transfer. Just as with respect to UCC Section 2-​7 18, the common-​law rule that makes expectation the preferred remedy for sellers whose performance has failed to satisfy the perfect tender requirement of Section 2-​601) allows the parties to “reach an agreement as to the type of cure provided … and … [to] agree that a money allowance against the price is to be given by the seller to compensate for defects in the goods” (Lord 2010). Moreover, some courts have observed that a money allowance may be established not only by express agreement but also by trade usage. See N Am Steel Corp v Siderius, Inc, 254 NW2d 899, 904 (Mich Ct App 1977). These rules adopt a version of the dual performance hypothesis, in effect making transfer of part of the value of trade (coupled with imperfect trade) into an alternative form of performance. Admittedly, no cases were found in which transfer of the full value of the trade was treated as a “cure” for the “imperfect tender” of not tendering trade at all. In addition, the dual performance hypothesis receives support from doctrines that make it easier for sellers to recover the price than for buyers to receive specific performance of trade terms. Compare UCC § 2-​709 (2011), with id. § 2-​7 16. If contracts contained only simple promises to trade, and expectation damages were merely substitutionary, then this difference would find no justification in the basic structure of contract law but would appear, rather, to reflect a special privilege for sellers (justified, perhaps, because it is cheaper to enforce promises to pay money than to do other things). In the context of a dual performance promise, by contrast, the seller’s “action for the price” under § 2-​709 and the buyer’s “right to specific performance” under § 2-​7 16 are not perfect analogs. Compare UCC § 2-​709, with id. § 2-​7 16. Whereas sellers promise to trade or to transfer money equal to the buyer’s value of trade, buyers promise to pay the price or to transfer money equal to the seller’s value of receiving the price in exchange for the goods. The money to be transferred under the second prong of the buyer’s promise will—​for example, where the buyer has accepted the goods—​often be simply the price. Giving sellers the right to receive the price therefore does not reflect any preference for sellers or give sellers greater access than buyers to direct relief. Rather, the difference between these provisions reflects an asymmetry in the content of the parties’ substantive entitlements under their dual performance contracts. 27 

A liquidated damages clause states a number. The expectation-​interest remedy is a formula for creating a number (that is, value less price). Parties have an incentive to use the number when the formula cannot be applied because the buyer’s value is unverifiable. The parties’ goal does not change with the choice between a number and a formula, however. In both cases, the parties usually prefer to permit the promisor to choose between trade and transfer.

 

(In)Efficient Breach of Contract    37 breach of contract thus implies that contracts should be interpreted, pursuing the logic of our model, to make transfer of a sum equal to the promisee’s expectation an alternative form of performance. In taking this approach, the law interprets common agreements as being Contract Bs—​the contracts that reflect the dual performance hypothesis. The general legal treatment of the expectation remedy thus is best understood in terms of an implied-​in-​fact liquidated damages clause (fixed by the price term), along the lines suggested here. To be sure, decision makers and commentators sometimes characterize the expectation remedy as providing substitutionary relief (damages) for breach of a contractual obligation to trade. But both what courts do in administering the expectation remedy and the rules under which they do these things are best explained (indeed, understood at all) only if this remedy is seen to provide direct relief for breach of the transfer prong of a promise in the alternative.28 The law thus adopts the conceptual structure associated with the dual-​performance hypothesis, allowing the parties to treat transfer as an alternative form of performance, as that hypothesis recommends. Moreover, the law commonly adopts the interpretive presumption associated with the dual-​performance hypothesis: the rule that a liquidated damage clause is the promisee’s exclusive remedy (even though not expressly made exclusive) is understandable only if the clause establishes the transfer it specifies as an alternative form of performance; and the law’s underlying preference for the expectation remedy over specific performance makes doctrinal sense only if this remedy is understood through the model of liquidated damages.

2.7. Conclusion The expectation remedy is commonly cast as providing merely substitutionary relief of contractual promises to trade goods or services for a price. The conventional theory of efficient breach implicitly adopts this account and defends the remedy, so understood, by arguing that expectation damages provide efficient incentives for promisors to perform or to breach. Critics of the remedy, and of the theory, insist that breaching is wrong even if it is efficient. The dispute between the supporters and critics of efficient breach theory rests on a false premise, which is that efficient breaches exist. Rather, if contract terms are efficient, then breaches cannot be, because a breach is the failure to comply with a contract term. Courts suppose that sophisticated parties write efficient terms so the relevant question 28 

These are the natural explanada of functionalist and formalist accounts of law, respectively. No prominent approach to understanding law devotes substantial attention to editorial characterizations of the law. Indeed, such editorializing (even coming from courts) does not belong to the primary law at all but rather is itself a secondary effort to explain the law—​an alternative legal theory rather than a new datum for theory to explain.

 

38    DANIEL MARKOVITS AND ALAN SCHWARTZ is interpretive: what contract did the parties write? Thus, if the contract required the promisor to trade a specified item (called Contract A above), then the failure to deliver the item would be a breach; but if the contract required the promisor either to trade the item or to transfer to the promisee his expectation (called Contract B above), then the failure to trade the item would not be a breach. But the failure either to trade or to transfer would be. And neither breach would be efficient. Parties write Contract B rather than Contract A  when they are symmetrically informed and do not use the contract to induce relation-​specific investment.29 Both contracts yield ex post efficiency—​parties trade only when trade is efficient—​but Contract B is less costly to write and to enforce. Contract law effectively enacts Contract B as the default because the law protects only the promisee’s expectation. Hence, the expectation interest is an efficient remedy.30 And when a court confronts a true breach of a dual promise—​Contract B—​and awards expectation damages, the court provides not substitutionary but rather direct relief. The expectation remedy, properly understood, is thus a form of specific performance. Critics of efficient breach theory therefore are right and wrong. They are right that contract breaches violate promisee rights and may be inefficient. But they are wrong because they see breaches where none exist. Supporters of efficient breach theory also should give up. Insofar as contracts are efficient, breaches are not.

Acknowledgments The arguments in this chapter borrow from our prior work, including, Markovits and Schwartz (2011) (from a symposium in honor of the thirtieth anniversary of the publication of Contract as Promise and (2012)) and Markovits (2014).

References Adras Building Material Ltd v Harlow & Jones Gmbh, (1995) 3 Restitution L Rev 235, 240, 242, 246, 249, 261, 268. Ajaxo Inc v E*Trade Grp, Inc, 37 Cal Rptr 3d 221, 248 (Cal Ct App 2005). Anderson, Roy Ryden. 2015. “The Compensatory Disgorgement Alternative to Restatement Third’s New Remedy for Breach of Contract.” Southern Methodist University Law Review 68, n. pag. Beck v Mason, 580 NE2d 290, 293 (Ind Ct App 1991). Beery v Plastridge Agency, Inc, 142 So 2d 332, 334 (Fla Dist Ct App 1962).

29 

When information is asymmetric, parties prefer either liquidated damages or specific performance, which is Contract A. Contracts would sometimes best protect investment with penalties, but penalties are difficult to enforce. See Edlin and Schwartz (2003). 30  That the expectation remedy is efficient is unsurprising. Common-​law contract rules that have survivorship value commonly are efficient. See Schwartz and Scott (2016).

 

(In)Efficient Breach of Contract    39 Birmingham, Robert L. 1970. “Breach of Contract, Damage Measures, and Economic Efficiency.” Articles by Maurer Faculty. Paper 1705. Available at:  [Accessed 18 July 2016]. Botterell, Andrew. 2010. “Contractual Performance, Corrective Justice, and Disgorgement for Breach of Contract.” Legal Theory 16(3), pp. 135–​160. Brooks, Richard R.W. 2006. “The Efficient Performance Hypothesis.” (2006). Faculty Scholarship Series. Paper 1698. Available at:  [Accessed 18 July 2016]. Calabresi, Guido and Melamed, Douglas A. 1972. “Property Rules, Liability Rules, and Inalienability: One View of the Cathedral.” Faculty Scholarship Series. Paper 1983. Available at:  [Accessed 18 July 2016]. Cunnington, Ralph. 2008. “The Assessment of Gain-​Based Damages for Breach of Contract.” Modern Law Review 71(4), pp. 559–​560. Daily v Gusto Records, Inc, No 3:94-​1090, 2000 US Dist LEXIS 22537, at *6 (MD Tenn Mar 31 (2000). Dastgheib v Genentech, Inc, 438 F Supp 2d 546, 552 (ED Pa 2006). Davis v Isenstein, 100 NE 940 (Ill 1913). Duckwall v Rees, 86 NE2d 460, 462 (Ind App 1949) (en banc). EarthInfo, Inc v Hydrosphere Res Consultants, Inc, 900 P2d 113, 119–​20 (Colo 1995). Edlin, Aaron S. and Schwartz, Alan. 2003. “Optimal Penalties in Contracts” 78 Chi.-​Kent Rev. 33. Available at:  [Accessed 18 July 2016]. Ferriell, Jeff. 2009. Understanding Contracts. 2nd ed. LEXISNEXIS. Friedman, Daniel. 1989. “The Efficient Breach Fallacy.” Journal of Legal Studies 18(1), pp. 1–​24. Gassner v Lockett, 101 So 2d 33, 34 (Fla 1958). Gilman Yacht Sales, Inc v FMB Invs, Inc, 766 So 2d 294, 296 (Fla Dist Ct App 2000). Hillman, Robert A. 2009. Principles of Contract Law. 2nd ed. Eagan, MN: West. Kull, Andrew. 2001. “Disgorgement for Breach, the ‘Restitution Interest,’ and the Restatement of Contracts.” Texas Law Review 79, pp. 2021–​2053. Lord, Richard A. 2010. 18 Williston on Contracts § 52.28. 4th ed. (“Nature of Cure”). McFadden v Fuentes, 790 SW2d 736, 738 (Tex Ct App 1990). Markovits, Daniel. 2011. “Promise as an Arm’s Length Relation.” In: Hanoch Sheinman, ed. 2011. Promises and Agreements: Philosophical Essays. New York: Oxford University Press. Markovits, Daniel and Schwartz, Alan. 2011. “The Myth of Efficient Breach: New Defenses of the Expectation Interest.” Virginia Law Review 97, pp. 1939, 1959–​1970. Markovits, Daniel and Schwartz, Alan. 2012. “The Expectation Remedy and the Promissory Basis of Contract.” Suffolk Law Review 45, pp. 799–​825. Mkt Place P’ship v Hollywood Hangar, LLC, No CA 06-​749, 2007 WL 1086573 (Ark Ct App Apr 11, 2007). Moss & Raley v Wren, 120 SW 847 (Tex 1909). N Am Steel Corp v Siderius, Inc, 254 NW2d 899, 904 (Mich Ct App 1977). N Ill Gas Co v Energy Coop, Inc, 461 NE2d 1049, 1056 (Ill App Ct 1984). O’Shield v Lakeside Bank, 781 NE2d 1114 (Ill App Ct 2002). Pomeroy, John Norton. 1879. A Treatise on the Specific Performance of Contracts. New York: Banks and Brothers. Posner, Richard A. 1972. Economic Analysis of Law. New York: Little Brown. Ray Farmers Union Elevator Co v Weyrauch, 238 NW2d 47, 50 (ND 1975). Restatement (Second) of Contracts §344, cmt a (1981) American Law Institute.

 

40    DANIEL MARKOVITS AND ALAN SCHWARTZ Restatement §235(2). Restatement (Third) of Restitution and Unjust Enrichment § 39(1) (2011). Schwartz, Alan. 1990. “The Myth that Parties Prefer Supracompensatory Remedies:  An Analysis of Contracting for Damage Measures.” Yale Law Journal 100, pp. 369. Schwartz, Allan and Scott, Robert E. 2016. “The Common Law of Contract and the Default Rule Project.” Virginia Law Review 102, pp. 1523. Shavell, Steven. 1980. “Damage Measures for Breach of Contract.” Bell Journal of Economics 11, pp. 466–​90. Shiffrin, Seana Valentine. 2007. “The Divergence of Contract and Promise.” Harvard Law Review 120, pp. 708–​753. Shiffrin, Seana Valentine. 2009. “Could Breach of Contract be Immoral?” Michigan Law Review 107, pp. 1551–​1568. Shiffrin, Seana Valentine. 2012. “Must I Mean What You Think I Should Have Said?” Virginia Law Review 98, pp. 159–​176. Snepp v United States, 444 US 507, 515–​516 (1980). Thel, Steve and Seigelman, Peter. 2009. “Willfulness versus Expectation: A Promisor-​Based Defense of Willful Breach Doctrine.” Michigan Law Review 107, pp. 1517, Cumulative Supplement 233–​241 Thel, Steve and Seigelman, Peter. 2011. “You Do Have to Keep Your Promises: A Disgorgement Theory of Contract Remedies.” William & Mary Law Review 52, p. 1181. UCC § 1-​305(a). UCC §2-​7 19. UCC Article 2, in § 2-​7 16(1). UCC (2011, § 1-​201). UCC §2-​7 19(1)(b) (2004). UCC § 2-​7 19 cmt 2 (2004). Univ of Colo Found, Inc v Am Cyanamid Co, 342 F3d 1298, 1311 (Fed Cir 2003). X-​It Prods, LLC v Walter Kidde Portable Equip, Inc, 155 F Supp 2d 577, 658 (ED Va 2001). Yorio, Edward. 1989. Contract Enforcement:  Specific Performance and Injunctions. New York: Little Brown. 439–​448.

 

Chapter 3

EC ONOM ICS OF TORT  L AW Jennifer Arlen

3.1. Introduction Every day, people take actions for their own benefit that impose a risk of harm on others.1 They use the health and safety of others as a means to their own ends. Many risky activities benefit both the risk-​imposer and society; others only benefit the risk-​imposer, and impose a net cost on society as a whole. Thus, most societies seek to regulate, rather than prohibit, activities that create risk. Social welfare is maximized when society deters risk-​producing activities whose expected costs exceed the social benefits and induces risk-imposers to invest optimally in precautions needed to reduce risk (see generally Coase 1960; Calabresi 1970; Shavell 1980). Tort liability is one of the most potentially effective mechanisms for optimally deterring risk. When properly structured, tort liability can use the threat of liability for harm caused to deter socially harmful activities and induce optimal precautions by both injurers and victims (e.g., Shavell 1980; Landes and Posner 1980, 1987). The central aim of the economic analysis of law is to identify the rules governing liability, damages, and procedure that induce optimal investment in risk reduction and optimal activity levels. Yet in order to determine the structure of optimal tort liability, it is important to identify the central challenges that tort liability is designed to address. Tort liability is far from costless. The tort system itself imposes social costs in the form of litigation costs, the fixed costs of a court system, and the costs associated with liability-​induced distortions in incentives (see, generally, Calabresi 1970). It also is not the only mechanism available to deter people from creating excessive risk of harm. Others mechanisms include private ordering, ex ante regulatory duties and monitoring, and ex post regulatory enforcement. 1  Jennifer Arlen is Norma Z. Paige Professor of Law, New York University School of Law. I benefited from the helpful comments of Giuseppe Dari-​Mattiacci, Mark Geistfeld, Jennifer Reinganum, Francesco Parisi, and Abraham Wickelgren, as well as from the financial support of the D’Agostino/​Greenberg Fund of the New York University School of Law.

 

42   JENNIFER ARLEN This suggests that, in order to be optimal, tort liability should be employed to induce optimal precautions only when it is a welfare-​enhancing alternative or supplement to alternative mechanisms, such as private ordering or regulation (see, e.g., Coase 1960; Calabresi and Melamed 1972; Epstein 1976; Spence 1977; Shavell 1980, 1984; Landes and Posner 1987; Polinsky 1980). Thus, economic analysis of torts must address both the optimal domain and structure of tort liability. Indeed, these two inquiries are inextricably linked. Tort liability is needed in certain circumstances and not others, and these circumstances determine the goals that optimal liability rules should be designed to achieve. Tort liability thus is not justified by the presence of risk-​producing activities alone. Instead, tort liability is justified by the combination of risk-​producing activities and substantial information and transactions costs. Tort liability is not needed when information is costless and transactions costs are low because, in this situation, private ordering will induce optimal risk-​taking. When transactions costs are low and parties are perfectly informed, private bargaining or market forces should induce optimal precautions (Coase 1960; Calabresi and Melamed 1972; see, e.g., Spence 1977; Shavell 1980; Polinsky 1980). Private ordering will not suffice when transactions costs are high (Coase 1960). But even here tort liability may not be needed when everyone is perfectly informed. In this situation, the state may be more effectively able to induce the desired precautions through regulation: it could specify the optimal precaution, and use its costless observation of actual conduct to intervene to sanction noncompliance (e.g., Kaplow 1992; Shavell 1984; Shavell 2014; see Epstein 2013). Tort liability is needed, and is potentially superior to regulation, when information is sufficiently costly that regulators cannot optimally determine and specify the optimal precaution for each activity ex ante, adequately monitor conduct to ensure compliance, or detect (and sanction) breach ex post (e.g., Kaplow 1992; Shavell 1984, 2014; see Epstein 2013).2 The conclusion that tort liability is needed to optimally deter risk when information costs are substantial has two important implications for the economic analysis of tort liability. First, it reveals that economic analysis of optimal tort rules generally should 2 

This discussion only touches on the considerations that determine the use of liability versus regulation. Other considerations include potential insolvency, interest group capture, and differences between private and social incentives to sue. Whatever are the considerations affecting the decision to use tort liability, these considerations should be taken into account when determining the optimal structure of tort liability and the features of the models used to analyze it. One potential consideration that is not seriously addressed in this chapter is the fact that tort liability provides compensation to victims (Shavell 1987, 206–​27). There are several reasons not to focus on this distinction. First, it is not an essential difference: regulators, in theory, could (and in some cases do) impose restitutionary sanctions in order to compensate injured victims. Second, when negligence functions optimally, injured victims are not compensated in equilibrium because injurers take due care (Shavell 1980; but see Arlen and MacLeod 2005a). Third, tort liability generally cannot be justified primarily on compensation grounds if lower cost first-​party insurance is readily available, because liability is a very costly and time-​consuming way of getting compensation to victims. What justifies tort liability, instead of insurance or no-​fault, is deterrence. When we look at deterrence-​based arguments for liability, information costs (and related considerations) tend to lie at the foundation of arguments for imposing tort liability in addition to (or instead of) regulation.

 

ECONOMICS OF TORT LAW    43 employ economic models where information is costly. Second, it reveals that the goals of optimal tort liability should extend beyond optimal precautions (and activity levels), as classically understood. In many situations, people undertaking risky decisions are imperfectly informed about the available precautions and their costs and benefits. They can, but may not, invest in obtaining information needed to improve their decisions. In these situations, a central purpose and effect of tort liability should be to induce optimal acquisition of information needed to identify the optimal level of care. In addition, when information is costly, people conducting complex activities that naturally require multiple participants are more likely to conduct these activities through organizations—​generally corporations—​than through contracts (Coase 1937). In this situation, information costs alter the scope of optimal tort liability by introducing an additional player, the organization, whose incentives and capacity to influence care must be considered and guided toward optimal decisions. Tort liability also can enhance deterrence by providing risk-​imposers incentives to make decisions regarding care within and through an organization, when this would reduce total expected accident costs (including the expected cost of information acquisition). This chapter presents economic analysis of optimal tort liability focusing on the use of liability where information is costly, and private parties cannot provide optimal incentives through private ordering. It shows that the central motivations for tort liability—​ imperfect contracting and costly information—​alter its optimal structure and incentive effects, both as applied to individual injurers and risk-​imposers operating within organizations. Section 3.2 examines optimal tort liability in the context where liability is most obviously needed: where injurers and victims are “strangers,” in that they cannot plausibly bargain over optimal risk prevention or responsibility for losses. This section begins with the foundational analysis of accidents between strangers when all parties possess perfect costless information. This analysis shows that many liability rules—​negligence (with and without contributory or comparative negligence) and strict liability with contributory negligence)—​induce optimal behavior by both potential injurers and victims (e.g., Brown 1973; Shavell 1980; Landes and Posner 1987). It then examines optimal liability when information is costly: when injurers or courts must incur costs to determine optimal (or actual) precautions and can do so only imperfectly. This section examines how information costs affect the goals of optimal deterrence, the optimal structure of liability and damages rules, and the effect of liability on injurers’ and victims’ incentives to deter risk optimally. Section 3.3 examines tort liability when injurers and victims are in a market relationship, as is the case with product liability and medical malpractice. This part identifies the factors justifying the liability in market settings. The analysis of strict liability focuses on product liability; the analysis of negligence liability focuses on medical malpractice. Again, information costs are shown to affect the optimal structure and incentive effects of liability and damage rules. Section 3.4 examines the argument that liability in market settings need not be imposed by the state because it will be optimally supplied by contract. It shows that this

 

44   JENNIFER ARLEN argument generally is not correct because information costs and other impediments to optimal contracting will usually render contracting over liability inefficient. Finally, Section 3.5 extends the analysis of optimal tort liability to the situation where risk-​imposers respond to information costs by conducting their activities through organizations, such as corporations. It shows how liability should be structured to induce optimal behavior by organizations and the agents who work for them, and it discusses limitations with existing liability rules on these issues.

3.2.  Accidents between Strangers This section examines the incentive effects and the optimal structure of tort liability when the injurer and victim are “strangers” with no market relationship and no ability to contract or bargain with each other. This section first presents the classic model of liability with perfect information and zero litigation costs, and then it examines tort liability when information about optimal and actual precaution is costly.

3.2.1. Classic Model of Accidents between Strangers: Full Information Many, if not most, accidents involve two parties, each of whom can take precautions that affect the probability that the accident occurs and/​or the magnitude of the harm caused. These accidents fall into one of two categories. The most common are “bilateral-​risk” cases, where two parties each undertake an activity that presents a risk of harm to themselves and the other. Automobile accidents involving either two cars, or a car and a pedestrian, are classic examples of bilateral risk accidents (Diamond 1974; Arlen 1990a, b; 1992). The other standard type of case involves “unilateral-​risk and bilateral-​care” accidents (hereinafter bilateral care): where one person (the injurer) undertakes an activity that imposes a unilateral risk of harm on the other (the victim), but both parties can affect the probability that the accident occurs (Brown 1973; Shavell 1980; Landes and Posner 1987). Product defects that injure nonconsumers and activities that produce environmental risks are examples of potential torts with well-​defined potential injurers and victims where both parties may be able to affect the probability or magnitude of the harm, but only one side imposes a risk of harm on the other (absent liability). In both types of cases, social welfare is maximized when both care and activity levels are optimal. Specifically, every risk-​imposing party should undertake the activity only when, and up to the point that, the social benefit of the activity equals or exceeds the expected social cost. In addition, when they conduct the activity, they should invest optimally in precautions to reduce the risk of harm to others or themselves (or both).

 

ECONOMICS OF TORT LAW    45 In each type of case, the motivation for tort liability is the same: rational, self-​ interested, utility maximizers will not behave optimally absent liability because they set care and activity levels without considering to the risk of harm to others. Tort liability can induce risk-​imposers to take optimal care and activity levels by making them bear the social costs of the accidents they cause. This is easily accomplished by holding injurers strictly liable for all harms caused. Yet in order to induce optimal care, tort liability also must ensure that victims invest in optimal precautions to protect themselves. Victims will not do this if all accidents are paid for by injurers (Brown 1973; Shavell 1980). To induce optimal care by both parties—​whether by two injurers in the bilateral risk case or the injurer and a victim in the bilateral care case—​tort liability must be structured to ensure that each expects to bear expected accident costs if they fail to take optimal care. This will ensure existence of an equilibrium in which each takes optimal care. Optimal tort liability rules that satisfy this requirement include pure negligence, negligence with contributory (or comparative) negligence, and strict liability with contributory negligence (Brown 1973 [bilateral care]; Shavell 1980 [bilateral care]; Arlen 1990a [bilateral risk]; Arlen 1992 [bilateral risk]; Cooter and Ulen, 1986 [comparative negligence]). By contrast, strict liability with full compensation damages will not induce optimal care by both injurers and victims because it insulates victims from expected accident costs, eliminating their incentive to take due care. These conclusions hold whether injurers impose risk unilaterally on victims (Brown 1973; Shavell 1980; Landes and Posner 1987) or both parties to the accident impose risk on each other (Diamond 1974; Arlen 1990, 1992).

3.2.1.1. Formal Analysis: Bilateral Care The classic model assumes that information and litigation is costless; expected accident costs depends on a single input by each party, “care,” given by x for injurers and y for victims. The cost of care is given by C(x) and c(y), respectively, with C′(x), c′(y) >0; and C″(x), c″(y) > 0. Expected accident costs are given by p(x,y)H, where H is the harm to the victim and 1 > p(x,y) > 0 is the probability that the harm occurs, where pi(x,y) < 0, pii(x,y) > 0, i = x, y. All parties and the court know the optimal level of care for each party ex ante; each party’s actual level of care is verifiable ex post. Social welfare is maximized when injurers and victims invest in the level of care (x and y, respectively) that minimize expected accident costs:

C(x ) + c ( y ) + p(x , y )H (1)

This implies that social welfare is maximized when injurers take care x* and victims take care y* as given by:

C ′(x ) = − px (x , y )H (2) c ′( y ) = − p y (x , y )H (3)

 

46   JENNIFER ARLEN Thus, each party should invest in care up to the point where the marginal cost of care equals the social marginal benefit of care, where the latter is given by the reduction in expected accident costs resulting from a marginal increase in care. Absent tort liability, injurers will not take optimal care. Injurers minimize their own expected accident costs, given by C(x), and thus take no care. Victims do take optimal care (given injurers’ care levels) because victims bear their own accident costs and thus select care to minimize c(y) + p(0,y)H.3 As a result, they select care to minimize the total private (and here social) cost of accidents, given the injurer’s suboptimal investment in care. Nevertheless, their care-​taking is not first best. If the marginal reduction in the probability of an accident resulting from victim care-​ taking is lower when injurers take no care than when they take optimal care, then victims will take less care than is first-​best optimal because injurers under invest in care. By contrast, all of the standard liability rules, except pure strict liability, can induce injurers and victims to take due care. Pure strict liability is inefficient because it produces the mirror result of no liability. Under strict liability, each injurer bears all expected accident costs; but each victim does not bear any of his own damages, assuming that damages equal the victim’s harm, H. In this case, each victim minimizes c(y) + p(x,y) (H–​D) =  c(y). Victims take no care because they have no reason to spend money to avoid accidents that injurers inevitably pay for. Each injurer thus selects care to minimize C(x) + p(x,0)D = C(x) + p(x,0)H, and thus selects the level of care at which: C ′(x ) = − px (x , 0)H (4)



The injurer will select the level of care that is optimal given the victim’s suboptimal behavior (i.e., second best optimal care), but will not select first-​best optimal care, x*. Under pure negligence, injurers are liable only if they fail to take “due care.” Thus, an injurer can, by selecting care x* instead of zero care, avoid all liability for the victim’s expected accident costs. We know that an injurer who expects the victim to take due care will do so as well. The definition of optimal care implies that C(x*) + p(x*,y*)H < C(x) + p(x,y*)H which in turn implies that the injurer incurs lower expected costs if he takes optimal care, C(x*), than if he does not and faces C(x) + p(x,y*)H. A victim who expects the injurer to take due care knows he must bear his own accident costs. He thus faces expected cost of c(y) + p(x*,y)H and takes optimal care (Equation (2)). This equilibrium is unique. An equilibrium will not arise in which both parties decide to be negligent. Both parties cannot find it in their best interests to be negligent because we know that C(0)+ c(0) + p(0, 0)H > C(x*) + c(y*)+ p(x*,y*)H, thus it cannot be the case that the injurer and victim would both be better off being negligent. Under negligence with contributory negligence, injurers are liable only if they failed to take due care and the victim took due care. Victims bear losses whenever injurers took due care or the victim did not. This rule also induces optimal care by both parties. 3 

In the bilateral-​risk case, where each party can injure the other and be injured by them, then neither party takes optimal care.

 

ECONOMICS OF TORT LAW    47 Victims will always take due care because they are better off taking due care whether they expect injurers to do so. If they expect injurers to be negligent, then victims will take due care in order to shift their own expected accident costs to injurers. If they expect injurers to take due care, then victims will take due care because they bear the full costs of accidents, and we know that their expected costs c(y) + p(x*,y)H are minimized when y = y*. Injurers in turn will be nonnegligent because they know victims will take due care. In this case, injurers face expected costs of C(x) + p(x,y*)H if they are negligent and expected costs of C(x*) if they take due care. Thus they take due care to minimize their costs. Negligence with comparative negligence also yields an equilibrium where both parties take optimal care (Cooter and Ulen 1986). Strict liability with contributory negligence also can induce optimal care by both parties. In this situation, it is the victim who always has incentives to invest in care (no matter what he expects the injurer to do), because the victim, by taking due care, shifts all accident costs onto the injurer. The injurer in turn will take due care because he expects to bear expected costs of C(x) + p(x,y*)H, which are minimized when he takes optimal care (see Equation 2).4

3.2.1.2. Discussion and Analysis Several features of this model and equilibrium are worth noting. First, under the assumptions of this model, negligence liability (with or without contributory negligence) would be the lowest cost rule were there are litigation costs, because injurers always take due care, and thus no one ever sues anyone. Therefore, negligence accomplishes its task without a single tort action ever being filed (Shavell 1980). Second, and related, in this model, all negligence is deliberate: no one ever fails to take due care accidentally (compare with Arlen and MacLeod 2003, 2005a). Third, strict liability only induces optimal care when damages precisely equal the harm caused because strict liability in effect sets a price on risk-​taking. The rule provides optimal incentives only when the expected price equals the expected social cost of accidents (Cooter 1984). By contrast, negligence can induce optimal care (by both parties) even when damage awards deviate from the social cost of harm, H. Excessive damages do not distort care-​taking when everyone is perfectly informed because excessive damages simply provide injurers with additional motivation to take due care. They do not incentivize excessive care. Injurers take due care leaving victims with expected costs of c(y) + p(x*,y)H, which is minimized at y*. Damages less than H also can induce optimal care as long as damages equal or exceed the level at which C(x*) < C(x) + p(x,y)D for all x 0, q″(e) < 0. Injurers who are not informed select suboptimal precaution. The cost of expertise is given by C(e), where C′(e) > 0, C″(e) > 0. Investment in expertise is unverifiable. We assume that ex post the court can accurately determine both optimal care and the injurer’s actual precaution. Social welfare is given by Equation (5), assuming that uninformed injurers always select the suboptimal precaution x0:

b − q(e) {c(x i ) + p(x i )H } − (1 − q(e)){c(x 0 ) + p(x 0 )H } − C(e ) (5)

Social welfare is maximized when two conditions are met. First, informed injurers (who exist with probability q(e)), should select the precaution that minimizes the total cost of accidents, c(x) – ​p(x)H. Thus, they should select optimal precaution, x*. Second, injurers should invest optimally in expertise. Social optimal expertise is the expertise that maximizes social welfare, assuming that informed injurers select optimal precautions and uninformed ones do not. It is given by the expertise such that:

q ′(e){[ p(x 0 ) − p(x * )]H − [c(x * ) − c(x 0 )]} = C ′ (e ) . (6)

Absent liability, the potential injurer does not bear any accident costs. Accordingly, he takes too little care and under-invests in information. Indeed, given that he would take

 

54   JENNIFER ARLEN no precaution even if informed, the injurer will not invest in information about precautions at all. Consider now the effect of negligence on both the ex ante decision to invest in expertise and the precaution decision. In most circumstances, courts can observe precautions but cannot observe expertise. Thus, one form of precaution is directly governed by negligence liability, but the other (expertise) is not. Nevertheless, negligence liability can induce both optimal precaution and optimal investment in expertise. Consider first the incentives of an informed injurer to select precaution under a negligence regime. The informed injurer will select the precaution that minimizes expected costs:

c(x i ) + p(x i )D i , (7)

where Di is the damage award, with i = *,0; D* = 0 and D0 = H. We know from previous analysis that an informed injurer facing damages of H for losses caused by negligence will minimize expected costs by selecting optimal care, x*, thereby avoiding liability. As for expertise, the injurer invests in the level of expertise that maximizes his expected welfare, assuming that he selects optimal care when informed. Thus, he selects expertise to maximize

b − q(e) {c(x * )} − (1 − q(e)){c(x 0 ) + p(x 0 )D} − C(e). (8)

He thus sets expertise such that the marginal cost of expertise equals the net marginal benefit of expertise, where the latter equals the reduction in liability minus the added cost of selecting optimal care instead of negligent care (accidentally):

q′(e){ p(x 0 )D − (c(x * ) − c(x 0 ))} = C ′(e). (9)

Accordingly, the injurer will invest optimally in information and e​ xpertise so long as the expected damage award for accidental negligence, p(x0)D, precisely equals the benefit to victims of expertise: p(x 0 ) − p(x * )H . Actual optimal damages for accidental harm thus equal:

D =

( p0 − p* )H < H. (10) p0

This implies that the optimal damages for injuries resulting from a knowing decision to select suboptimal care exceed the optimal damages for accidental negligence, even when liability is imposed for all injuries attributable to injurer negligence. By contrast, setting damages equal to H for both types of cases induces excessive investment in expertise (Arlen and MacLeod 2003; 2005).

 

ECONOMICS OF TORT LAW    55

3.2.2.2.3. Formal Analysis: Strict Liability Unlike negligence, strict liability can induce both optimal precaution and optimal investment in information when damages equal the harm caused in all cases. We know that damages equal to H suffice to induce informed injurers to select the optimal precaution. Thus, under strict liability with full compensation damages the injurer faces expected costs of:

b − q(e) {c(x * ) + p(x * )H } − (1 − q(e)){c(x 0 ) + p(x 0 )H } − C(e) (11)

A comparison of Equation (11) and (5) reveals that injurers expected costs of care and expertise equal the social costs; thus they invest optimally in expertise and care.

3.2.2.3. Courts Err When Determining Optimal Care The preceding analysis assumed that courts are perfectly and costlessly informed. But courts also must incur costs to obtain information, and regularly are only imperfectly informed. This affects both the cost and the accuracy of the litigation process.10 The possibility that courts will be imperfectly informed affects both the relative efficacy of strict liability and negligence and the impact of negligence on incentives to invest in precautions. Strict liability can provide optimal incentives, even if courts have no information on optimal care, as long as courts set damages correctly and the parties have (or can optimally obtain) information on optimal care (Cooter 1984). By contrast, negligence is potentially (but not always) vulnerable to court error in determining optimal care (Cooter 1984; Craswell and Calfee 1986) and damages (Arlen and MacLeod 2005a). To assess the impact of error, we must first identify more precisely the type of court error, as well as whether the affected party can predict the court’s decision. Assume first that courts err in determining due care, but injurers accurately predict the error. If courts predictably set due care too low, then injurers also will select suboptimal care, assuming damages are set correctly (Shavell 2007, 161). This under-​investment in care also can affect other precautions. For example, when due care is suboptimal, injurers may under-​invest in expertise—​and obtain it on the “wrong” precautions (from society’s perspective) because injurers gain nothing by determining optimal care. Thus, if damages are set second best optimally, then injurers will select suboptimal precaution when informed and will have an excessive risk of accidental error. If, by contrast, courts set due care too high, court error will not necessarily induce excessive care. Should courts set due care so high that the cost of compliance exceeds the reduction in expected liability, then injurers will not try to comply. In this case, 10  Ex ante information costs do not inevitably translate into costly or even inaccurate adjudication. After all, if the parties invest optimally in information about optimal precautions ex ante, ex post adjudication would not entail additional costs to obtain information if the parties can credibly share their information about optimal care and information on actual conduct is readily available. Nevertheless, often the information available to courts is not only costly but also imperfect.

 

56   JENNIFER ARLEN negligence in effect becomes strict liability: injurers will take optimal care if damages equal the harm caused (Shavell 2007, 161). Injurers also will invest optimally in information and expertise, provided damages equal the harm caused (see Section 3.2.2.2.3). By contrast, if courts set due care too high, but not so high that injurers would prefer to be deliberately negligent, then injurers will take excessive care (Shavell 2007, 161). This excessive care will also distort investment in information, causing injurers to seek information on due care (excessive care), not optimal care. It also will lead to less investment than is optimal because the injurer gains less marginal benefit from taking due care when due care is set too high than does society when care is set optimally. Alternatively, courts may err in setting damages. Strict liability is very vulnerable to error in setting damages because liability in effect imposes a price that the injurer must pay for his activities. Charging a price less than the social cost will induce too little care and excessive activity levels (Cooter 1984). By contrast, the traditional analysis finds that negligence is less vulnerable to errors in setting damages. As long as damages exceed the damages at which the injurer is better off taking optimal care—​which is the D at which c(x*) = c(0)+p(x*)D—​then negligence will induce due care. Damages thus often can be less than H and still induce optimal care. Damages also can substantially exceed H without inducing excessive care-​taking because injurers can avoid all expected liability by taking due care (Cooter 1984). These conclusions do not hold, however, when injurers can be negligent accidentally and can reduce the risk of accidental negligence by investing in expertise. In this situation, all errors in setting damages affect expected accident costs through the impact on expertise. Negligence liability holds injurers liable for accidental negligence. In effect, this operates to “price” the injurer’s decisions to invest in information. As we have seen, this price must be set precisely equal to the social marginal benefit of expertise in order to induce optimal expertise (Arlen and MacLeod 2003, 2005). Finally, courts may err in setting the standard of care in ways that cannot be predicted ex ante. In this case, under negligence liability, court error may induce injurers to take excessive care, even if, on average, courts set due care equal to optimal care (Calfee and Craswell 1986; see Diamond 1974). To see this, assume that injurers expect the due care to be set equal to x * ± ε , where ε is an error term. Assume that half the time ε > 0 and the other half ε < 0. In this situation, negligence liability will induce each injurer to take excessive care. An injurer who takes excessive care x* + ε incurs unnecessary expected care costs since there is a 50% probability the court will set care too low: .5{c(x* + ε )–​ c(x* – ε​ )}. By contrast, when the court sets due care too high, injurers obtain a large benefit from taking excessive care: they can avoid any liability for injuries caused. Whenever the expected liability savings exceeds the marginal increase in care, injurers will over-​invest in care: p(x* –​ ε )D > c(x* + e)-​c(x* – ​e) (Calfee and Craswell 1986). The conclusion that court error induces excessive care-​taking does not hold in all situations, however. Negligence may not induce excessive care—​or the effect will be less pronounced—​in bilateral care cases that include a defense of comparative negligence, provided the uncertainty or risk of error affects both parties (Cooter and Ulen 1986;

 

ECONOMICS OF TORT LAW    57 see Dari-​Mattiacci and Hendriks [2014], finding comparative negligence is welfare-​ enhancing when due care is set above optimal care). In addition, negligence may not induce excessive care if courts accurately apply causation rules (Kahan 1989). The preceding analysis assumes that small deviations between the injurer’s level of care and due care can have large consequences—​causing the injurer to bear liability of H for all harms caused. Under negligence liability, injurers are not liable for harms caused unless it is more likely than not that, had the injurer taken due care, the injury would not have occurred. Accordingly, if the court sets due care at x*+ε , but the defendant only took care, x*, he nevertheless will not be held liable unless the plaintiff can show that it is more likely than not that the injury would not have happened if the injurer took due care (x*+ε ) rather than allegedly negligent care, x*. In other words, if we consider all the ways the harm could have happened given care x*, more than half the risk of harm would need to be removed by an additional marginal investment of ε . Given that x* is optimal and the marginal benefit of care is declining, in most circumstances, failure to take excessive care rather than optimal care will rarely be the legal cause of the harm, unless courts err by a wide margin. To the extent that injurers anticipate this, uncertainty will not induce them to take excessive care (Kahan 1989; see Grady 1983; see, generally, Hylton 2013, 100–​103).

3.2.2.4. Implications for Future Research Comparing Section 3.2.2 and Section 3.2.1 we see that tort liability is more effective at minimizing the expected cost of accidents the lower the cost to injurers, courts, and victims (see infra Section 3.2.4) of determining both optimal and actual care and the greater the marginal benefit of information relative to the marginal cost. This suggests that a productive avenue for scholarship and tort reform includes mechanisms for improving information flows both into the tort system, as well as out of the tort system, to those making decisions about due care and whether to litigate (e.g., Daughety and Reingaman [2005], discussing secrecy and settlement).

3.2.3. Risk-​Averse Victims When victims are risk averse but injurers are not, the social cost of accidents is higher when victims bear their own losses than when injurers do. This can affect optimal liability rules. By definition, a risk-​averse person would prefer to receive $X with certainty than to accept a risky bet with an expected value of $X. This implies that the social cost of accidents is higher when losses fall on a risk-​averse person than when they fall on someone who is risk neutral because an accident introduces variance in expected outcomes. Thus, when injurers are risk neutral (e.g., large corporations) and victims are risk averse, tort liability can reduce the social cost of accidents simply by shifting losses from victims to injurers, all else equal. This risk-​spreading benefit potentially provides an argument for favoring strict liability (with contributory negligence) over any form of negligence

 

58   JENNIFER ARLEN liability, as strict liability provides victims compensation for losses even when injurers took due care (Shavell 1987, Chapter 9). By contrast, when victims and injurers are each risk averse—​as with automobile accidents—​then tort liability cannot increase or decrease expected social welfare by shifting the loss from one party to the other. Nevertheless, even when only victims are risk averse the social benefit of using strict liability to insulate victims from accident costs purely for risk-​spreading reasons may not be as large is it might at first appear. First, tort liability is not the only mechanism for insulating victims from expected accident costs. First-​party insurance, such as health insurance and life insurance, provides coverage against many costs associated with accidents more quickly and with lower administrative costs compared with tort liability (see, e.g., Weiler 1991). Insurance is particularly attractive because even under strict liability, victims generally would have to purchase first-​party insurance. First, victims inevitably face a risk of suffering harms for which they will not receive tort compensation (e.g., because the harm was caused by an act of nature, causation cannot be shown, or the injurer is judgment proof). Second, tort recovery rarely arrives in time to cover the mounting medical bills of a seriously injured person. Thus, even with tort liability, victims need to purchase first-​ party insurance, the cost of which may not be significantly lower under strict liability than under negligence if the risk of non-​tort-​based accidents and the cost to insurers of pursing claims against injurers is high (see Shavell 2004, 268–​269; see also Epstein [1985] for additional discussion of limitations of tort liability as insurance). Thus, while victims can benefit from the risk-​spreading benefits of tort liability, in practice, the risk-​spreading benefits of strict liability versus negligence may not be sufficiently large, on net, to justify favoring the former or the latter independent of the deterrence benefits of liability.

3.2.4. Litigation Costs and Settlement Litigation costs also affect the optimal choice of liability rule. As between several equally effective rules, the optimal liability rule is the one that optimally deters at the lowest cost (in terms of administrative and litigation costs). Strict liability (with or without contributory negligence) would appear to be more costly because it generates more litigation (Posner 2011). After all, in a perfect world, no one sues under negligence because injurers always take due care (Shavell 2007, 155). Nevertheless, once we expand the analysis to include information costs and other imperfections, injurers will fail to take due care even under an optimal negligence rule (see supra Section 3.2.2). In this case, relative administrative costs are less clear. Negligence generates fewer suits than strict liability does. Nevertheless, each negligence action may be substantially more expensive because each party may incur enormous expenditures to establish optimal or actual care.11 Negligence also may reduce 11 

This conclusion may not hold if we compare negligence with strict liability with contributory negligence if courts applying the latter must calculate optimal care by injurers to determine optimal

 

ECONOMICS OF TORT LAW    59 parties’ willingness to settle if parties are more likely to have divergent expectations of outcomes under negligence (Spier 1997; Shavell 2007, 155; Wickelgren 2013).12 Finally, negligence liability may undermine optimal deterrence if victims cannot easily determine ex ante whether injurers took due care. In this case, victims may be less likely to sue under negligence than under strict liability, even when litigation would be publicly and privately optimal under full information, because litigation costs exceed expected recovery when victims are uncertain. In this situation, injurers who fail to take care may face a lower risk of liability under negligence than strict liability, leading them to take too little care.13 This result is particularly likely if victims know they were injured, but cannot identify who injured them without paying for an investigation (as with many environmental wrongs). Accordingly, reforms that increase victims’ upfront litigation costs—​as arbitration appears to do14—​may undermine the deterrent effect of the tort system if it reduces litigation and expected damage awards below the level needed to induce injurers to take optimal care.

3.3.  Products Liability and Medical Malpractice Much of the risk of harm that people face is not imposed by strangers. Instead, people often, in effect, invite risk of harm into their lives through the goods and services (e.g., medical care) that they purchase, seeking to improve their lives. This situation differs from strangers cases in two respects. First, when potential injurers sell risky goods and services to consumers, both producers and consumers are better off ex ante when potential injurers are able to credibly commit that they will invest optimally in care. Optimal care maximizes the value of the product to the consumer and

care by victims, as is the case in the traditional model (Shavell 1987, 16–​17), but this might not be the case when courts can optimally set victim’s due care without calculating the injurer’s optimal care. For example, courts may optimally conclude pedestrians should not jaywalk or rush out into traffic, without determining the optimal care of drivers. 12   Negligence also may increase the risk of underdeterrence if information to determine the injurer’s care is costly and victims rationally fail to incur this cost, even when it would be optimal for them to do so, because the harm is one that could have occurred even absent negligence. 13  Tort liability could induce optimal litigation in this situation through supercompensatory damage awards. Yet actual tort damages rules for accidental injuries are not supercompensatory. Punitive damages tend to be reserved for harms that result from deliberate (or wanton and willful) breaches. 14  In litigation, the victim often can hire a lawyer on a contingency fee basis and does not pay for the judge. By contrast, under arbitration, each party not only must pay his own litigation costs, but also must pay the arbitrator. Thus, even when arbitration reduces each person’s own lawyer’s fees by shortening the process, it may increase the victim’s litigation costs by requiring him to pay for a decision maker who otherwise would be provided by the state. This may reduce incentives to sue, particularly if arbitrators tend to award lower expected damages.

 

60   JENNIFER ARLEN thus the consumer’s willingness to pay. Second, in this context the justification for tort liability does not root in traditional transactions costs: transactions costs generally are low since the producers and consumers are in a market relationship (Spence 1977; Polinsky 1980). Instead tort liability is needed when information costs render markets inefficient (Spence 1977) and impediments to optimal contracting render contracting over liability inefficient (Arlen 2010). This section examines the justifications for and optimal structure of tort liability to govern these market relationships. It highlights the importance of victim information costs in both justifying liability in the first place and victim ​or injurer information costs in justifying the use of mandatory liability instead of contractual liability. Specifically, this section shows that tort liability is preferred to contractual liability when, as is often the case, the parties cannot contract optimally because consumers lack the required information (e.g., Geistfeld 1994) or contracting is plagued by inefficiencies, such as collective goods, adverse selection, or time-​inconsistency problems largely arising from injurers’ imperfect information (Arlen 2010). Tort liability thus emerges as a potentially valuable supplement to contract, enabling the parties to reduce inefficiencies produced by information costs.

3.3.1. All Parties Are Perfectly Informed Informed producers of goods and services (e.g., medical care) will invest optimally in precautions that affect the risk of harm to consumers or patients if purchasers are perfectly informed about either investments in care or the expected accident costs associated with each individual injurer (Spence 1977; Polinsky 1980). Tort liability is not needed. The intuition behind this result is straightforward. Consider producers operating in a competitive market making goods that are identical on all dimensions except expected accident costs. It might appear that producers would seek to minimize their expected costs by not investing in care. But producers know that consumers want to purchase the product with the lowest total expected cost to them, which includes the expected accident costs associated with the product. Accordingly, using the notation from Section 3.2, consumers select the producer of the product with the lowest P + p(x)H = c(x) + p(x)H, where P is the competitive market price (which equals the marginal cost of the product, c(x)). A producer seeking to minimize the total cost of his product thus must invest in care to minimize the consumer’s total expected costs. He thus selects care to minimize c(x) + p(x)H, which is optimal care (see Section 3.2.1) (Spence 1977; Polinsky 1980; see Daughety and Reinganum 2013, 70–​73). Activity levels also are efficient. Activity levels are efficient when producers only produce up to the point where the social marginal benefit of the last unit made (as given by the marginal consumer’s willingness to pay for that unit) equals the marginal cost to society of making the product (c(x) + p(x)H). When consumers are informed, each consumer will only demand the product if his willingness to pay equals or exceeds the

 

ECONOMICS OF TORT LAW    61 cost of the product to him, which equals c(x) + p(x)H. Thus, activity levels are optimal (see Polinsky 1980; Shavell 1980).

3.3.2. Product Liability When Consumers Cannot Observe Quality Consumers of products and purchasers of medical care rarely can accurately determine the expected risk associated with purchasing a good or service from a particular producer. Consumers of products often can get information on the relative reliability (in terms of repair rates) of products made by different producers, but they rarely have good information on either the precise probability that they will be injured by a product or the expected magnitude of the harms caused (e.g., Geistfeld 1994, discussing information problems).15 When consumers cannot directly observe the producer’s choice of care before purchase,16 each producer knows that his investment in care will not help him compete for consumers (but see Section 3.3.4, discussing signaling). Producers in competitive markets thus will not invest in care as low care enables them to compete by lowering prices. Thus, absent liability, producers make suboptimal-​quality products and expected accident costs are too high (Polinsky 1980; Shavell 1980). Activity levels, however, may be second-​best optimal—​optimal given actual risk. Although consumers cannot observe care, they can predict that producers will take zero care. Consumers thus will be able to correctly calculate the total cost to them of the product, including expected accident costs, provided they can estimate p(0) and H (Daughety and Reinganum 2013, 75). By contrast, if consumers underestimate product risk, then they will underestimate the total cost to them of the product, and may purchase the product even when the total expected cost of the product is less than its benefit. This results in excessive activity levels (see Spence 1977; Shavell 1980).

15 

For example, consumers of pharmaceuticals are provided information through warning labels about harms that may occur, but rarely receive good information about the precise probability of each adverse event. They have even less information about the risks they face given their personal health characteristics. Similarly, patients seeking medical care can obtain some information on their care-​ givers. But they cannot obtain accurate information on the probability of medical error associated with receiving treatment from any particular physician or hospital because this information is not publicly disclosed (Glied 2000; Arlen and MacLeod 2003; see, generally, Arrow 1963). In addition, even if it were disclosed, the market is unlikely to function optimally as patients’ choice of provider often is distorted by both health insurers—​which favor some providers over others—​and by geography, as patients often cannot obtain long-​term care too far from home. 16  Specifically, we now consider the situation where information asymmetry produces a moral hazard problem. Information asymmetry also can produce an adverse selection problem when unobservable ex ante investments result in some firms being higher-​quality than others. In this case, strict liability is superior to no liability. For a discussion of how adverse selection affects the optimal liability rule see Daughety and Reinganum (2013, 76–​79).

 

62   JENNIFER ARLEN Strict tort liability for product-​related harms can enhance the ex ante welfare of both producers and consumers by inducing producers to take optimal care. This enables consumers to get the quality of product that they want and are most willing to pay for. It also induces optimal activity levels. The analysis is similar to the analysis in Section 3.2.1. Under strict liability, each producer bears the full social cost of his product—​both production costs and expected accident costs. Accordingly, to minimize total expected costs, they each take optimal care. Each producer then sets his price equal to his marginal cost of making the product, including expected liability: c(x*) + p(x*)H. Consumers thus face a price equal to the full social cost, including their expected accident costs. This leads to optimal activity levels as consumers who bear the full expected social cost of the product will only purchase it if their willingness to pay equals or exceeds the social cost of the product (Polinsky 1980; Shavell 1980). Here, unlike in Section 3.3.1, the price leads to optimal activity levels even if the consumer under-​estimates expected accident costs. Negligence liability also can induce optimal care; whether it induces optimal activity levels depends on the source of the consumers’ imperfect information. Negligence liability will induce optimal care if due care equals optimal care and damages equal the harm caused (or the minimum amount needed to induce due care) (Shavell 1980; Polinsky 1980). Activity levels also are optimal, but only if consumers accurately estimate product risk. Under negligence, all producers take due care and thus do not pay damages. Accordingly, they each charge a price equal to the marginal cost of the product (P = c(x*)). Activity levels nevertheless will be efficient if, but only if, consumers accurately estimate expected accident costs when care is optimal. In this case, they will anticipate expected costs per unit equal to the price plus their expected accident costs, even if consumers cannot observe actual quality. In this case, activity levels will be optimal because consumers purchase the product only if their willingness to pay exceeds the total expected cost to society of making the product, c(x*) + p(x*)H. By contrast, if consumers have no information on product quality, then under negligence consumers will predicate their purchasing decision on the price plus their inaccurate guess of expected risk. As a result, activity levels will be inefficient.

3.3.3. Producers Also Are Imperfectly Informed: Medical Malpractice Information costs can affect both producers and consumers. This is particularly an issue with medical care.17 Evidence shows physicians regularly provide erroneous treatment;18 17  The present analysis focuses on Arlen and MacLeod (2003, 2005a). For an analysis of products liability where producers can invest in research and development related to product safety see Daughety and Reinganum (1995, 2006). 18  See Institute of Medicine (2007). Patients seeking medical treatment face a serious risk that they will be severely injured or be killed by the treatment they receive. Studies suggest that 4%–​18% of patients seeking care in hospitals are the victims of preventable medical error, with many suffering serious injury.

 

ECONOMICS OF TORT LAW    63 moreover, medical error generally is accidental (see Mello and Studdert 2008; see also Studdert et. al 2006). Medical providers often err accidentally because they lacked the information or expertise needed to properly diagnose the patient, identify the correct treatment,​ or provide the treatment correctly (Arlen and MacLeod 2003, 2005a; see Mello and Studdert [2008], identifying inadequate knowledge as an important cause of medical negligence).19 Physicians can reduce the probability of accidental error by investing in expertise and systems that reduce the risk error (such as health care technology). Thus, expertise constitutes a form of precaution that affects patients’ safety (Arlen and MacLeod 2003, 2005). Optimal deterrence thus requires that liability induce optimal investment in expertise by uninformed physicians, as well as optimal care when physicians are informed about how to best care for their patients. In many ways, the analysis of malpractice liability with expertise is similar to Section 3.2.2.2 supra. But there are two differences that warrant giving it separate attention in the formal section. First, in the medical care context, information and expertise are collective goods. Physicians’ investments in the expertise needed to properly diagnose patients and select treatments, and in the systems and health care technology needed to ensure treatment is properly delivered, constitute “collective goods” because they affect the quality of care a physician provides to all of his patients. As a result, liability that enhances expertise also is a collective good, as a physician’s incentive to invest in expertise to protect each patient depends on his expected liability for accidental harm to all of his patients (Arlen and MacLeod 2003, 2005a; Arlen 2010, 2013). We develop the implications of this in Section 3.3.4 infra. Second, medical markets usually involve an additional party—​the health insurer—​ who is in a contractual relationship with both the patient and the medical provider. Insurers alter the analysis because they bear most of the cost of medical services ex post and enter into contracts ex ante with physicians designed to alter treatment choices often to reduce costs relative to the treatment the patient would often prefer ex post.

(Weiler et al. [1993], reviewing written hospital records, found that 3.7% of the patients were victims of an error that caused significant harm; Andrews 2005, finding that 17.7% of patients were the victim of at least one error that extended their hospital stay). About 1% of hospital patients suffer errors that constitute medical negligence (Brennan et al. 1991). Indeed, medical error increases average hospital costs by $1,246 per patient admission, and, increases average costs in the riskiest hospitals by $4,769 per patient admission (Mello et al. 2007, 847). Moreover, medical error reaches beyond hospital walls: physicians routinely provide their patients less than medically recommended care (McGlynn et al. [2003, 2641] finding that patients on average receive only about 55% of recommended care; Schuster et al. [1998, 521] finding that, for chronic conditions, only “60% [of patients] received recommended care and 20% received contraindicated care”). Overall, medical error costs about $17–​29 billion per year (Institute of Medicine 1999). 19  In addition, even when physicians want to provide optimal treatment, their ability to do so depends on whether they and the medical institutions within which they practice invested optimally in the systems and health care technology needed to reduce optimally the risk of medical error (Abraham and Weiler 1994; Mello and Brennan 2002; Arlen and MacLeod 2003).

 

64   JENNIFER ARLEN Section 3.3.3.1. evaluates optimal negligence liability for medical malpractice assuming that patients pay a fixed price for treatment and physicians bear both all subsequent treatment costs, and the cost of obtaining information on optimal treatment for the patient’s condition. Section 3.3.3.2 considers the situation where the health insurer covers some or all of the patient’s treatment costs.

3.3.3.1. Formal Analysis: No Insurance20 Assume that each patient seeks treatment from a physician who he must rely on entirely to select the medically appropriate treatment. For simplicity, we assume that the physician can select one of two treatments: (1) high-​quality and higher cost treatment, t*, and (2) lower cost treatment, t0.21 Lower cost treatment imposes a higher risk of harm on the patient. We assume that the expected benefit to the patient of receiving treatment is given by b minus expected accident costs, which are given by p(ti)H, where i is an index of treatment choice and p(t0) > p(t*). It is assumed that the net social benefit of high cost treatment is higher than the net social benefit of low cost treatment: (p(t0)–​ p(t*))H > c(t*)–​c(t0), where c(t) is the cost of treatment. Thus, treatment t* is both the optimal treatment and the treatment that the patient would prefer, whether or not he bears his own medical costs. We assume that treatment quality is unobservable and noncontractible. Thus, it is assumed that the patient pays a price for treatment that is based on expected quality, but the patient and physician cannot contract in advance to ensure the physician selects t*. Physicians do not automatically know which treatment is optimal. Instead, each physician must invest e in expertise at a cost of C(e).21 This investment determines the probability, given by q(e), that the physician correctly identifies the optimal treatment for each patient’s condition.22 Thus q(e) is the probability that the physician provides “informed” treatment. We assume that the physician invests in patient safety after contracting with the patient. The analysis applies, as well, to unobservable investments made pre-​contract when investment is nonverifiable and the physician bears this cost directly. Physicians who are “informed” select the treatment that maximizes their welfare. When uninformed, physicians are assumed to provide suboptimal treatment.23 Patients cannot observe either the amount the physician invests in patient safety or the probability he selects suboptimal treatment. 20   The following analysis is based on Arlen and MacLeod (2003, 2005a) who provide a model of medical malpractice in which patient welfare depends on both physician’s ex ante investment in expertise and information on the patient’s condition and the physician’s treatment choice when informed. 21 C′(e)>0, C″(e)>0. 22 q′(e)>0, q″(e) 0 (with R″(q) < 0). Assume that each unit imposes a risk of harm on third parties, given by p(x)H, where H is the harm caused, p(x) is the probability of the accident, and x is the employee’s level of care (x is unobservable ex ante). Assume that employees bear the cost of effort, c(x). Social welfare is given by the joint welfare of the employee, firm, victim, and consumer R(q) − (c(x ) + p(x )H )q. (17)



Differentiating with respect to care and activity levels, we see that social welfare is maximized when employees take the level of care at which the marginal cost of care equals the marginal benefit of care: c ′(x ) = − p ′(x )H , (18)

and activity levels are such that

R ′ (q ) = c ( x ) + p ( x ) H . (19)

Assume now that strict liability is imposed solely on individual employees. The principal maximizes profits subject to two constraints. First, it must ensure the employee nets his reservation wage (which we set equal to zero). Second, the firm cannot dictate effort; instead, the agent selects the effort that maximizes his welfare (incentive compatibility constraint). Accordingly, the principal and the agent will solve the following problem (per unit sold) assuming competitive product and labor markets:

P − w (20)

subject to

w = c(E(x )) + p(E(x ))H ; (21)



E(x ) is the x that maximizes w − c(x ) − p(x )H , (22)

where E(x) is the employee’s expected level of care, P is the price per unit (which equals marginal cost), w is wages, and tort damages are given by D=H. Employees will select the level of care that minimizes their expected costs as given by Equation (22). Differentiating by x reveals that they take optimal care. Given this, the individual rationality constraint, Equation (21), implies that the firm must pay wages equal to the c(x*) + p(x*)H. In turn the firm will set the price equal to c(x*) + p(x*)H. This induces optimal activity levels:  consumers purchase up to the point where the marginal benefit of the last unit equals the price, which in turn equals the social cost of

 

ECONOMICS OF TORT LAW    81 the product (Kornhauser 1982; Sykes 1984; Segerson and Tietenberg 1992; Polinsky and Shavell 1993).38 Assume now that only the firm is strictly liable for harms caused by employees. The firm initially bears liability of p(x)H. To minimize its price, which includes the cost of its expected liability, it needs to incentivize its employees to take care. It can do this by imposing a sanction on employees who cause harm. Accordingly, the firm will solve the following maximization problem, where s is the sanction imposed on employees:

P − w − p(x )(H − s) (23)

subject to

w = c(E(x )) + p(E(x ))s; (24)



E(x ) is the x that maximizes w − c(x ) − p(x )s. (25)

Combining Equations (23) and (24), we see that in equilibrium, the firm bears expected cost of c(x) + p(x)H regardless of what sanction it selects. It will incorporate these costs into its price. Firms operating in competitive markets seek to minimize prices and thus expected costs. As a result, the firm will seek to induce its employees to take optimal care. To achieve this goal, it will select the sanction that induces optimal care. Equation (25) implies that the firm will set the sanction equal to H, the harm caused, thereby inducing optimal care. Equation (24) implies that the firm will have to pay wages of c(x*) + p(x*)H. Thus, it sets prices equal to the social cost of the product, inducing optimal activity levels (Kornhauser 1982; Sykes 1984; Segerson and Tietenberg 1992; Polinsky and Shavell 1993). Thus, corporate and individual liability each induce optimal care and activity levels in the simple model.

3.5.2.  Nonneutrality: Economic Justification for Corporate Liability Although individual liability can induce optimal behavior by firms and employees in the simple model, in most real-​world situations individual liability will not produce optimal care and activity levels. Individual liability will not induce optimal activity levels when liability for accidents is governed by negligence and not strict liability, unless consumers are perfectly informed about product quality. In addition, individual liability alone will not induce optimal behavior when optimal deterrence depends on firms taking actions that affect (but are 38  Firms facing wage payments equal to the expected cost of crime have optimal incentives to invest in measures to deter crime. Thus, in this mode, when the state imposes optimal individual liability, firms have optimal incentives to prevent crime. This implies that they will not act because individual liability is sufficient to deter crime optimally, in this model.

 

82   JENNIFER ARLEN independent of) employees’ care-​taking care and are neither observable nor contractible ex ante. These investments include systemic investments that reduce employees’ risk of accidental error, monitoring, and self-​reporting (Arlen 2012; see Shavell 2007, 171–​172). Finally, pure individual liability will not induce optimal behavior by agents or firms if agents cannot pay optimal damages under pure individual liability (Kornhauser 1982; Sykes 1984).

3.5.2.1.  Agents Liability Governed by Negligence Pure individual liability will not induce optimal behavior by firms if agent liability is governed by negligence. Pure individual negligence liability does not induce optimal corporate activity levels because employees take optimal care and thus are not liable. As a result, the firm does not bear expected accident costs and thus does not incorporate them into product prices. This leads to excess activity levels if accidents fall on strangers or consumers who under-​estimate the risk of harm (Polinsky and Shavell 1993). By contrast, firms will have optimal activity levels if individual negligence liability is combined with strict liability for firms (Polinsky and Shavell 1993).

3.5.2.2.  Firms’ Noncontractible Conduct Directly Affects Expected Accident Costs Expected accident costs regularly depend on investments by both the employee and the firm in care. For example, the firm often can reduce expected accident costs by investing in technology, equipment, or systems. We can represent this formally by assuming that expected accident costs are given by p(x,y), where y is the firm’s investment in safety. This investment may affect expected accident costs directly and also may make employees’ care investments more effective, pxy(x,y)>0. Employees often cannot observe or contract over the firm’s investment. When organizations can take actions that directly affect expected accident costs, the organizations (or the individuals making those decisions) must be held liable. Liability can either be imposed for all harms or for harms resulting from the organization’s failure to invest optimally in care. Relying entirely on individual liability targeted at employees who directly determine care, x, will not induce optimal organizational precautions when these precautions are unobservable because under individual liability the firm only bears expected liability through employees’ wages. These wages will be based on employees’ expectations of organizations’ investment in precautions. When firms are not liable and precautions are unobservable, employees will expect firms not to take precaution. They will insist on higher wages no matter what the firm does. Organizational investment also is suboptimal if employees do not understand how organizational precautions affect their expected liability (Shavell 2007, 170–​171). In either case, both firms and employees would be better off if firms also are held liable, as this enables them to credibly commit to optimal precautions, thereby lowering wage payments (see Arlen 2012; Arlen and MacLeod 2005a; see also Shavell 2007, 170–​171).

 

ECONOMICS OF TORT LAW    83

3.5.2.3.  Employee Asset Insufficiency Corporate liability also is an essential prerequisite to achieving optimal deterrence if employees do not have sufficient wealth to pay the optimal damage award. In this situation, employees take too little care under individual liability. As a result, firms’ activity levels are too high. Under individual liability, firms bear employees’ expected care and liability costs through wages. Since asset constrained employees take too little care and bear less than full liability, firms do not internalize the full expected costs associated with their activities (Kornhauser 1982; Sykes 1984). Moreover, under pure individual liability, firms have incentives to create an insolvency issue by strategically selecting asset-​constrained employees in order to lower costs, as when firms hire thinly-​capitalized outside contractors (Arlen and MacLeod 2005b). Corporate liability promotes optimal deterrence by inducing both optimal corporate activity levels (Polinsky and Shavell 1993) and corporate actions that induce optimal (or second-​best optimal) care-​taking by employees (Kornhauser 1982; Sykes 1984; see Arlen 2012). Activity levels are (second-​best) optimal because the firm bears the full cost of accidents and employees’ care costs; thus, prices will result in the full social cost of the product. Corporate liability also can improve employee care-​taking. Under corporate liability, firms bear their employees’ expected accident costs and thus have optimal incentives to select employees who are less likely to be asset constrained (Arlen and MacLeod 2005b). They also may intervene to induce optimal care by using a combination of direct mandates, ex ante monitoring, and non-​harm contingent sanctions imposed on any employee who takes suboptimal care even if no harm occurs. The latter approach increases employees’ care-​taking by increasing employees’ expected sanction for selecting suboptimal care from p(x)W to W (if all negligence is detected) (see Kornhauser 1982; Sykes 1984). Third, if corporate liability is properly structured (see Section 3.4.3), it can be used to induce firms to render individual tort liability more effective by reporting tortious harms and helping victims sanction negligent employees (see Arlen 1994; Chu and Qian 1995).

3.5.3.  Optimal Corporate Liability Respondeat superior holds organizations, and other principals, strictly liable for their employees’ torts committed in the scope of employment. Strict entity-​level liability can be used to induce efficient activity levels, as we have already seen. It also can induce optimal investment in the type of precautions that reduce the probability of an accident without affecting the probability that harm or negligence is detected. These actions include screening employees, super-​compensatory wages, and firm-​level precaution. Strict corporate liability with full compensation damages will not induce firms to undertake optimal precautions that take the form of corporate

 

84   JENNIFER ARLEN “policing”—​precautions that increase the probability that a harm is detected, negligence is proved, or causation is established (see Arlen 1994; Chu and Qian 1995; Arlen and Kraakman 1997; Arlen 2012). This is important because in many situations (e.g., pharmaceuticals), the organization that produced the risk is the least cost provider of information about both the tortious nature of the harm and the identity of the individual(s) responsible. Moreover, in the case of environmental harms, product defects, and other widespread harms, corporate action is needed to induce firms to both detect tortious conduct and report them to authorities so that they can warn potential victims and reduce harm (see Arlen 1994; Arlen and Kraakman 1997). Firms held liable through respondeat superior for all harms will not optimally invest in these “policing” measures because policing enhances the firm’s expected liability for any employee torts that do occur. Thus, when damages equal H, the net benefit to the firm of policing does not equal the social benefit of the harms deterred, as is needed to provide optimal incentives to police. Instead, the firm’s benefit equals its benefit from all harms deterred minus the increase in its expected liability for any harms that do occur resulting from its policing efforts (Arlen 1994; Arlen and Kraakman 1997; see Chu and Qian [1995], discussing negligence liability). Indeed, when liability is high and policing detects many torts that otherwise would remain hidden, respondeat superior may deter policing (see Arlen 1994; see Arlen and Kraakman 1997; cf. Shavell [1994], examining the effect of mandatory disclosure on incentives to acquire information about product risks).39 Entity liability can be structured to induce firms to invest optimally in corporate policing—​e.g., measures to detect tortious conduct and identify responsible individuals. But liability must be structured so that firms that undertake optimal policing bear lower expected liability than those who do not. When optimal deterrence requires optimal investment by the firm in prevention and corporate policing, then the optimal corporate liability regime imposes respondeat superior liability on the firm with expected liability equal to the harm caused, coupled with enhanced regulatory sanctions imposed only on firms that fail to optimally monitor, self-​report, and cooperate by providing evidence. These regulatory duties will induce optimal investment in information and self-​reporting so long as firms that breach these duties are subject to enormous sanctions, whereas firms are not subject to enhanced sanctions if they comply with their duties and harm nevertheless results. In turn, the residual liability for harm caused imposed on all firms ensures that firms will induce optimal care-​taking and that activity levels are optimal (see Arlen and Kraakman 1997; Arlen 2012). 39 

This analysis focuses on strict corporate liability with a fixed sanction equal to H, or, in the case of imperfect detection, H/​P, where P is the expected probability of detection. For a discussion of strict corporate liability when liability equals H/​P(c), with P(c) equaling the actual probability of liability given the firm’s behavior, see Arlen and Kraakman (1997).

 

ECONOMICS OF TORT LAW    85

3.5.4.  Limitations of Existing Corporate Liability In addition to the inefficiencies discussed in the preceding section, corporate liability may fail to induce optimal corporate behavior as a result of legal rules: specifically, limited liability and the independent contractor rule.

3.5.4.1.  Limited Liability Organizational liability induces optimal care-​taking by people who own (or control) firms only if those in control bear the expected costs of their failure to induce optimal investment in care. Even when tort liability rules are optimal, corporate law allows owners to avoid bearing the full expected cost of the risks they create if they conduct their risky activities through corporations and do not personally engage in the risky conduct. Corporations enjoy limited liability: a tort plaintiff can seek recovery from the firm, but cannot go after the personal assets of the owners (as long as the owner follows corporate formalities). Moreover, owners can retain the benefits of limited liability even if they enhance the asset insufficiency problem by removing almost all profits from the firm each year. Limited liability thus will lead smaller, less-well-capitalized firms to under-​ invest in care (see Hansmann and Kraakman 1991).

3.5.4.2.  Independent Contractor Rule Under respondeat superior, firms are liable for harms caused by their agents if the agent and principal had a “master–servant” relationship, giving the principal the right to control the conduct of the agent. By contrast, the principal generally is not liable if the agent is an independent contractor. This rule has been justified on the grounds that there is little reason to hold firms liable for agents whose conduct the firm does not directly control (Sykes 1988; see Epstein and Sykes 2001). Yet a dynamic perspective reveals that corporate liability for independent contractors often is needed to induce firms to take optimal steps to deter risk-​taking by their independent contractors. Firms can affect care-​taking by their independent contractors in a host of ways including (1) their choice of agent, (2) compensation structure, and (3) sanctions for harm caused. The independent contractor rule fails to provide optimal incentives to firms and can actually deter socially valuable corporate actions. To see this, consider a firm that has hired an independent contractor to undertake a risk-​producing activity. Under U.S. law, the independent contractor would be liable for any torts he causes and will charge the firm for his expected liability (and care) costs. Absent corporate liability, the firm can reduce its costs by hiring independent contractors who are asset constrained because they will have lower expected liability costs and thus will charge less (Arlen and MacLeod 2005b). In addition, under the independent contractor rule, a firm that would have optimally hired an agent as an employee in order to exert control and oversight may choose instead to hire the employee as an independent contractor in order to avoid liability. This is a serious issue if the employee is asset constrained. Consider a firm who must

 

86   JENNIFER ARLEN hire an agent to perform an important task. Assume the firm can hire the agent as an employee or an independent contractor. Assume further that all available agents would be judgment proof. In this situation, it often will be socially optimal for the firm to hire the agent as an employee in order to increase care by monitoring. Yet, under the independent contractor rule, firms may rationally eschew control, even when control is optimal, because they can avoid being held liable for all expected accident costs if they hire asset constrained agents as independent contractors instead of as employees (Arlen and MacLeod 2005b; see Sykes [1988], suggesting the independent contractor rule should depend not on the actual relationship but instead on whether the agent would optimally be hired as an independent contractor or an employee).

3.6. Conclusion Tort liability is vital to our society’s ability to prosper because it deters excessive risk-​ taking by those seeking personal gain at the expense of others. Indeed, as our society has become more complex—​with new technologies for risk-creation and risk-​reduction emerging constantly—​tort liability is likely to play an ever more central role. Rapid technological change, and globalization of production, has placed severe strains on efforts to deter risk through regulation. The more rapid the development of new risk-​producing activities, and the farther flung the production locations, the more difficult it is for regulators to develop reasonable safety standards and oversee compliance. Tort liability with actions brought by private plaintiffs is needed to provide adequate deterrence to those who would benefit at a potential cost to others. Tort liability can enhance welfare and protect safety if it is structured to optimally deter. Yet the tort systems must seek optimality in a sea of imperfection arising, in large part, from information costs. Indeed, the same information costs that justify the use of tort liability as a supplement to regulation also affect the operation of the tort system. They affect risk-​imposers, victims, and courts in ways that alter the purposes and effects of tort liability. Each of these parties must invest in information to determine and implement optimal care. Accordingly, a tort system that seeks to optimally deter should treat optimal information acquisition and use by the parties and courts as a core goal of optimal deterrence to be taken into account when structuring liability and damage rules. Information costs also are present in less obvious ways. Information costs produce the litigation costs that drive a wedge between optimal tort liability and our existing system. Contingency fees paid to obtain legal expertise and payments for investigation and experts reflect, in part, litigators’ asymmetric information (including expertise). Information costs help explain the numerous risk-​imposing activities done by corporations through their own employees, instead of by hiring outsiders (Coase 1937). They also help limit the firm’s control over its own activities and introduce an additional role for the tort system.

 

ECONOMICS OF TORT LAW    87 Economic analysis of tort liability has provided considerable insights on how rules governing liability, damages, and procedures should be structured in order to optimally deter in a world of costly information, as this chapter has highlighted. This chapter also provides insights on productive avenues for both future scholarship and optimal tort reform. It suggests that optimal tort reforms may include measures that reduce the distortion of costly information, whether by lowering information costs or by extending liability to those (e.g., organizations) who can better address them. Finally, many fundamental and important questions about liability in a world of costly and dynamic information remain to be addressed by future scholarship.

Bibliography Abraham, Kenneth and Weiler, Paul C. 1994. “Enterprise Medical Liability and the Evolution of the American Health Care System.” Harvard Law Review 108, pp. 381–​436. Akerlof, George A., 1970. “The Market for ‘Lemons’:  Quality Uncertainty and the Market Mechanism.” Quarterly Journal of Economics 84, pp. 488–​500. Andrews, Lori. 2005. “Studying Medical Error in Situ: Implications for Malpractice Law and Policy.” Depaul Law Review 54, pp. 357–​392. Andrews, Lori B. et al. 1997. “An Alternative Strategy for Studying Adverse Events in Medical Care.” Lancet 349, pp. 309–​313. Arlen, Jennifer. 1985. “An Economic Analysis of Tort Damages for Wrongful Death.” New York University Law Review 60, pp. 1113–​1136. Arlen, Jennifer. 1990a. “Reconsidering Efficient Tort Rules for Personal Injury: The Case of Single Activity Accidents.” William & Mary Law Review 32, pp. 41–​103. Arlen, Jennifer. 1990b. “Reexamining Liability Rules When Injurers As Well As Victims Suffer Losses.” International Review of Law & Economics 10, pp. 233–​239. Arlen, Jennifer. 1992. “Liability for Physical Injury When Injurers As Well As Victims Suffer Losses.” Journal of Law, Economics & Organization 8, pp. 411–​426. Arlen, Jennifer. 1994. “The Potentially Perverse Effects of Corporate Criminal Liability.” Journal of Legal Studies 23, pp. 833–​867. Arlen, Jennifer. 2000. “Tort Damages.” In: Boudewijn Bouckaert and Gerrit De Geest, eds., Encyclopedia of Law & Economics. Northampton, MA: Edward Elgar Press. 682–​734. Arlen, Jennifer. 2006. “Private Contractual Alternatives to Malpractice Liability.” In Rogan Kersh and William M. Sage, eds., Medical Malpractice and the U.S. Healthcare System: New Century, Different Issues. Cambridge, MA: Cambridge University Press. 245–​266. Arlen, Jennifer. 2010. “Contracting Over Liability:  Medical Malpractice and the Cost of Choice.” University of Pennsylvania Law Review 158, pp. 957–​1023. Arlen, Jennifer. 2012. “Economic Analysis of Corporate Criminal Liability:  Theory and Evidence.” In: Keith Hylton and Alon Harel, eds., Research Handbook on the Economics of Criminal Law. Northampton, MA: Edward Elgar Press. 144–​203. Arlen, Jennifer. 2013. “Economic Analysis of Medical Malpractice and Its Reform.” In: Arlen, Jennifer, ed., Research Handbook on the Economics of Torts. Northampton, MA: Edward Elgar Press. 33–​68. Arlen, Jennifer and Kahan, Marcel. Forthcoming. “Corporate Governance Regulation through Nonprosecution,’ University of Chicago Law Review.

 

88   JENNIFER ARLEN Arlen, Jennifer and MacLeod, Bentley W. 2003. “Malpractice Liability for Physicians and Managed Care Organizations.” New York University Law Review 78, pp. 1929–​2006. Arlen, Jennifer and MacLeod, Bentley W. 2005a. “Torts, Expertise and Authority: Liability of Physicians and Managed Care Organizations.” Rand Journal of Economics 36, pp. 494–​519. Arlen, Jennifer and MacLeod, Bentley W. 2005b. Beyond Master-​Servant: A Critique of Vicarious Liability. In: M. Stuart Madden, ed., Exploring Tort Law. New York: Cambridge University Press. 111–140. Arrow, Kenneth J. 1963. “Uncertainty and the Welfare Economics of Medical Care.” American Economic Review 53, pp. 941–​973. Avraham, Ronen. 2007. “An Empirical Study of the Impact of Tort Reform on Medical Malpractice Settlement Payments.” Journal of Legal Studies 36, pp. S183–​S229. Avraham, Ronen, Dafny, Leemore S., and Schanzenbach, Max M. 2012. “The Impact of Tort Reform on Employer-​ Sponsored Health Insurance.” Journal of Law, Economics, and Organization 28, pp. 657–686. Babcock, Linda and Furgeson, Joshua. 2013. “Bounded Rationality in the Settlement Process: Empirical Evidence on the Causes of Settlement Failure in Litigation.” In: Jennifer Arlen, ed., Research Handbook on the Economics of Torts. Northampton: Edward Elgar Press. Baker, Tom. 2005. The Medical Malpractice Myth. Chicago: Chicago University Press. 360–382. Baker, Tom and Lytton, Timothy. 2010. “Allowing Patients to Waive the Right to Sue for Medical Malpractice: A Response to Thaler and Sunstein.” Northwestern University Law Review 104, pp. 233–​252. Bakos, Yannis, Marotta-​Wurgler, Florencia, and Trossen, David. 2014. “Does Anyone Read the Fine Print? Testing a Law and Economics Approach to Standard Form Contracts,” Journal of Legal Studies 43, pp. 9–​40. Bar-​Gill, Oren and Ben-​Shahar, Omni. 2003. “The Uneasy Case for Comparative Negligence.” American Law and Economics Review 5, pp. 433–​469. Bernardo, Antonio, Talley, Eric, and Welch, Ivo. 2000. “A Theory of Legal Presumptions.” Journal of Law, Economics and Organization 16, pp. 1–​49. Black, Bernard et al. 2005. “Stability Not Crisis: Medical Malpractice Outcomes in Texas, 1988–​ 2002.” Journal of Empirical Legal Studies 2, pp. 207–​259. Brennan, Troyen et  al. 1991. “Incidence of Adverse Events and Negligence in Hospitalized Patients. Results of the Harvard Medical Practice Student I.” New England Journal of Medicine 324, pp. 370–​376, Brown, John Prather. 1973. “Toward an Economic Theory of Liability.” Journal of Legal Studies 2, pp. 323–​349. Calabresi, Guido. 1970. The Cost of Accidents: A Legal and Economic Analysis. New Haven: Yale University Press. Calabresi, Guido A., and Melamed, Douglas. 1972. “Property Rules, Liability Rules, and Inalienability: One View of the Cathedral.” Harvard Law Review 85, pp. 1089–​1128. Calfee, John and Craswell, Richard. 1986. “Deterrence and Uncertain Legal Standards.” Journal of Law, Economics & Organization 2, pp. 279–​303. Chu, C. Y.  Cyrus and Qian, Yingy. 1995. “Vicarious Liability under a Negligence Rule.” International Review of Law and Economics 15, pp. 302–​322. Choi, Albert and Spier, Kathryn 2014. “Should Consumers be Permitted to Waive Products Liability? Product Safety, Private Contracts, and Adverse Selection.” Journal of Law, Economics, & Organization 30, pp. 734–​766. Coase, Ronald H. 1937. “The Nature of the Firm.” Economica 4, pp. 386–​405.

 

ECONOMICS OF TORT LAW    89 Coase, Ronald H. 1960. “The Problem of Social Costs.” Journal of Law and Economics 3, pp. 1–​44. Cooter, Robert D. 1982. “Economic Analysis of Punitive Damages.” Southern California Law Review 56, pp. 79–​101. Cooter, Robert D. 1984. “Prices and Sanctions.” Columbia Law Review 84, pp. 1523–​1560. Cooter, Robert D. 1989. “Punitive Damages for Deterrence: When and How Much?.” Alabama Law Review 40, pp. 1143–​1196. Cooter, Robert D. and David DePianto. 2013. “Damages for Incompensable Harm.” In: Arlen, Jennifer, ed., Research Handbook on the Economics of Torts. Northampton, MA:  Edward Elgar Press. 439–​459. Cooter, Robert D. and Ulen, Tom. 1986. “An Economic Case for Comparative Negligence.” New York University Law Review 61, pp. 1067–​1110. Craswell, Richard and Calfee, John E. 1986. “Deterrence and Uncertain Legal Standards.” Journal of Law, Economics, and Organization 2, pp. 279–​303. Currie, Janet and W. Bentley MacLeod. 2008. “First Do No Harm? Tort Reform and Birth Outcomes,” Quarterly Journal of Economics 123, pp. 795–​830. Danzon, Patricia M. 1997. “Tort Liability: A Minefield for Managed Care?” Journal of Legal Studies 26, pp. 491–​519. Danzon, Patricia M. 1998. “Medical Malpractice, 624.” In: Peter Newman, ed., The New Palgrave Dictionary of Economics and the Law. Vol. 2. Hampshire: Palgrave Macmillan. Danzon, Patricia M. 1985. Medical Malpractice: Theory, Evidence And Public Policy. Cambridge, MA: Harvard University Press. Danzon, Patricia M. 2000. “Liability for Medical Malpractice.” In:  Anthony J. Culyer and Joseph P. Newhouse, eds., Handbook of Health Economics. Vol. 1. Amsterdam:  North Holland. 1339–​1404. Danzon, Patricia and Lillard, Lee. 1983. “Settlement Out of Court: The Disposition of Medical Malpractice Claims.” Journal of Legal Studies 12, pp. 345–​377. Dari-​Mattiacci, Giuseppe. 2005. “The Optimal Scope of Negligence.” Review of Law and Economics 1, pp. 331–​364. Dari-​Mattiacci, Giuseppe. 2009. “Negative Liability.” Journal of Legal Studies 38, pp. 21–​60. Dari-​ Mattiacci, Giuseppe and Hendriks, Eva S. 2013. “Relative Fault and Efficient Negligence: Comparative Negligence Explained.” Review of Law and Economics 9, pp. 1–​40. Dari-​Mattiacci, Giuseppe, Lovat, Bruno, and Parisi, Francesco. 2014. “The Optimal Scope of Negligence.” Journal of Institutional and Theoretical Economics 170, pp. 571–​598. Dari-​Mattiacci, Giuseppe and Franzoni, Luigi Alberto. 2014. “Innovative Negligence Rule.” American Law & Economics Review 16, pp. 333–365. Daughety, Andrew F. and Reinganum, Jennifer. 1995. “Product Safety:  Liability, R&D, and Signaling.” American Economic Review 85, pp. 1187–​1206. Daughety, Andrew F. and Jennifer Reinganum. 2005. “Secrecy and Safety,” American Economic Review 95, pp. 1074–​1091. Daughety, Andrew F. and Jennifer Reinganum. 2006. “Markets, Torts, and Social Inefficiency,” RAND Journal of Economics 37, pp. 300–​323. Daughety, Andrew F. and Jennifer Reinganum. 2008. “Communicating Quality:  A  Unified Model of Disclosure and Signaling.” RAND Journal of Economics 39, pp. 973–​989. Daughety, Andrew F. and Reinganum, Jennifer. 2013. “Economic Analysis of Products Liability.” In: Jennifer Arlen, ed., Research Handbook on the Economics of Torts. Northampton, MA: Edward Elgar Press pp. 69–96.

 

90   JENNIFER ARLEN Diamond, Peter A. 1974. “Single Activity Accidents.” Journal of Legal Studies 3, pp. 107–​164. Epstein, Richard A. 1976. “Medical Malpractice:  The Case for Contract.” American Bar Foundation Research Journal 1, pp. 87–​149. Epstein, Richard A. 1985. “Products Liability as an Insurance Market.” Journal of Legal Studies 14, pp. 645–​669. Epstein, Richard A. 2005. Contractual Principle versus Legislative Fixes:  Coming to Closure on the Unending Travails of Medical Malpractice.” DePaul Law Review 54, pp. 503–​526. Epstein, Richard A. and Sykes, Alan O. 2001. “The Assault on Managed Care:  Vicarious Liability, ERISA Preemption, and Class Actions.” Journal of Legal Studies 30, pp. 625–​659. Farber, Henry S. and White, Michelle J. 1991. Medical Malpractice: An Empirical Examination of the Litigation Process.” Rand Journal of Economics 22, pp. 199–​217. Farber, Henry S. and White, Michelle J. 1994. “A Comparison of Formal and Informal Dispute Resolution in Medical Malpractice.” Journal of Legal Studies 23, pp. 777–​806. Frakes, Michael. 2012. “Defensive Medicine and Obstetric Practices.” Journal of Empirical Legal Studies 9, pp. 457–​481. Frakes, Michael. 2013. “The Impact of Medical Liability Standards on Regional Variations in Physician Behavior: Evidence from the Adoption of National Standard Rules.” American Economic Review 103, pp. 257–276. Frank, Richard G. and Zeckhauser, Richard. 2007. “Custom-​Made versus Ready-​to-​Wear Treatments: Behavioral Propensities in Physicians’ Choices.” Journal of Health Economics 26, pp. 1101–​1127. Fudenberg, Drew and Tirole, Jean. 1990. “Moral Hazard and Renegotiation in Agency Contracts.” Econometrica 58, pp. 1279–​1319. Gandhi, Tejal K. et al. 2006. “Missed and Delayed Diagnosis in the Ambulatory Setting: A Study of Closed Malpractice Claims.” Annals of Internal Medicine 145, pp. 488–​496. Gawande, Atul. 2003. Complications:  A  Surgeon’s Notes on an Imperfect Science. New York: Picador. Geistfeld, Mark. 1994. “The Political Economy of Neocontractual Proposals for Products Liability Reform.” Texas Law Review 72, pp. 803–​847. Geistfeld, Mark. 2013. “Fault Lines in the Positive Economic Analysis of Tort Law.” In: Arlen, Jennifer, ed., Research Handbook on the Economics of Torts. Northampton, MA: Edward Elgar Press. 149–​168. Gilles, Stephen G. 1992. “Rule-​Based Negligence and the Regulation of Activity Levels.” Journal of Legal Studies 21, pp. 319–​363. Glied, Sherry. 2000. “Managed Care.” In:  Anthony J. Culver and Joseph P. Newhouse eds., Handbook of Health Economics. Vol. 1A. Amsterdam: North Holland. 707–​753. Glied, Sherry A. 2005. “The Employer-​ Based Health Insurance System: Mistake or Cornerstone?” In: David Mechanic et al., eds., Policy Challenges In Modern Health Care. New Brunswick, NJ: Rutgers University Press. 46–​47. Glied, Sherry and Zivin, Joshua G. 2002. “How Do Doctors Behave When Some (But Not All) of Their Patients Are in Managed Care?” Journal of Health Economics 21, pp. 337–​353. Grady, Mark F. 1983. “A New Positive Economic Theory of Negligence.” Yale Law Journal 92, pp. 799–​829. Grady, Mark F. 1988. “Why Are People Negligent?: Technology, Nondurable Precautions, and the Medical Malpractice Explosion.” Northwestern University Law Review 82, pp. 293–​334.

 

ECONOMICS OF TORT LAW    91 Grossman, Sanford. 1981. “The Informational Role of Warranties and Private Disclosure About Product Quality.” Journal of Law & Economics 24, pp. 461–​484. Haddock, David, McChesney, Fred S. and Spiegel, Menahem. 1990. “An Ordinary Economic Rationale for Extraordinary Legal Sanctions.” California Law Review 78, pp. 1–​51. Hansmann, Henry and Kraakman, Reinier. 1991. “The Uneasy Case for Limiting Shareholder Liability in Tort.” Yale Law Journal 100, pp. 1879–1934. Havighurst, Clark C. 1986. “Private Reform of Tort-​Law Dogma: Market Opportunities and Legal Obstacles.” Law and Contemporary Problems 49, pp. 143–​172. Havighurst, Clark C. 1995. Health Care Choices. Washington, DC: American Enterprise Institute. Havighurst, Clark C. 2000. “Vicarious Liability: Relocating Responsibility for the Quality of Medical Care.” American Journal of Law and Medicine 26, pp. 7–​29. Hylton, Keith N. 1990. The Influence of Litigation Costs on Deterrence under Strict Liability and Negligence. International Review of Law and Economics 10, pp. 161–​171. Hylton, Keith N. 2000. “Agreements to Waive or to Arbitrate Legal Claims:  An Economic Analysis.” Supreme Court Economic Review 8, pp. 209–​263. Hylton, Keith N. 2013. “Causation in Tort Law:  A  Reconsideration.” In:  Jennifer Arlen, ed., Research Handbook on the Economics of Torts. Northampton, MA:  Edward Elgar Press. 97–​113. Hyman, David A., Black, Bernard, Zeiler, Kathryn, Silver, Charles, and Sage, William. 2007. “Do Defendants Pay What Juries Award? Post Verdict Haircuts in Texas Medical Malpractice Cases.” Journal of Empirical Legal Studies 4, pp. 3–​68. Institute of Medicine. 1999. To Err Is Human: Building A Safer Health System. Washington, DC: National Academies Press. Institute of Medicine. 2007. Preventing Medication Errors. Washington, DC:  National Academies Press. Kahan, Marcel. 1989. Causation and Incentives to Take Care under the Negligence Rule. Journal of Legal Studies 18, pp. 427–​447. Kaiser Family Foundation/​Agency for Health Care Research and Quality. 2000. National Survey on Americans as Health Care Consumers: An Update on the Role of Quality Information. http://​kff.org/​health-​costs/​poll-​finding/​national-​survey-​on-​americans-​as-​ health-​care/​. Kaplow, Louis. 1986. “Private versus Social Costs in Bringing Suit.” Journal of Legal Studies 15, pp. 371–​385. Kaplow, Louis. 1990. “Optimal Deterrence, Uninformed Individuals, and Acquiring Information about Whether Acts Are Subject to Sanctions.” Journal of Law, Economics, and Organization 6, pp. 93–​128. Kaplow, Louis. 1992. “Rules versus Standards.” Duke Law Journal 42, pp. 557–​629. Kaplow, Louis. 1994. “The Value of Accuracy in Adjudication: An Economic Analysis.” Journal of Legal Studies 23, pp. 307–​401. Katz, Avery. 1998. “Standard Form Contracts.” In:  Peter Newman, ed., The New Palgrave Dictionary of Law and Economics. New York: Stockton Press. Kessler, Daniel and McClellan, Mark. 1996. “Do Doctors Practice Defensive Medicine.” Quarterly Journal of Economics 111, pp. 353–​390. Klerman, Daniel. 2012. “Personal Jurisdiction and Product Liability.” Southern California Law Review 85, pp. 1551–​1596.

 

92   JENNIFER ARLEN Kolstad, Charles D., Ulen, Thomas S., and Johnson, Gary V. 1990. “Ex Post Liability for Harm vs. Ex Ante Regulation:  Substitutes or Complements?” American Economic Review 80, pp. 888–​901. Korobkin, Russell. 1999. “The Efficiency of Managed Care ‘Patient Protection’ Laws: Incomplete Contracts, Bounded Rationality, and Market Failure. Cornell Law Review 85, pp. 1–​88. Kornhauser, Lewis A. 1982. “An Economic Analysis of the Choice between Enterprise and Personal Liability for Accidents.” California Law Review 70, pp. 1345–​1392. Kornhauser, Lewis A. 2013. “Economic Analysis of Joint and Several Liability.” In: Jennifer Arlen, ed., Research Handbook on the Economics of Torts. Northampton, MA: Edward Elgar Press. 199–233. Kornhauser, Lewis A. and Revesz, Richard. 1993. “Settlements under Joint and Several Liability.” New York University Law Review 68, pp. 427–​493. Kornhauser, Lewis A. and Revesz, Richard. 1994. “Multi-​Defendant Settlements: The Impact of Joint and Several Liability.” Journal of Legal Studies 23, pp. 41–​76. Landes, William M. and Posner, Richard A. 1985. “A Positive Economic Analysis of Products Liability.” Journal of Legal Studies 14, pp. 535–​67. Landes, William M. and Posner, Richard A. 1987. The Economic Structure of Tort Law. Cambridge, MA: Harvard University Press. Leshem, Shmuel and Miller, Geoffrey. 2009. “All-​or-​Nothing versus Proportionate Damages.” Journal of Legal Studies 38, pp. 345–​382. Marotta-​Wurgler, Florencia. 2012. “Does Disclosure Matter?” Journal of Institutional and Theoretical Economics 168, pp. 94–​119. McGlynn, Elizabeth A. et al. 2003. “The Quality of Health Care Delivered to Adults in the United States.” New England Journal of Medicine 348, pp. 2635–​2645. Mello, Michelle et al. 2007. “Who Pays for Medical Errors: An Analysis of Adverse Event Costs, the Medical Liability System, and Incentives for Patient Safety Improvement.” Journal of Empirical Legal Studies 4, pp. 835–​860. Mello, Michelle M. and Brennan, Troyan A. 2002. “Deterrence of Medical Errors: Theory and Evidence for Malpractice Reform.” Texas Law Review 80, pp. 1595–​1637. Mello, Michelle M. and Studdert, David M. 2008. “Deconstructing Negligence: The Role of Individual and System Factors in Causing Medical Injuries.” Georgetown Law Journal 96, pp. 599–​623. Mello, Michelle M. and Zeiler, Kathryn. 2008. “Empirical Health Law Scholarship: The State of the Field.” Georgetown Law Journal 96, pp. 649–​702. Miller, Geoffrey. 2013. “Group Litigation in the Enforcement of Law.” In: Jennifer Arlen ed., Research Handbook on the Economics of Torts. Northampton, MA: Edward Elgar Press. 262–278. Newhouse, Joseph P. 2008. “The Economics of Health Insurance.” In:  Steven N. Durlauf and Lawrence E. Blume, eds. The New Palgrave Dictionary of Economics. 2nd ed. Vol. 3. New York: Palgrave Macmillan. 872–​875. Oi, Walter. 1973. “The Economics of Product Safety.” Bell Journal of Economics and Management Science 4, pp. 3–​28. Paik, Myungho, Black, Bernard, and Hyman, David. 2012. “Will Tort Reform Bend the Cost Curve? Evidence from Texas.” Journal of Empirical Legal Studies 9, pp. 173–​216. Palmieri, Patrick A. et al. 2008. “The Anatomy and Physiology of Error in Adverse Health Events.” Advances in Health Care Management 7, pp. 33–​68.

 

ECONOMICS OF TORT LAW    93 Parisi, Francesco and Fon, Vincy. 2004. “Comparative Causation.” American Law and Economics Review 6, pp. 345–​368. Peters, Philip G., Jr. 2002. “The Role of the Jury in Modern Malpractice Law.” Iowa Law Review 87, pp. 909–​969. Peters, Philip G., Jr. 2007. “What We Know About Malpractice Settlements.” Iowa Law Review 92, pp. 1783–​1833. Polinsky, A. Mitchell. 1980. “Strict Liability vs. Negligence in a Market Setting.” American Economic Review 70, pp. 363–​367. Polinsky, A. Mitchell and Rogerson, William P. 1983. “Products Liability, Consumer Misperceptions, and Market Power.” Bell Journal of Economics 14, pp. 581–​589. Polinsky, A. Mitchell and Shavell, Steven. 1988. “Punitive Damages: An Economic Analysis.” Harvard Law Review 111, pp. 869–​962. Polinsky, A. Mitchell, and Shavell, Steven. 2010. “The Uneasy Case for Product Liability.” Harvard Law Review 123, pp. 1437–​1492. Polinsky, A. Mitchell and Shavell, Steven. 2012. “Mandatory versus Voluntary Disclosure of Product Risks.” Journal of Law, Economics, and Organization 28, pp. 369–​370. Posner, Richard. 2011. Economic Analysis of Law. 8th ed. New York: Wolters Kluwer Law & Business. Priest, George. 1981. “A Theory of Consumer Product Warranty.” Yale Law Journal 90, pp. 1297–​1352. Priest, George. 1987. “The Current Insurance Crisis and Modern Tort Law.” Yale Law Journal 96, pp. 1521–​1590. Robinson, Glen O. 1986. “Rethinking the Allocation of Medical Malpractice Risks between Patients and Providers.” Law and Contemporary Problems 49, pp. 173–​199. Rolph, John E., et al. 2007. “Identifying Malpractice-​Prone Physicians.” Journal of Empirical Legal Studies 4, pp. 125–​153. Rothschild, Michael and Stiglitz, Joseph E. 1976. “Equilibrium in Competitive Insurance Markets:  an Essay on the Economics of Imperfect Information.” Quarterly Journal of Economics 80, pp. 629–​649. Rubin, Paul, 1993. Tort Reform by Contract. Washington, D: American Enterprise Institute for Policy Research. Rubinfeld, Daniel. 1987. “The Efficiency of Comparative Negligence.” Journal of Legal Studies 16, pp. 375–​394. Sage, William. 1997. “Enterprise Liability and the Emerging Managed Health Care System.” Law and Contemporary Problems 60, pp. 159–​210. Sage, William. 1999. “Regulating through Information: Disclosure Laws and American Health Care.” Columbia Law Review 99, pp. 1701–​1829. Segerson, Kathleen and Tom Tietenberg, 1992. “The Structure of Penalties in Environmental Enforcement:  An Economic Analysis.” Journal of Environmental Economics and Management 23, pp. 179–​200. Schuster, Mark A., et al. 1998. “How Good is the Quality of Health Care in the United States.” Millbank Quarterly 76, pp. 517–​563. Schwartz, Alan. 1988. “Proposals for Products Liability Reform: A Theoretical Synthesis.” Yale Law Journal 97, pp. 353–​419. Schwartz, Gary. 1998. “Medical Malpractice, Tort, Contract, and Managed Care.” Illinois Law Review 3, pp. 885–​907.

 

94   JENNIFER ARLEN Sharkey, Catherine M. 2005. “Unintended Consequences of Medical Malpractice Damage Caps.” New York University Law Review 80, pp. 391–​512. Sharkey, Catherine M. 2013. “Economic Analysis of Punitive Damages.” In:  Jennifer Arlen, ed., Research Handbook on the Economics of Torts. Northampton, MA: Edward Elgar Press. 486–​512. Shavell, Steven. 1980. “Strict Liability versus Negligence.” Journal of Legal Studies 9, pp. 1–​25. Shavell, Steven. 1982. “The Social versus the Private Incentives to Bring Suit in a Costly Legal System.” Journal of Legal Studies 11, pp. 333–​339. Shavell, Steven. 1984. “Liability for Harm versus Regulation of Safety.” Journal of Legal Studies 13, pp. 357–​374. Shavell, Steven. 1987. Economic Analysis of Accident Law. Cambridge, MA:  Harvard University Press. Shavell, Steven. 2004. Foundations of Economic Analysis of Law. Cambridge, MA: Harvard University Press. Shavell, Steven. 2007. “Liability for Accidents.” In: A. Mitchell Polinsky and Steven Shavell, eds., The Handbook of Law and Economics. Amsterdam: North Holland. 139–​182. Shavell, Steven. 2014. “A Fundamental Enforcement Cost Advantage of the Negligence Rule over Regulation.” Journal of Legal Studies 42, pp. 275–​302. Sloan, Frank A., Whetten-​Goldstein, Kathryn, Githens, Penny B., and Entman, Stephen S. 1995. “Effects of the Threat of Medical Malpractice Litigations and Other Factors on Birth Outcomes.” Medical Care 33, pp. 700–​7 14. Sloan, Frank A. and Chepke, Lindsey M. 2008. Medical Malpractice. Cambridge, MA: MIT Press. Spence, Michael. 1977. “Consumer Misperceptions, Product Failure and Producer Liability.” Review of Economic Studies 44, pp. 561–​572. Spier, Kathryn E. 1997. “A Note on the Divergence Between the Private and the Social Motive to Settle Under a Negligence Rule.” Journal of Legal Studies 26, pp. 613–​621. Spier, Kathryn. 2007. “Litigation.” In:  A. Mitchell Polinsky and Steven Shavell, eds., The Handbook of Law and Economics. Amsterdam: North Holland. 259–​342. Spier, Kathryn. 2011. “Product Safety, Buybacks and the Post-​Sale Duty to Warn.” Journal of Law, Economics, & Organization 27, pp. 515–​539. Stein, Alex. 2012. “Toward a Theory of Medical Negligence.” Iowa Law Review 97, pp. 1201–​1257. Stremitzer, Alex and Tabbach, Abraham. 2014. “The Robustness Case for Proportionate Liability.” B.E. Journal of Theoretical Economics 14, pp. 371–​395. Studdert, David M. et al. 2000. “Negligent Care and Malpractice Claiming Behavior in Utah and Colorado.” Medical Care 38, pp. 250–​260. Studdert, David et  al. 2006. “Claims, Errors, and Compensation Payments in Medical Malpractice Litigation.” New England Journal of Medicine 354, pp. 2024–​2033. Sykes, Alan O. 1984. “The Economics of Vicarious Liability.” Yale Law Journal 93, pp. 1231–​1280. Sykes, Alan O. 1988. “The Boundaries of Vicarious Liability: An Economic Analysis of the Scope of Employment Rule and Related Legal Doctrines.” Harvard Law Review 101, pp. 563–​609. Talley, Eric L. 2013. “Law, Economics, and the Burden(s) of Proof.” In:  Jennifer Arlen, ed., Research Handbook on the Economics of Torts. Northampton, MA:  Edward Elgar Press. 305–​329. Thaler, Richard H. and Sunstein, Cass R. 2008. Nudge:  Improving Decisions about Health, Wealth, and Happiness. New York: Penguin.

 

ECONOMICS OF TORT LAW    95 Vicusi, W. Kip. “Empirical Analysis of Tort Damages.” In:  Jennifer Arlen, ed., Research Handbook on the Economics of Torts. Northampton, MA: Edward Elgar Press. Weiler, Paul C. 1991. Medical Malpractice on Trial. Cambridge, MA: Harvard University Press. Weiler, Paul C. et  al. 1993. A Measure of Medical Malpractice:  Medical Injury, Malpractice Litigation, and Patient Compensation. Cambridge, MA: Harvard University Press. Wickelgren, Abraham. 2006. “The Inefficiency of Contractually-​Based Liability with Rational Consumers.” Journal of Law, Economics & Organization 22, pp. 168–​183. Wickelgren, Abraham. 2013. “Law and Economics of Settlement.” In: Jennifer Arlen ed., Research Handbook on the Economics of Torts. Northampton, MA: Edward Elgar Press. 330–359. Zeiler, Kathryn. 2005. “Turning from Damage Caps to Information Disclosure: An Alternative Approach to Tort Reform.” Yale Journal of Health Policy, Law, and Ethics 5, pp. 385–​400. Zeiler, Kathryn and Hardcastle, Lorian. 2013. “Do Damage Caps Reduce Medical Malpractice Insurance Premiums? A Systematic Review of Estimates and the Methods Used to Produce Them.” In: Jennifer Arlen, ed., Research Handbook on the Economics of Torts. Northampton, MA: Edward Elgar Press. 551–​587. Zeiler, Kathryn, Black, Bernard, Silver, Charles, Hyman, David, and Sage, William. 2007. “Physicians’ Insurance Limits and Malpractice Payments: Evidence from Texas Closed Claims, 1990–2003.” Journal of Legal Studies 36, pp. S9–​S45.

 

Chapter 4

ESTIMATI NG PA I N-​ AND-SU FFERING DA MAG E S Ronen Avraham

4.1. Introduction When a person is injured, tort law recognizes several types of losses: the victims’ economic loss (actual medical costs and diminished earning capacity) and noneconomic loss, which serves as a catchall for many losses, such as pain and suffering, mental anguish, emotional distress, and loss of enjoyment of life. Historically, the common law recognized pain-​and-​suffering losses only in intentional torts. In negligence claims, by contrast, the common law implicitly, and sometimes explicitly, expected the plaintiff to get over it. Over the years as courts have started to award pain-​and-​suffering damages in negligence cases, courts have struggled to clearly define and distinguish different aspects of these damages (Whaley 1992).1 One of the consequences of this struggle is the emergence of a confusing array of terms meant to elucidate the content and scope of pain-​and-​suffering damages. Courts refer, among other terms, to mental anguish, emotional losses, emotional distress, loss 1 

Part of this confusion arises from the distinction between contract cases and tort cases, and part arises from the differences in intentional and negligent torts. The negligent performance of a contractual duty may be both a tort and a breach of contract. Similarly, bad-​faith breach of contract may be deemed a tort, thus opening the door for emotional-​distress damages. Regarding intent, courts often allow actions for emotional distress in intentional torts, but only the minority of courts allows negligent infliction of emotional distress claims. As courts have gradually widened emotional distress claims, the demarcations between precedent related to intentional and negligent infliction have likewise been obfuscated. There is even confusion arising from a distinction between psychological damages and the more physical damages that make up pain and suffering. While pain and suffering actually covers many categories of nonpecuniary loss, including physiological pain, anguish and fear, emotional distress from losing someone you love, and the enduring loss of enjoyment of life, all of these are psychological losses. However, some types of physical loss are also included under pain and suffering. For instance, Chamallas and Wriggins (2010) point to reproductive harm as a physical loss that still represents nonpecuniary loss.

 

ESTIMATING PAIN-AND-SUFFERING DAMAGES    97 of consortium, hedonic damages, and psychic damages (Croley and Hanson 1995). With so many different terms, it is no wonder that confusion has been rampant. For example, damages for loss of enjoyment of life, alternatively known as hedonic damages, are intended to compensate for the loss of quality of life (Schwartz 2004). Some courts have had significant difficulties interpreting this term, and have even expressed doubts about “whether loss of enjoyment of life is compensable at all, and if so, whether it is part of pain and suffering, mental anguish, or physical impairment, or is a separate, independent category of damages” (Golden Eagle Archery, Inc. v Jackson, 116 S.W.3d 757, 768 [Tex. 2003]; Schwartz and Silverman 2004). While most jurisdictions treat loss of enjoyment of life, or hedonic loss, as a part of pain and suffering, other jurisdictions allow recovery of hedonic damages as a separate category of damages. In yet other states, either there is no clear ruling or hedonic damages are allowed in some instances and not in others (Schwartz and Silverman 2004).2 With respect to their content, some courts have indicated that these awards compensate for “the inability to perform activities which had given pleasure to this particular plaintiff ” (McGarry v Horlacher, 775 N.E.2d 865, 877-​78 [Ohio Ct. App. 2002]). Other courts have treated hedonic damages not as affirmative distress or suffering, but forgone gains, such as being unable to engage in activities that the victim valued, such as athletics or sex (Day v Ouachita Parish School Bd., 823 So. 2d 1039, 1044 [La ct. App. 2002]; Allen v Wal-​Mart Stores, Inc., 241 F.3d 1293, 1297 [10th Cir. 2001]; Varnell v Louisiana Tech University, 709 So. 2d 890, 896 [La. Ct. App. 1998]). Some courts have anchored them in the loss of something recognizable such as a limb or mental capacity (Pierce v N.Y. Cent.R.R.Co., 409 F.2d 1392 [Mich. 1969]; Matos v Clarendon Nat. Ins. Co., 808 So. 2d 841, 849 [La. Ct. App. 2002]; Kirk v Wash. State Univ., 746 P.2d 285, 292 [Wash. 1987]; Nemmers v United States, 681 F. supp 567 [C.D. Ill. 1988]). With so much inconsistency and indeterminacy, one should not wonder that pain-​ and-​suffering damages are under constant attack and are a major component of every tort reform. Opponents of pain-​and-​suffering damages argue that, unlike pecuniary damages, pain-​and-​suffering damages are hard to quantify accurately (King 2004). They argue that victims exaggerate their losses to receive higher damage awards, so that awarding pain-​and-​suffering damages may frustrate the function of tort law by compensating the victims too highly and arbitrarily. Another argument against pain-​and-​suffering damages is that the difficulty in objective measurement leaves the potential for enormous variance in awards at the discretion of individual judges and juries (Diamond 1998). This individual discretion may create a lack of horizontal equity and thus impede consistency among awards for like victims (Bovbjerg 1989).

2  States that treat hedonic damages as part of pain and suffering include Kansas, Nebraska, New York, Ohio, Pennsylvania, California, Minnesota, and Texas. States allowing hedonic damages in addition to pain and suffering include Maryland, New Mexico, South Carolina, and Wyoming.

 

98   RONEN AVRAHAM Avraham (2015) recently argued that from a law and economics perspective, pain-​ and-​suffering damages should be fully compensated and should receive the same “respect” that economic damages receive. I provided several arguments for that view. First, Avraham (2105) argued that the lack of horizontal equity might not represent a problem with damage calculations. Factfinders may treat like cases differently, but there is another possible explanation for heterogeneity in awards for injuries that, on the surface, appear to be “the same.” Judges and juries may be aware that there is great variation in how individuals subjectively experience pain and suffering (Chamallas and Wriggins 2010). In other words, factfinders in individual cases are aware of facts that are not available to researchers afterward. Moreover, evidence shows that observable seriousness of the physical injury is a reliable predictor of the size of award (Sloan and Hsieh 1990; Vidmar 1999; Vidmar, 1993; Diamond 1998). Another indication that pain-​ and-​suffering awards are not as wildly variant as some critics claim is that even though European countries have implemented rules to minimize horizontal inequity in damage awards, European pain-​and-​suffering awards are similar to those in the United States (Sugarman 2005). Still, even if there is real horizontal inequity in pain-​and-​suffering awards, the conclusion that they should be abolished does not follow. Rather, the inequity merely indicates the need for a more determinate process for factfinders to follow. Just because damage is difficult to quantify does not mean it should be ignored. The difficulty of quantification and the lack of horizontal equity it might cause do not justify setting damages at zero. Second, pain-​and-​suffering damages are not the only kind that prove difficult to quantify. Economic damages are often equally difficult (Rabin 1993). Lost wages and future medical expenses, for instance, have been drastically miscalculated by courts, showing that even these pecuniary losses are difficult to estimate precisely (Seffert v Los Angeles Transit Lines, 364 P.2d 337 [Cal. 1961]). There is no justification to single out for elimination pain-​and-​suffering damages from among all the difficult-​to-​quantify categories of damages. Third, the mere fact that some plaintiffs may exaggerate their symptoms provides no justification to eliminate pain-​and-​suffering damages. Courts have developed methods of correcting for this possibility. They have been estimating pain-​and-​suffering losses in intentional tort cases for centuries. Fourth and perhaps most importantly, scientific developments in the field of neuroimaging are creating the possibility of understanding and even verifying both pain and some emotional conditions (Kolber 2007). One recent study successfully used fMRI processes to identify specific neurologic signatures that identified pain sensation with very high accuracy levels (Wager et al. 2013). It is possible that science will soon be able to verify an accident victim’s subjective level of pain. Exaggeration may soon be impossible because judges and juries may have access to objective, scientific methods of quantifying pain and suffering as accurately as the quantification methods available for physical damages.

 

ESTIMATING PAIN-AND-SUFFERING DAMAGES    99 This kind of scientific progress is especially important as a potential response to the problem of malingering, which is when an individual continues to report pain even after it has stopped for the purpose of extending disability benefits or inflating a damage award (Cunnien 1997). It is hard to estimate malingering, but it may be a factor in 34% to 40% of chronic pain cases (Mittenberg 2002; Gervais 2001; Kolber 2007). Further, studies show that treatment outcomes are worse for people involved in personal injury litigation than for those who do not seek compensation (Mendelson 1997). Even if the litigation is a cause of malingering, it could be because the stressors and delays of the tort system retard the healing process. Or, people who are prone to poor treatment may also be more likely to file a lawsuit. Recent scientific advances may obviate the need for this type of speculation by creating objectively verifiable pain metrics. In sum, the best solution for dealing with the high cost of administrating pain-​and-​ suffering damages and the alleged variation in horizontal quality is to reduce the cost of administration and to increasing uniformity, not limit plaintiffs’ recoveries. The best way to accomplish this is via simplifying ways to estimate pain-​and-​suffering losses. In this chapter, I survey a number of solutions discussed in the literature on how to simplify the estimation of pain-​and-​suffering damages to cut administrative costs. My goal is to demonstrate the feasibility of the task of estimating the loss more than to recommend any specific path to it.

4.2.  How To Estimate Pain-​ and-​Suffering Damages As the court in Botta v Brunner said, “For hundreds of years, the measure of damages for pain and suffering following in the wake of a personal injury has been ‘fair and reasonable compensation.’ This general standard was adopted because of universal acknowledgment that a more specific or definitive one is impossible. There is and there can be no fixed basis, table, standard, or mathematical rule that will serve as an accurate index and guide to the establishment of damage awards for personal injuries” (Botta v Brunner, 138 A.2d 713, 718 [N.J. 1958]). Defining and monetizing “fair and reasonable” has been a matter of protracted difficulty. Indeed, the problem of standardizing and achieving horizontal and vertical equity in pain-​and-​suffering damages has been a hot topic for both scholars and courts for decades, but no common understanding has emerged. (Horizontal equity means similar compensation for injuries of similar severity, and vertical equity means more severe injuries receive more compensation than less severe ones do. These principles are not well served by the current tort system.) This chapter will demonstrate that there have been various attempts to provide some structure to the definition of “fair and reasonable compensation,” and it will critically

 

100   RONEN AVRAHAM survey a number of mechanisms used to quantify and monetize damages. Some of these are tailored specifically to pain and suffering, and others are used to determine a total award without distinguishing the amount awarded (if any) for pain and suffering. Some are individualized and some are not. For some of these systems (such as Worker’s Compensation and per diem), calculations have been used by courts, whereas for others (such as Quality-​Adjusted Life Year [QALY]) such methods have been used only by nonjudicial institutions. In short, this chapter details available methods for analyzing pain-​and-​suffering damages. I start by discussing the individualized approaches: the golden rule, per diem, medical costs as a basis for pain and suffering, utilizing a loss of pleasure of life scale, and finally, QALYs. Then, I discuss the non-​individualized (routinized) approaches: fixed amounts of pain-​and-​suffering damages, fixed caps on noneconomic damages, flexible caps on noneconomic damages, fixed amounts for the entire injury, legislated damages schedules, court-​based damages schedules, and commission-​created damages schedules. While I  do not necessarily recommend a specific approach, my main goal is to demonstrate that contemporary complaints about the high costs associated with valuing pain and suffering cannot serve as a sound reason to limit or eliminate pain-​ and-​suffering damages from tort law: there are many cheaper and more feasible alternatives to estimating them.

4.2.1. Individualized Approaches This section looks at individualized approaches to estimating pain-​and-​suffering damages. Most of these approaches have been examined by courts in one form or another.

4.2.1.1. The Golden Rule One way to operationalize the idea of fair compensation is via the “Golden Rule” jury instruction. The Golden Rule approach to valuation asks tort juries to place themselves in the shoes of the plaintiff when determining damages. That is, jury members should estimate the amount of money they would require as (ex post) compensation for having to experience the victims’ pain and suffering (Avraham 2006). The Golden Rule has been widely rejected by courts because, to achieve the goal of compensating for the injuries suffered, courts have observed that we need “fair and impartial jurors who are not governed by sympathy or bias” (American Law Reports 2d, 1964). There are several possible variations of the Golden Rule. The one that has been examined and rejected by courts looks at the plaintiff ’s willingness to accept (WTA) compensation (ex post) for her injury. Other variations exist that potentially have a greater chance of success in courts. One possible variation is to ask the jury to estimate how much the plaintiff would have been willing to accept (ex ante) to bear the

 

ESTIMATING PAIN-AND-SUFFERING DAMAGES    101 risk of injury. These two variations of plaintiff ’s WTA can be contrasted with her willingness to pay (WTP) to prevent the harm, or a certain chance of a particular harm. In the debate on how to optimize regulation, these two concepts (WTA versus WTP) have received considerable attention. The debate, however, is thousands of years old.3 Asking people what they would be willing to accept in payment for incurring the loss might be problematic because some evidence suggests people will demand more to give up a good than they would be willing to pay to obtain that good in the first instance. (Thaler 1991; Sunstein 1986). Endowment effect theory has been used to explain why studies have found that WTP (for something you do not have) is routinely lower than WTA (to depart with something you have). Yet, failures in the design of the experiments that found the gap probably provide the best explanation for this phenomenon (Plott and Zeiler 2005). The WTP/​WTA gap often disappears when experimental procedures are altered to eliminate alternative explanations (Klass and Zeiler 2013). In a similar fashion, some suggest that the WTP metrics should be used to measure hedonic loss (Viscusi 1998). Geistfeld (1995) has considered using a similar approach as a method for determining pain-​and-​suffering damages. He asks what price would a consumer pay to avoid a 1/​10,000 chance of injury, rather than asking what they would pay to avoid certain injury. He uses a methodology employed by regulatory agencies, and believes that it can provide an objective basis for determining pain-and-suffering damages. This approach, however, is prohibited in tort cases because the question of damages is not decided by valuing risks, but instead by valuing the damages after they occur (Viscusi 1988; Sunstein 1993). Some argue that the WTP method also inflates the valuation because of the underestimation of hedonic adaptation, which is people’s tendency to adjust to their new disability and restore their level of overall happiness. Hedonic adaptation suggests that the plaintiff ’s WTP has the potential to be substantially higher immediately following the injury than it will be later, at the conclusion of the lawsuit (Bronsteen, Bucafusco, and Masur 2008). In addition, WTP is determined by the use of hypothetical questions about experiences the respondents have never had. Further, WTP varies with income and wealth, which would lead us to conclude wrongly that poor people suffer less pain (Abel 2006).

3  The Mishna (Judaism’s first major book describing Jewish law written around 200 A.D.), has instructions on compensations for pain. It says, “they assess how much a person such as [the injured] would be willing to be paid in order to suffer such pain” (mBK 8.2). This is a WTA approach. The Talmud (Judaism’s second major book completed in the 5th century A.D., which is comprised of commentaries on The Mishna), contemplates a WTP approach. It says that pain-and-suffering damages should be appraised by estimating one’s willingness to pay to have an anesthetic drug for the amputation of his arm, which by the decree of the government must be amputated by a sword (tBK 8).

 

102   RONEN AVRAHAM Even though the Golden Rule is prohibited in the United States, in practice, plaintiffs’ lawyers openly advocate and attempt to find ways around the Golden Rule to encourage the jurors to sympathize with the plaintiff. In a case where a plaintiff lost an arm, some jurisdictions allow counsel to ask the jury how much the plaintiff would have sold the arm for (McElheney 1987). Because the jurors have no way of knowing what the plaintiff would do in this situation, this selling-​price perspective on damages encourages the jurors to consider how much money they personally would accept in return for the injury (McCaffery 1995). This is just one of many techniques used by plaintiff ’s attorneys to encourage the jury to empathize with, or put themselves in the place of, the plaintiff. Yet, plaintiff ’s lawyers making Golden Rule arguments risk a mistrial, as in Velocity Express Mid-​Atl., Inc. v Hugen, 585 S.E.2d 557 (2003).

4.2.1.2. Per Diem Courts have considered the per diem approach for estimating pain and suffering. Per diem arguments are a mechanism allowed by some states (e.g., Arizona) whereby a lawyer argues that the ongoing pain and suffering should be calculated according to a short period, such as a day, an hour, or a minute, and then multiplied by the victim’s remaining life expectancy (Totaro 2006). Most states have deemed the use of a mathematical formula in determining an award for pain and suffering to be a per diem if the amount is derived from a multiplication of a unit of time and a dollar amount (Giant Food Inc. v Satterfield, 603 A.2d 877 [Md. 1992]). Many states have ruled that is improper for counsel to suggest per diem valuations, and that such arguments are prohibited as a matter of law. But some states have decided that the argument is appropriate and can be given at any time. Some states have decided that the use of the per diem argument is within the discretion of the trial court judge (King 2003), while still other states permit per diem arguments only when accompanied by a caution from the judge that the arguments are not “facts” (Debus v Grand Union Stores, 621 A.2d 1288 [Vt. 1993]). Reasons against allowing per diem arguments have been advanced by both scholars and courts alike. Among many others, they include lack of evidentiary basis for the conversion of pain and suffering into monetary damages, misleading the jury into making larger awards, and disadvantaging the defendant because the defendant must rebut an argument with no basis in evidence (Giant Food, 603 A.2d at 879). Scholars have also argued that per diem arguments promote cognitive anchoring (assigning a value according to the only suggested number you have heard and adjusting slightly from that number), in the face of “the impossible task of converting in some rational way pain and suffering into a sum of money” (King 2003). Overall, it seems plausible that a juror would have an easier time estimating the value of a day’s worth of suffering than a lifetime’s worth, and, since so much subjectivity is involved, this seems like an appropriate technique. On the other hand, some argue that damage awards generated by per diem calculations are generally inflated because they are suggested by plaintiff ’s counsel for this purpose and thus do

 

ESTIMATING PAIN-AND-SUFFERING DAMAGES    103 not serve the goal of a “fair and reasonable compensation,” as is the goal of tort law. The answer to this claim is that defendant’s lawyers are always free to offer a different number or a different way to get at the number than that proposed by plaintiff ’s lawyers. Indeed a recent study that analyzed real jury deliberations suggests that plaintiff ’s lawyers requests do not serve as an anchor for the jury damages determination (Diamond et al. 2011).

4.2.1.3. Medical Costs as Basis for Pain and Suffering In an effort to streamline and simplify pain-​and-​suffering damage calculations, one suggestion has been to use a victim’s economic loss (also known as special damages) as a starting point. A common approach sometimes used by insurance adjusters treats special damages as a reference point, which is then multiplied by a factor of three or more to arrive at the settlement amount. This approach is known as “three times specials.” However, sometimes “three times specials” means three times the medical bills, exclusive of damages for loss of income. Eliminating loss of income from the calculation is sensible on the ground that the legal system should not recognize differences in pain and suffering based on the victim’s earning capacity. A more sophisticated approach will be to use a multiplier of medical costs that is not fixed but rather increases as medical costs increase (Avraham 2006). The specific multiplier could be determined by precedent or by state or federal law, and could be different for different categories of torts. For example, in a system of Non-​binding Age-​Adjusted Multipliers (NBAAM) multipliers would adjust for the age of the victim, and calculated according to the anticipated remaining years of life. These multipliers will not be binding, leaving discretion in the hands of the jury, while providing non-​binding guidelines that increase both horizontal and vertical equity. Because medical costs can be determined by evidence, this decreases the guesswork for which tort awards of pain-​and-​suffering damages are widely criticized. The main problem in adopting NBAAM is a lack of evidence relating to what the multipliers should be, and what factors should be considered when calculating these multipliers. Despite these difficulties, NBAAM has been used in courts. One court applied a multiplier of 1.5 to medical costs (Rhoades v Walsh, 2009 U.S. Dist. LEXIS 75784, 47). Another court would have used NBAAM if medical cost information had been available (Henderson v Atl. Pelagic Seafood, LLC, 2011 U.S. Dist. LEXIS 33727, 34). Recent empirical work lends some support, as well as suggesting some possible guidelines for NBAAM calculations. A recent study of Taiwanese district court accident cases shows that nonpecuniary damages are strongly correlated with amounts awarded for medical expenses and the level of injury (Chang et al. 2013). The authors of that study also find that the U.S. tort system awards nonpecuniary damages at roughly the same percentage of total damages as Taiwan does, indicating that nonpecuniary damages in the United States may also be influenced by the same factors. These findings not only indicate that judges and juries intuitively anchor their nonpecuniary awards to the

 

104   RONEN AVRAHAM amount of medical expense but also may provide a reasonable mechanism by which to calculate the precise multiplier that should be used as a guideline for triers of fact in the United States. Implementing a formal mechanism for NBAAM could simultaneously help to eliminate the irrational tendency cognitively to anchor nonpecuniary damages to the amount requested by the victim’s lawyer, while preserving the rational tendency of judges and juries cognitively to anchor nonpecuniary damages to the amount of medical damages. It would also allow introducing other relevant factors into the calculation, such as the age of victim. NBAAM can thus serve as both a preservation of and an improvement upon the current system.4

4.2.1.4. Loss of Pleasure of Life Scale Another suggested method for rating hedonic losses is the Lost Pleasure of Life (LPL) Scale, which asks a psychologist to evaluate a person’s postinjury lifestyle in comparison to her preinjury lifestyle (Berlá 1989). The psychologist is asked to gauge the ability of the person to experience pleasure in four areas: practical functioning, emotional functioning, social functioning, and occupational functioning. A zero rating is an individual who has lost no ability to function, whereas a 100 rating means that the individual is unable to function and cannot derive any pleasure from a particular area. Each of these four functionalities is judged upon the percentage loss from the preinjury state. Losses are divided according to those percentages into minimal, mild, moderate, severe, extreme, and catastrophic. This method would also consider age and length of time that a person’s loss of pleasure is likely to persist (Berlá 1989). The goal is to provide an objective, consistent, and expert recommendation about the degree of LPL. One weakness in the LPL measure is that it does not solve the problem of translating these measures into dollars. Indeed, courts’ citations to the LPL are relatively scarce, although courts have used it to estimate damages (Schwatz and Silverman 2004; Brookshire and Smith 1992–​1993; Hunt v K-​Mart Corp. 981 P. 2d 277 [June 1999]).

4.2.1.5. Quality Adjusted Life Years QALY is a concept imported from the field of health economics, and it refers to the value of living a year of life with a certain health impairment. In fact, it is an umbrella term for various different ways of comparing diseases and injuries to one another and to a condition of perfect health. It is a form of cost-​effectiveness analysis where health states that persist for a period of years have been ranked by patients who experienced the states, by physicians who treat the states, or by members of the general population (Adler 2006). The rankings are on 0–​1 scale, where 0 is death and 1 is perfect health. The “life year” concept enables the rankings to address both the severity and the duration of health impairments. 4  One problem with this approach is that it might create incentives to inflate medical costs in order to obtain higher pain and suffering damages and higher contingency fees. For a rejoinder to this concern, see Avraham (2006).

 

ESTIMATING PAIN-AND-SUFFERING DAMAGES    105 QALY’s are increasingly used by regulatory agencies as the basis for their rulemaking. Agencies such as the FDA, EPA, OMB, and Public Health Service have all used the QALY cost-effectiveness analysis in varying capacities.5 QALY is preferred by many to a traditional cost–​benefit analysis because cost–benefit analyses require placing dollar valuations on the outcome of any program or intervention, whereas QALY only requires choosing between options based on their relative returns in QALY.6 Using QALY requires controversial decisions, such as determining the appropriate dollar amount of each QALY and the appropriate QALY scale to determine the weight each injury should receive. Still, with more research and analysis, QALY may provide a reliable reference point for calculating pain-​and-​suffering damages in courts (Karapanou and Visscher 2010). Yet, somewhat surprisingly, despite the expansive use of QALY in evaluating health programs and medical treatments, the concept has not penetrated tort law. Only a couple of papers have proposed that QALY research could be converted for use in tort cases by basing pain-​and-​suffering awards on the impact of the health impairment (Miller 2000; Karapanou and Visscher 2010). Whereas Miller (2000) proposed choosing QALY scales on a case-​by-​case basis, Karapanou and Visscher (2010) proposed using general QALY weights as a starting point for a more individualized assessment of the plaintiff ’s injury. Either of these two methods provides some objective measurement in which pain-​and-​suffering damages could be grounded.

4.2.2. Non-​Individualized Approaches: Routinization Routinization is defined as an attempt to establish a form of “orderliness and predictability that is conducive to individuals’ planning their lives” (Goldberg and Zipursky 2010). As third-​party insurance has expanded, so have the numbers of repeat players in the tort system: even small-​time defendants are represented by insurance companies with vast experience in settling tort claims (Robinette 2013). Generically, a routinized system is one that permits a victim to file a claim, dispenses with a prolonged inquiry into fault, and results in a very large percentage of victims receiving compensation without having to deal with the difficult issues of individualized monetizing of pain-​and-​suffering damages. Such systems—​of which Worker’s Compensation is a prime example—​generally erase the distinction between pecuniary and nonpecuniary components of harm (Goldberg and Zipursky 2010).

5  Food and Drug Administration, Environmental Protection Agency, Office of Management and Budget, Public Health Service. 6  “To many in the worlds of medicine and public health, any attempt to place a value on human life is anathema. Thus most ‘economic’ evaluations of health care have applied [cost effectiveness] analysis, which limits the analyst’s responsibility to providing information about the efficiency with which alternative strategies achieved health effects” (Garber 1999).

 

106   RONEN AVRAHAM Routinization can make compensation more predictable, efficient, and administrable: benefiting all of the repeat players in the system. Defendants prefer predictable liability payments because they reduce defendants’ liability premium. Insurers prefer predictable payments because they facilitate premium and profit calculations that are more accurate. Plaintiffs’ lawyers working on a contingent fee prefer a large judgment or settlement in relation to the amount of hours spent on the case. All of these interests favor the prompt resolution of cases. There are, of course, major drawbacks to routinizing pain-​and-​suffering damages. Too much routinization puts people into classes while ignoring relevant individual circumstances. For example, a violinist who loses a finger and a law professor who loses a finger do not experience the same suffering. If their recovery were routinized so that they received the same compensation, the violinist would be undercompensated whereas the law professor would be over compensated. In addition, routinization fixes damages for years without updating them for changes in victim’s experience. Depending on the type of routinization, this problem could possibly be overcome; for example, by regular inquiries into appropriate values. Section 4.2.2.1 will show that if unpredictability is the primary objection to pain-​and-​suffering damages, there are viable possibilities to allay the concern. Section 4.2.2.1 surveys the major forms of routinization.

4.2.2.1. Fixed Amount of Pain-​and-​Suffering Damages When one event has many victims, it sometimes makes sense to award each one the same amount or very close to the same amount. The most famous example is the 9/​11 Victim’s Compensation Fund, which set pain and suffering at $250,000 for most cases, giving little consideration to individual circumstances. While the plan drew heavy criticism from plaintiff ’s lawyers and victims, it was implemented and executed quickly (Glaberson 2001). As with all other routinization mechanisms, nuance and individualization are sacrificed in favor of administrative efficiency. Awarding fixed amounts is an example of the biggest such sacrifice. If the amount is low, everybody might be undercompensated. If the amount is high, everybody might be overcompensated. A medium amount, might cause the severely injured to be undercompensated and the least severely injured to be overcompensated. Even though a predetermined payout system does eliminate the need for litigation, some determinations must still be made. For instance, in the 9/​11 Victim’s Compensation Fund, there were arguments that people who had suffered extraordinary harm should be paid a higher amount for their pain and suffering (Steenson and Saylor 2009). The claimants were entitled to a hearing on request to make a case for higher than average pain-​and-​suffering awards, but in practice, the awards remained uniform. In addition, the administrators of the Fund represented the final word on claims. Once victims chose to make a claim with the Fund, they waived any tort claim against any potential tortfeasor (except the terrorists) so that litigation was no longer a possibility and their only option for compensation was through the Fund administration. There was literally no appeals process, which on the one hand promoted streamlined payouts but on the

 

ESTIMATING PAIN-AND-SUFFERING DAMAGES    107 other hand provided no recourse for an undercompensated individual to seek review of his case. The 9/​11 Victim’s Compensation Fund is an example of how routinized payouts setting a fixed amount for pain and suffering might play out. Despite its weaknesses, the legislature succeeded in what it set out to accomplish by establishing the Fund. Litigation against the airlines (and other possible tortfeasors) was minimized, and claimants were paid swiftly and surely. The 9/​11 Fund was created at a time of unprecedented patriotism and cooperation in the United States (Steenson and Saylor 2009). It remains to be seen if a project of similar magnitude under other circumstances could be administered as efficiently. Moreover, it is unlikely that routine claims and payouts from a fund would be useful outside the context of major catastrophes with a large number of victims and one entity (or a very few) potentially liable. However, the 9/​11 Fund might represent a possible solution to avoid the high administrative costs of claim payouts when liability is not a question.

4.2.2.2. Noneconomic Damage Fixed Caps Damage caps represent a hybrid approach between individualized and entirely routinized damages. More than half of the fifty states have passed legislation imposing caps on noneconomic damages (Avraham 2014). State caps have varying degrees of flexibility. Some impose an absolute cap on damages, irrespective of any background circumstances. For example, California capped noneconomic damages for medical malpractice actions at $250,000 in 1975, even though the real value of that amount has declined dramatically over the last 40 years (Cal. Civ. Code §3333.2 [2004]). Other states have caps that are adjusted for inflation each year. For example, Idaho adjusts its noneconomic damages cap annually based on average wage data (Idaho Code §6-​1603 [2004]). There are also states that have attempted to eliminate some of the rigidity in their damage caps, by allowing the judge or jury to waive the damages cap if aggravating circumstances justify a higher award. Caps maintain the traditional duty of the jury to assess pain-​and-​suffering damages (and in that respect provide no help on how to do it), but they limit the discretion so that any harm above the cap is truncated and thus undercompensated. Thus, caps solve the problem of factfinders erroneously overcompensating the least severely injured but leave unresolved the problem of undercompensating the most severely injured. Moreover, caps do not help courts determine damages for injuries below the caps. In sum, caps do not reduce administrative costs, and they do not increase equity by much or at all. And yet, caps are the most common tort reform in the United States.

4.2.2.3. Noneconomic Damage Flexible Caps In Australia, judges take the role occupied by juries in the United States (Masada 2004). Judges are finders of fact and assess damages.7 Perhaps the fact that judges 7 

Australia also requires fees for legal services to be payable on an hourly basis, instead of contingency fees which discourage attorneys from pursuing speculative claims.

 

108   RONEN AVRAHAM are repeat players in assessing damages is why New South Wales, the largest state in Australia, passed a somewhat complicated liability reform for health care providers in 2001. The law imposes a cap of AU$350,000, which is adjusted each year for market conditions, but it is much more nuanced compared with the hard caps used by states in the United States. It delineates a damages estimation scheme and imposes a minimum loss requirement, as well as a loss threshold below which nonpecuniary losses are not awarded. Specifically, the New South Wales law compares all cases to a hypothetical most extreme case. For example, a hand injury would be compared with losing the whole hand. The judge is asked to find a specific percentage of the “most extreme” case for each injury seeking compensation. No noneconomic damages are awarded if the severity of the noneconomic loss is less than 15% of the most extreme case. For injuries where the noneconomic loss is between 15% and 33%, a table provides the maximum recovery as a percentage of the maximum possible recovery. For example, when the severity of the noneconomic loss as a proportion of the most extreme case is 15%, the damages (as a proportion of the maximum amount that may be awarded) is 1% of the AU$350,000 cap. When the severity is 25%, the damages are 6.5% of the cap, etc. For any injury where the injury surpasses 34% of the most extreme case, noneconomic damages will be the same percentage of the maximum amount. That is, an injury that is 50% as severe as when the most extreme case will result in AU$175,000 in damages for nonpecuniary harm (Masada 2004). Australia’s tort reform measures, including the “most extreme case” rule, has been successful in achieving some of its aims. Unlike statutory damages limitations in the United States, in Australia there are no constitutional questions causing insurers to be uncertain about the stability of tort reform (King 2010). And perhaps for that reason, the Australian damages reform has had the unambiguous effects of reducing litigation and of lowering insurance premiums. In his 2010 article, King argues that despite these apparently desirable effects, the tort reform made consumers economically worse off by forcing them to bear the risk of suffering an injury resulting in losses that exceed those limits. However, if reducing litigation and lowering insurance premiums are main policy goals in the United States, the Australian system, which has been in place since 2002, might provide a good model (Sugarman 2005; Sullivan 2005).

4.2.2.4. Fixed Amount for the Entire Injury While the tort system has experienced considerable informal routinization in recent years, trust systems provide a more formal model of routinization. For example, Agent Orange was an herbicide used by the U.S. Army in Vietnam (to kill plants that provided cover to the Vietnamese). It was widely believed to cause a variety of forms of cancer (Weinstein 2006). A class action was filed against the manufacturers of Agent Orange, and Judge Weinstein approved a settlement that set up a trust fund to provide compensation for anybody who could show a minimum possible connection to exposure. There was a list of diseases that were compensated according to matrices based upon age, time

 

ESTIMATING PAIN-AND-SUFFERING DAMAGES    109 since exposure, nature of the disability, and similar conditions. No distinction was made between pecuniary and nonpecuniary harm; the compensation was for the entire injury. Judge Weinstein believed that this system provided a mechanism to avoid the costs and delays of litigation for victims, while maximizing the impact of available funds by providing insurance policies instead of direct recovery, which covered a range of diseases for anybody who became ill.8 Another example is the British Petroleum (BP) Deepwater Horizon Oil Spill. The 2010 oil spill was the worst in American history, and it precipitated thousands of lawsuits (Mullenix 2011). Lawyers for BP and the plaintiffs entered into a settlement that relies upon the Specified Medical Conditions Matrix.9 In addition to listed baseline amounts for certain conditions, the variables on the matrix include length of hospital stay, geographic location at the time of exposure, and status of the victim (whether resident or clean-​up worker). The matrix is open to the public, but actual payment amounts are determined by a court-​appointed claims administrator and are subject to challenges. The settlement is open-​ended. Therefore, even though BP put $20 billion in trust, it must still pay out claims according to the matrix after the trust fund runs out. The matrix and the court-​approved settlement are available online (Deepwater Horizon Claims Administrator 2012). A similar solution was used for compensating victims of injuries resulting from the Dalkon Shield birth control device (Vairo 1992). The Dalkon Trust offered three settlement options with differing payouts and standards of proof. Option 1 awarded each claimant $725 (plus $300 for the husband) if the claimant merely stated she had used the device and suffered injury. Option 2 was for claimants with medical proof of use and injuries, but alternative causation complications, and it awarded claimants between $850 and $5500. Option 3 offered settlements based on prepetition claim values (excluding all transaction costs). The Dalkon Trust worked very well by almost all accounts (Vairo 1992; Chamblee 2004). It has even been hailed as a “paradigm of fairness and efficiency” because it “gave individual claimants a voice” (Chamblee 2004). The claims were paid out 9 years ahead of schedule, and the Trust ended up with extra funds. Despite minor criticism, the Trust 8  Asbestos trusts are similar. The ARPC (Analysis Research Planning Corporation) manages an aggregated trust from nearly all the large-​scale asbestos defendants, and uses a web-​based claims-​ processing system, which has processed millions of claims and paid billions of dollars with “cost to benefit ratios that are far lower than comparable settlement approaches” (U.S. Government Accountability Office, 2011). 9  Matrices have been used in other aggregate settlements; notably, the Vioxx litigation settlement included developing a matrix. Vioxx was a painkiller that was associated with increased risk of heart attacks and strokes. Merck, the manufacturer, settled with the plaintiffs for $4.85 billion. A third-​party claims administration used a points system to “score” plaintiffs based upon factors such as severity of the injury, age, and the presence of other risk indicators. The Vioxx settlement, like the Agent Orange settlement, differs from the BP settlement matrix in that it was for a fixed amount of money. Also, the Vioxx plaintiffs had no way of estimating how much their award would be before they had to opt in or out of the settlement, because the dollar value per point was not determined until after all the claims had been made (Erichson and Zipursky 2011).

 

110   RONEN AVRAHAM was considered successful. Keeping in mind its limited application, the Dalkon Shield Trust has the potential to serve as a model for large-​scale distribution of payouts for tort claimants (Vairo 1992). Another example of fixed amounts for the entire injury is the Hokie Spirit Memorial Fund, which was funded by private donations to compensate victims of the 2007 Virginia Tech shooting. The Hokie Spirit Fund paid out fixed amounts for the entire injury, and this was accomplished with minimal administration. Families of victims who died received lump-​sum payments of $180,000. The victims who survived received payments of either $40,000 or $90,000, depending on how long their hospital stay was. There was no discretion to vary this payment schedule at all. The Hokie Spirit Memorial Fund was very different from the 9/​11 Fund because it was created solely by charitable contributions, and the claimants were not required to waive their tort remedies. In fact, the Hokie Spirit Fund was not created or administered with tort liability in mind, and it had no legal effect on subsequent litigation (Feinberg 2007). Still, it demonstrates that pain-​and-​suffering losses can be rolled into a lump-​sum amount compensating victims for their entire injuries.

4.2.2.5. Legislated Damage Schedules Although damage schedules have been widely adopted overseas, they have never been adopted in the United States.10 In the U.S., courts have hinted at scheduling but have no authority to adopt such a mechanism (Sebok 2005; Rabin 2005; Seffert v Los Angeles Transit Lines, 364 P.2d 337 [Cal. 1961]). Indeed, the only damage schedules in the United States were adopted by legislation, such as in the Worker’s Compensation system. The U.S. Worker’s Compensation system offers a trade-​off wherein pain-and-suffering damages are not awarded in any case, and there is no option to sue in the tort system, but in return, there is no question of fault, and damages are awarded to every injured worker. The Worker’s Compensation system schedules damages in a predictable way (Larson and Larson 2008). This is designed to eliminate the transaction costs, delays, and uncertainty inherent in the tort system. The Workers’ Compensation model thus provides a no-​fault alternative that displaces tort law within a major field of accidents (Rahdert 1995). For example, the Longshore and Harbor Worker’s Compensation Act provides for four tiers of damages: Permanent Total Disability, Temporary Total Disability, Permanent Partial Disability, and Temporary Partial Disability (33 U.S.C. §908 [1988]; 5 U.S.C. §8105 [a]‌Total Disability [2011]). Total Disability is defined as loss of any two of hands, arms, feet, legs, or eyes. Total Disability benefits are 66.67% of average weekly wages to be paid during the continuance of the total disability. Partial disability is defined where loss of specific limbs, senses, and digits, and other impairments are scheduled according to severity. To illustrate the high and low end of the schedule, 10  Comparison studies of European and U.S. recovery levels for pain and suffering indicate that the United States provides more damages that are nonpecuniary: at least a 10:1 ratio. Differences in the strength of the social safety net, loser pay rules, and stronger regulatory rules resulting in fewer injuries might explain part of this difference.

 

ESTIMATING PAIN-AND-SUFFERING DAMAGES    111 the loss of an arm awards 312 weeks compensation at 66.67% of salary, and loss of a fourth finger provides just 15 weeks compensation. The schedule gives specific compensation periods for loss of phalanges—​a rating scale for where a digit or limb must be amputated—​and a variety of other metrics. All of the damages are awarded according to the schedule and for 2/​3 of wages before the accident. While the Workers’ Compensation scheme eliminates the messiness and expense inherent in the current system of determining and awarding pain-​and-​suffering damages, it does so at the cost of awarding no pain-​and-​suffering damages, even to the most deserving victims. This trade-​off has been seen every time the law has expanded liability itself: in such expansions, general (nonpecuniary) damages are always reduced or eliminated (Diller 2003; Abel 2006). The Worker’s Compensation scheme is relevant to the monetization of pain-​and-​ suffering damages because it demonstrates that various disabilities can be scheduled and quantified. One way to understand the Worker’s Compensation scheme is that the damages allocated include compensation for pain-​and-​suffering losses. Under this understanding, pain-​and-​suffering losses are understood to depend on one’s salary. As mentioned, in the context of NBAAM, it is hard to justify such an approach. Another way to understand the Worker’s Compensation scheme is that it quantifies all pain-​and-​ suffering damages at zero. Even under this understanding, there is nothing that prevents policy makers from treating pain-​and-​suffering losses in the same way, as a fixed amount according to the severity of the injury. Indeed, there has recently been some legislative action moving toward damage schedules. In 2004, the Washington legislature established a task force to promulgate a schedule for noneconomic damages (Washington State Legislature 2004). The purpose of the advisory schedule was to increase “the predictability and proportionality of settlements and awards.” The task force report authors believed that a schedule would promote both horizontal and vertical equity across noneconomic compensation awards (State of Washington Task Force 2005). The task force suggests that creating a schedule will take two steps: delineation of levels of severity and assignment of dollar values to the tiers. They suggest using a severity scale developed by the National Association of Insurance Commissioners, which is the best-​known and most widely used quantitative scale for classifying injury severity (Sowka 1980; Studdert 2011). As for assigning dollar values to the tiers, the task force report indicates that there is a significant lack of data available, but that the best source of data would be precedent. In a 2011 article by the same authors, which was written to supplement and emphasize the feasibility of the task force report recommendations, they referenced a 3-​year study done for the Canadian government called the Ontario Noneconomic Loss Study (ONELS) (Studdert 2011). The ONELS study conducted face-​to-​face interviews, where participants were asked to rate injury severity with thirty-​eight explanatory variables. This produced ratings for loss of quality of life for a variety of benchmark conditions (Sinclair and Burton 1994, 1997; Studdert 2011). The authors then used the weights to each injury derived from the ONELS study to compare with a Harvard Malpractice

 

112   RONEN AVRAHAM Insurers Medical Error Prevention Study (MIMEPS) to assign a monetary value to the categories of injuries (Ghandi 2006; Rogers 2006; Studdert 2006). Trained physician specialists reviewed claim files for malpractice insurance companies and used the payouts in those cases to determine a monetary value. Thus, the National Association of Insurance Commissioners table serves as the severity scale; the ONELS and MIMEPS studies assign a monetary value to the different severities; and together they form a schedule of damages. While scheduling damages has not yet materialized in legislation in the United States, the task force report for the State of Washington and the follow-​up research done to supplement this report indicate that it is conceivable to see some legislated damage schedules in the future.

4.2.2.6. Court-​Based Damage Schedules The previous section discussed legislated damages schedule. In Britain, a Judicial Studies Board publishes guidelines every few years for general damages in personal injury cases (McKay 2010). Designed for British courts when settling personal injury claims, these guidelines are non-​binding, but they are generally adhered to absent extenuating circumstances. The recommendations purport to cover “general damages,” which would include all nonpecuniary damages unrelated to medical costs. The guidelines give a narrow range for awards based on a variety of factors. Age and life expectancy are included in virtually all of the calculations. Other factors include the existence and extent of pain, depression, range of movement, and other physical limitations. The numbers are based on data derived from precedent and are adjusted according to statistical methodologies. The guidelines are increasingly referenced in scholarly works on the issue of pain-​and-​ suffering damages in the United States. They are thorough (with the exception of small claims with slight injury), and increasingly authoritative not just in Britain but in other European jurisdictions.11 The guidelines use the amount of damages awarded in reported cases as indicators in slotting particular injuries into its framework. They are intended to “distil the conventional wisdom contained in the reported cases,” and they provide consistency between awards as required by justice. The categories of damages include Paralysis, Head Injuries, Psychiatric Damage, Sensory Damage, Organ Damage, and many others (McKay 2010). Each is broken down into specific kinds of injuries (e.g., location and type of trauma), and divided into minor, moderate, moderately severe, and very severe on the severity scale. The guidelines are now in their thirteenth edition and are considered a must-​read by all personal injury lawyers in Britain because they serve as a jumping-​off

11  The introduction to the Guidelines describes the voluntary nature of the adoption of these standards by the courts. It further explains that the lack of data is the primary reason that small injury awards are not included. The introduction also explains that contesting the amount of the award has become much less prevalent in Britain because both plaintiff and defendant are very often willing to accept the Guidelines’ suggested award, so that litigation is limited to material disagreement about facts and evidence (McKay 2010).

 

ESTIMATING PAIN-AND-SUFFERING DAMAGES    113 point for damage negotiations in most disputes. Similar guidelines exist in other countries such as Ireland.12 Other countries that use a form of non-​binding schedules are Germany and the Netherlands. They utilize past pain-​and-​suffering awards to decide how much to award in a particular case, so that there is more horizontal and vertical equity. In Germany, official tables are used, and in the Netherlands, there is an overview of cases published regularly that serves as a guideline. Like the British system, courts in Germany and the Netherlands still enjoy a certain amount of discretion, but they are expected to consult the compilations of past awards. This process produces modest awards compared with other methods. Italy and Belgium also rely on tables to inform tort damage awards, although the tables were not created by precedent. Italy’s tables consist of awards corresponding to percentage-​based invalidity points. The first nine percentage points have legislatively mandated award amounts, but injuries that are more serious are left to the discretion of the court. Similarly, Belgium has a table specifying amounts to award for personal injuries, basing the percentage of invalidity on a victim’s inability to work following the injury. As in the other systems discussed in the chapter, these tables are advisory to the courts and are not mandated.

4.2.2.7. Damage Schedules Created by a Commission In Sweden, compensation for iatrogenic injury is provided through voluntary, contractual Patient Compensation Insurance that provides compensation without proof of provider fault through an administrative mechanism (Danzon 1994).13 Although patients retain the right to sue in tort, this administrative mechanism has largely displaced tort claims. The system is funded by a tax on medical care, internalizing costs to the medical care system, but without specific deterrence or fault assigned to particular institutions. Pain-​ and-​ suffering damages are determined according to the schedule defined by the Traffic Injuries Board and approved by the courts. The Road Traffic Injuries Commission (Commission) deals with loss of income and noneconomic compensation. Noneconomic compensation is partly composed of pain and suffering during the period of acute sickness, and it is determined according to standardized tables. Two other forms of noneconomic compensation are provided; inconveniences in daily life, and, for those who return to work, inconveniences at work. In an attempt to ensure impartiality, the Commission’s chair is appointed by the government, and the three categories of members are a panel of judges, a panel of insurance representatives, and a panel of laypeople recommended by Labor Organizations. This panel is tasked with creating a schedule of damages, which, like the British system noted in Section 4.2.2.6, 12 See http://​www.jsbni.com/​Publications/​personal-​injury-​guide/​Documents/​Green%20Book.pdf.

13  Danzon’s (1994) article lays out the structure, incentives, and constraints individuals face, along with the success and limitations of the Swedish Tort System. The Danish system is modeled after the Swedish system (Studdert and Mello 2011). Finally, Studdert, et al. (1997), discussing the Swedish model and the similar New Zealand model, concluded through their research that the Swedish model would result in lower costs and a greater number of compensated victims than would the current systems in Utah and Colorado.

 

114   RONEN AVRAHAM is based on past precedent, but also adjusted according to inflation and the opinions of experts. The Commission conducts investigations into each individualized case and gives a recommendation of damages based on the findings of the investigation and the damage schedule. The recommendations are then sent to both the injured individual and the insurance company. The Commission thus serves as a “guiding force regarding the development of the law in the field of personal injury compensation in Sweden” (Sweden Road Traffic Injuries Commission 2008).

4.3. Conclusion Pain-​and-​suffering damages in the United States are an unsettled and volatile area of the law. There is great disparity both in the types of damages allowed and in the judicial theories justifying the award, or denial, of damages. The lack of an objective metric and the resulting inconsistencies are so troubling that they have led some commentators to conclude that pain-​and-​suffering damages should be eliminated from tort law altogether. However, pain-​and-​suffering damages in tort law represent legitimate transfers compensating injured victims for very real losses and deterring potential tortfeasors from taking insufficient care. The mere fact that pain-​and-​suffering damages—​like almost every other form of tort damages—​are difficult to quantify is not a reason to set them at zero. Instead, the inconsistency that opponents of pain-​and-​suffering damages have identified is a reason to address the problem of finding or creating a simplified theoretical framework or structure for awarding pain-​and-​suffering damages. Once the case for simpler pain-​and-​suffering damages is established, it remains to be seen how best to estimate them. There are a number of approaches available. Individualized approaches include the Golden Rule, per diem, and multiplier of medical costs. Although some individualized approaches have generally been rejected by courts, it may behoove the legal system to examine them anew. If the goal of reducing disparity and administrative costs of pain-​and-​suffering damages is recognized, judges may see these approaches as a viable method of increasing equity and predictability while still compensating victims for their individual and unique injury. The non-​individualized approaches, both formal and informal, may be more appealing to the critics who value predictability and simplicity above other considerations. These approaches include fixed amounts of pain-​and-​suffering damages, fixed caps, flexible caps, fixed amounts for the entire injury, legislated damages schedules, court-​ based damage schedules, and damage schedules created by a commission. Both the informal routinization that is currently spreading through the tort system and the more formal mechanisms represented by the trust examples and the 9/​11 Fund can serve as a model for reducing the delays and transaction costs associated with tort claims. While pain-​and-​suffering damages have traditionally represented an expensive and volatile area of the law, they need not continue to do so. Although none of the alternatives explained herein is flawless, the very existence of such a number and variety of

 

ESTIMATING PAIN-AND-SUFFERING DAMAGES    115 methods to simplify pain-​and-​suffering denies the contention that it is best to eliminate them altogether. Maybe some of these methods could be used in conjunction with others, or maybe one method is appropriate in certain types of cases, whereas another method is appropriate in others. Either way, future research and judicial consideration of pain-​and-​suffering damages can and should approach the problem from a solution-​ based perspective, taking into account the availability and viability of numerous alternatives. This is especially true now, when objective measures of pain and suffering are becoming more realistic owing to recent scientific breakthroughs in brain science.

References 33 U.S.C. §908 (1988). 5 U.S.C. §8105 (a) Total Disability (2011). Abel, Richard. 2006. “General Damages are Incoherent, Incalculable, Incommensurable, and Inegalitarian (But Otherwise a Great Idea).” DePaul Law Review 55, pp. 253, 300–​317. Adler, Matthew. 2006. “QALYS and Policy Evaluation:  A  New Perspective.” Yale Journal of Health Policy Law and Ethics 1, pp. 4–​11. American Law Reports. 1964. “Instructions in a Personal Injury Action Which, in Effect, Tell Jurors that in Assessing Damages They Should Put Themselves in Injured Person’s Place.” American Law Reports 96, p. 760. American Psychiatric Association. 2013. Diagnostic and Statistical Manual of Mental Disorders. 5th ed. Washington, DC: American Psychiatric Association. 271–​274. Avraham, Ronen. 2006. “Putting a Price on Pain-​and-​Suffering Damages: A Critique of the Current Approaches and a Preliminary Proposal for Change.” Northwestern Law Review 87, pp. 90–​110. Avraham, Ronen, 2014. “Dataset on Tort Law Reforms (5th). http://​papers.ssrn.com/​sol3/​ papers.cfm?abstract_​id=902711 [Accessed 15 August 2016]. Avraham, Ronen, 2015. “Does The Theory of Insurance Support Awarding Pain and Suffering Damages in Torts?” In: Daniel Schwarcz and Peter Siegelman, eds., Research Handbook on the Economics of Insurance Law. Cambridge, MA: Edward Elgar. 94–​123. Berlá, Edward P., et al. 1989. “Hedonic Damages and Personal Injury: A Conceptual Approach,” Journal of Forensic Economics. Dec. 1989. http://​valueoflife.com/​Assets/​Documents/​Pub_​ Hedo_​Jour.pdf [Accessed 18 August 2016]. Botta v Brunner, 138 A.2d 713, 718 (N.J. 1958). Bovbjerg, Randall R., Sloan, Frank A., and Blumstein, James F. 1989. “Valuing Life and Limb in Tort: Scheduling Pain and Suffering.” Northwestern University Law Review 83, pp. 908–​924. Bronsteen, John, Buccafusco, Christopher, and Masur, Jonathon. 2008. “Hedonic Adaptation and the Settlement of Civil Lawsuits,” Columbia Law Review 108, pp. 1516–​1549. Brookshire, Michael L. and Smith, Stan V. (1992–​1993). Economic/​Hedonic Damages:  The Practice Book for Plaintiff and Defense Attorneys. Cincinnati: Anderson Publishing. 168–​174. Cal. Civ. Code Ann. §3333.2 (West 2004). Chamallas, M., and Wriggins, J. 2010. The Measure of Injury:  Race, Gender, and Tort Law. New York: New York University Press. 179. Chamblee, Elizabeth L. 2004. “Unsettling Efficiency: When Non-​Class Aggregation of Mass Torts Creates Second-​Class Settlements.” Louisiana Law Review 65, pp. 157, 318–​321.

 

116   RONEN AVRAHAM Chang, Yun-​Chien, Eisenberg, Theodore, Li, Tsung Hsien, and Wells Martin T. 2013. An Empirical Study of Pain and Suffering Damages in Personal Injury Cases. Paper presented at the 2013 Conference on Empirical Legal Studies. Philadelphia: University of Pennsylvania Law School. 15–​36. Clark, Stuart and Harris, Christina. 2005. “Tort Reform in Australia:  Fundamental and Potentially Far-​Reaching Change.” Defense Counsel Journal 72, pp. 16–​23. Croley, Steven P. and Hanson, Jon D. 1995. “The Non-​Pecuniary Costs of Accidents: Pain-​and-​ Suffering Damages in Tort Law,” Harvard Law Review 108, pp. 1785–​1789, fn. 11. Cunnien, Alan J. 1997. “Psychiatric and Medical Syndromes Associated with Deception.” In:  Richard Rogers, ed., Clinical Assessment of Malingering and Deception. 2nd ed. New York: Guilford Press. Chap. 2. Danzon, Patricia. 1994. “The Swedish Patient Compensation System: Lessons for the United States.” Journal of Legal Medicine 15(2), pp. 199–​247. Debus v Grand Union Stores, 621 A.2d 1288 (Vt. 1993). Deepwater Horizon Medical Benefits Claims Administrator. 2012. https://​deepwaterhorizonmedicalsettlement.com/​en-​us/​home. aspx [Accessed 15 August 2016]. Diamond, Shari Seidman, Saks, Michael J., and Landsman, Stephen. 1998. “Juror Judgments about Liability and Damages:  Sources of Variability and Ways to Increase Consistency.” DePaul Law Review 48, pp. 301–​317. Diamond, Shari Seidman, Rose, Mary R., Murphy, Beth, and Meixner, John. 2011. “Damage Anchors on Real Juries.” Journal of Empirical Legal Studies 8(s1), pp. 148–​178. Diller, Matthew. 2003. “Tort and Social Welfare Principles in the Victim Compensation Fund,” DePaul Law Review 53, p. 719. Erichson, Howard M. and Zipursky, Benjamin C. 2011. “Consent versus Closure.” Cornell Law Review 96, pp. 279–​280. Feinberg, Kenneth. 2007. “Compensating the Victims of Catastrophe:  The Virginia Tech Victims’ Assistance Program.” Virginia Law Review in Brief 93, p. 181. Gandhi, Tehal K. et  al. 2006. “Missed and Delayed Diagnoses in the Ambulatory Setting: A Study of Closed Malpractice Claims.” Annals of Internal Medicine 145, p. 488. Garber, Alan M. 1999. “Advances in Cost-​Effectiveness Analysis of Health Interventions.” Working Paper No. 7198, NBER. Geistfeld, Mark. 1995. “Placing a Price on Pain and Suffering: A Method for Helping Juries Determine Tort Damages for Nonmonetary Injuries.” California Law Review 83, 773. Gervais, Roger O. et al. 2001. “Effects of Coaching on Symptom Validity Testing in Chronic Pain Patients Presenting for Disability Assessments.” Journal of Forensic Neuropsychology 2, pp. 13–​14. Giant Food Inc. v Satterfield, 603 A.2d 877 (Md. 1992). Glaberson, William. 2001. “Federal Plan for an Aid Formula is Criticized.” New York Times. Nov 7. B7. Goldberg, John C. P. and Zipursky, Benjamin C. 2010. Oxford Introduction to U.S. Law: Torts. Oxford: Oxford University Press. 394. Golden Eagle Archery, Inc. v Jackson, 116 S.W.3d 757, 768 (Tex. 2003) (compiling cases). Henderson v Atl. Pelagic Seafood, LLC, 2011 U.S. Dist. LEXIS 33727, 34, fn. 9. Hunt v K-​Mart Corp. 981 P. 2d 277 (June 1999). Idaho Code Ann. § 6-​1603 (West 2004). Karapanou, Vaia and Visscher, Louis. 2010. “Towards a Better Assessment of Pain and Suffering Damages.” Journal of European Tort Law 1(1), pp. 48–​74.

 

ESTIMATING PAIN-AND-SUFFERING DAMAGES    117 Kelly, Carly N. and Mello, Michelle M. 2005. “Are Medical Malpractice Damages Caps Constitutional? An Overview of State Litigation.” Journal of Law, Medicine & Ethics 33, pp. 515–​517. King, Joseph. 2003. “Counting Angles and Weighing Anchors:  Per Diem Arguments for Noneconomic Damages,” Tennessee Law Review 17(1), pp. 13–​14. King, Joseph H. Jr. 2004. “Pain and Suffering, Noneconomic Damages, and the Goals of Tort Law,” S.M.U. Law Review 57, p. 163. King, Stephen P. 2010. “Does Tort Law Reform Help or Hurt Consumers?” Economic Record 86(275), pp. 563–​577. Kirk v Wash. State Univ., 746 P.2d 285, 292 [Wash. 1987]. Klass, Gregory, and Zeiler, Kathryn. 2013. “Against Endowment Theory:  Experimental Economics and Legal Scholarship.” UCLA Law Review 61, pp. 2–​64. Kolber, Adam J. 2007. “Pain Detection and the Privacy of Subjective Experience.” American Journal of Law & Medicine 33, pp. 433–​435. Larson, Lex K. and Larson, Arthur. 2008. Workers’ Compensation Law: Cases, Materials and Text. Newark: LexisNexis/​Matthew Vender. §5 at 1. Maletic, V., Robinson, M., Oakes, T., Lyengar, S., Ball, S. G., and Russell, J. 2007. “Neurobiology of Depression: an Integrated View of Key Findings.” International Journal of Clinical Practice 61(12), pp. 2030–​2040. Masada, Steven. 2004. “Australia’s ‘Most Extreme Case’: A New Alternative for U.S. Medical Malpractice Liability Reform.” Pacific Rim Law and Policy Journal 13, pp. 181–​184. McCaffery, Edward J., Daniel J. Kahneman, and Matthew L. Spitzer. 1995. “Framing the Jury: Cognitive Perspectives on Pain and Suffering Awards.” Virginia Law Review 81, pp. 1341–​1378. McElheney, James W. 1987. McElheney’s Trial Notebook. 2nd ed. Chicago:  American Bar Association. 515–​516. McGarry v Horlacher, 775 N.E.2d 865, 877–​878 (Ohio Ct. App. 2002). McKay, Colin and Great Britain Judicial Studies Board. 2010. Guidelines for the Assessment of General Damages in Personal Injury Cases. Oxford: Oxford University Press. Mendelson, George. 1997. ‘Outcome-​Related Compensation:  In Search of a New Paradigm’ in Richard Rogers, ed., Clinical Assessment of Malingering and Deception. 2nd ed. New York: Guilford Press. 220, 222. Miller, Ted R. 2000. “Valuing Nonfatal Quality of Life Losses with Quality-​Adjusted Life Years: The Health Economist’s Meow.” Journal of Forensic Economics 13, p. 145. Mittenberg, Wiley et al. 2002. “Base Rates of Malingering and Symptom Exaggeration.” Journal of Clinical and Experimental Neuropsychology 24, pp 1094–​1096. Mullenix, Linda S. 2011. “Prometheus Unbound: The Gulf Coast Claims Facility as a Means for Resolving Mass Tort Claims-​A Fund Too Far.” Louisiana Law Review 71, 819–​912. Pierce v N.Y. Cent.R.R.Co., 409 F.2d 1392 [Mich. 1969]. Plott, Charles R., and Zeiler, Kathryn. 2005. “The Willingness to Pay–​Willingness to Accept Gap.” The American Economic Review 95(3), pp. 530–​545. Rabin, Robert L. 2005. “Pain and Suffering and Beyond:  Some Thoughts on Recovery for Intangible Loss.” DePaul Law Review 55, 359–​376. Rahdert, Mark. 1995. Covering Accident Costs:  Insurance, Liability, and Tort Reform. Philadelphia: Temple University Press. 19. Rhoades v Walsh, 2009 U.S. Dist. LEXIS 75784, 47 fn. 26. Robinette, Christopher J., 2013. “Two Roads Diverge for Civil Recourse Theory.” Indiana Law Journal 88, pp. 543–​550.

 

118   RONEN AVRAHAM Rogers, Selwyn O., et al. 2006. “Analysis of Surgical Errors in Closed Malpractice Claims at 4 Liability Insurers.” Surgery 140, p. 25. Rubin, Paul. 1993. Tort Reform by Contract. Washington, DC: AEI Press. Schwartz, Victor and Silverman, Cary. 2004. “Hedonic Damages:  The Rapidly Bubbling Cauldron.” Brooklyn Law Review 69, pp. 1037–​1043. Sebok, Anthony J. 2005. “Translating the Immeasurable: Thinking about Pain and Suffering Comparatively.” DePaul Law Review 55, pp. 379–​392. Sinclair, Sandra and Burton, John F. 1994. “Measuring Non-​economic Loss: Quality-​of-​Life Values versus Impairment Ratings.” Worker’s Compensation Monitor, July–​Aug. 5. Sinclair, Sandra and Burton, John F. 1997. “A Response to the Comments by Doege and Hixson.” Workers’ Compensation Monitor, July–​Aug. 15. Sloan, Frank and Hsieh, Chee Ruey. 1990. “Variability in Medical Malpractice Payments: Is the Compensation Fair?” Law and Society Review 24, p. 997. Sowka, M. Patricia, ed. 1980. Malpractice Claims: Final Compilation. Brookfield, WI: National Association of Insurance Commissioners. State of Washington Task Force on Noneconomic Damages. 2005. Report to the Legislature. 15–​70. Available at http://​www.ofm.wa.gov/​rmd/​publications/​nedfinalrpt.pdf.29. [Accessed 15 August 2016]. Steenson, Mike and Sayler, Joseph Michael. “The Legacy of the 9/​11 Fund and the Minnesota I-​35w Bridge-​Collapse Fund: Creating A Template for Compensating Victims of Future Mass-​Tort Catastrophes.” William Mitchell Law Review 35, 524–598. Studdert, David M. et  al. 1997. “Can the United States Afford a “No Fault” System of Compensation for Medical Malpractice Injury?” Law and Contemporary Problems 1, pp. 32–​33. Studdert, David, et  al. 2006. “Claims, Errors, and Compensation Payments in Medical Malpractice Litigation.” New England Journal of Medicine 354, p. 2024. Studdert, David et  al. 2011. “Rationalizing Noneconomic Damages:  A  Health-​ Utilities Approach.” Law & Contemporary Problems 74, pp. 57–​70. Sugarman, Stephen D. 2005. “A Comparative Law Look at Pain and Suffering Awards.” DePaul Law Review 55, pp. 399–​413. Sugarman, Stephen D. 2005. “Tort Reform through Damages Law Reform:  An American Perspective.” Sydney Law Review 27, p. 507. Sullivan, Mathew C. 2005. “Mississippi Tort Reforms as Compared to Other Jurisdictions Abroad—A Sensible Treatment Protocol for the US Tort System Ills or Not.” Temple International and Comparative Law Journal 19, pp 507–​519. Sunstein, Cass R. 1986. “Legal Interface with Private Preferences.” University of Chicago Law Review 35, pp. 1129, 1146. Sunstein, Cass R. 1993. After the Rights Revolution:  Reconceiving the Regulatory State. Cambridge, MA: Harvard University Press. 161. Sweden Road Traffic Injuries Commission (Trafikskadenämnden). 2008. “The Activities of the Road Traffic Injuries Commission.” www.trafikskadenamnden.se/​upload/​Om%20TSN/​ Om%20n%C3%A4mnden%20p%C3%A5%20engelska.pdf [Accessed 15 August 2016]. Thaler, Richard. 1991. Quasi Rational Economics. New York: Russel Sage Foundation. Totaro, Martin. 2006. “Modernizing the Critique of Per Diem Pain and Suffering Damages.” Virginia Law Review 92, pp. 289–​290.

 

ESTIMATING PAIN-AND-SUFFERING DAMAGES    119 U.S. Government Accountability Office. 2011. “Asbestos Injury Compensation: The Role and Administration of Asbestos Trusts,” 16 GAO-​11-​819. http://​www.gao.gov/​assets/​590/​585380. pdf. [Accessed 15 August 2016]. Vairo, Georgene, M. 1992. “The Dalkon Shield Claimants Trust: Paradigm Lost (Or Found)?” Fordham Law Review 61, pp. 617–​633. Velocity Express Mid-​Atl., Inc. v Hugen, 585 S.E.2d 557 (2003). Vidmar, Neil and Jeffrey Rice. 1993. “Assessments of Noneconomic Damages Awards in Medical Negligence: A Comparison of Jurors with Legal Professionals.” Iowa Law Review 78, pp. 883–​898. Vidmar, Neil et al. 1999. “Jury Awards for Medical Malpractice and Post-​Verdict Adjustments of Those Awards.” DePaul Law Review 48, pp. 265–​269. Viscusi, W. Kip. 1988. “Pain and Suffering in Product Liability Cases: Systematic Compensation or Capricious Awards?” International Review of Law and Economics 8, p. 203. Viscusi, W. Kip. 1991. Reforming Products Liability. Cambridge, MA: Harvard University Press. 50. fn. 183. Wager, Tor D., Atlas, Lauren Y., Lindquist, Martin A., Roy, Mathieu, Woo, Choong-​Wan, and Kross, Ethan. 2013. “An fMRI-​Based Neurologic Signature of Physical Pain.” New England Journal of Medicine 368(15), pp. 1388–​1397. Washington State Legislature, Chapter 276, Laws of 2004, Section 118, ESHB 2459. Weinstein, Jack B. 2006. “Preliminary Reflections on Administration of Complex Litigations.” Cardozo Law Review 2009(1), pp. 6–​9. Whaley, Douglas J. 1992. “Paying for the Agony: The Recovery of Emotional Distress Damages in Contract Actions.” Suffolk University Law Review 26, p. 935.

 

Chapter 5

MEDICAL MAL PRAC T I C E Ronen Avraham and Max M. Schanzenbach

Medical malpractice reform has been the focus of significant attention from policy makers in the United States for decades. Hundreds of reforms to medical malpractice, including caps on noneconomic and punitive damages, reform of joint and several liability rules, and changes to evidentiary standards, have been enacted, struck down, and reenacted at the state level since the 1970s.1 A national cap on noneconomic damages in medical malpractice cases has been proposed numerous times and very nearly became a part of the Affordable Care Act.2 Physician groups, private health insurers, and both Presidents Bush and Obama have argued that medical malpractice reform will reduce healthcare costs.3 The impulse for limiting medical malpractice liability rests in part on a broader dissatisfaction with the U.S. healthcare system. The U.S. healthcare system is widely criticized, not only because of the costs associated with treatment but also because of the overuse and misuse of medical treatment (Becher and Chassin 2001). In fact, the high cost of healthcare in the United States is often attributed to the “overconsumption” of medical treatment. An Institute of Medicine (IOM) report estimated that unnecessary medical services cost $210 billion dollars a year (Smith et al. 2012). Surprisingly, although Americans are consuming healthcare at high rates, they often receive substandard care and medical errors abound (Andel et al. 2012; Graham et al. 2011; Van Den Bos et al. 2011). 1 

At first, many of these reforms were general reforms to tort law. Since the 1990s, most reforms have been targeted only to medical malpractice. 2  Patient Protection and Affordable Care Act of 2010, Pub. L. No. 111-​148, 124 Stat. 119. 11th Congr. (2010). 3  See the president’s remarks at the 2004 President’s Dinner (Obama 2009). America’s Health Insurance Plans (AHIP) buys advertisements promoting tort reform, arguing that medical malpractice liability has increased the cost of providing health insurance. AHIP asserts that “the current litigation system for compensating patients injured by medical negligence is expensive, slow, and does little to benefit the injured patients.” Available at: http://​www.ahip.org/​content/​default.aspx?bc=39|341|320.

 

MEDICAL MALPRACTICE   121 Proponents of limiting medical malpractice liability argue that the threat of liability is one cause of the systematic overtreatment given to Americans, and reducing liability pressure will help reduce healthcare costs without affecting, and perhaps improving, patient care. This claim rests on several potentially testable assumptions. The first is that medical malpractice liability imposes significant costs on the healthcare system through direct litigation costs and, more importantly, by inducing healthcare providers to treat patients too aggressively (so-​called defensive medicine). The second assumption is that limitations to malpractice will not induce offsetting behavior that also imposes significant medical and social costs. Most obviously, limiting providers’ liability could increase the rate of medical errors, increasing pain and suffering and the costs of remedial care. Some scholars have also recently suggested that medical malpractice limitations may free providers to pursue riskier, but more profitable, procedures. This phenomenon is called induced demand or offensive medicine, and it is of particular concern when providers may choose between alternative procedures and interventions, as in obstetric and cardiac care (Currie and McLeod 2006; Shurtz 2014; Avraham and Schanzenbach 2015). Though the effect of damage limitations has received the most attention in the empirical literature, other reforms to medical malpractice may actually be more relevant to changing provider behavior. Both the adoption of national standards of care (as opposed to local standards) and the increasing importance of Clinical Practice Guidelines in determining standards of care have been largely overlooked until recently. Additionally, communication and disclosure programs, though not directly a creature of legal reform and not yet widely implemented, offer some promise of systemic improvement of the medical system. The purpose of this chapter is to assess the theories and evidence regarding the efficacy of medical malpractice liability and limitations to it in achieving better healthcare outcomes, and to identify the unresolved issues that merit further attention from scholars. First, we will explore the theoretical and legal background on medical malpractice. Our discussion here is just a sketch of the issues, but it lays out the basic framework from which most scholars have worked. We next turn to the available evidence by focusing on three basic areas of study:  (1)  the effect of malpractice limitations on payouts and litigation; (2) the effect of malpractice limitations on overall healthcare costs; and (3) the effect of malpractice on two major cost drivers in the healthcare system: cardiac and obstetrics practice. We conclude that limitations on liability did not (and likely cannot) significantly reduce healthcare costs. Finally, we discuss new and important trends in the literature regarding reforms to standards of care and the role of Clinical Practice Guidelines, as well as communication and disclosure programs. There is some evidence that healthcare providers are very sensitive to the legal standard of care, and correctly setting the standard of care is one of the greatest challenges faced by the healthcare system. Thus, standard setting in conjunction with medical liability has potential to improve medical treatment.

 

122    RONEN AVRAHAM AND MAX M. SCHANZENBACH

5.1.  Theoretical and Legal Background From an economic perspective, the relationship between patients and providers can be characterized as that of principal and agent. The goal of malpractice liability is, therefore, to align the interests of healthcare providers and patients by making providers internalize the costs of their behavior. Providers externalize costs in three ways. First, there are large information asymmetries between providers and patients regarding treatment options and provider performance. Second, providers are largely reimbursed for their costs by third-​party health insurers, mostly based on fee for services provided. This means that medical problems caused or exacerbated by substandard care do not result in a loss to providers unless they face liability for them. In fact, to the extent victims of substandard care require further medical treatment, the result is a gain to providers who are reimbursed for the subsequent care they provide to their victims. Finally, physicians and some hospitals are in practice fully insured against medical malpractice liability, and for various reasons, insurance companies do not engage in significant experience rating of physicians(Weiler 1993). To align the interests of patients and providers, U.S.  law has developed a highly detailed body of case law (often overlaid by statutes) that imposes malpractice liability.4 This liability is imposed ex post for the harm arising out of violations of a standard of care. Standard of care refers not merely to typical examples of medical malpractice such as negligent misdiagnosis or careless errors during surgery. Standard of care also induces choice of treatment and diagnostic regimes (e.g., when to seek an MRI; whether and what type of surgery to perform). If the standard is set correctly, malpractice liability will produce better patient outcomes. Moreover, private law has advantages over regulatory frameworks by providing flexibility through an organic, bottom-​up approach, accounting for local variance and patient-​specific factors. In order for the medical malpractice system to cost-​effectively reduce harm, it must set the standard of care correctly while detecting negligent acts and sifting out unwarranted litigation. If the tort system is functioning as intended, negligent care should be thwarted and there should be little incentive to undertake defensive medicine. Moreover, if the standard of care is set correctly, offensive medicine should also be deterred because overtreatment is also negligent. But is the medical malpractice system functioning as intended? Advocates of limitations to medical malpractice liability have argued that (1) the tort system, especially the jury system, has difficulty setting the proper standard of care; (2) the tort system is not well tuned to detecting medical errors; and (3) the system is costly to apply because of high administrative costs and because liability induces defensive medicine. In short,

4  Although victim compensation is often a goal of tort law, in the case of medical malpractice, it is a relatively unimportant feature because of rampant underclaiming and the unusually high costs of medical malpractice litigation.

 

MEDICAL MALPRACTICE   123 medical malpractice litigation may be haphazard, costly, ineffective, and overdeterring, meaning that damage limitations could improve social welfare by reducing the litigation rate and increasing certainty in damages. Opponents of limitation, in contrast, argue that the tort system is actually rational and that most malpractice litigation is well founded. On this view, the real failure in the system is that the rate of claiming is too low because malpractice is often undiscovered or too expensive to prosecute. We now discuss these ideas in turn.

5.1.1. The Standard of Care Defining the standard of care required by professionals in the medical context is fraught. Historically, most jurisdictions eschewed a reasonable physician standard set ex-​post by courts. Instead, they have applied a standard of care determined by actual custom. The Second Restatement of Torts states that the physician must “exercise the skill and knowledge normally possessed” by other physicians (§ 299A [1965]). This effectively allows the medical industry to set the standard of care, a privilege unique to the profession. In the rest of tort law’s domain, the standard of behavior is “reasonableness,” a legal fiction that essentially enables courts to set the standard of care. Although custom may be evidence of reasonable behavior, it is not dispositive of the question. Initially, states applied the customary standard by reference to physicians in the state or locality. Over the past few decades, most states have redefined custom by national standards of care reflected in growing numbers of “Clinical Practice Guidelines” or CPGs. The majority of states now follow a national standard of practice.5 Nonetheless, there is ample evidence that physician practices can vary dramatically by location (Chandra et al. 2012; Skinner 2012). The use of the “customary” standard is likely a response to the view that juries and judges will have great difficulty implementing a “reasonableness” standard. However, the customary standard, whether national or local, is criticized on several grounds. First, the customary standard may inhibit innovation in medical practice, including the adoption of promising new procedures, because it encourages doctors to hold to past practice. Second, determining liability by custom effectively narrows the jury’s role to that of a mere factfinder, eliminating its normative function. Finally, under either the custom or reasonable physician standard, the standard of care is often a contested issue of fact. The lack of a uniform standard of negligence that is easily ascertainable prior to trial creates uncertainty that could inhibit the implementation of sound practices. The custom standard has been the majority rule since the late nineteenth century. Indeed, some states have explicitly enacted statutes that establish the standard of the profession as the test for negligence. A small number of states apply the “respectable minority rule,” which exempts physicians from liability when they follow a practice used 5 

Frakes (2013) surveys the law and reports that sixteen states retain some reference to local custom in their medical malpractice standard of care law.

 

124    RONEN AVRAHAM AND MAX M. SCHANZENBACH even if by only a small number of respected practitioners. Similarly, some states use the “error in judgment rule,” which exempts physicians from liability if they committed only an error in judgment in choosing among several therapeutic approaches. A few jurisdictions have moved toward a “reasonable physician” standard of care for at least certain types of medical practice (Peters 2000). The inquiry under the reasonable physician standard is not what other physicians would do, but what a reasonable physician would do.6 The industry custom is relevant to the question of whether the physician acted reasonably, but it is not conclusive. On the one hand, courts can find that a physician acted negligently even if the physician fully complied with customary standards. On the other hand, the reasonable physician standard frees doctors to make decisions based on recent research, even where that decision is not yet in keeping with the industry custom. On their face, the two standards may not seem very different. What a “reasonable” physician would do is likely informed, at least in part, by what other physicians practice. However, the custom standard may be more relevant if a state retains its locality rules. Local custom can deviate from national norms, and national norms informed by expert committees are more likely to be reasonable. Do providers pay attention to standards of practice? There is ample anecdotal evidence that they do, and Frakes (2013) shows systematically that changes in standards can produce large changes in provider behavior. Frakes examined what happens to geographical treatment variation in obstetrics and cardiac care when negligence standards were altered from local standards to national standards. He found that when states move to a national standard, local practices converge to that standard. In states that had intense treatment regimens relative to the national standard, treatment intensity was reduced. By contrast, in states with less intensive treatment, treatment intensity increased following a move to a national standard. A move to national standards reduced the overall gap between state and national treatment rates by 30%–​45%, without any measurable impact on patient outcomes.

5.1.2. Detecting Errors Even assuming that the standard of care is set correctly, a cost-​effective system must still detect enough medical error to create some incentive for caretaking, while not making too many false positives. Medical errors are well known to be expensive and common, possibly affecting 1.5  million persons annually and causing over 100,000 premature deaths (see Andel et al. [2012]; Van Den Bos et al. [2011]; Institute of Medicine [2000]; collecting studies and applying various methodologies). Given the volume of error, imposing significant malpractice liability holds out much promise for reducing overall healthcare costs and other social costs. However, there 6  For example, even if prevailing customs did not provide for a glaucoma screen, it was unreasonable for a doctor to fail to recommend to the patient this safe, cost-​effective test for a condition that, if caught early, is treatable. See Helling v Carey, 519 P.2d 981,981-​82 (Wash. 1974).

 

MEDICAL MALPRACTICE   125 is some reason to doubt malpractice liability’s efficacy. Despite widespread medical error, very few negligent medical injuries result in a paid damage claim. The National Practitioner Data Base reports between 10,000 and 15,000 positive payouts in medical malpractice cases (trials and settlements) per year over the last 20 years, a tiny fraction relative to the universe of possible claims. Studdert et al. (2006) examined two hospitals, audited the medical records of patients, and found that only 2.5% of negligent injuries resulted in malpractice litigation. Contrary to the conventional wisdom in the popular press, most researchers conclude that claims of rampant frivolous litigation are overblown. There is evidence that many weak claims are filed, but these also receive much lower average payouts (see Studdert and Mello [2007] for a review of the evidence). The disparity between injury rates and successful litigation raises grave doubts about the efficacy of medical malpractice liability as a means to reduce errors. However, the system’s weak propensity to detect error is not dispositive of the debate. We do not observe the errors the system avoids and hence must be cautious about inferring a counterfactual from the rate of litigated injuries. Moreover, the perceived liability pressures, even if unfounded in evidence, could still create incentives to standardize and assess practices and create internal risk-​management infrastructures to the benefit of patients. As discussed below, doctor surveys suggest liability may be a salient factor in treatment decisions.

5.1.3. Defensive Medicine, Offensive Medicine, and Administrative Costs The cost of the malpractice system is open to some debate. The direct costs of medical malpractice to the healthcare system, in terms of insurance, payouts, and litigation expenses, have been reckoned to be between 0.4% (Mello et al. 2010) and 1.5% (CBO 2004) of all medical spending.7 Thus, reducing medical malpractice liability, even if it reduced direct liability costs by half, would have little effect on overall health spending. Moreover, such small savings could easily be overwhelmed if reduced liability exposure causes an increase in medical errors. Most advocates of tort reform are therefore left to argue that, although the direct costs of the liability system are not large, the system incentivizes doctors to undertake expensive forms of defensive medicine. Defensive medicine is care that is not medically advisable but prescribed nonetheless in the belief that treatment may prevent liability for malpractice. Most likely, this unneeded care comes in the form of hospital admissions and noninvasive procedures such as medical imaging. The theory of defensive medicine is in some tension with the aforementioned fact that malpractice litigation is not very common, even relative to the universe of 7 

How the Congressional Budget Office estimate was arrived at is not fully explained.

 

126    RONEN AVRAHAM AND MAX M. SCHANZENBACH potentially negligent injury. Moreover, in practice, most physicians are fully insured against malpractice claims. Medical malpractice insurance and the bankruptcy remoteness of many assets (such as homes and retirement savings) generally protect physicians from having to pay large awards. Almost all claims are settled and fully paid out through malpractice insurance.8 Nonetheless, one survey found that 93% of Pennsylvania doctors admitted that they sometimes or often engage in defensive medicine practices (Studdert 2005). Certain professions are particularly susceptible to medical malpractice litigation, in particular obstetrics, neurosurgery, and cardiac care (Jena et al. 2011), and one would consequently expect to see greater response to liability limitations among these practitioners. A recent survey of neurosurgeons found that most admit to practicing defensive medicine and that their propensity to do so is correlated with their perception of legal risk (Smith et al. 2012). More importantly, the survey found that perception of litigation risk correlated with the actual legal environment and likelihood of litigation in the state in which the neurosurgeon practiced (Smith et al. 2012). These survey results, which evidence significant doctor sensitivity to liability pressures, are somewhat puzzling. Why would doctors practice defensive medicine if the probability of facing a suit is low and they are, in practice, fully insured? An easy answer is that doctors both strategically exaggerate their plight and have irrationally high fears of litigation. There are several additional possible answers that do not rely on fear or subterfuge. The first is that patients are poorly informed about healthcare choices and the cost of defensive medicine is largely externalized to third-​party payors, so even the small incentive doctors have to reduce exposure to liability can elicit a large response in increased treatment intensity. Second, the probability of a lawsuit, rather than the magnitude of awards, may be a key motivator. The incentives to avoid litigation include the desire to avoid reputational harm, the stress of litigation and bad publicity, and the lost time associated with defending a claim. Thus, even with a small risk of successful suits and the fact that losses are covered by liability insurance, the threat of malpractice litigation influences doctor behavior. Dranove et al. (2012) provide the clearest evidence of physician harm from litigation, in particular a finding that physicians take less remunerative patients after being sued. Moreover, hospitals, clinics, and practice groups are key players in setting practice standards and monitoring physician behavior and these providers can also reduce their exposure through defensive medicine. These groups may self-​insure to some extent or carry higher deductibles. Thus, defensive medicine may provide significant benefits to healthcare providers by increasing revenue and decreasing their exposure to various costs associated with malpractice liability. Third, it is possible that defensive medicine is actually effective at avoiding litigation, and, consequently, the level of litigation is endogenous to the practice of defensive medicine. The threat of liability induces defensive action by providers, which in turn reduces the probability of litigation. Thus, doctors prescribe aggressive treatments or admit 8 

This is a widely held view. Zeiler et al. (2007) provide evidence of this from the Texas close-​claimed database.

 

MEDICAL MALPRACTICE   127 patients, even if not medically necessary, because it provides evidence that all possible steps were taken to protect the patient. But even if defensive medicine is important, it is not likely to be the whole story regarding overtreatment. A counterpart to defensive medicine is “induced demand,” or “offensive medicine.” Some have posited that medical malpractice may deter offensive medicine-​related overtreatment, particularly in the case of invasive procedures more likely to result in complications (Currie and MacLeod 2008). Excessive care in medical treatment is believed to be a widespread problem in healthcare. Studies on the effect of reimbursement schemes, for example, have found that moves away from fee-​for-​service care can reduce costs without harming patient care. Moreover, when different procedures are available and context permits physician discretion, as in birth delivery method or cardiac care, procedure choice appears sensitive to reimbursement rates (Dranove and Wehner 1994; Gruber and Owings 1996; Gruber et al. 1999; Clemens and Gottlieb 2014; Avraham and Schanzenbach 2015). It may be relatively difficult to win a malpractice claim based on a procedure being inappropriate or unnecessary. Informed consent authorizing the procedure and challenges of proof regarding procedure choice, which is often discretionary, are significant barriers. However, medical malpractice likely provides a deterrent nonetheless because more remunerative procedures are likely to be more invasive (e.g., Cesarean sections and cardiac bypass), and they may be more likely to involve complications leading to claims for breach of the standard of care. Given the degree of overtreatment thought to be present in U.S. healthcare, reducing overtreatment is a significant potential benefit of medical malpractice liability. In sum, there are reasonable arguments that damage limitations could increase, decrease, or have little effect on treatment intensity. The empirical measurement of the overall effects of liability limitations, under a variety of reimbursement schemes and in different medical care contexts, is therefore essential.

5.2.  The Effect of Damage Limitations Creating an effective system of malpractice liability is a daunting task. The system must set standards correctly and detect errors. There are reasons to think it does not do either well. Moreover, policy makers have not been very interested in directly improving standard setting and detection, with the exception of moving toward a national, custom-​ based standard. Instead, most reforms have been aimed at indirectly reducing the incentives provided by medical malpractice liability by capping or otherwise reducing damages. The assumption of reform, then, has been that the perverse incentives of the system to provide defensive medicine should be tackled by limiting liability. Opponents of damage limitations argue that this undermines deterrence. Given this policy background, the empirical study of medical malpractice has been focused on the following questions: Do doctors, in fact, practice defensive or offensive medicine? If they do, is the

 

128    RONEN AVRAHAM AND MAX M. SCHANZENBACH practice sensitive to liability pressures? Can tort reform significantly limit liability pressures? Are there offsetting effects to malpractice limitations, in terms of patient care and medical injuries, observed? To answer these questions, empirical studies have focused primarily on the effect of medical malpractice limitations on (1) damage awards and claiming; (2) general healthcare utilization; and finally, (3) cardiac and obstetric practice in particular.

5.2.1. Are Damage Limitations Effective at Reducing Awards and Claiming? The answer to the question of whether damage limitations affect litigation pressure should be straightforward. In particular, it is reasonable to expect that low caps on noneconomic damages should reduce payouts, because noneconomic damages constitute a large percentage of medical malpractice jury awards (Hyman et al. 2009; Studdert et al. 2004; Pace et al. 2004). Caps on damages and other limitations reduce expected payoffs, which should reduce the probability of litigation and damages paid. It is also possible, however, that caps would have little effect if caps were set too high to be reasonably effective or if lawyers had significant flexibility to shift around damages between types of damages that are statutorily capped and types that are not. To study the effect of tort reform on litigation and awards, scholars have used the National Practitioner Data Base, which purports to report all positive payouts and settlements of medical malpractice claims. Despite a comprehensive national database and clearly defined reforms (usually passed by a state legislature), a number of early studies from the 1980s and 1990s examining claim frequency, payout amounts, and malpractice premiums reached mixed conclusions on the effect of various tort reforms, some finding the expected outcome that limitations reduced payouts and lawsuit probabilities, but others finding little or no effect.9 Nonetheless, two surveys concluded that tort reform—​particularly caps on noneconomic damages—​reduce payouts and claims. Holtz-​Eakin (2004) concluded that caps on damages are consistently found to reduce the number of lawsuits and the size of awards. Mello’s (2006) survey concluded that, in addition to caps on noneconomic damages, collateral source reforms and joint and several liability reforms are also associated with reductions in awards and (more weakly) with physician supply.10 Other surveys of the literature also concluded that there was limited evidence that rates of growth in medical malpractice premiums have been slower on average in states that enacted caps on noneconomic damages than in states with more limited reforms (GAO 2003; Danzon et al. 2004). 9  For a survey, see Sloan and Chepke (2008, 100), collecting studies and detailing their varying findings. 10  Joint and several liability reform generally required some apportionment of liability between the physician and the hospital or practice group, whereas the common law rules generally allowed collection

 

MEDICAL MALPRACTICE   129 It is now clear that several data issues and a misunderstanding of legal institutions interfered with many earlier studies. First, malpractice premiums are particularly influenced by the investment performance of insurance funds (Sloan and Chepke 2008; Baker 2004). As a consequence, premiums are a very noisy measure of malpractice pressure. Born et al. (2009) consider long-​run medical malpractice insurers’ loss ratios (the extent to which payouts are below premiums collected). This measure has the advantage of removing the noise contained in investment returns, and, using it, Born et al. (2009) found that tort reform significantly reduced insurers’ losses. Second, the dataset coding tort reforms employed by most researches was produced by American Tort Reform Association, and this dataset contained significant errors (Avraham 2007; Currie and MacLeod 2008). Avraham (2006) produced and has since maintained a comprehensive tort reform database, which has been subject to improvements and checks from other scholars as well (Currie and MacLeod 2008). Finally, previous research did not take into account that while statutory reforms are prospective, such reforms, especially damage caps, were often struck down by state supreme courts, and such strike downs were retroactive. Thus, the law governing observed settlements and claiming rates was sometimes significantly miscoded. Moreover, many reforms, particularly caps on noneconomic damages, are struck down within the first few years of enactment. The legal uncertainty surrounding reforms that faced credible state constitutional challenges would tend to depress the effect of reform, and some scholars suggested removing reforms that were quickly struck down from the analyses (Frakes 2012; Avraham et al. 2012). Additional work, with a more careful appreciation for timing of reforms and legal frameworks, has confirmed that tort reform is effective in reducing physicians’ liability exposure in both the amount of payouts and the probability of a positive payout. Avraham (2007) examines medical malpractice settlements in the National Practitioner Data Base using his own tort reform coding and accounting for the retroactive application of reforms. He finds that some noneconomic damage caps decrease the number of claims by roughly 5% to 13% and total annual payouts by more than half. Likewise, Paik et al. (2013) allow for the gradual nature of reform phase in owing to prospective enactment and find strong evidence that caps reduce claiming and payouts, in some cases by over half. The effect was particularly pronounced (as would be expected) for large claims. A study of Texas claims, based on that state’s unique closed claim database, similarly showed a dramatic reduction in payouts a result of caps on noneconomic damages (Paik et al. 2012).

from any party found liable. Thus, reform to the joint and several liability rule would tend to reduce recoveries because insurance caps for physicians are reached much sooner, though it may also have some incentive effects. Collateral source reform reduces recoveries for any collateral source that paid out for injuries. The common law rule generally disregards the injured party’s insurance payouts in calculating damages. Thus, collateral source rules tend to reduce recoveries, as well, by preventing a double recovery if disability and medical insurance payments were paid out as a result of the injury.

 

130    RONEN AVRAHAM AND MAX M. SCHANZENBACH

5.2.2. Are Damage Limitations Effective at Reducing Healthcare Utilization? An important line of studies has used broad datasets on costs or procedures to study how medical practice changes after malpractice reform. These studies then make inferences about defensive medicine based on these changes. The challenge inherent in the work is that both the outcome variable (general healthcare utilization) and the independent variable of interest (liability pressure) are difficult to measure. General healthcare utilization studies have relied heavily on data from Medicare expenditures, which is limited to a subset of the population less likely to generate liability pressure. Studies that focus on a very specific procedure may have an excellent measure of that procedure but lose generalizability. Most studies employ a difference-​in-​difference strategy around the adoption of damage caps and other liability limitations to measure the effect of tort reform, but some studies look to other measures of liability pressure such as average or total payouts. Both approaches have drawbacks. Limitations to tort liability, particularly damage caps, are not uniform, but they are often treated as such in the identification. For example, caps on noneconomic damages vary tremendously (from $250,000 to more than a $1 million), and some states have exceptions for serious injuries. Using malpractice payouts as a measure of liability pressure is also problematic. Payouts may be endogenous to the legal environment. For example, a cap on noneconomic damages creates savings that may be partially offset by greater payouts caused by less caretaking. Increasing malpractice liability may lead to provision of more or better treatment, which, if effective, would tend to reduce payouts. It is important to keep these limitations in mind when evaluating the studies.

5.2.2.1. Medicare Studies Several studies use Medicare datasets, likely because Medicare spending is well tracked and the procedures are coded in detail. Baicker et al. (2007) found that a 10% increase in average malpractice liability payments per physician within a state was associated with a 1.0% increase in Medicare payments for total physician services and a 2.2% increase in the imaging component of these services. Baicker et al. did not find that higher malpractice liability costs were associated with reductions in total or disease-​specific mortality, potentially indicating that these increases in payments are the result of defensive medicine. Sloan and Shaddle (2009) look across a variety of ailments and find little evidence that tort reforms reduce Medicare costs, though they have a relatively small sample size.11 Likewise, studying the 2003 Texas noneconomic damage cap, Paik et al. (2013) concluded that Medicare spending did not decline as a result of reform. They employ a quasi-​experimental design in which they consider baseline litigation rates at the county level, reasoning that tort reform should have a larger impact in counties that were more litigious. 11 

For example, they have fewer than 3,000 heart attacks.

 

MEDICAL MALPRACTICE   131 The work on Medicare spending suggests that there are no or very small effects of tort reform on healthcare utilization for a large and important group of healthcare consumers. However, we doubt that the studies of Medicare patients are generalizable to non-​ Medicare patients. Indeed, there is good reason to think that the Medicare population’s treatment should be little affected by tort reform. First, the median age of a Medicare patient in the Sloan and Shaddle study is almost 80 years. Given shorter life expectancies even without medical error, the lack of significant economic damages, and the naturally higher risk of procedures involving elderly patients creating evidentiary problems, it seems unlikely that a great deal of medical malpractice pressure comes from this group. Indeed, studies have found that those aged 65 and older comprise around 10% of all damage payouts for medical malpractice (Paik et al 2012; GAO 1993) and they are far less likely to sue (Studdert 2000) even though per-​capita healthcare spending on those 65 and older is 4.5 times that of those under 65 (Health and Human Services 2005). Second, almost one-​quarter of Medicare spending is done during end-​of-​life care, usually in the course of a terminal illness and repeated hospitalizations (Calfo et al. 2012; Hogan et al. 2001). Such spending seems unlikely to be influenced by the minimal threat of potential liability in such cases.

5.2.2.2. Non-​Medicare Studies Studies outside of the Medicare context have found small but measureable reductions in healthcare expenditures after tort reform. Using a proprietary insurance premium database with highly detailed plan information, Avraham et al. (2012) found that some tort reforms, primarily caps on noneconomic damages, reduce health insurance premiums by up to 2%. Moreover, the reduction appears to come entirely from health insurance plans that are not managed care plans. The authors interpret the price decline as evidence of reduced treatment intensity and that more tightly managed-​care plans reduce defensive medicine even in the absence of tort reform. Avraham and Schanzenbach (2010) using the Current Population Survey, found that caps on noneconomic damages increase health insurance coverage for price-​sensitive groups. The authors interpret their findings as evidence that tort reform reduces healthcare costs, which are ultimately reflected in health insurance prices. Cotet (2012) uses county-​level data on procedures and admissions and finds that caps on noneconomic damages lead to a 3.5% decrease in surgeries, a 2.5% decrease in admissions, and a 4.5% decrease in outpatient visits, but no significant effect on emergency room visits. Cotet suggests that decisions to visit the emergency room are largely made by patients and, consequently, the finding that emergency visits are not responsive to liability pressure makes sense. There are several consistent findings across this work. The first is that the effect of tort reform on healthcare prices or usage is small, approximately a 1% to 2% decrease outside the Medicare context, and an even smaller or no effect within the Medicare context. The second is that there are heterogeneous treatment effects from limitations on damages. For example, elective procedures are more affected than emergency procedures are because there is less discretion in emergency procedures, and the cost savings may depend on whether individuals are covered in managed care–​type plans. Heterogeneous

 

132    RONEN AVRAHAM AND MAX M. SCHANZENBACH effects are also suggested by the lack of findings from the Medicare patients; the effect of tort reform is likely to be smaller among the elderly. But do these limited results finding small to modest reductions in treatment intensity provide evidence of defensive medicine? Answering that question would require knowledge of whether the spending reductions occurred among cost-​justified procedures or non-​cost-​justified procedures. This question could be answered by looking to changes in the quality of care post tort reform. To the extent the studies discussed above measured quality, they did so using crude measures such as hospital mortality, and therefore cannot fully answer the question. However, it is possible to make an informed guess about the likely effect of reduced treatment intensity. For the most part, the reimbursement schemes in the U.S. healthcare system tend to push providers toward more, not less, spending. The marginal procedure is likely unnecessary, and if the marginal procedure is affected by tort reform, then the procedures forgone were not cost-​justified. As a theoretical matter, then, it seems likely that reductions in treatment intensity post reform are coming from unnecessary treatment, but that assertion has not been clearly established empirically. The only incentive for providers to reduce care is for care that is unlikely to be adequately reimbursed. This may occur particularly among the marginally insured or uninsured, or those for whom reimbursement rates are very low, such as Medicaid recipients. Thus, the inadequately insured or uninsured could face reductions in cost-​justified care after liability limitations. But there is an alternate story too. It is possible that medical malpractice liability, as an added cost, leads to avoidance of serving this patient population, and liability reductions might encourage providers to give more, not less, treatment to them. The effect of tort reform on costs is small, but it does not follow that defensive medicine is therefore a small component of health spending. The studies here measure the response of treatment intensity to tort reform on the margin. It could be that defensive medicine is an important component of healthcare costs, but that providers’ behavior is insensitive to the scope of damage limitations measured here. Provider insensitivity to liability levels could be due to a high level of physician risk aversion to litigation combined with third-​party payment systems that provide strong incentives to treat. The conclusion to be drawn from the empirical work on general healthcare utilization is that, on net, tort reform reduces treatment intensity slightly. For very important groups such as the elderly, there is likely no effect on treatment intensity from damage limitations. Given the magnitude of the results, interpreting them in the social welfare context is difficult. The small reductions in measured treatment intensity could easily be outweighed by additional but unmeasured uncompensated pain and suffering caused by medical errors. However, the results at least demonstrate that reductions to treatment intensity are not outweighed by the medical costs of increased medical errors or a greater freedom among providers to induce demand. This finding is consistent with the broader health economics literature that suggests that there are much stronger demandand supply-​side factors than defensive medicine that drive increases in health spending, including new technologies, a system of third-​party payers, tax incentives, and the financial incentives of providers (for a careful survey, see Skinner [2012]). The law is

 

MEDICAL MALPRACTICE   133 not terribly important to the complex story of high U.S. health costs. Indeed, astonishingly large variations in patient expenditures occur between counties within a state, and therefore within a single legal regime (Gawande 2009).

5.2.2.3. Cardiac and Obstetric Care Even if tort reform does not reduce overall healthcare costs much, it can still reduce healthcare costs in some contexts. Certain practice areas face very high liability pressures, and two have been studied extensively in this regard: cardiac and obstetric care.12 In addition to the intensity of liability pressure faced by these specialties, they have been the focus of study for three additional reasons. First, they involve common procedures that consume a significant amount of medical expenditures. Cardiac care has historically accounted for as much as one-​seventh of all healthcare spending (Cutler et  al. 1998). In the United States, births are dramatically more expensive than they are in many other countries, and the United States has a high rate of Cesarean sections (Rosenthal 2013). The second reason that these procedures have been studied is because outcomes can be reliably assessed. Infant and maternal mortality is well recorded and, because of a well-​known infant health scoring system (APGAR), there are validated measures of birth outcomes. Heart treatment success can be measured at least partially by complication and mortality rates. Finally, there is the possibility of substitution between procedures in both cases, a fact that helps researchers identify the motivation behind treatment choices. In the case of cardiac care, medical management, Percutaneous Transluminal Coronary Angioplasty (PTCA), and CABG (Coronary-​Artery Bypass Graft) are available treatment options. In the case of births, the choice between vaginal births and Cesarean section has been the topic of much debate and concern, with evidence of widely varying practices and increasing rates over time (Baicker et al. 2006). In a well-​cited study, Kessler and McClellan (1996) examine Medicare beneficiaries treated for serious heart disease in 1984, 1987, and 1990 and found that malpractice reforms that directly reduce provider liability pressure lead to reductions of 5% to 9% in medical expenditures without substantial effects on mortality or medical complications. Based on these results, Kessler and McClellan concluded that liability reforms reduce defensive medical practices for patients with heart diseases. In a later study, Kessler and McClellan (2002) repeated their 1996 work but controlled for costs containment achieved by managed care. The magnitude of the results of their 1996 study were reduced by half, likely because tort reform was correlated with increases in managed care. This result is consistent with Avraham et al. (2012) who found the effects of tort reform were lower for non-​Medicare managed care plans as well. Clouding the picture further, a 2004 study by the Congressional Budget Office, applying the methods used in the Kessler and McClellan study to a broader set of ailments, as well as heart disease, could not replicate their results (Beider and Hagen 2004). Another 12  Neuro and cardiovascular surgeons face the greatest likelihood of litigation, with a roughly 18% chance of having a claim filed each year. Obstetricians face high pressure as well, though not as high compared with general surgeons (Jena et al. 2011).

 

134    RONEN AVRAHAM AND MAX M. SCHANZENBACH study on Medicare heart patients by Sloan and Shadle (2009) also failed to find significant effects of tort reform on costs or outcomes for cardiac patients. Unfortunately, however, Sloan and Shaddle had significantly smaller sample sizes compared with the Kessler and McClellan sample, sometimes as little as 1% of the Kessler and McClellan sample size, so Sloan and Shaddle’s failure to replicate could be attributed to sample-​size differences. 13 These competing studies suffer from the significant drawbacks, detailed earlier, of using Medicare samples. Avraham and Schanzenbach (2015) used a nationwide inpatient sample of those coded as having had acute myocardial infarctions. The authors exclude those 85 years and older, yielding an average age in their sample of 68 years, which encompasses many non-​Medicare patients in contrast to prior studies. They find that the probability of PTCA interventions decline by as much as 2 percentage points after the enactment of caps on noneconomic damages. However, this decline is partially offset by a slight increase in the rate of CABG of roughly .5 percentage points.14 The CABG rate increases only for those with insurance, which the authors interpret as evidence consistent with tort reform facilitating induced demand (or offensive medicine). A competing explanation, however, is that before tort reform doctors were inefficiently deterred from treating risky patients who needed CABG. As a highly invasive procedure, liability is more likely to follow from CABG, even if CABG in the long-​run yields better outcomes. Even though practice behavior changed, Avraham and Schanzenbach find no evidence of changes in mortality from heart disease; if anything heart disease mortality declined after caps on noneconomic damages. Obstetricians face the highest risk of medical malpractice liability of any medical specialty. A number of studies from the 1990s examined the relationship between malpractice insurance rates and obstetric practice. Localio (1993) found an association between malpractice claims risk and the rate of Cesarean delivery in the state of New York in 1984, and Dubay and Waidmann (1999) found that greater malpractice pressure leads to a higher probability of Cesarean delivery for the period 1990–​1992, without any significant improvement in health outcomes. Sloan et al. (1995) found no systematic improvement in birth outcomes because of medical malpractice pressure, and Baldwin (1995) found no association between the malpractice exposure of individual physicians and an increase in the use of prenatal resources or Cesarean deliveries for the care of low-​risk obstetric patients. The results of these studies are not consistent, which may be a result of 13  Dhankhar, Khan, and Bagga (2007) examined heart conditions with the data employed herein, but only considered 1 year of data; hence, their analysis was cross sectional. They considered the more limited question of the effect of medical malpractice costs on resource use and mortality of heart patients, finding that an increase in medical malpractice risk leads to a reduction in resource use and improvement in health outcome. 14  The authors do not find robust evidence of expenditure declines, but in models with state-​specific trends, they find a roughly 4% price decline, a magnitude similar to that estimated by Kessler and McClellan (2002).

 

MEDICAL MALPRACTICE   135 using medical malpractice insurance or award levels as the source of variation. Awards, payments, and malpractice insurance levels are quite noisy as they depend on many factors apart from actual liability pressure. Moreover, it is possible that award levels are endogenous if higher liability leads to more avoidance of risky patients or other loss-​ avoidance techniques. Currie and MacLeod (2008) revisited Cesarean sections using tort reform as the source of variation. They found that caps on noneconomic damages increased unnecessary Cesarean sections and led to worse outcomes for mothers. They attribute these effects to reduced care taking and more offensive medicine resulting from limitations on liability. Cesarean sections have higher complication rates for mothers relative to vaginal births when not medically indicated, but they are much more remunerative than vaginal births. The authors take this as evidence that caps on noneconomic damages can raise medical costs and damage outcomes not through less care taking but rather through induced demand. Indeed, there is prior evidence of induced demand in the Cesarean context (Gruber et al. 1999; Gruber and Owings 1996). In contrast to their finding regarding caps on noneconomic damages, Currie and MacLeod find that reform of the rule on the legal doctrine of joint-​and-​several liability has reduced Cesarean section rates while improving delivery outcomes. Reform of joint and several liability rules tends to place more liability on the doctor relative to hospitals, and Currie and MacLeod attribute the reduction in Cesarean section rates and improved outcomes to more liability being placed on the obstetrician, the primary decision maker. Frakes (2012) uses the National Hospital Discharge Survey to examine Cesarean section rates over a much broader timeframe (1979–​2005) than that of Currie and MacLeod, encompassing many more reforms. Frakes fails to find any measureable impact of caps on noneconomic damages or joint and several liability reform on Cesarean section rates. He attributes the differing results to the fact that only four caps on noneconomic damages were enacted within the Currie and MacLeod study. However, Frakes does find some evidence of decreased use of other procedures, such as episiotomies, after the enactment of caps on noneconomic damages. Iizuka (2013) used the Nationwide Inpatient Survey and found that reform to the collateral source rule15 worsened birth outcomes, reform to the joint and several liability rule improved birth outcomes, and that noneconomic damage caps had no measureable effect, though the sign of the effect suggested birth outcomes may have improved after non-​economic damage caps were enacted. The collateral source results, however, rely on only one enactment, which makes Iizuka’s results questionable.16 Frakes and Jena (2014)

15  The collateral source rule prevents deductions made from other sources of recovery, such as medical and disability insurance. Reform to the rule creates an offset to recovery from insurance, thereby reducing plaintiff ’s overall recovery. 16  Two other reforms were enacted but were done so simultaneously with Joint and Several Liability reform. See Iizuka (Table 1, p. 169).

 

136    RONEN AVRAHAM AND MAX M. SCHANZENBACH question Iizuka’s results in light of the fact that caps on noneconomic damages usually are found to have the largest impact. Shurtz (2013), using Texas data and examining the 2003 Texas cap on noneconomic damages, provides a possible avenue toward reconciling some of the differing results on tort reform and obstetric practice. Shurtz considered the financial incentives providers faced post reform for privately insured mothers versus mothers insured by Medicaid, which reimbursed at lower rates for Cesarean sections. Cesarean rates increased for those with private insurance but appeared to decrease for those with Medicaid coverage. The net effect of reform on Cesarean sections was quite small, but the net effect masked underlying changes that depended upon doctor’s financial incentives stemming from the patients’ level of insurance. Shurtz’s finding that Cesarean sections increased for those with private insurance post tort reform is consistent with Currie and MacLeod’s theory that caps on noneconomic damages may facilitate induced demand in the obstetric context. The studies of cardiac and obstetric care suggest that tort liability may affect treatment intensity, but in a potentially ambiguous fashion. In particular, Currie and MacLeod (2008); Shurtz (2013); and Avraham and Schanzenbach (2015) have demonstrated the need to investigate tort reform’s interaction with reimbursement schemes and the possibility that increases in offensive medicine may offset decreases in defensive medicine. Nonetheless, the failure to find consistent effects across periods and data sets, particularly in the case of obstetric care, prevents any firm conclusions from being drawn.

5.2.3. Tort Reform and the Quality of Care As discussed previously, studies of specific treatments have narrowly analyzed quality of care after tort reform. There is at present little work on the broader effect of tort reform on quality of care, though this effect is a key part of the assessment of tort reform’s social impact. Using cause of death records, Cotet (2012) finds that mortality from “medical error” increased after tort reform, and infers that treatment quality declined. But this could also be an endogenous coding response, with providers more willing to admit errors in the face of reduced liability. Baicker et al. (2007), finding a small decrease in spending in the Medicare data, also found no effect of liability pressure on mortality. Mortality, in any case, is a fairly blunt measure, and, as discussed, the elderly may not be the population most affected by changes in liability standards. The most comprehensive study of the effect of medical malpractice on patient outcomes is by Frakes and Jena (2014). The authors examine a number of clinically validated measures of healthcare quality, including inpatient mortality rates and avoidable hospitalization rates, and find a precisely estimated null effect of caps on noneconomic damages and most other tort reforms on these quality measures. No published study, to our knowledge, has otherwise explored the impact of tort reform on patient outcomes.

 

MEDICAL MALPRACTICE   137

5.3.  Future of Reform: Medical Malpractice and Clinical Practice Guidelines With insurance coverage expanding in the United States under the Affordable Care Act, cost and quality control have become the primary challenges of the U.S. healthcare system. Most observers agree that provider incentives will be the key factor in improving the system’s performance (Kinney 2011). It is clear that state-​level tort reforms, relying primarily on caps on noneconomic damages, have not brought about economically significant reductions in healthcare cost or utilization. Damage limitations also run the risk of creating offsetting costs and increased utilization in some circumstances. On the other hand, there is little evidence that the current system of medical malpractice, with or without damage caps, is close to optimal. The system is characterized by underclaiming and perverse incentives created by uncertain and potentially incorrect standards of care. Improving the incentives provided by medical malpractice liability could hold some promise for improving care. In particular, two ideas have regained the attention of scholars: (1) the use of Clinical Practice Guidelines (CPGs) to set the standard of care and thus direct medical practice norms ex ante, and (2) Disclosure and Resolution Programs (DRPs), which provide incentives to disclose and address errors and thereby address the underclaiming in the system.

5.3.1. Clinical Practice Guidelines and Medical Malpractice CPGs are statements that detail the optimal clinical practices that physicians should apply in providing care for patients. CPGs are developed by groups of medical professionals who review the scientific evidence and assess the benefits and harms of alternative care options. A number of different organizations have drafted CPGs (such as private organizations and public organizations, including advocacy groups, state agencies, and commercial entities). Currently, more than 350 groups have drafted more than 2,700 CPGs in the National Guidelines Clearinghouse (run by the U.S. government) (Avraham 2014). CPGs, if well drafted, may be combined with the medical malpractice system to change the norms of medical practices. If providers are sensitive to medical malpractice liability standards, then by tying liability standards to validated best practices, CPGs can encourage higher-​quality care. Moreover, to the extent that CPGs become safe harbors from malpractice litigation, they will reduce the incentive to practice defensive medicine. Moreover, CPGs may avoid the potential deadening influence of the reliance on custom to determine liability by instead relying on flexible and evolving standards. A variety of commentators has suggested the CPGs have promise for distilling the

 

138    RONEN AVRAHAM AND MAX M. SCHANZENBACH overwhelming information on optimal care and communicating it effectively to providers (Havighurst 1991; Rosoff 2001; Avraham 2011; Frakes 2014). The uncertainty of the standard of negligence deters hospitals and providers from pursuing the creation of collectively valuable medical practices, adversely affecting patient safety. Further, in cases where the legal system assesses medical negligence based on customary practices, generally accepted defensive medical procedures may lock in suboptimal standards. Healthcare providers already lack sufficient incentive to invest privately in optimizing care because they do not fully capture the benefits of their innovations as other providers learn from technique. There are several other reasons to think courts are not optimally positioned to fix the healthcare system, especially given that their main “tool” is rendering ex post judgments about the provider’s practice. First, as discussed previously, malpractice detection rates are low, and even when malpractice is detected, the high costs of medical malpractice litigation mean that only the high damages cases are brought. Because so few negligence cases are pursued, the legal system’s signal to doctors and hospitals is weak (Hyman and Silver 2006). Second, even among those cases that are litigated, audits indicate that courts render correct judgments on average about three-​ quarters of the time (Studdert et al. 2006). Courts’ errors might be biased (toward or against defendants) because courts are not exposed to all the pertinent facts or because of numerous cognitive biases such as the “hindsight bias,” or because defendants have “deep pockets,” as well as for a host of other reasons. Moreover, the problem of offensive medicine (induced demand) is largely outside of tort law’s radar because in the vast majority of cases in which offensive medicine occurred, there are no direct observable injuries and the costs are externalized to the insurance pool or the government, which bear the costs of excessive care. In sum, the signal provided by court judgments regarding care is infrequent, muddled, and reflects only a subclass of medical errors. The signal is far too weak to inform determinations about patients’ healthcare. When the tort system does not work well, direct regulation of the activity becomes more attractive (Shavell 1984). However, regulatory agencies can face the same information problems faced by the courts, and they are subject to political pressures from which the courts are at least partly insulated. As a way out of all this mess, several scholars have advocated for a more extensive and nuanced use of CPGs to both harness state-​of-​the-​ art research about best practice without directly regulate the standard of practice itself (Havighurst 1991; Rosoff 2001; Avraham 2011; Frakes 2014). From a legal perspective, the promise behind CPGs is to provide physicians with a safe harbor against liability, effectively replacing the custom standard with a published scientific standard that is open to change as science advances. This would move the profession much closer to the ideal of “reasonable” standard of conventional tort law. Implementing CPGs would serve one of the major purposes of medical malpractice liability by providing incentives for physicians to follow guidelines and choose correct treatments, while reducing incentives to overtreat through the practice of defensive medicine.

 

MEDICAL MALPRACTICE   139 CPGs are more likely to reduce the number of unnecessary and incorrect medical procedures because of their direct focus on providing informed guidelines for delivering proper care to patients rather than focusing on providing incentives for the physicians to find the proper care themselves. It is impossible for a physician (especially a solo practitioner) to stay abreast of all of the developments in medical research occurring constantly. For example, a cardiologist would have to read ten articles per day, 365 days a year to stay current (Ayres 2007). With the continuous expansion and development of technological devices such as smart phones, tablets, and laptops, CPGs become even more effective because physicians can pull up up-​to-​date CPGs on their technological devices instantaneously. While there are many benefits associated with CPGs, implementation of them faces several challenges. First, physicians often do not follow even widely accepted guidelines. One example is the guideline promulgated by the U.S. Preventative Services Force for the process of screening for prostate cancer, which recommended against Prostate-​ Specific-​Antigen (PSA) based screening for prostate cancer for a number of evidence-​ based reasons (e.g., false positives and complications arising from follow-​up biopsies) (U.S. Preventative Task Force 2012). After the recommendation against the screening was issued in 2012, a survey found that while 49% of physicians agreed with the recommendation, only 1.8% planned to stop using the test. Among the reasons given for failure to comply with the recommendation were the following: physicians believed their patients expected to receive the screening; physicians thought patients would feel their healthcare was being rationed; and others felt they did not have time to explain such changes to their patients (Kliff 2012). Additionally, there are serious challenges in drafting CPGs. CPGs are currently produced by various entities and organizations, including professional medical organizations, hospitals, healthcare maintenance organizations (HMOs), liability insurers, and government agencies. Until recently, the guideline authors could have conflicts of interest. A publication by the American Medical Association found that conflicts of interest were present for 71% of the chairpersons and 91% of the committee co-​chairpersons of guideline producers (Kung 2012). Sometimes indirect connections with medical device and pharmaceutical companies remain. For example, since 2010, pharmaceutical companies can no longer fund the creation of CPGs, but they can still pay for their distribution and updating, thereby retaining some influence on the process (Avraham 2011b). Consequently, many doctors question the guidelines’ ability to deliver unbiased and evidence-​based advice (Cabana et al. 1999). Even if they avoid conflicts of interest, the organizations that produce CPGs do not face financial liability for their guidelines and therefore may lack sufficient incentives to promulgate optimal CPGs. Even if unbiased, the entities producing guidelines do so with differing goals. CPGs produced by medical associations focus intently on improving patient outcomes; CPGs produced by third-​party payers such as HMOs are focused on cost-​containment; and CPGs produced by malpractice insurers focus on reducing the risk of malpractice. Therefore, “[i]‌f physicians try to follow every CPG, they may find themselves trying to serve potentially conflicting goals:  to provide the best quality of care for their patients, to secure

 

140    RONEN AVRAHAM AND MAX M. SCHANZENBACH reimbursement for their services, and to avoid the risk of malpractice liability” (Recupero 2008). Moreover, there is no method of ensuring that the CPGs are updated and that obsolete recommendations are removed from circulation. Organizations that produce CPGs often struggle to update them because of a lack of resources and financial incentives to invest in incorporating quickly evolving scientific research (Avraham 2011b). The results are guidelines that are lacking in adequate scientific support: at least one study found that only a small percentage of CPGs are based on scientific evidence (Shekelle et al. 2001). To deal with problems in drafting CPGs, at Congress’s direction the Institute of Medicine promulgated eight standards with respect to the optimal development of CPGs (IOM 2011).17 These standards are meant to address problems with transparency, conflicts of interest, external review, and updating the guidelines as needed, which are critical to ensuring the reliability of CPGs. The IOM also proposed that a public–private mechanism examine the procedures used to produce CPGs and certify whether these organizations’ CPG development procedures comply with the standards for trustworthy CPGs (IOM 2011). The IOM’s proposed model requires that a public–​private entity approve the process the existing guideline developers follow so that they meet these standards. This is not the only possible model for promulgation of CPGs. While the IOM’s approach maintains that the CPGs should be produced in an unbiased manner and should be expert and convergent, this is likely not achievable. The IOM itself recognizes the myth of neutrality surrounding current CPGs, and acknowledges that CPG authors inevitably bring their personal and professional biases to the table (Graham et al. 2011). The IOM attempts to resolve these conflicts of interest through regulating procedure—​that is, ensuring that the chairs of guideline development groups will not have conflicts of interest individually, and keeping the other members of the group separate from financial investments. An alternative approach would be to accept the unavoidability of bias and the diversity among CPGs and substitute regulation for a market-​based system: On the one hand, letting CPG developers profit from their product by allowing them to license it, but on the other, holding them liable for the outcomes of the CPGs rather than for the process of producing them. In short, the law could treat CPG developers the same way the law treats manufacturers. CPGs produced under this system could be more scientifically accurate and trustworthy because of transparency combined with legal accountability (Avraham 2014).

5.3.2. Communication and Resolution Programs Communication and Resolution Programs (CRPs) involve voluntary disclosure and resolution of errors by the healthcare provider itself. In essence, these programs hope to 17 

The standards were as follows: (1) transparency; (2) management of conflicts of interest; (3) the composition of guideline development groups; (4) the intersection of CPGs and systematic review of technology; (5) evidence foundations for guidelines and rating the strength of recommendations; (6) articulation of recommendations; (7) external review of developed guidelines; and (8) updating the guidelines.

 

MEDICAL MALPRACTICE   141 increase the abysmal error-​detection rate evident in conventional medical malpractice, while reducing hospital costs by addressing problems early on and preserving the relationship with the patient. The current malpractice system fails to produce adequate information about errors for two reasons. First, there is rampant underclaiming, likely exacerbated by enactments of damages caps. Second, the threat of liability leads providers to adopt a deny-​and-​ defend approach, such that errors are not subject to the internal review that one would hope. Communication and Resolution programs attempt, among other things, to increase the flow of information to healthcare providers by encouraging open communication and compensation where it is appropriate. The basic design of a CRP is as follows: After an adverse event, the care facility initiates an investigation to determine whether the harm was caused by substandard care. The conclusion of this investigation is then disclosed to the patient. If the facility determines that care was not substandard, the facility will defend the physician in any lawsuit. On the other hand, if the investigation determines that the care was substandard, the facility communicates with the patient about her needs and voluntarily offers appropriate compensation and remedial measures if needed. Some argue that the amount of compensation under a CRP should be based upon the patient’s needs rather that upon an approximation of a possible jury verdict (Sage et al 2014). Perhpas most importantly, CRPs could facilitate the implementation of higher standards of care. Changing standards of care during the course of litigation could be evidence of substandard care. If a CRP succeeds in quickly procuring a resolution after an adverse event, the facility is free to immediately set and instill better standards of care and avoid a recurrence of the harm. To be effective, CRPs must do three things: (1) actually communicate information and not engage in cover-​ups; (2) save money for providers, which is possible through reduced liability; and (3) instill better standards of care which will reduce future rate of errors. The effect of these programs has not been conclusively assessed. In 2009, the federal government authorized $25 million to planning and demonstration grants to develop CRPs (Sage et al. 2014). Early results from these projects are promising, but longer studies would be necessary to produce truly reliable data (Mello 2014). Early information from the grant projects suggests that CRPs have the potential to reduce costs and medical mistakes. Like CPGs, they will require more study but offer some promise of an improvement over the traditional system of medical malpractice liability.

5.4. Conclusion Medical malpractice reform has become a persistent political issue, and it will remain on the political agenda. Damage limitations are popular with healthcare providers, particularly physicians in high-​risk specialties. Yet, evaluations of the impact of damage limitations, particularly caps on noneconomic damages, suggest that policy makers should be

 

142    RONEN AVRAHAM AND MAX M. SCHANZENBACH cautious about such reforms. Damage limitations are highly unlikely to affect healthcare costs significantly, and they have potential offsetting effects that may outweigh any savings. Our conclusion from various studies using different approaches is that there is no clear answer to whether caps and other damage limitations are detrimental or improving to social welfare. What clearly emerges from the studies, however, is the conclusion that damage limitations will have small effects on outcomes such as provider behavior, costs, and patient welfare, even if the direction of these effects are uncertain or vary by practice area. Despite the relatively low impact of damage limitations on the healthcare system, medical malpractice is still highly relevant. Providers pay close attention to standards of care and change practices accordingly. Future reforms should focus not on limitations to damages, but rather on direct ways to help the system function more effectively. Clinical practice guidelines and conflict resolution programs offer some promise of reducing costs and improving care, and future research should tend toward evaluating these more relevant avenues of reform.

References Andel, Charles et al. 2012. “The Economics of Health Care Quality and Medical Errors.” Journal of Health Care Finance 39.1, p. 39. Avraham, Ronen. 2009a. “Private and Competitive Regulation of Medicine.” The Economists’ Voice 6(8), pp. 1–​5. Avraham, Ronen. “Private Regulation of Medicine: A Win-​Win-​win For Doctors, Patients and Public.” Huffington Post (2009b) http://​www.huffingtonpost.com/​ronen-​avraham/​private-​ regulation-​of-​med_​b_​242937.html. [Accessed 16 August 2016]. Avraham, Ronen. 2011. “A Market Solution for Malpractice.” The New York Times 2011c http://​ www.nytimes.com/​2011/​03/​29/​opinion/​29Avraham.html?_​r=1&ref=opinion. [Accessed 16 August 2016]. Avraham, Ronen. 2011a. “Clinical Practice Guidelines:  The Warped Incentives in the U.S. Healthcare System.” American Journal of Law & Medicine 37, pp. 7–​40. Avraham, Ronen. 2011b. “Private Regulation.” Harvard Journal of Law & Public Policy 34, pp. 543–​638. Avraham, Ronen. 2014. Database of State Tort Law Reforms (DSTLR 5th), 1980–​2012. SSRN. Avraham, Ronen. 2014. “Overlooked and Underused:  Clinical Practice Guidelines and Malpractice Liability for Independent Physicians.” Connecticut Insurance Law Journal 20.2, pp. 273–​333. Avraham, Ronen and Schanzenbach, Max. 2015. “The Impact of Tort Reform on Intensity of Treatment Evidence from the Heart Patients.” Journal of Health and Economics 39, pp. 273–​288 Ayres, Ian. 2007. Super Crunchers: Why Thinking-​by-​Numbers Is the New Way to Be Smart. New York: Bantam. Baicker, Katherine, Buckles, Kasey S., and Chandra, Amitabh. 2006. “Geographic Variation in the Appropriate Use of Cesarean Delivery.” Health Affairs 25.5, pp. w355–​w367.

 

MEDICAL MALPRACTICE   143 Baicker, Katherine, Fisher, Elliott S., and Chandra, Amitabh. 2007. “Malpractice Liability Costs and the Practice of Medicine in the Medicare Program.” Health Affairs 26.3. pp. 841–​852. Baldwin, L. 1995. “Defensive Medicine and Obstetrics.” Journal of American Medical Association 274, p.1606. Baker, Tom. 2004. “Medical Malpractice and the Insurance Underwriting Cycle.” DePaul Law Review 54, p. 393. Becher, Elise C., and Chassin, Mark. 2001. “Improving the Quality of Health Care: Who Will Lead?” Health Affairs 20, pp. 164–​166. Beider, Perry and Hagen, Stuart. 2004. Limiting Tort Liability for Medical Malpractice. Congressional Budget Office. http://​www.cbo.gov/​ftpdocs/​49xx/​doc4968/​01-​08-​ MedicalMalpractice.pdf. [Accessed 3 October 2016]. Ben-​Shahar, Omri and Logue, Kyle D. 2012. “Outsourcing Regulation: How Insurance Reduces Moral Hazard.” Michigan Law Review 111, 197–​248. Born, Patricia, Viscusi, W. Kip, and Baker, Tom. 2009. “The Effects of Tort Reform on Medical Malpractice Insurers’ Ultimate Losses.” Journal of Risk and Insurance 76.1, pp. 197–​219. Cabana, Michael et  al., 1999. “Why Don’t Physicians Follow Clinical Practice Guidelines? Framework for Improvements.” JAMA 282, p. 1458. Calfo, Steve, Smith, Jonathan, and Zezza, Mark. 2012. “Last Year of Life Study.” Baltimore, MD: Centers for Medicare & Medicaid Services, Office of the Actuary. Clemens, Jeffrey and Gottlieb, Joshua D. 2014. “Do Physicians’ Financial Incentives Affect Medical Treatment and Patient Health?” The American Economic Review 104(4), pp. 1320–​1349. Congressional Budget Office. 2004. “Limiting tort liability for medical malpractice.” Washington, DC: CBO; Jan. 8, 2004 [cited July 1, 2010]. http://​www.cbo.gov/​ftpdocs/​49xx/​ doc4968/​01-​08-​MedicalMalpractice.pdf [Accessed 16 August 2016]. Cotet, Anca M. 2012. “The Impact of Noneconomic Damages Cap on Heath Care Delivery in Hospitals.” American Law and Economics Review 14(1) pp. 192–​234. Chandra, Amitabh, Cutler, David, and Song, Zirui. 2012. “Who Ordered That? The Economics of Treatment Choices in Medical Care.” In: A. J. Culyer and J. P. Newhouse, eds., Handbook of Health Economics. 2nd ed. Elsevier. 397–​432. Currie, Janet, and W. Bentley MacLeod. 2008. First Do No Harm? Tort Reform and Birth Outcomes. Quarterly Journal of Economics 123, 795-​830. Cutler, D. M., McClellan, M., Newhouse, J. P., and Remler, D. 1998. “Are medical prices declining? Evidence from heart attack treatments.” Quarterly Journal of Economics 113, pp. 991–​1024. Dhankhar, Khan M., and Shalini Bagga. 2007. “Effect of Medical Malpractice on Resource Use and Mortality of AMI Patients.” Journal of Empirical Legal Studies 4(1), pp. 163–​183. Dranove, David, and Wehner, Paul. 1994. “Physician-​Induced Demand for Childbirths.” Journal of Health Economics 13.1, pp. 61–​73. Dranove, David, Subramaniam, Ramanarayanan, and Watanabe, Yasutora. 2012. “Delivering Bad News: Market Responses to Negligence.” Journal of Law and Economics 55.1, pp. 1–​25. Dubay, Kaestner R., and Waidmann, Timothy. 1999. “The Impact of Malpractice Fears on Cesarean Section Rates.” Journal of Health Economics 18(4), p. 491. Frakes, Michael. 2013. “The Impact of Medical Liability Standards on Regional Variations in Physician Behavior: Evidence from the Adoption of National-​Standard Rules.” American Economic Review 103(1), pp. 257–​276.

 

144    RONEN AVRAHAM AND MAX M. SCHANZENBACH Frakes, Michael. 2015. “The Surprising Relevance of Medical Malpractice Law” University Of Chicago Law Review 82, pp. 317–​391. Frakes, Michael, and Anupam, B. Jena. 2014. Does Medical Malpractice Law Improve Health Care Quality? Working Paper, no. w19841, NBER. Gawande, Atul. 2009. “The Cost Conundrum.” New Yorker (June). General Accounting Office. 1993. “Medical Malpractice:  Medicare/​Medicaid Beneficiaries Account for a Relatively Small Percentage of Malpractice Losses,” GAO/​HRD-​93-​126. http://​ www.gao.gov/​assests/​220/​218311/​pdf [Accessed 16 August 2016]. Graham, R., Mancher, M., Wolman, D. M., Greenfield, S., and Steinberg. E., eds. 2011. Clinical Practice Guidelines We Can Trust 146. Washington, DC: National Academies Press. Gruber, Jonathan and Owings, Maria. 1996. “Physician Financial Incentives and Caesarean Section Delivery.” RAND Journal of Economics 27(1), pp. 99–​123. Gruber, Jon, Kim, John, and Mayzlin, Dina. 1999. “Physician Fees and Procedure Intensity: The Case of Cesarean Delivery.” Journal of Health Economics 18.4, pp. 473–​490. Hall, Mark A. 1991. “The Defensive Effect of Medical Practice Policies in Malpractice Litigation.” Law and Contemporary Problems (Spring), pp. 119–​145. Havighurst, Clark. 19991. Practice Guidelines as Legal Standards Governing Physician Liability. 54 Law & Contemporary Problems 87, pp. 651–​699. Helling v Carey, 519 P.2d 981,981-​82 (Wash. 1974). Hogan, Christopher, Lunney, June, Gabel, Jon, and Lynn, Joanne. 2001. “Medicare Beneficiaries’ Costs of Care in the Last Year of Life.” Health Affairs 20.4, pp.188–​195. Hotz-​Eakin, Douglas. 2004. The Effects of Tort Reform: Evidence from the States. Washington, DC: Congressional Budget Office. Hyman, David A. and Silver, Charles. 2006. “Medical Malpractice Litigation and Tort Reform: It’s the Incentives, Stupid.” Vanderbilt Law Review 59, 1085–​1090. Hyman, D. A., Black, B., Silver, C., and Sage, W. M. 2009. “Estimating the effect of damages caps in medical malpractice cases: evidence from Texas.” Journal of Legal Analysis 1, pp. 355–​409. Iizuka, Toshiaki. 2013. “Does Higher Malpractice Pressure Deter Medical Errors?” Journal of Law and Economics 56(1), pp. 161–​188. Institute of Medicine. 2000. To Err Is Human: Building a Safer Health System. Washington, DC: The National Academies Press. Institute of Medicine. 2011. Clinical Practice Guidelines We Can Trust. Ed. Robin Graham, Michelle Mancher, Dianne Miller Wolman, Sheldon Greenfield, and Earl Steinberg. Washington, DC: The National Academies Press. Jena, Anupam B., et al. 2011. “Malpractice Risk According to Physician Specialty.” New England Journal of Medicine 365.7, pp. 629–​636. Kessler, Daniel, and McClellan, Mark. 1996. “Do Doctors Practice Defensive Medicine?” Quarterly Journal of Economics 111, p. 353. Kessler, Daniel, and McClellan, Mark. 2002. “Malpractice Law and Health Care Reform: Optimal Liability Policy in an Era of Managed Care.” Journal of Public Economics 84(2), pp. 175–​197. Key, Charles M., 2007. “Toward a Safer Health System:  Medical Injury Compensation and Medical Quality.” University of Memphis Law Review 37, p. 459. Kinney, D. Eleanor. 2011. “Comparative Effectiveness Research under the Patient Protection and Affordable Care Act: Can New Bottles Accommodate Old Wine?” American Journal of Law and Medicine 37(4), pp. 522–​566.

 

MEDICAL MALPRACTICE   145 Kliff, Sarah. 2012. “Many Doctors Think PSA Tests Don’t Work. But They’ll Keep Doing Them Anyway.” New York Times, May 29, 2012. http://​www.washingtonpost.com/​blogs/​ezra-​klein/​ post/​many-​doctors-​think-​psa-​tests-​dont-​work-​but-​theyll-​keep-​doing-​them-​anyway/​ 2012/​05/​29/​gJQAOl0qyU_​blog.html. [Accessed 16 August 2016]. Kung, Justin et al., 2012. “Failure of Clinical Practice Guidelines to Meet Institute of Medicine Standards.” Archives of Internal Medicine 172, p. 1628. Localio, A. R. 1993. “Relationship between Malpractice Claims and Cesarean Delivery.” Journal of American Medical Association 269, p. 336. Medical Liability Reform and Patient Safety: Demonstration Grants. 2010. Agency for Healthcare Research and Quality website. http://​www.ahrq.gov/​professionals/​quality-​ patient-​safety/​patient-​safety-​resources/​resources/​liability/​demogrants.html. Updated June 2010. [Accessed 16 July 2014]. Mello, Michelle M. 2006. Medical Malpractice:  Impact of the Crisis and Effect of State Tort Reforms. Research Report, Robert Wood Johnson Foundation. Mello, Michelle M., et al. 2010. “National Costs of the Medical Liability System.” Health Affairs 29.9, pp. 1569–​1577. Mello, Michelle M., Studdert, David, and Kachalia, Allen. (2014). “The Medical Liability Climate and Prospects for Reform,” Special Communication. Journal of the American Medical Association. http://​jama.jamanetwork.com/​ on 11/​19/​2014 [Accessed 16 August 2016]. Obama, Barack. 2009. United States of America, Remarks by the President in Town Hall Meeting on Health Care (June 11). http://​www.whitehouse.gov/​the_​press_​office/​Remarks-​ by-​t he-​President-​in-​Town-​Hall-​Meeting-​on-​Health-​C are-​in-​Green-​B ay-​Wisconsin/​ [Accessed 16 August 2016]. Pace, Nicholas M., Golinelli, Daniella, and Zakaras, Laura. 2004. Capping Non-​Economic Awards in Medical Malpractice Trials: California Jury Verdicts under MICRA (RAND). http://​rand.org/​pubs/​monographs/​MG234/​. [Acessed 16 August 2016]. Paik, Myungho, et al. 2012. “How Do the Elderly Fare in Medical Malpractice Litigation, Before and After Tort Reform? Evidence from Texas.” American Law and Economics Review 14, pp. 561–​600. Paik, Myungho, Black, Bernard S., Hyman, David A., and Silver, Charles. 2012. “Will Tort Reform Bend the Cost Curve? Evidence from Texas.” Journal of Empirical Legal Studies 9.2, pp. 173–​216. Paik, Myungho, Black, Bernard, and Hyman, David. 2013. “The Receding Tide of Medical Malpractice Litigation: Part 2—​Effect of Damage Caps.” Journal of Empirical Legal Studies 10(4), pp. 639–​669. Philip G. Peters, Jr., 2000. “The Quiet Demise of Deference to Custom: Malpractice Law at The Millennium.” Washington and Lee Law Review 57, p.163. Recupero, Patricia R. 2008. “Clinical Practice Guidelines as Learned Treatises: Understanding Their Use as Evidence in the Courtroom.” Journal of the American Academy of Psychiatry and the Law 36, pp. 290–​301. Richards, Nathan S. 2012. “Judicial Resolution of Emtala Screening Claims at Summary Judgment.” NYU Law Review 87, pp. 591–​636. Rosenthal, Elisabeth. 2013. “American Way of Birth, Costliest in the World.” New York Times (July 1) A1. Rosoff, A. J. 2001. “Evidence-​Based Medicine and the Law:  The Courts Confront Clinical Practice Guidelines.” Journal of Health Politics & Law 26, pp. 327–​368.

 

146    RONEN AVRAHAM AND MAX M. SCHANZENBACH Roth, Jonathan H. 2001. “Regulating Your Medical History Without Regulations: A Private Regulatory Framework to Electronic Health Record Adoption.” Boston University Law Review 91, pp. 2103–​2122. Sage, William M., Gallagher, Thomas H., Armstrong, Sarah, Cohn, Janet S., McDonald, Timothy, Gale, Jane, Woodward, Alan C., and Mello, Michelle M. 2014. “How Policy Makers Can Smooth The Way For Communication-​And-​Resolution Programs.” Health Affairs 33(1), pp. 11–​19. Shavel, Steven. 1984. “Liability for Harm versus Regulation of Safety.” Journal of Legal Studies 13, p. 357 Shekelle, Paul G. et  al., 2001. “Validity of the Agency for Healthcare Research and Quality Clinical Practice Guidelines: How Quickly Do Guidelines Become Outdated?” JAMA 286, pp. 1461–​1464. Shurtz, Ity. 2014. “Malpractice Law, Physician’s Financial Incentives, and Medical Treatment: How Do They Interact?” Journal of Law and Economics 57(1), pp. 1–​29. Skinner, Jonathan. 2012. “Causes and Consequences of Regional Variations in Health Care.” Handbook of Health Economics 2, pp. 45–​93. Sloan, F. A., Whetten-​Goldstein, K., Githens, P. B., & Entman, S. S. (1995). Effects of the threat of medical malpractice litigation and other factors on birth outcomes. Medical Care, 33(7), 700–​7 14. Sloan, Frank A. and Chepke, Lindsey M. 2008. Medical Malpractice. Cambridge, MA: MIT Press. Sloan, Frank A., and Shadle, John H. 2009. “Is There Empirical Evidence for “Defensive Medicine”? A Reassessment.” Journal of Health Economics 28(2), pp. 481–​491. Smith, Mark, Saunders, Robert, Stuckhardt, Leigh, and McGinnis, J. Michael, eds. Institute of Medicine. 2012. Best Care at Lower Cost: The Path to Continuously Learning Health Care in America. Ed. Washington, DC: The National Academies Press. Smith, Timothy R., et al. 2015. “Defensive Medicine in Neurosurgery: Does State-​Level Liability Risk Matter?” Neurosurgery 76(2), pp. 105–​114. Stein, Alex. 2012. “Toward A  Theory of Medical Malpractice.” Iowa Law Review 97, pp. 1201–​1257. Studdert, D. M., Thomas, E. J., Burstin, H. R., Zbar, B. I., Orav, E. J., and Brennan, T. A. 2000. “Negligent Care and Malpractice Claiming Behavior in Utah and Colorado.” Medical Care 38(3), pp. 250–​260. Studdert, David M., Yang, Y. Tony, and Mello, Michelle M. 2004. “Are Damage Caps Regressive? A Study of Malpractice Jury Verdicts in California.” Health Affairs July/​Aug., pp. 54–​67. Studdert, David M. et  al., 2006. “Claims, Errors, and Compensation Payments in Medical Malpractice Litigation.” New England Journal of Medicine 354, pp. 2024–​2033. Studdert, David M., and Michelle M. Mello. 2007. “When Tort Resolutions Are “Wrong”: Predictors of Discordant Outcomes in Medical Malpractice Litigation.” Journal of Legal Studies 36(S2), pp. S47–​S78. U.S. Department of Health and Human Services. 2012. “Long Term Growth of Medical Expenditures—​Public and Private.” ASPE Issue Brief, May 2005. U.S. Preventive Services Task Force, Screening for Prostate Cancer: Current Recommendation. 2012. http://​www.uspreventiveservicestaskforce.org/​prostatecancerscreening.htm [Accessed 16 August 2016].

 

MEDICAL MALPRACTICE   147 Van Den Bos, J., Rustagi, K., Gray, T., Halford, M., Ziemkiewicz, E., and Shreve, J. 2011. “The $17.1 Billion Problem: The Annual Cost of Measurable Medical Errors.” Health Affairs 30(4), pp. 596–​603. Van Tassel, Katharine. 2012. “Blacklisted:  The Constitutionality of the Federal System for Publishing Reports of ‘Bad’ Doctors in the National Practitioner Data Bank.” Cardozo Law Review 33, pp. 2032–​2095. Weiler, Paul C., Hiatt, Howard H., Newhouse, Joseph P., Johnson, William G., Brennan, Troyen A., and Leape, Lucian L. 1993. A Measure of Malpractice: Medical Injury, Malpractice Litigation, and Patient Compensation. Cambridge, MA: Harvard University Press. Zeiler, Kathryn, Silver, Charles, Black, Bernard, Hyman, David, and Sage, William. 2007. “Physicians’ Insurance Limits and Malpractice Payments: Evidence from Texas Closed Claims, 1990–​2003.” Journal of Legal Studies 36, pp. s9–​s45.

 

Chapter 6

EC ONOMI C S OF PROPERT Y L AW Henry E. Smith

6.1. Introduction The economic analysis of property has rarely been applied to the first question of property theory: what is property?1 Law and economics has made impressive strides in the area of property, but further progress will come only when we apply economic tools to analyzing the institution of property as a system. An economic analysis of property law as a system reveals it to be the law of “near separation” of clusters of activities of private actors into interacting components, or modules. Among the most important of these modules are the legal things that mediate many of the relations between actors. Like its sister areas in private law—​torts and contracts—​property law is a partial solution to the problem of the horizontal interactions of actors. From an economic point of view, moving away from total chaos, much less achieving efficiency, in such a complex system can be a tall order. Consider the set of all actors and every action each could take with respect to each other. A system of detailed rules that would directly ensure optimal sets of actions would be intractable—​at least for interesting and useful definitions of “optimal.” Such top-​down solutions are not the only possible ones. From the ground level, it is not hard to see how given some starting point, actors might want to make mutually beneficial deals (contracts) and forbear from injuring each other (torts). Where does this leave property? It is mostly left to the philosophical literature to speculate about the origins of property in the service of justifying it as an institution. One modest theory along these lines is that of David Hume (1739–​1740). On Hume’s theory, property emerges in conventions of possession that people recognized as mutually 1  Henry Smith is Fessenden Professor of Law, Harvard Law School, and Director of the Project on the Foundations of Private Law. Email: [email protected]. I would like to thank participants at a Workshop on Private Law at the William & Mary Law School for helpful comments. All errors are mine.

 

ECONOMICS OF PROPERTY LAW    149 beneficial and to which they converged based on what we would now call salience. Robert Sugden (2004 [1986]) has extended this theory using the tools of modern game theory. Building on Hume and Sugden, I will argue that basic possession and many of the more refined features of property law approach the private law problem—​reconciling potentially conflicting complex horizontal interactions among actors—​using a strategy of incomplete separation. “Separation” here refers to taking chunks of this world of free interaction and treating them as semifreestanding groups that can be managed in partial isolation from other groups of interactions. This separation is an example of modularity, a well-​known way of managing complexity (Simon 1981; Baldwin and Clark 2000, 58–​59, 236–​37, 257). Decomposing a network into (possibly overlapping) components in which interactions are dense inside the components but relatively sparse between the components allows complexity to be tamed and evolution to occur without major shocks to the system. This separation or modularization happens on many levels in property. First, by defining things that can be possessed, complementary attributes are grouped together under the control of the possessor, obviating the need for rules dealing with the interactions of the complementary attributes. When soil nutrients and moisture are part of the same “thing,” property law need not deal directly with their complementarity. Second, possession in larger social settings becomes more in rem: the rights and duties between the possessor and others are mediated through the modular thing. The duty (to keep off) need not make direct reference to the possessor and her uses, or the details of the internal attributes of the resource. Third, along the spectrum from possession to ownership, exclusion strategies, supplemented by governance norms and rules of proper use, allow information to be managed at lower cost than in a less modular system. Legal thinghood also allows for easier alienability. Fourth, the law of remedies often suppresses information about uses and attributes of owners, thereby enhancing the protective function of property. Fifth, devices for managing the system keep property more formal and more compartmentalized, again using the notion of a thing as a springboard. Sixth, other forms of separation include complex divisions of property like trusts and corporations—​forms of entity property—​and divisions between property regimes, like common and private property intertwined in a semicommons. What all these forms of separation have in common—​thinghood, in rem status, formalization and standardization, and divided rights and regimes—​is that they permit specialization at the cost of strategic behavior. By focusing decision makers on some information to the exclusion of other information, actors can specialize in maximizing the value of given assets or subsets of attributes. The stability afforded by property to the expectations of owners allows them to plan, invest, and develop information about their assets. Closely related to these benefits of property is the problem of strategic behavior: overly narrow-​minded decision-​making by owners (and other interest holders) will sometimes fail to maximize value for others, and overall. Classes of such situations involving exploitation often travel under familiar labels. Thus, “externalities” involve spillovers of uses that affect someone beyond the scope of one’s property

 

150   HENRY E. SMITH rights. The tragedy of the commons involves arbitrage between two systems: the fish in the pond is common (including its ability to help regenerate the resource) but the fish taken from the pond is private. Worse still, highly informed actors will engage in strategic behavior: they may take deliberate advantage of the modular system in order to arbitrage off its structure for wealth-​destroying private gain. When regimes of common and private property come together, people will act as commoners with too much regard for their interests in the private property system: this is true from medieval open fields (common grazing and private grain growing) to joint ventures and standard setting organizations (common project and private intellectual property) (Smith 2000a). Property faces a trade-​off: property law characteristically tackles the private law problem of complex horizontal interactions through the device of separation, but this separation makes possible a range of strategic behavior. To deal with this potential strategic behavior, property law employs more targeted rules and standards—​governance structures like nuisance, the interface of property with contract law in covenants, the interventions of equity, and organizational law.

6.2.  Property as a Collection or as a System It is easier to see how economic tools can be used to analyze property as a system if we have a benchmark for the analysis. The “private law problem” of generalized horizontal interactions among actors in society (or the state of nature) is one such reference point. Closely related are benchmarks from economics, including the zero transaction cost world of the Coase Theorem and the “full” property system that would be possible in such a world (Coase 1960). As we will see, disaggregated views of property—​the so-​called “bundle of rights” or “bundle of sticks” —​take these benchmarks as more realistic than they actually are. But this has things backward. Positive transaction costs point to why some bundles (and rather lumpy ones that) characterize our world and to why property law is a system rather than (as the bundle view tends to see it) a collection of detachable rules.

6.2.1. The Full Property Benchmark Property law, along with contracts, torts, and restitution, forms the traditional bedrock of private law. In the United States and the Commonwealth, these subjects sport a large common law component, although legislation has played an important role, especially in property. In civil law systems, these areas, and property in particular, constitute the core of the civil code. When we turn to economics, its practitioners appear to focus intensely on “property rights,” and economists do engage in a lot of talk about property.

 

ECONOMICS OF PROPERTY LAW    151 But economic analysis in general and law and economics in particular employ very different notions of property from legal ones. Starting with Coase, law and economics has adopted an extreme version of the bundle of rights picture of property (Merrill and Smith 2001b, 2011). The idea of property as no more than an aggregate of rights, duties, privileges, and so on, availing between the owner and others, especially if defined in terms of a list of uses, has been familiar since the Realists adapted Hohfeld’s scheme of jural relations to their ends (Hohfeld 1913, 1917). The bundle of rights picture is both an analytical device and, for many, a substantive claim about property: property is not more than a bundle of sticks, with no unifying theme. As such, there are few presumptions about how sticks are collected or structured, and the bundle picture typically contains no glue holding the notion of property together. In particular, property is not about things at all, and serious social scientists and policy makers have supposedly got beyond the myth that it is (Grey 1980; see also Symposium 2011). The bundle of rights got a big boost from Ronald Coase, and later from law and economics. This choice was an understandable one for Coase, whose main goal was to show the impact of the law on the economy. For this purpose, the bundle of rights picture is made to order. Coase’s purpose was not to explain the law of property—​or the law at all. His target was neoclassical economics, with its unrealistic approach of assuming away the institutional framework governing the interactions of economic actors (Coase 1988, 174). To make his point he shows (in what has come to be known as the Coase Theorem) that if transaction costs were zero, the same pattern of resource use would occur regardless of the set of initial entitlements (or in a weaker version maximum efficiency given a set of initial entitlements) (Coase 1960). In our positive transaction cost world, this guarantee disappears, and the lesson is that we must do comparative institutional analysis to figure out which is the least bad arrangement from an economic point of view (Allen 1991; Eggertsson 1990, 101–​116; McCloskey 1998; see, generally, Posner and Parisi 2013). The analytical convenience of the bundle of rights picture can be carried too far. If property is a collection of rights, duties, privileges, immunities, and so on, with no inherent content, no interaction between them, and no glue holding them together, then one can vary a stick—​Who has the right to create vibrations or prevent them? Who has the right to let cattle graze or to prevent them from doing so?—​and the effect of this variation can be traced out on economic behavior, hypothetically under zero transaction costs and more realistically under various institutional arrangements. The usefulness of the bundle picture for analytical convenience accounts for some of its popularity. “All else equal” is easier to achieve (or to assume) when the various legal relations and legal rules are detachable from their context. One can ask, as law and economics, especially in its early phase, was inclined to do, whether a given “legal rule” was efficient. This procedure is familiar from contracts and torts, where studies of damages rules in contracts and negligence versus strict liability in torts became central topics. For the most part, legal economists treated property in a similar fashion, even though property is—​as I will argue—​less well captured by a rule-​by-​rule analysis than are contracts and torts. It is precisely the system aspects of property that are left out in this approach.

 

152   HENRY E. SMITH Once positive transaction costs come into the picture, the bundle of rights picture shows weaknesses, as well as strengths. For example, one could ask whether various good faith purchaser rules and exceptions for trespass (like necessity) increase efficiency. Notably, these applications of economics to property law occurred especially where property overlaps with torts and contracts. More systemic aspects of property such as land surveying and recording systems received less attention in the formative period of law and economics. I will argue that these other aspects of property are less amenable to the rule-​by-​rule style of analysis popular in first generation law and economics.

6.2.2. Property as a Modular System Economic analysis of property law as a system does not come naturally. System effects and emergent properties are more difficult to measure than rules in isolation. Perhaps because the isolation of legal rules was easier in areas of property that overlapped with torts and contracts, much of the first wave of law and economics focused on parts of property law that are the most contractual or tort-​like, rather than property itself. Indeed, property was treated as the working out of contract and tort principles (Merrill and Smith 2001b). To borrow a civilian term, the view of property implicit in law and economics was an “obligational” one.2 Damages rules in contracts or negligence-versusstrict liability, it seemed, could be treated in isolation for modeling purposes. When it came to property, the analysis tended to be an extension of these strands of analysis, with the law of nuisance taking center stage. To get at what is unique about property law, we can return to the theoretical benchmark of the world of zero transaction costs. We can derive a Coase Corollary from the Coase Theorem:  in a zero transaction cost world, the form of property entitlements would not matter to resource allocation (Merrill and Smith 2011). Or to put it more concretely, in the absence of transaction costs, much of what is done by the law of property could indeed be accomplished using only tort and contract. Taking the true lesson of the Coase Theorem (and the Coase Corollary) seriously, we can ask what is the “essential role” of property in a positive transaction cost world?: what does property law make possible that could not be accomplished by contract?3 Property law owes its actual contours to positive transaction costs. Return to the basic problem in private law—​the problem of potentially conflicting activities by members of society. Think of all the actors and all the resource attributes,

2  The law of obligations embraces contracts, torts, and restitution, and in civil law, an “obligation” is “a two-​ended relationship which appears from the one end as a personal right to claim and from the other as a duty to render performance.” (Zimmermann 1996, 1). 3  I borrow the “essential role” terminology from Hansmann and Kraakman (2000), who use it to analyze what contribution organization law makes that cannot be replicated by contract. They conclude that organizational law is property law.

 

ECONOMICS OF PROPERTY LAW    153 and all the actions each might take that could possibly impact the others. We could theo­ retically define legal relations for each pair of actors, each resource attribute, and each action. If we stopped there, we would be assuming very strict separation between activities and attributes, but this is not realistic. Certain collections of attributes go together (in a compositional dimension of property) (Barzel 1997; Smith 2001). For example, someone with the right to determine how soil nutrients are used might need control over the moisture level, etc. To handle these interdependencies, the rules governing the super thin slices of the world of our thought experiment would themselves have to be very complex and interdependent, or we would need rules of priority among the rules. In a zero transaction cost world, this would all be costless, but in our world, it would be prohibitively costly. Property law economizes on transaction costs by providing massive shortcuts over this fully articulated or “complete” property system. In our world, property law provides a first cut at this problem that aggregates some of these slices, along various dimensions (Lee and Smith 2012; Smith 2012). Complementary resource attributes are collected into things or assets. Rights are defined over many uses by using exclusion strategies over these collections, supplemented as needed by governance strategies referring to particular uses (Smith 2002; see also Field 1989; Rose 1991). The more specific governance rules and standards partially override the background exclusion regime (which applies to the more heterogeneous “otherwise” or “elsewhere” pattern of situation). Thus, licenses or (more robustly) easements displace the exclusion rule (trespass) because they refer more specifically to uses and users. (When two devices or rules apply, the more specific applies over the more general. Very general rules that have many disparate exceptions (because they are displaced by many specific devices) apply “otherwise” or “elsewhere” (Smith 2014a)). In an external shortcut, owners of these collections of attributes have rights defined against other generally—​in rem rights—​that deal with duty bearers in a wholesale fashion. Property is a shortcut over the “full property” that could be achieved in the zero transaction cost world. In an analogy to the incomplete contracts literature, we can call property inherently “incomplete” in our world.4 The degree of this shortcut can be captured using algorithms for finding community structure in a network. There is a large body of literature on finding optimal structures of modules in a network (Newman and Girvan 2004; Newman 2006). We can model emergent legal relations by considering actors as the nodes and their activities as the edges (possibly of varying strengths) (Sichelman and Smith ms.). Then a family of algorithms instructs one on how to remove the edges of most “betweenness”—​the ones that are on the most paths between nodes (based, for example, on shortest walk or random walk), leaving the semi-​isolated modules. The virtue of these models is that they do not prejudge the structure of the system: we can model the emergence of clustering of interactions around legal things if they really do tend to lend the system a modular structure, 4  In the incomplete contracts literature, contracts are incomplete because of positive transaction costs. Property’s incompleteness likewise stems from the costs of more complete property. Note too that it is recognized that in a world of complete contracts, there would be no role for ownership (Maskin and Tirole 1999).

 

154   HENRY E. SMITH i.e., with more interactions internal to them and relatively sparse (but important) interactions in between. For example, the law of trespass tracks modular parcels, whereas nuisance focuses to a greater extent on a few important relations between the users of adjacent parcels. Second, algorithms for finding community structure do not prejudge the level of grain that the law should implement. The structure-​finding algorithm can tell one where modularity is maximized, and it can be combined with the costs and benefits of delineation effort to predict how fine-​grained legal relations should be if they respond to efficiency concerns, either through evolutionary pressures or by design. The use of network theory and measurable modularity carry the potential to open up avenues of empirical work on property as a system.

6.3.  An Ontology for Property Unlike much of the rest of property, economics has offered a bottom-​up theory of possession. The word possession covers many different phenomena, ranging from prelegal notions of control to rights of possession protected by trespass and adverse possession.

6.3.1. Salience and Convention Robert Sugden (2004 [1986]) has reinterpreted Hume’s theory of property using salience and focal points. Thus, when two actors might want to use a resource, a convention may emerge based on who is nearer to the resource or has control over it. According to Sugden, the relevant game is Hawk-​Dove,5 but what level of cooperation of conflict can be left open, as long as it makes sense for people to pick an equilibrium based on salience. Hume and (to some extent) Sugden see the salience that breaks the symmetry between actors approaching a resource as a matter of psychology and inductive reasoning. As David Friedman (1994) points out, the norms of property that emerge from such a bottom-​up process based on salience will reflect a very local version of morality and efficiency, because it must work in pairwise encounters of actors. The question remains open how much this ground-​level morality and efficiency scale up to society. Nonetheless, conventions characterized by this local morality and efficiency place relatively modest informational demands on actors, compared to norms that reflect bigger-​ picture notions of moral desert, distribution, or efficiency (Gold and Smith 2016; Merrill and Smith 2007). A rule of property that required actors to optimize the overall use of resources would not be useful to guide behavior (or litigation). Yet, salience may relate to economic usefulness, in the sense of the benefits of one actor’s use or his or her control over the resource. In Yoram Barzel’s theory of property 5 

In the Hawk-​Dove game, each player’s best outcome would be to play hawk (assert oneself) if the other plays dove (yields), but the worst outcome results for each when each plays hawk.

 

ECONOMICS OF PROPERTY LAW    155 (Barzel 1997), resources are analyzed into their constituent attributes and changes in property rights tend toward efficiency if those with a greater ability to affect the mean return of a collection of attributes get the residual claim over them.6 Contracting parties can be expected to move toward this result (transaction costs permitting), and even nonconsensual activities will have this tendency under some conditions. Barzel disclaims an explanation of how property got started (likening it to a Big Bang), but his approach is a useful supplement to the theory of Hume and Sugden. In a world of actors encountering undifferentiated resources, it often makes sense for those who have the ability to affect the mean return of collections of resources to have de facto possession over them, which includes a legitimate claim that persists beyond the moment. Breaking resources into attributes highlights two points about a bottom-​up theory of property (Smith 2014a). First, notions of possession and thing definition go hand in hand. The law of accession most directly reflects the process of thing definition. “Accession” can refer to a principle of lesser assets (or attributes) going to the owner of related greater assets (or attributes). This principle is reflected in a wide variety of doctrines, including the law of increase (the calf goes to the owner of the mother cow) and fixtures (chattels go with the land they are attached to). In common law, “accession” also refers to what was called “specificatio” in Roman law: someone who innocently mixes labor with material and either transforms it or adds most of the value can keep the new object and pay damages for the things worked upon (or replace it with an equivalent). This branch of accession deals with the persistence of things over time (Newman 2009; Smith 2007). Accession overall can be interpreted as the law of thing definition and claim scope or as an acquisition principle (compare Newman [2009] and Smith [2007], with Merrill [2009]). As we have seen, both Hume and Sugden explain accession on the basis of salience. At any rate, in a possessory claim we need to know the scope of the claim. Accession can be thought of as basic thing definition and maintenance, which feeds into possession. Accession has been analyzed from an economic point of view, mainly as an acquisition principle, as has possession (Merrill 2009). When it comes to first possession, rules that designate a clear winner early in the process tend to be efficient (Lueck 1995; see, also, Posner 2000), and we can note the role of salience and control in establishing a good candidate for someone uniquely well positioned to compete (and use) the resource. Likewise, accession can head off costly competition for unclaimed resources by designating a clear (salient) winner—​the owner of salient resource X gets lesser resource Y. By contrast, where potential appropriators are equally well positioned to access the resource, a rule of first possession can easily lead to the tragedy of the commons. Here homogeneity of appropriators can help maintain rules of property use, in a governance regime (Lueck 1995; Libecap 1989). What we need for the rest of property law is a basic ontology for purposes of social and legal norms. An ontology separates out the basic elements and their relations, including who counts as an actor, what a thing is, and how actors may act with respect to one 6 

Barzel’s theory has some parallels in the theory of asset ownership (Grossman and Hart 1986; Hart and Moore 1990), but Barzel explicitly endogenizes the assets.

 

156   HENRY E. SMITH another.7 Some of these relations are mediated by the things of the ontology: if A possesses resource R, then others have a duty to forbear from entering it, touching it, and so on, depending on the nature of the resource. What attributes are grouped—separated— into things and who has possession of them both respond to considerations of salience and utility.

6.3.2. Possessory Things and Ownership For property law, a key shortcut over “complete” property constructed from contracts is the legal thing. The relation of those with possessory rights and those with corresponding duties is mediated through the thing. When one walks through a parking lot, one need only avoid taking or damaging the cars if one does not own them: the duty bearer need not know anything about the owner’s characteristics or planned uses (Penner 1997, 75–​76). Particularly as possession becomes more formalized into law, there is less need to assess the qualities of the possessor or the potential challenger in order to follow the Humean custom of mutual forbearance. Parcel definition greatly simplifies the law of trespass, particularly when the law develops persistent rights to possess such that the owner need not be actually present in order to be in possession and therefore have the right to possess protected by trespass. Customs of possession are formalized in the process of being adapted into the law. Thus, in the mining camps, miners had a right to work a spot without interference, in a custom called pedis possessio. When the custom was adopted by courts and in legislation, the boundaries of the spot were identified with the formal boundaries of the claim, even when this meant expansion (Smith 2009, 2014a). (More recently, courts have resisted extending it beyond the boundaries of a single claim, as the uranium industry has long wished.) That is, possession itself separates the possessory norm from a great deal of personal information of the actors. As possession becomes “more legal” and more widely applicable, the thing plays a greater role, such that even details of the thing itself do not matter to the duty. As mentioned earlier, when custom and law afford rights to possess to those not actually physically controlling (or even present), the “possessory” right must rely on formal thing definition because the possessor may or may not be around to voice objections. Well-​demarcated boundaries in the case of land serve this function. Extending possession in this way makes sense: property becomes much more useful in terms of investment, specialization, and autonomy by persisting even when actual control or proximity is attenuated, at the relatively low cost of being clear about demarcating the connection of (extended) “possessors” and their “things.” Within possessory norms, we can see nested defaults at work. For example, as we saw, norms of possession arise in particular groups, like whalers on the open seas (Ellickson 7  I am using ontology here as computer scientists do, to refer to the basic set-​up, without making deep metaphysical claims. Here we at least need legal persons, activities, things, and relations between persons with respect to things and activities.

 

ECONOMICS OF PROPERTY LAW    157 1989, 1991). The famous “fast fish loose fish” rule provided that a whale belonged to the first one to harpoon it as long as the harpoon was attached to the whaler’s boat. The physical connection is salient and close to de facto possession. For particularly valuable and dangerous sperm whales, the rule was that of the “first iron,” which gave the first harpooner exclusive rights as long as fresh pursuit continued. The “first iron” rule is less salient, more costly to promulgate, and more specific. It partially displaces “fast fish loose fish,” which was the general rule. One would expect a new species of whale with no special dangerousness to fall under the general rule. Likewise, Ellickson’s findings in Shasta County are consistent with a theory of nested defaults (Ellickson 1991). He found that regardless of whether the formal law is fencing in (ranchers have responsibility for damage caused by their animals on land belonging to others) or fencing out (no such responsibility), the informal norm was for animal owners to take responsibility. In the area of informal norms, the basic possessory norm has a lot of pull. We might further hypothesize that this basic exclusionary regime would be more widespread than expected based on which activity was more valuable in a given small area (Merrill and Smith 2001b). It is the more general default. Layered on top of basic possession and rights to possess are further rules culminating in ownership. At first blush it seems puzzling to have two notions of property-like “control” and to sometimes protect one without the other—​vindicating the rights of nonowner possessors some of the time and nonpossessing owners at others. By layering ownership and title rules on top of possession, property law is able to effect further types of separation, at some cost. The thing in ownership is even more formal than in possession. Indeed possession is often used for low-​stakes everyday interactions: one does not do title searches on the pen one buys from a store. With the layering of ownership and title rules on possession, we see a further example of nested defaults, or the specific over general principle. Rules of title and ownership are more specific (and used for higher stakes) and displace possessory rules that would “otherwise” apply. This means that possession applies whenever it is not displaced, in a very heterogeneous pattern (called “elsewhere”) (Smith, 2014a). This means that finding a unifying theme of possession, other than being a more general partially displaced stratum of law is unlikely to succeed. Indeed, ownership is so formal that some property theorists outside of economics have proposed that property is an office (Essert 2013; Hart 1982: 208; Katz 2012). The idea is that duties avail to the owner qua owner without the need for personal information. Further, when the owner’s property right is transferred to another, the right need not change in content: the new owner just steps into the old owner’s shoes. There is no need to say that the right is different even though an old duty bearer is now a right holder and the old right holder is now a duty bearer. Ownership as an office captures the impersonality of the right but may be overkill. In keeping with the bottom-​up theory, the “legal thing” itself is impersonal, and ownership need not rise to the level of a full-​blown office for separability and alienability of the right to be possible. The key to alienability is depersonalization of the right, which happens through the definition of the thing.

 

158   HENRY E. SMITH What the impersonality of duties and the ease of alienability require is not that ownership be an office but that legal things be depersonalized (Smith 2012). Again, think of the parked cars in the parking lot. The duty bearer need not know whether the car is on loan to the owner’s sister, or whether the owner is a non-​natural person like a corporation. (These divisions of rights stem from further forms of separation that will be taken up in Section 6.6.) The separation of the legal thing from its context allows for this degree of in rem simplicity (Smith 2012). It also allows transfers to take place more smoothly: potential purchasers have less to inquire into, and the transfer happens automatically in many respects. The more formal ownership and title rules allow for more elaborate forms of separation than does simple possession. Divided rights are not really possible when all we have is a notion of possession. Either one is in possession or not. For divided rights, something more is needed including a method of keeping track of more connections between persons and things in order to divide rights. At the very least, we need rights to possess that persist despite their holder not being in possession. Title is one method of keeping track of these connections. Furthermore, ownership allows for divisions over time and by use, such as easements.

6.4.  Delineating Property Rights In moving from possession to ownership and the division of rights, we are in the realm of delineating property rights. Economists going back to Demsetz (1967) and carried through in the New Institutional Economics (e.g., Barzel 1997) have focused a great deal of attention on the delineation of property rights, including the descriptive question of how they evolve in response to background conditions and what efficiency properties we should expect (or not) from systems of property rights. Demsetz’s theory was a demand side theory: the Demsetz Thesis holds that property rights will evolve in order to deal with externalities as they arise (Symposium 2002). Other branches of the literature ask how property rights are supplied, often very imperfectly relatively to an ideal baseline (Alston, Harris, and Mueller 2012; Libecap 1989; Ostrom 1990; Wyman 2005).

6.4.1. Exclusion Versus Governance One aspect of property rights definition is the problem of separating out clusters of activities and attributes and then dealing with the problems of strategic interaction that such separation gives rise to. Property rights are typically delineated with some combination of exclusion and governance (Smith 2001, 2004a). An exclusion strategy is low cost and low precision: it defines basic modules and takes care of problems wholesale. The law of trespass reflects an exclusion strategy, and the message is a simple one: keep off unless you have permission. This leaves important spillovers unaddressed, and

 

ECONOMICS OF PROPERTY LAW    159 where the stakes are high enough, they are sometimes worth dealing with through a governance strategy. A governance strategy focuses in on a narrower class of uses, as in covenants, easements, and nuisance law. This part of property law enriches the interface between parcels in real property law. Because most personal property has no fixed location, restrictions on use generally are achieved exclusively through in personam contract, tort law, product standards, and safety regulations, rather than through the law of servitudes. Separation of attributes and their associated activities into clusters gives rise to problems of strategic behavior. The separation into modules is not complete. Modularization allows for specialization—​each owner can become more informed and skilled than others at using his or her asset—​but property rights also lead to myopia (or worse). Some of this goes under the heading of “spillover” or “externality.” Some of these spillovers can even be created to extort others, as, for example, opening a horse stable in order to be paid by neighbors to shut it down, or emitting pollutants to receive subsidies for stopping (Kelly 2011). Further, what counts as a thing for property has to be stable in the face of strategic behavior. Gathering attributes together into “units” needs to be done in order to reduce measurement costs (Barzel 1982), and what counts as a unit may change depending on the legal treatment: taxes and legal rules that burden or benefit units or things can call forth effort at changing the underlying unit or thing (Barzel 1976; Smith 2000b). Thus, taxes on cigarettes have resulted in longer cigarettes, and access to riparian rights has resulted in “bowling alley” parcels. Property law needs doctrines to prevent strategic reconfiguring of things. Various combinations of exclusion and governance may be better at containing strategic behavior. Putting the effects of an entire interaction within the scope of a parcel would be one method—​by removing the module boundary that causes the externality (see Ellickson 1993, 1322–​1335). This comes at the cost of any specialization that would be facilitated by a more fine-​grained parcel definition with a boundary through the activity. Another method to contain strategic behavior is to supplement the partial exclusion with a governance regime that addresses the potential opportunism. Thus could be by easement, contract, nuisance, or regulation. Or various forms of entity property could be used to govern the wider interaction—​as, for example, in a common interest community.

6.4.2. Law Versus Equity One pervasive governance-​like device aimed at containing opportunism is equitable decision making by courts (or arbitrators or administrators). Separate courts of law and equity are mostly a thing of the past, but law and equity persist as partially separate when it comes to substantive law and remedies. Economic analysis mostly treats the law versus equity divide as irrelevant, and treats the distinction as faintly reflected in the choice between rules and standards (Kaplow 1992), and property rules versus liability rules (Calabresi and Melamed 1972; Ayres 2005). I argue that equity is a functionally distinct

 

160   HENRY E. SMITH decision-​making mode that is particularly necessary when it comes to cleaning up the strategic behavior made possible by incomplete separation. One problem with formal law, including the part of property law that defines property rights, is that some actors with a lot of information will take unforeseen advantage of these systems. This is a problem very familiar from tax law (Weisbach 1999; Lawsky 2009). Economists have not been of one mind about the usefulness of opportunism as a category for analysis. Oliver Williamson relies heavily on opportunism in his version of New Institutional Economics, in which he shows how a variety of contractual and organizational devices can be used to prevent opportunism that would otherwise prevent cooperation. Williamson adopts a very broad definition of opportunism as “self-​interest seeking with guile” (1985, 47). This is broad in the sense that Williamson includes all manner of rule violations and promise breaking as opportunism. It is narrow in the sense that it is not clear that full-​blown deception should be required for opportunism. Others have argued that opportunism is simply self-​interest, and all a system should do is its best to provide an environment where self-​interest is consistent with efficiency (compare Barzel [1985, 10–​11] with Williamson [1993]). On this skeptical view, failure to do so does not call for hand wringing about the immorality of the actors within the system. Elsewhere I have defined opportunism as: […] behavior that is undesirable but that cannot be cost-​effectively captured—​ defined, detected, and deterred—​by explicit ex ante rulemaking…. It often consists of behavior that is technically legal but is done with a view to securing unintended benefits from the system, and these benefits are usually smaller than the costs they impose on others. (Smith 2010, 10–​11)

The key from the point of view of legal design is to come up with proxies that can trigger presumptions against the potential opportunists, along with further rules of thumb for evaluating behavior once the equitable safety valve is triggered. Here too, it is not surprising that the law tracks widely accepted morality. Taking unforeseen advantage of a situation in order to appropriate more of a smaller pie does meet with disapproval. More importantly, an institutional analysis will keep on the table the full range of responses to such opportunism. These include better ex ante rules, ex post equitable standards, and tolerating some residual opportunism. It should also be noted that social norms do much of the work in getting people not to act opportunistically (a point to which I return below). Much of what goes under the heading of equity (and used to fall under equity jurisdiction) and which further drew on an Aristotelian strand of thinking about equity, sees a need for an ex post, morally infused discretionary standard that would be targeted in personam) to people engaged in opportunism (Ayotte, Friedman, and Smith 2013; Feldman and Smith 2014; Smith 2010, 2014b). Crucially from an institutional design perspective, this decision-​making mode is implemented in a system of proxies and presumptions that constrain the operation of equity in order to prevent it from chilling legitimate behavior. Consider as a prototypical example a building encroachment. It is

 

ECONOMICS OF PROPERTY LAW    161 sometimes said that building encroachments (at least in the old days) would automatically call forth a harsh remedy of injunctions but that modern courts worried about hold-outs will now usually give damages. This is not totally wrong, but it gives an incomplete picture. Continuing trespass (of which a building encroachment is an example) gives rise to a rebuttable presumption for an injunction. However, we are worried about overcompliance (Craswell and Calfee 1986; Sterk 2008), and, from a “less economic” point of view, ex post unfairness. So the law allows a shift to damages where the encroacher has made a good faith mistake and the encroachee would face disproportionate hardship. This meant that the injunction would harm the enjoined party far more than it would benefit the moving party—​a torn down building would be a classic example. It did not mean equipoise or some other kind of cost–benefit test that would ask whether the particular proposed injunction’s benefits would exceed its costs. Nevertheless, if the encroacher is in bad faith, an injunction should issue. The scheme of proxies and presumptions, keyed to good versus bad faith and to disproportionate (or “undue”) hardship, is designed to solve a two-​sided opportunism problem that is set up by the system of boundaries between parcels and the ad coelum rule.8 We have to worry about the hold-out on one side and the land-​thief on the other. Here equity works though the different remedies on offer in the various situations that carry more or less dangers of opportunism. I return to remedies in the next subsection. Equity is thus a safety valve largely defined by the proxies and presumptions that trigger this kind of inquiry. Like governance (versus exclusion) and ownership (versus possession), equity is a more specific regime that overrides the law in particular circumstances (it is second-​order “law about law”). The proxies that trigger the presumption against the possible opportunist are more specific than the general rules. Where does this happen? The idea that some situations are so pregnant with the danger of opportunism that a class of transactions will not be enforced or will be subjected to searching scrutiny is familiar across private law. Unconscionability in contract law, another outgrowth of equity, was traditionally keyed to “constructive” or “near” fraud. The dangers of opportunism and error can be given an economic interpretation as in Epstein’s (1975) theory of unconscionability: some transactions have indicia of fraud that cannot be directly proven and, given the possibilities of type 1 and type 2 errors, it sometimes makes sense to refuse enforcement of transactions (even if this will prevent a few positive deals) (see, also, Leff 1967). He draws an analogy to the statute of frauds: withholding enforcement of certain kinds of contracts not in writing could be better than enforcing them all or picking and choosing (this is an empirical question).9

8  Under the ad coelum rule (short for the maxim “cujus est solum, ejus est usque ad coelum et ad inferos,” which translates as “one who owns the soil owns also to the sky and to the depths”), the boundaries of a parcel extend upward and downward (presumptively, with adjustments, for example, for overflights). 9  Interestingly, the statute of frauds was an early alternative to the system of land records, another device to prevent shady dealings in property—​a topic to which I return in Section 6.5.

 

162   HENRY E. SMITH Equity relies on a type of local morality, in a fashion similar to the one we saw in connection with possession. Equity historically had a role in enforcing custom, and was the vehicle by which some customs could be adapted into the law. The theory of social norms suggests that close-​knit groups will come up with wealth-​maximizing norms (Ellickson 1989, 1991). There may be outgroup externalities, as there were with whaling norms, which were good at dealing with (pairwise) conflicts between whaling ship crews but led to the overhunting of whales (Ellickson 1989; Friedman 1994). Further, some customs will scale up better than others will. There is the danger of enforcing in an end game situation customs that presuppose an ongoing interaction and of thereby distorting the process of norm formation (Bernstein 1996). Nevertheless, custom has been and can be a source of new legal norms under the right conditions (Parisi 2000). As we saw, part of the process of making a custom more widely applicable is to formalize it (Smith 2009). Despite its reputation as discretionary and detailed, equity probably had a role in simplifying custom for use by the law. In particular, rough judgments about whether one is violating a custom could inform determinations of good faith. There is as always a danger here too of distorting the norm or the law in the process of “enforcing” it. Equity acts in personam in the remedies it offers, and normally it intervenes in a targeted fashion. It acts as a (higher-​order) safety valve, in which a court can intervene—​ based on the proxies and presumptions—​against a specific instance of opportunism. On the theory offered here, equity’s contours reflect the strategic interaction of parties with each other and with courts. But the existence of the system is announced to the world. Indeed, the possibility of equitable intervention leads to the possible chilling effect, much emphasized by the opponents of equity throughout its history. Whether the net effect is chilling, or reassurance on the part of ordinary non-​opportunistic actors is an empirical question. Legal designers also face the question of how the possibility of equitable intervention, especially when it is couched in sometimes-​ambiguous moral-​sounding terms (such as good faith), will reach its target audience, or audiences. When the law sends a different message to different audiences, legal theorists call this “acoustic separation” (Dan-​ Cohen 1984). The term was introduced by Meir Dan-​Cohen to describe the possibility that conduct rules directed at primary actors might differ from decision rules used by judges or other officials. For example, the criminal law might tell people not to steal, but give judges or prosecutors discretion not to enforce, or to show leniency, in certain situations. Equitable intervention may work similarly. The same moral-​sounding directives that tell highly informed opportunists that courts will come down on their misdoing, may sound like everyday morality and provide reassurance to those with no nefarious intent. In “behavioral equity,” the equitable message may promote compliance and prevent evasion by the opportunists, while not interfering with the intrinsic moral motivation of ordinary people (Feldman and Smith 2014). Insights from behavioral decision theory can be integrated into the familiar rules versus standards paradigm, in this as in other areas of property law.

 

ECONOMICS OF PROPERTY LAW    163

6.4.3. Damages Versus Injunctions We have already seen in building encroachments that conventional law and economics uses the lens of property rules versus liability rules to analyze questions of remedies in property and other areas. Calabresi and Melamed (1972) defined a property rule as protection for an entitlement that aims at forcing a would-​be taker to obtain the holder’s consent. If an entitlement is protected by a liability rule, it can be taken, with the only consequence being the payment of officially determined damages. (Calabresi and Melamed also introduced inalienability rules, which forbid the transfer of the entitlement.) The primary example for Calabresi and Melamed, as for Coase, was the law of nuisance. Calabresi and Melamed applied their criteria of efficiency and fairness (and, in principle, “other justice considerations”) to this question and argued that where transaction costs are low property rules should be used, in order to ensure that takings of entitlements are welfare increasing. Where transaction costs are high, because of potential holdouts among those affected by the nuisance—​think of a factory and residents—​or because of free riders among those who might pay to shut down the nuisance, liability rules could improve matters over the corresponding property rule. The polluter would take external harm into account because of the damages, and those affected by the pollution would be compensated. Of course, the determination may be incorrect, and subjective values are left out of the picture. Calabresi and Melamed (1972) derived four rules from a two-​ by-​ two choice. Following Coase’s (1960) notion of reciprocal causation,10 they posited that the “entitlement”—​for example, to emit smoke or to enjoy clean air—​could be “given” to either party, for example a factory owner and a resident (or residents). The entitlement (in either party) could be protected by a property rule or a liability rule. This yields four possibilities. Under Rule 1, the resident has the entitlement to clean air protected by a property rule, and so can get an injunction to shut down a polluting factory. Rule 2 affords the resident the entitlement but only protected by a liability rule, namely damages. Flipping things around so that the factory owner has the “entitlement,” they say that the entitlement (to pollute) can be protected by a property rule or a liability rule. They identify the property rule in the polluter, Rule 3, as the resident’s inability to get an injunction and the factory’s ability to continue to pollute. In the most innovative move, they pointed out that the polluter could be protected by a liability rule, in Rule 4, where the resident would have the right to take the entitlement to pollute from the factory owner but the resident would have to pay damages (the factory owner’s shut down costs). Others have pointed out that Rules 2 and 4 are like call options and have explored put options (forced 10 

Coase argued that causation was reciprocal in the sense that in any resource conflict, the conflict would not arise but for the presence of each of the parties. Thus, the crops contribute to the trampled crops as do the cattle, and in the example under consideration, the resident causes the conflict, as well as the factory.

 

164   HENRY E. SMITH sales) (Morris 1993; Ayres 2005). Additional rules are possible (e.g., higher-​order rules) (Ayres 2005). It is the entitlement structure itself that typically receives too little attention in the economic analysis of property law. One symptom of the problem is that the law against theft, hardly controversial in general, has been considered difficult to capture in law and economics. What if thieves value entitlement more than “victims” do? Part of the explanation is that the law of theft obviates elaborate precautions by owners, which would otherwise be wasteful (Hasen and McAdams 1997). The same can be said for broad use of undercompensatory liability rules: holders will use self-​help to avoid the taking, and property rules can save on some of the costs of self-​help (Hylton 2011; Smith 2004b). More generally, entitlements are not as thin and malleable as this literature assumes (Rose 1997, 2178–​2179; Merrill and Smith 2001b). Again, for Coasean transaction cost reasons, the entitlement structure is lumpy and asymmetric (Lee and Smith 2012; Merrill and Smith 2001, 2011; Rose 1997; Smith 2004a, 2012; see, also, Fennell 2012). We do not ask against a blank slate whether pollution should or should not occur. The background set of entitlement includes a right to be free from more than de minimis pollution, depending on the nature of locality, and, subject to safety valve exceptions, this entitlement is protected by a property rule. Crucially, the general right to exclude sweeps in this type of right without it needing to be separately delineated. As noted earlier, property solves these types of problems wholesale, subject to retail-​level adjustments for situations of bad faith, extreme holdouts, and the like. And because entitlements are lumpy, they are far from being symmetric. Consider again the factory-​resident example. If an entitlement in the resident protected by a property rule is Rule 1, and an entitlement in the resident protected by a liability rule is Rule 2, it is problematic to switch things around and expect symmetry as Calabresi and Melamed do. Thus, “Rule 3” is not the mirror image of Rule 1: there is in the default package of entitlements no right to pollute, in the sense of being able to sue the resident for not accepting the pollution (for example, if the resident blows it back). At most, Rule 3 means the denial of an injunction such that the polluter can exercise a liberty to pollute. A polluter may have a right to pollute if it has an easement to pollute (or some non-​property tradable permit relating to pollution), by either grant or prescription, but these are adjustments to the interface between the modules, the legal things corresponding to the adjacent parcels. They are a step on the road to governance and modify basic exclusion. Thus, if we move from Rule 1 (injunction for the resident) to Rule 2 (damages for the resident) in certain situations of extreme holdout behavior (roughly), there is no corresponding need to go from Rule 3 to Rule 4, because the nature of the “entitlements” is very different. Because the default package of rights does not include a right to pollute, there is no corresponding need to “soften” it with a Rule 4. The famous coming-​to-​the-​nuisance case of Spur Industries, Inc. v. Del E. Webb Development Co.,11 in which the court required the victim side to pay the costs of shutting down the feedlot,

11 

494 P.2d 700, 708 (Ariz. 1972) (en banc).

 

ECONOMICS OF PROPERTY LAW    165 is anomalous—​and indeed such a result has not occurred again (Epstein 1997; Smith 2004a). A subsequent (vast) literature further developed the framework of property and liability rules, and much of the tenor of this work is pro-​liability rule. Ever more elaborate schemes of liability rules can do better than a simple estimate of market value of the entitlement taken (Ayres 2005). One interesting result in the liability rule literature is that in contexts like nuisance, an average harm rule can be better than a property rule (Kaplow and Shavell 1996). If the liability rule is based on average harm and the courts’ estimate of liability is unbiased, then ex ante the polluter and the victim will be presented with a correct expected value, and their incentives will also be correct. The same cannot be said for a property rule, which will deter some takings where the potential taker values the entitlement more than does the holder (potential takee). There is, however, a problem of separation involved in this strand of liability rule literature. The assumption that liability rules can be based on an unbiased estimate of value (or harm) is stronger than it appears, because the actuarial classes involved may not be stable (Ortiz 1994: 403–​06; Smith 2004b). The problem is the familiar one of partial separation and consequent strategic behavior. Once a liability rule is in place, knowledgeable takers may be able to cherry pick assets that are more valuable than the court is likely to find them to be, based on its imperfect proxies. This is a form of arbitrage, and it fits under the heading of opportunism discussed earlier. And not surprisingly, the injunction is used here, especially in situations of subjective value and putative game playing. Indeed, as discussed in the building encroachment situation, sometimes the problem in remedies is potential strategic behavior on both sides, which is accounted for on the traditional test for injunctions (Gergen, Golden, and Smith 2012). For example, traditional injunction jurisprudence was keyed to good faith. It also was based not on a cost–benefit test but on disproportionate or undue hardship, not equipoise. Thus, in the central examples to the liability rule literature like Boomer v. Atlantic Cement Co.,12 the law of injunctions is flattened out in both the more recent law and in the conventional economic literature (Gergen, Golden, and Smith 2012; Laycock 2011). Older law, far from giving automatic injunctions, was attuned to the two-​sided opportunism problem, and it provided a safety valve for holdouts while doing less damage to the entitlement to be free from nuisance. Generally, we may say that the term property rule was chosen advisedly after all, and there is an information–​cost rationale for the prevalence—​surprising on the conventional approach—​of property rules. In addition to the abatement of self-​help, more generally, broad entitlements protected by property rules solve some problems of strategic behavior, while causing others. Property rules and liability rules correspond to sanctions and prices, respectively, in the terms of Cooter (1984), and we can see why. Protection of a set of entitlements through a property rule is a sanction in the sense that expected liability takes a jump at a certain signal (e.g., crossing the boundary of a parcel

12 

257 N.E.2d 870 (N.Y. 1970).

 

166   HENRY E. SMITH or taking away a chattel), which corresponds to the notion of doing something wrong. It is not a price, which is a charge for doing something permitted, and which varies continuously with harm (assuming harm is continuous). As Cooter points out, sanctions are more appropriate where we know the correct standard of behavior and know less about the extent of harm in a given case, whereas prices are more appropriate where we know more about marginal harm than about the optimal level of an activity. The sanction does not vary with variance in behavior, making them less subject to those forms of strategic behavior like the cherry-​picking problem. Correspondingly, owners can develop information about their assets without having to worry whether they can prove harm to a court. When holdout behavior becomes too strong, the rationale for the sanction is diminished. The cost of a property rule is the strategic behavior of holdouts and the problem of externalities that span the boundary of the legal thing and the rest of the system. It is for this reason that we do not have only the law of trespass or a rule of automatic injunctions. Governance strategies and more tailored remedies—​conditions on and exceptions to injunctions—​are designed to deal with remaining strategic behavior at some cost. Again, the question is how the system overall deals with horizontal interactions, including the strategic behavior presented by various forms of partial separation into modules.

6.5.  Managing the System Property features many principles, doctrines, and institutions that aim to manage the problem of separation and strategic behavior. Property law is known (and often criticized) for its formalism, but it is not completely formal or equally formal across the board. Formalism is a type of separation—​the relative invariance of a rule, statement, or system—​to its context (Heylighen 1999). Thus, the language of first-​order logic is more formal than natural language, and written language tends to be more formal than spoken language. Likewise, everyday mathematical notation is less formal than the language used in proofs, because it is a sort of shorthand for those who are in the know (including the writer). The separation involved in the formalism of property involves making property more (but partially) invariant to contextual information. This is helpful where in rem duty bearers are involved (Smith 2003). All legal systems, and property law in particular, face a basic communication trade-​off: at the same cost, one can communicate in an informationally intense way (much information per unit of effort) to a socially closer audience or in a more stripped down formal way to a more impersonal and extended audience (Smith, 2003). Elliptical communication, as when someone asks that a window be closed by mentioning it is cold, is more possible in contexts that are more personal. Correspondingly, rights can presuppose background information in contract law (in personam) more than in property (in rem). More can be expected by neighbors as duty bearers (nuisance) than by the world at large

 

ECONOMICS OF PROPERTY LAW    167 (trespass). Even more can be expected in a scheme of covenants, although familiarity will decrease over time (changed conditions). Property law is more formal in some contexts than in others. More in rem aspects have to reach an audience of large and less informed parties. Making them responsible for idiosyncratic information would make for high processing costs on their part. This is a rationale for the numerus clausus, or the closed list of basic property forms (Merrill and Smith 2000, 2001b). Those creating idiosyncratic property rights will tend not take into account informational externalities. It might be thought that property records remove the need for standardizing property (Epstein 1982; Hansmann and Kraakman 2002). This does not follow. What the impact of property records is on standardization is an empirical question. For one thing, registration systems appear to require a stricter numerus clausus because the registrar, who pronounces valid title, cannot be expected to process a great deal of idiosyncrasy (Arruñada 2003, 2012; Smith 2010). (The registrar does a mini quiet title action and stands in for the set of in rem duty bearers.) In other areas, having a standardized format for information is beneficial even if the information is readily available (e.g., court documents). Standardization happens sometimes spontaneously, sometimes through private actors with a stake (trade associations), and sometimes by the state. There is reason to think that where the state is already enforcing property rights there are economies of scale and scope in the state taking on the standardization function as well (Barzel 2002). Interestingly, the standardization function can be separated from property and placed with special private actors, in standard setting organizations, and we worry there about strategic behavior. Equitable intervention is used to prevent misuse of patents in standard setting organizations and could be used even more systematically (Smith 2013). For areas that fall in between property and contract, we need to find intermediate levels of standardization (Merrill and Smith 2001a). Thus, in bailments, landlord-​tenant, security interests, and trusts, major aspects of the law are not fully in rem or fully in personam, and those aspects that are more toward the in rem side of the spectrum tend to be more formal and standardized: reaching a more indefinite and/​or more impersonal audience requires more formalization and standardization. Also, in general, where contracts are treated as property and made alienable, they are treated more like things. Traditionally, it was equity that lent personal rights a property character and the prerequisite for propertarian treatment was that the contract be unconditional, specifically enforceable, and tied to identifiable property (Worthington 1996; Penner 1997). In other words, property treatment required separation from context. Negotiability is the most extreme form of separation, with cash being the strongest example. One can get good title to cash even with a thief in the chain of title. As elsewhere, separation promotes alienability, but at the cost of potential strategic behavior. Sometimes this behavior is handled by the criminal law, and sometimes (as with a cashier’s check), the drawer chooses to take the risk in the interest of minimizing the need for inquiry by the payee. The costs and benefits of separation and prevention of opportunism all vary by the type of resource and the situations parties are likely to find themselves in.

 

168   HENRY E. SMITH If strategic behavior is enough of a problem, rules can become mandatory. The mandatory rule can provide for notice or protection. Where the externalities from alienation are too great, we sometimes find inalienability rules, such as for human organs, votes, and the like. If there is a mandatory core in fiduciary law, it is justified by very hard to anticipate strategic behavior. And the enforceability of equitable interests against third parties requires notice, and this not surprisingly cannot be contracted around. One benefit of standardization that receives little attention is the interconnection problem. If we had many idiosyncratic property rights, the question would be how they combine. As it is, when two parcels are unified, their basic feature—​the nature of the boundary, the rules of co-​ownership and the like—​automatically do not clash. This is by no means guaranteed to occur without any effort. An indication of the dangers averted by standardization can be found in the area of intellectual property licensing (Van Houweling 2008, 938–​939). If someone is trying to make a work that involves more than one piece of copyrighted material, it is important that the licenses not clash. This problem surfaced in open-​access licenses, which went through several generations. The important aspect of such licenses is that the obligations travel to remote users. Different generations of open-​access licenses specified inconstant duties. This is the type of problem that, on a much larger scale, the numerus clausus and other standardizing aspects of property law are aimed at preventing. Standardization benefits can also be seen in the realm of land surveying and parcel definition. In a series of articles Libecap and Lueck (2011) have shown that the rectangular survey system leads to higher land values and less conflict than the metes and bounds system (Libecap, Lueck, and O’Grady 2011).13 In Libecap and Lueck (2011), they present the results of a natural experiment based on two regions of Ohio, which for exogenous reasons received rectangular survey versus metes and bounds treatment, and found better alignment of parcels, 18 times fewer land disputes in the nineteenth century, 20%–​30% higher per acre value in flat terrain through at least the middle of the twentieth century, and higher population densities, urbanization, and investment in industry in the areas on the rectangular survey side of the boundary. The rectangular system achieves a greater separation of parcel definition from local context, which is more stable over time and allows for better modularization of parcels.

6.6.  Extended Property Rights Property comes in more complicated and flexible forms than the basic set (“estates and future interests”), even supplemented by forms of co-​ownership. Separation can be pursued further to create entity property, the foundation of many organizational forms 13  Under the metes and bounds system, descriptions of parcel boundaries are based on angles and measurements, often using markers like rocks and trees as fixed points. The rectangular survey prescribes a grid within which rectangular parcels can be defined.

 

ECONOMICS OF PROPERTY LAW    169 (Merrill and Smith 2010, 123–​158; 2012, 646–​806). These include the trust, but also the corporation, common interest communities, and the like. The familiar types of property regimes themselves—​private, common, and public—​can also be mixed together in a variety of ways. Both the extension of property into forms of organization and the mix of property regimes raise issues of separation and strategic behavior.

6.6.1. Entity Property As discussed earlier, personal obligations can be treated as property, and equity courts played a large role in this process. The trust is a systematic treatment of personal obligations as property. In a trust, a settlor transfers legal title of the property (the corpus of the trust) to a trustee, who is obligated to manage it for one or more beneficiaries. The beneficiary has equitable title, which means that the beneficiary has rights protected against the trustee, most notably through the fiduciary duties of loyalty and care. The beneficial interest (equitable property) receives some, but limited protection against third parties. If the trustee transfers the property to a third party wrongfully, the beneficiary can claim the property back from the transferee unless the transferee is a good faith purchaser for value. No one with notice can be in good faith, but the law does not impose much of a duty of inquiry on potential transferees. There is another kind of separation involved, which Hansmann and Kraakman (2000) call “asset partitioning.” In affirmative asset partitioning, a pool of assets is designated that is free from the personal creditors of the holders of the asset. In a trust, the corpus is not subject to the personal creditors of the trustee (and through the concept of a fund, the claims of personal creditors of the beneficiary are subordinate to those of the creditors dealing with the trustee qua trustee). The trust involves separation and requires special devices to contain consequent strategic behavior. Conventionally, it is said that a trust involves the separation of legal and equitable title and the fiduciary duties are designed to contain the extreme danger of trustee misbehavior. Beneficiaries are often vulnerable and have difficulty monitoring trustees. The duty of loyalty prohibits self-​dealing and conflicts of interest with a flat ban; good faith and substantive fairness are no defense. The trust is a complex mixture of contract and property, with a different emphasis in different jurisdictions (Lau 2011; Sitkoff 2011). It needs to be in order to respond to this special kind of separation. This is a lot like the equitable safety valve aimed at countering opportunism, except that in response to defined highly dangerous situations, fiduciary law also uses broad ex ante rules as well as targeted ex post intervention (Smith 2013). (Broad ex post intervention would be the most destabilizing of private arrangements.) Organizations like corporations and partnerships also involve affirmative asset partitioning (Hansmann and Kraakman 2000). Some organizations also feature defensive asset partitioning—​known as limited liability in organizational law. Under defensive asset partitioning, the holders of the designated asset pool (the corpus, the firm’s assets) are protected in their personal assets from claims of creditors of the entity. Entity assets

 

170   HENRY E. SMITH can be treated as semiautomatous, which allocates information costs among the various actors and can allow certain actors, e.g., creditors of a firm, to specialize in monitoring a given pool of assets. Unlike defensive asset partitioning, affirmative asset partitioning cannot be replicated by contract (Hansmann and Kraakman 2000). To designate a pool of assets and shield them from the personal creditors of the owners would require a complex set of covenants that would have to be updated, and which would be hard to enforce. The transaction costs would be prohibitive, in a fashion reminiscent of the transaction costs that property law obviates by taking the shortcuts involved in delineating rights by way of legal things. Hansmann and Kraakman identify asset partitioning as a contribution of property to organizational law. Indeed, we can go further: the conventional view of corporations as a nexus of contracts is limited in just the way that the bundle of rights picture of property misses the contribution made by thing definition and related devices in the law of property. Like property law generally, asset partitioning involves separation, here of the asset pool from certain classes of claims. In a way, this is module or thing definition, but of a more entity-​oriented sort. Like other forms of separation, asset partitioning gives rise to potential strategic behavior. In trusts and in organizational law, albeit to different extents, one major method of constraining opportunism is through fiduciary duties. Monitoring and competition may also serve this function, and the greater possibilities of the latter probably explain why fiduciary duties are weaker and more subject to variation by contract in organizational law than in trust law. Separation and encapsulation of information in organizations, with consequent specialization, are a fundamental property-​like contribution to firms in general. Daniel Spulber (2009a, 2009b) applies the Fisher Separation Theorem to define a firm as an organization in which the decision making about the firm’s objectives can be separated from the personal preferences of its residual owners. If possible, such separation facilitates the famous “separation of ownership and control” in corporations, which can be seen as a prominent version of entity property. When Berle and Means (1932) introduced the separation of ownership and control, they regarded it as a challenge to the validity of private property. They emphasized what we would now call agency cost problems. Problems like these are, however, the flip side of the separation that allows for specialization (e.g., management and capital provision). The real question is whether such strategic behavior can be cost-​effectively contained, through fiduciary duties, market competition, and other devices (see, e.g., Jensen and Meckling 1976; Easterbrook and Fischel 1991). Far from undermining the notion of property, the separation of ownership and control in corporations is a paradigm case of separation, specialization, and strategic behavior that is a leitmotif through all of property law. Entity property allows for more complex mixes of exclusion and governance than do the basic property forms. What role asset definition plays in the theory of the firm is still an open question. Problems of measurement and strategic behavior can be dealt with by exclusion strategies in moving firm boundaries (e.g., making instead of buying, vertical

 

ECONOMICS OF PROPERTY LAW    171 integration) or with more elaborate governance rules, both off-​the-​rack and contractual, for specific types of opportunism.

6.6.2. Mixed Regimes Similar complex mixtures of types of rights can also be achieved by mixing different property regimes—​public, common, and private. Private, common, and public property can be mixed together. One system may abut another, in the sense that a thing will be treated differently as it moves from one regime to the other. The tragedy of the commons occurs because the stock is common but units of flow (e.g., fish from a pond) are private if taken by first possession (Cheung 1970; Gordon 1954; Warming 1911). If the fish and pond were common, there would be no tragedy but perhaps there would be insufficient incentive to fish. If both fish and pond were private, there would likewise be no overfishing, but the risk spreading and cheap definition of common property in the pool would be forgone. Which combination is most efficient is an empirical question. Theoretically a property regime, by creating separation, can give rise to strategic behavior and externalities. Thus, if in a private property regime, too many actors have exclusion rights or overfragmented rights given likely use, the multiple-​veto can lead to underuse, in an anticommons (Buchanan and Yoon 2000; Heller 1998; Parisi, Schulz, and Depoorter 2005). There is a higher-​order question of how to motivate actors to create or modify a property system in the first place (Krier 1992; Fennell 2004). In some situations, disparate stakes can make some actors find it worthwhile to create such a system with a combination of positive and negative externalities for others (Levmore 2002; Wyman 2005). Private, common, and public property can be combined in the micro level. In a semicommons, common and private property cover the same things and interact (Bertacchini, de Mot, and Depoorter 2009; Fennell 2011; Smith 2000a). The problem with this type of separation is again strategic behavior. An example is the open fields of medieval and early modern England. Peasants owned long strips that were cultivated as private property for grain growing but would be thrown open for common grazing after harvest and in fallow periods. This allowed for specialization and internalization in grain growing and operation on a larger scale for grazing, which seems to have involved greater scale economies. The problem with this temporal interleaving of private and common property was the potential for strategic behavior: actors could appropriate goods (manure) and fend off bads (excessive trampling) in the common use with regard to how it impacted their private parcels. These scattered strips can be seen as a method of containing the strategic behavior: at some cost of inconvenience and externalities in the grain growing, the strategic picking and choosing, with steering of cattle for trampling and manure, would be too difficult. Similar problems, requiring special solutions, can be seen in a wide variety of areas, especially where the things of property are hard to define, as in water, the Internet, and intellectual property (Grimmelmann 2010; Heverly 2003; Smith 2007, 2009).

 

172   HENRY E. SMITH The public and private property regimes also interact. In addition to the externalities from one set of users to the other, we must also worry about rent seeking resulting in problematic transfers. The takings doctrine polices private to public transfers, and the public trust reins in some public to private transfers (Epstein 2003; Merrill 2011). The partial separation of the two regimes gives rise to strategic behavior that requires policing.

6.7. Conclusion The increasing complexity of property in response to new economic activity is not a reason to jettison formalism outright. In this, the Realists went overboard. Economic analysis, in adopting the bundle of rights and in treating problems in a detached fashion, runs the risk of entrenching this non sequitur. Instead, new developments in the world and in property law call now more than ever for an analysis of property as a system. Part of that project will involve exploring how property separates—​and how and whether it should separate—​chunks of the world of private interactions and how and whether it should deal with the resulting patterns of potential strategic behavior.

References Allen, D. W. 1991. “What are Transaction Costs?” Research in Law and Economics 14, pp. 1–​18. Alston, L. J., Harris, E., and Mueller, B. 2012. “The Development of Property Rights on Frontiers: Endowments, Norms, and Politics.” Journal of Economic History 72, pp. 741–​770. Arruñada, B. 2003. “Property Enforcement as Organized Consent.” Journal of Law, Economics, and Organization 19, pp. 401–​444. Arruñada, B. 2012. Institutional Foundations of Impersonal Exchange: The Theory and Policy of Contractual Registries. Chicago: University of Chicago Press. Ayotte, K., Friedman, E., and Smith, H. E. 2013. “A Safety Valve Model of Equity as Anti-​ Opportunism.” Northwestern Law & Econ Research Paper No. 13–​ 15. Available at SSRN: http://​ssrn.com/​abstract=2245098 [Accessed 17 August 2016]. Ayres, I. 2005. Optional Law: The Structure of Legal Entitlements. Chicago: University of Chicago Press. Baldwin, C. Y. and Clark, K. B. 2000. Design Rules: The Power of Modularity. Vol. 1. Cambridge, MA: MIT Press. Barzel, Y. 1976. “An Alternative Approach to the Analysis of Taxation.” Journal of Political Economy 84, pp. 1177–​1197. Barzel, Y. 1982. “Measurement Cost and the Organization of Markets.” Journal of Law and Economics 25, pp. 27–​48. Barzel, Y. 1985. “Transaction Costs:  Are They Just Costs?” Journal of Institutional and Theoretical Economics 141, pp. 4–​16. Barzel, Y. 1997. Economic Analysis of Property Rights. 2nd ed. Cambridge:  Cambridge University Press.

 

ECONOMICS OF PROPERTY LAW    173 Barzel, Y. 2002. A Theory of the State: Economic Rights, Legal Rights, and the Scope of the State. Cambridge: Cambridge University Press. Berle, A. A. and Means, G. C. 1932. The Modern Corporation and Private Property. New York: Macmillan. Bernstein, L. 1996. “Merchant Law in a Merchant Court:  Rethinking the Code’s Search for Immanent Business Norms.” University of Pennsylvania Law Review 144, pp. 1765–​1821. Bertacchini, E., de Mot, J., and Depoorter, B. 2009. “Never Two Without Three: Commons, Anticommons and Semicommons.” Review of Law and Economics 5, pp. 163–​176. Buchanan, J. M. and Yoon, Y. J. 2000. “Symmetric Tragedies: Commons and Anti-​commons.” Journal of Law and Economics 43, pp. 1–​13. Calabresi, G., and Melamed, A. D. 1972. “Property Rules, Liability Rules, and Inalienability: One View of the Cathedral.” Harvard Law Review 85, pp. 1089–​1128. Cheung, S. N. S. 1970. “The Structure of a Contract and the Theory of a Non-​Exclusive Resource.” Journal of Law and Economics 13, pp. 49–​70. Coase, R. H. 1960. “The Problem of Social Cost.” Journal of Law and Economics 3, pp. 1–​44. Coase, R. H. 1988. The Firm, the Market, and the Law. Chicago: University of Chicago Press. Cooter, R. 1984. “Proces and Sanctions.” Columbia Law Review 84, pp. 1523–​1560. Craswell, R. and Calfee, J. E. 1986. “Deterrence and Uncertain Legal Standards.” Journal of Law, Economics, and Organization 2, pp. 279–​303. Dan-​Cohen, M. 1984. “Decision Rules and Conduct Rules: On Acoustic Separation in Criminal Law.” Harvard Law Review 97, pp. 625–​677. Demsetz, H. 1967. “Toward a Theory of Property Rights.” American Economic Review 57 (Papers & Proceedings), pp. 347–​359. Easterbrook F. H. and Fischel, D. R. 1991. The Economic Structure of Corporate Law. Cambridge, MA: Harvard University Press. Eggertsson, T. 1990. Economic Behavior and Institutions. Cambridge:  Cambridge University Press. Ellickson, R. C. 1989. “A Hypothesis of Wealth-​Maximizing Norms:  Evidence from the Whaling Industry.” Journal of Law, Economics, and Organization 5, pp. 83–​97. Ellickson, R. C. 1993. “Property in Land.” Yale Law Journal 102, pp. 1315–​1400. Epstein, R. A. 1975. “Unconscionability: A Critical Reappraisal.” Journal of Law and Economics 18, pp. 293–​315. Epstein, R. A. 1982. “Notice and Freedom of Contract in the Law of Servitudes.” Southern California Law Review 55, pp. 1353–​1368. Epstein, R. A. 1997. “A Clear View of The Cathedral: The Dominance of Property Rules.” Yale Law Journal 106, pp. 2091–​2190. Epstein, R. A. 2003. “In and Out of Public Solution: The Hidden Perils of Forced and Unforced Property Transfer.” In: T. Anderson and F. McChesney, eds., Property Rights: Cooperation, Conflict, and the Law. Princeton: Princeton University Press. 307–​342. Essert, C. 2013. “The Office of Ownership.” University of Toronto Law Journal 63, pp. 418–​461. Feldman, Y. and Smith, H. E. 2014. “Behavioral Equity.” Journal of Institutional and Theoretical Economics 170, pp. 137–​159. Fennell, L. A. 2004. “Common Interest Tragedies.” Northwestern University Law Review 98, pp. 907–​990. Fennell, L. A. 2011. “Commons, Anticommons, Semicommons,” In:  K. Ayotte and H. E. Smith, eds., Research Handbook on the Economics of Property Law. Cheltenham: Edward Elgar. 35–​56.

 

174   HENRY E. SMITH Fennell, L. A. 2012. “Lumpy Property.” University of Pennsylvania Law Review 160, pp. 1955–​1993. Field, B. C. 1989. “The Evolution of Property Rights.” Kyklos 42, pp. 319–​345. Friedman, D. 1994. “A Positive Account of Property Rights.” Social Philosophy and Policy 11(2), pp. 1–​16. Gergen, M. P., Golden, J. M., and Smith, H. E. 2012. “The Supreme Court’s Accidental Revolution? The Test for Permanent Injunctions.” Columbia Law Review 112, pp. 203–​249. Gold, A. and Smith., H. E. (2016). “Sizing Up Private Law.” Available at SSRN: http://​ssrn.com/​ abstract=2821354 [Accessed 25 August 2016]. Gordon, H. S. 1954. “The Economic Theory of a Common Property Resource: The Fishery.” Journal of Political Economy 62, pp. 124–​142. Grey, T. C. 1980. “The Disintegration of Property.” In: J. R. Pennock and J. W. Chapman, eds., NOMOS XXII: Property. New York: New York University Press. 69–​86. Grimmelmann, J. 2010. “The Internet Is a Semicommons.” Fordham Law Review 78, pp. 2799–​2842. Grossman, S. J. and Hart, O. D. 1986. “The Costs and Benefits of Ownership:  A  Theory of Vertical and Lateral Integration.” Journal of Political Economy 94, pp. 691–​7 19. Hansmann, H. and Kraakman, R. 2000. “The Essential Role of Organizational Law.” Yale Law Journal 110, pp. 387–​440. Hansmann, H. and Kraakman, R. 2002. “Property, Contract, and Verification: The Numerus Clausus Problem and the Divisibility of Rights.” Journal of Legal Studies 31, pp. S416–​S420. Hart, H. L.  A. 1982. Essays on Bentham:  Studies in Jurisprudence and Political Theory. Oxford: Clarendon Press. Hart, O. and Moore, J. 1990. “Property Rights and the Nature of the Firm.” Journal of Political Economy 98, pp. 1119–​1158. Hasen, R. L., and McAdams R. H. 1997. “The Surprisingly Complex Case Against Theft.” International Review of Law and Economics 17, pp. 367–​378. Heller, M. A. 1998. “The Tragedy of the Anticommons: Property in the Transition from Marx to Markets.” Harvard Law Review 111, pp. 621–​688. Heverly, R. A. 2003. “The Information Semicommons.” Berkeley Technology Law Journal 18, pp. 1127–​1189. Heylighen, F. 1999. “Advantages and Limitations of Formal Expression.” Foundations of Science 4, pp. 25–​56. Hohfeld, W. N. 1913. “Fundamental Legal Conceptions as Applied in Judicial Reasoning.” Yale Law Journal 23, pp. 16–​59. Hohfeld, W. N. 1917. “Fundamental Legal Conceptions as Applied in Judicial Reasoning.” Yale Law Journal 26, pp. 710–​770. Hume, D. 1896 [1739]. A Treatise of Human Nature. Ed. L. A. Selby-​ Bigge. Oxford: Clarendon Press. Hylton, K. N. 2011. “Property Rules and Defensive Conduct in Tort Law Theory.” Journal of Tort Law 4(1), art. 5. Jensen, M. C. and Meckling, W. H. 1976. “Theory of the Firm: Managerial Behavior, Agency Costs and Ownership Structure.” Journal of Financial Economics 3, pp. 305–​360. Kaplow, L. 1992. “Rules versus Standards:  An Economic Analysis.” Duke Law Journal 42, pp. 557–​629. Kaplow, L. and Shavell, S. 1996. “Property Rules versus Liability Rules: An Economic Analysis.” Harvard Law Review 109, pp. 713–​790.

 

ECONOMICS OF PROPERTY LAW    175 Katz, L. 2012. “Governing Through Owners: How and Why Formal Private Property Rights Enhance State Power.” University of Pennsylvania Law Review 160, pp. 2029–​2059. Kelly, D. B. 2011. “Strategic Spillovers.” Columbia Law Review 111, pp. 1641–​1721. Krier, J. E. 1992. “The Tragedy of the Commons, Part Two.” Harvard Journal of Law and Public Policy 15, pp. 325–​347. Lau, M. W. 2011. The Economic Structure of Trusts. Oxford: Oxford University Press. Lawsky, S. B. 2009. “Probably? Understanding Tax Law’s Uncertainty.” University of Pennsylvania Law Review 157, pp. 1017–​1074. Laycock, D. 2011. “The Neglected Defense of Undue Hardship (and the Doctrinal Train Wreck in Boomer v. Atlantic Cement).” Journal of Tort Law 4(3) art. 2. Lee, B. A. and Smith, H. E. 2012. “The Nature of Coasean Property.” International Review of Economics 59, pp. 145–​155. Leff, A. A. 1996. “Unconscionability and the Code—​The Emperor’s New Clause.” University of Pennsylvania Law Review 115, pp. 485–​559. Levmore, S. 2002. “Two Stories about the Evolution of Property Rights.” Journal of Legal Studies 31, pp. S421–​S451. Libecap, G. D. (1989). Contracting for Property Rights. Cambridge: Cambridge University Press. Libecap. G. D, and Lueck, D. 2011. “The Demarcation of Land and the Role of Coordinating Property Institutions.” Journal of Political Economy 119, pp. 426–​467. Libecap. G. D., Lueck, D., and O’Grady, T. 2011. “Large-​Scale Institutional Changes: Land Demarcation in the British Empire.” Journal of Law and Economics 54, pp. S295–​S327. Lueck, D. 1995. “The Rule of First Possession and the Design of the Law.” Journal of Law and Economics 38, pp. 393–​436. McCloskey, D. 1998. “The So-​ Called Coase Theorem.” Eastern Economic Journal 24, pp. 367–​371. Maskin, E. and Tirole, J. 1999. “Unforeseen Contingencies and Incomplete Contracts.” Review of Economic Studies 66, pp. 83–​114. Merrill, T. W. 2009. “Accession and Original Ownership.” Journal of Legal Analysis 1, pp. 459–​510. Merrill, T. W. 2011. “Private Property and Public Rights.” In: K. Ayotte and H. E. Smith, eds., Research Handbook on the Economics of Property Law. Cheltenham: Edward Elgar. 75–​103. Merrill, T. W., and Smith, H. E. 2000. “Optimal Standardization in the Law of Property: The Numerus Clausus Principle.” Yale Law Journal 110, pp. 1–​70. Merrill, T. W., and Smith, H. E. 2001a. “The Property/​Contract Interface.” Columbia Law Review 101, pp. 773–​852. Merrill, T. W., and Smith, H. E. 2001b. “What Happened to Property in Law and Economics?” Yale Law Journal 111, pp. 357–​398. Merrill, T. W., and Smith, H. E. 2007. “The Morality of Property.” William and Mary Law Review 48, pp. 1849–​1895. Merrill, T. W., and Smith, H. E. 2010. The Oxford Introductions to U.S. Law:  Property New York: Oxford University Press. Merrill, T. W., and Smith, H. E. 2011. “Making Coasean Property More Coasean.” Journal of Law and Economics 54, pp. S77–​S104. Merrill, T. W., and Smith, H. E. 2012. Property: Principles and Policies. 2nd ed. New York: Foundation Press. Morris, M. 1993. “The Structure of Entitlements.” Cornell Law Review 78, pp. 822–​898.

 

176   HENRY E. SMITH Newman, C. M. 2009. “Patent Infringement as Nuisance.” Catholic University Law Review 59, pp. 61–​123. Newman, M. E. J. 2006. “Modularity and Community Structure in Networks.” Proceedings of the National Academy of Sciences 103(23), pp. 8577–​8582. Newman, M. E. J. and Girvan, M. 2004. “Finding and Evaluating Community Structure in Networks.” Physical Review E 69, 026113 pp. 1–​15. Ortiz, D. R. 1994. “Neoactuarialism:  Comment on Kaplow (1).” Journal of Legal Studies 23, pp. 403–​409. Ostrom E. 1990. Governing the Commons: The Evolution of Institutions for Collective Action. Cambridge: Cambridge University Press. Parisi, F. 2000. “Spontaneous Emergence of Law: Customary Law.” In: Boudewijn Bouckaert and Gerrit de Geest eds., Encyclopedia of Law & Economics. Vol. 5. Cheltenham: Edward Elgar. 603–​630. Parisi, F., Schulz, N., and Depoorter, B. 2005. “Duality in Property:  Commons and Anticommons.” International Review of Law and Economics 25, pp. 578–​591. Penner, J. E. 1997. The Idea of Property in Law. Oxford: Clarendon Press. Posner, R. A. 2000. “Holmes, Savigny, and the Law and Economics of Possession.” Virginia Law Review 86, pp. 535–​567. Posner, R. A. and Parisi, F., eds. 2013. The Coase Theorem. Cheltenham: Edward Elgar. Rose, C. M. 1991. “Rethinking Environmental Controls: Management Strategies for Common Resources.” Duke Law Journal 1991, pp. 1–​38. Rose, C. M. 1997. “The Shadow of The Cathedral.” Yale Law Journal 106, pp. 2175–​2200. Sichelman, T. and Smith, H. E. (ms.). “Modeling Legal Modularity.” Simon, H. A. 1981. The Sciences of the Artificial. 2nd ed. Cambridge, MA: MIT Press. Sitkoff, R. H. 2011. “The Economic Structure of Fiduciary Law.” Boston University Law Review 91, pp. 1039–​1049. Smith, H. E. 2000a. “Semicommon Property Rights and Scattering in the Open Fields.” Journal of Legal Studies 29, pp. 131–​169. Smith, H. E. 2000b. “Ambiguous Quality Changes from Taxes and Legal Rules.” University of Chicago Law Review 67, pp. 647–​723. Smith, H. E. 2002. “Exclusion versus Governance:  Two Strategies for Delineating Property Rights.” Journal of Legal Studies 31, pp. S453–​S487. Smith, H. E. 2003. “The Language of Property: Form, Context, and Audience.” Stanford Law Review 55, pp. 1105–​1191. Smith, H. E. 2004a. “Exclusion and Property Rules in the Law of Nuisance.” Virginia Law Review 90, pp. 965–​1049. Smith, H. E. 2004b. “Property and Property Rules.” New  York University Law Review 79, pp. 1719–​1798. Smith, H. E. 2007. “Intellectual Property as Property: Delineating Entitlements in Information.” Yale Law Journal 116, pp. 1742–​1822. Smith, H. E. 2009. “Community and Custom in Property.” Theoretical Inquiries in Law 10, pp. 5–​41. Smith, H. E. 2010. “An Economic Analysis of Law versus Equity.” (Oct. 22) (unpublished manuscript). http://​www.law.yale.edu/​documents/​pdf/​LEO/​Hsmith_​LawVersusEquity7.pdf [Accessed 17 August 2016]. Smith, H. E. 2011. “Standardization in Property Law.” In:  K. Ayotte and H. E. Smith, eds., Research Handbook on the Economics of Property Law. Cheltenham: Edward Elgar. 148–​173.

 

ECONOMICS OF PROPERTY LAW    177 Smith, H. E. 2012. “Property as the Law of Things.” Harvard Law Review 125, pp. 1691–​1726. Smith, H. E. 2013. “Property as Platform:  Coordinating Standards for Technological Innovation.” Journal of Competition Law and Economics 9, pp. 1057–​1089. Smith, H. E. 2014a. “The Elements of Possession.” In: Y.-​C. Chang, ed., The Law and Economics of Possession. Cambridge: Cambridge University Press. 65–102. Smith, H. E. 2014b. “Why Fiduciary Law Is Equitable.” In: A. S. Gold and P. B. Miller, eds., Philosophical Foundations of Fiduciary Law. Oxford: Oxford University Press. 261–​284. Spulber, D. F. 2009a. “Discovering the Role of the Firm: The Separation Criterion and Corporate Law.” Berkeley Business Law Journal 6, pp. 298–​347. Spulber, D. F. 2009b. The Theory of the Firm: Microeconomics with Endogenous Entrepreneurs, Firms, Markets, and Organizations. Cambridge: Cambridge University Press. Sterk S. E. 2008. “Property Rules, Liability Rules, and Uncertainty about Property Rights.” Michigan Law Review 106, pp. 1285–​1335. Sugden, R. 2004 [1986]. The Economics of Rights, Co-​operation and Welfare. New York: Palgrave Macmillan. 2002. “Symposium, the Evolution of Property Rights.” Journal of Legal Studies 31, pp. S1–​S330. 2011. “Symposium, Property:  A  Bundle of Rights?” Econ Journal Watch 8(3). Available at https://​econjwatch.org/​issues/​volume-​8-​issue-​3-​september-​2011. Van Houweling, M. S. 2008. “The New Servitudes.” Georgetown Law Journal 96, pp. 885–​950. Warming, J. 1911. “Om ‘Grundrente’ af Fiskegrunde.” Nationaløkonomisk Tidsskrift 49: 495–​506. Translated in Andersen, P. 1983. “‘On Rent of Fishing Grounds’:  A  Translation of Jens Warming’s 1911 Article, with an Introduction.” History of Political Economy 15, pp. 391–​396. Weisbach, D. 1999. “Formalism in the Tax Law.” University of Chicago Law Review 66, pp. 860–​87. Williamson, O. E. 1985. The Economic Institutions of Capitalism. New York: Free Press. Williamson, O. E. 1993. “Opportunism and its Critics.” Managerial and Decision Economics 14, pp. 97–​107. Worthington, S. 1996. Proprietary Interests in Commercial Transactions. Oxford: Clarendon Press. Wyman, K. M. 2005. “From Fur to Fish: Reconsidering the Evolution of Private Property.” New York University Law Review 80, pp. 117–​240. Zimmermann, R. 1996. The Law of Obligations: Roman Foundations of the Civilian Tradition. Oxford: Oxford University Press.

 

Chapter 7

C OM MONS AND ANTIC OMMONS Michael Heller

This chapter offers a concise overview of anticommons theory. The anticommons thesis states that when too many people own pieces of one thing, nobody can use it. Usually private ownership creates wealth. But too much ownership can have the opposite effect—​it leads to wasteful underuse. This is a free-​market paradox that shows up all across the global economy. If too many owners control a single resource, cooperation breaks down, wealth disappears, and everybody loses. Conceptually, underuse in an anticommons mirrors the familiar problem of overuse in a “tragedy of the commons.” The field of anticommons studies is now well established, with thousands of scholars detailing examples across the innovation frontier, including drug patenting, telecom licensing, climate change, eminent domain, oil field unitization, music and art copyright, and postsocialist transition. Addressing anticommons tragedy is a key challenge for any legal system committed to innovation and economic growth.

7.1.  Three Motivating Examples Some years ago, a drug company executive presented me with an unsettling puzzle. His scientists had found a potential treatment for Alzheimer’s disease, but they couldn’t develop it for the market unless the company bought access to dozens of patents. Any single patent owner could demand a huge payoff; some blocked the whole deal. This story does not have a happy ending. The drug sits on the shelf though it might have saved millions of lives and earned billions of dollars.1

1 

The fullest account of anticommons theory and solutions appears in Heller (2008). On the anticommons in drug patents, see (49–​78); in telecom, see (79–​106); in land, see (107–​142). See also

 

COMMONS AND ANTICOMMONS    179 Here’s a second high-​stakes puzzle: what’s the most underused natural resource in America? The answer may be a surprise. It’s the airwaves. Over 90% is dead air because ownership of the broadcast spectrum is so fragmented. As a result, America’s information economy is relatively hobbled; wireless broadband capacity lags far behind that in Japan and Korea. The cost of spectrum underuse may be in the trillions. And another puzzle: why do we waste weeks of our lives stuck in airports? Bad law for real estate assembly. In America, air travel was deregulated 35 years ago. The number of fliers has tripled. So how many new airports have been built since 1975? One. Denver. You can’t build new airports, not anywhere, because multiple landowners can block every project. Twenty-​ five new runways at America’s busiest airports would end most routine air travel delays in the country. Imagine that. All these puzzles share a common cause: when too many people own pieces of one thing, nobody can use it. The anticommons thesis is that simple: when too many people own pieces of one thing, nobody can use it. There has been an unnoticed revolution in how we create wealth. In the old economy, twenty or thirty years ago, you invented a product and got a patent; you wrote a song and got a copyright; you subdivided land and built houses. Today, the leading edge of wealth creation requires assembly. From drugs to telecom, software to semiconductors, anything high-​tech demands the assembly of innumerable patents. And it’s not just high tech that’s changed—​today, cutting-​edge art and music are about mashing up and remixing many separately owned bits of culture. Even with land, the most socially important projects, like new runways, require assembling multiple parcels. Innovation has moved on, but we are stuck with old-​style ownership that’s easy to fragment and hard to put together. Fixing anticommons tragedy is a difficult puzzle. Some solutions are entrepreneurial; for example, people can profit from finding creative ways to bundle ownership. Philanthropists can assemble patents for disease cures. Political advocacy and legal reform will be needed to secure solutions. But the first and most important step in solving anticommons tragedy is to name it and make it visible. This chapter takes that step by briefly introducing the anticommons lexicon. With the right language, anyone can spot links among anticommons puzzles, and all can come together to fix them.

7.2.  Commons and Anticommons To understand the dilemma of resource underuse in an anticommons, it is helpful to start with overuse in a commons. Aristotle was among the first to note how shared ownership can lead to overuse: “That which is common to the greatest number has the least

Heller (2011) (collecting and reprinting, in two volumes, the key scholarly articles on the theory and economics of commons and anticommons property).

 

180   MICHAEL HELLER care bestowed upon it … each thinks chiefly of his own, hardly at all of the common interest; and only when he is himself concerned as an individual” (1996, 33).2 Why do people overuse and destroy things that they value? Perhaps they are shortsighted or dim-​witted, in which case reasoned discussion or gentle persuasion may help. But even the clear-​headed can overuse a commons, for good reasons. The most intractable overuse tragedy arises when individuals choose rationally to consume a common pool of scarce resources even though each knows that the sum of these decisions destroys the resource for all. In such settings, reason cuts the wrong way, and gentle persuasion is ineffective. In other words, I do what’s best for me, you do what’s best for you, and no one pays heed to the sustainability of the shared resource. Ecologist Garrett Hardin captured this dynamic well when he coined the phrase tragedy of the commons (1968, 1243–​1244).3 In 1968 he wrote, “Ruin is the destination toward which all men rush, each pursuing his own best interest in a society that believes in the freedom of the commons. Freedom in a commons brings ruin to all” (1244). Since Hardin wrote these lines, thousands have identified additional areas susceptible to overuse and commons tragedy.4 In addition, Hardin’s metaphor inspired a search for solutions. Most solutions revolve around two main approaches: regulation or privatization. Suppose a common lake is being overfished. Regulators can step in and decide who can fish, when, how much, and with what methods. Such direct “command-​and-​control” regulation has dropped from favor, however, partly because it fails so often and partly because of disenchantment with socialist-​type regulatory control. These days, regulators are more likely to look for some way to privatize access to the lake. They know that divvying up ownership can create powerful personal incentives to conserve. Harvest too many fish in your own lake today, starve tomorrow; invest wisely in the lake, profit forever. Extrapolating from such experience, legislators and voters reason—​wrongly—​that if some private property is a good thing, more must be better. In this view, privatization can never go too far. Until now, ownership, competition, and markets—​the guts of modern capitalism—​ have been understood through the opposition suggested by Figure 7.1. Private property solves the tragedy of the commons. Privatization beats regulation. Market competition outperforms state control. Capitalism trounces socialism. But these simple oppositions mistake the visible forms of ownership for the whole spectrum. The assumption is fatally incomplete. 2  Before Aristotle, Thucydides noted that people “devote a very small fraction of time to the consideration of any public object, most of it to the prosecution of their own objects. Meanwhile each fancies that no harm will come to his neglect, that it is the business of somebody else to look after this or that for him; and so, by the same notion being entertained by all separately, the common cause imperceptibly decays” (1910, bk. 1, sec. 141). 3  The power of Hardin’s rhetoric sometimes exceeded the reach of his data. For example, Hardin’s work overlooks the important distinction between “open access” and “limited access commons.” On the implications of this distinction for common pool resource dilemmas, see note 10 and accompanying text. 4  See, e.g., The Digital Library of the Commons (2012).

 

COMMONS AND ANTICOMMONS    181

Commons Property

Private Property

Figure 7.1  The standard solution to commons tragedy.

Commons Property

Private Property

Anticommons Property

Figure 7.2  Revealing the hidden half of the ownership spectrum.

Privatizing a commons may cure the tragedy of wasteful overuse, but it may inadvertently spark the opposite. English lacks a term to denote wasteful underuse. To describe this type of fragmentation, I coined the phrase tragedy of the anticommons.5 The term covers any setting in which too many people can block each other from creating or using a scarce resource. Rightly understood, the opposite of overuse in a commons is underuse in an anticommons. This concept makes visible the hidden half of our ownership spectrum, a world of social relations as complex and extensive as any we have previously known (see Figure 7.2). Beyond normal private property lies anticommons ownership. As one commentator notes, “To simplify a little, the tragedy of the commons tells us why things are likely to fall apart, and the tragedy of the anticommons helps explain why it is often so hard to get them back together” (Fennell 2004). Making anticommons ownership visible is not easy. Let me give you an image that I have found helpful in crystallizing the abstract idea. One thousand years ago, the Rhine was one of the world’s great trade routes. Boatmen traded under the protection of the Holy Roman Emperor. When the Emperor weakened in the thirteenth century, freelance German barons built castles and began collecting their own illegal tolls. The growing gauntlet of “robber baron” castles, over 200 at one point, make for a wonderful bicycling holiday today, but made shipping impracticable back then. The river continued to flow, but for 500 years, boatmen would not bother making the journey. As one boatman’s plaintive song went: The Rhine can count more tolls than miles And knight and priestling grind us down The tollman’s heavy hand falls first, Behind him stands the greedy line Master of tolls, assayer, scribe Four man deep they tap the wine. (Chamberlain 1929) 5  Heller (1998). In turn, my term builds on earlier conceptual work by Frank Michelman. See ibid 667–​669, discussing evolution of the concept; see also Michelman (1982), discussing what he calls the “regulatory regime.”

 

182   MICHAEL HELLER Everyone suffered—​even the robber barons. The European economic pie shrank. Wealth disappeared. Too many tolls meant too little trade. To understand anticommons tragedy, just update this image. Phantom tollbooths today emerge whenever ownership first arises and property is being created all the time in ways many of us do not realize. Today’s robber barons are often private owners and public regulators, all the people holding vetoes on the path to innovation. Today’s missing river trade takes the form of crushed entrepreneurial energy and forgone investment. When too many public decision makers or private owners can block access to a resource, they harm us all. Often, we think that governments need only to create clear property rights and then get out of the way. So long as rights are clear, owners can trade in markets, move resources to higher valued uses, and generate wealth. But clear rights and ordinary markets are not enough. The anticommons perspective shows that the content of property rights matters as much as the clarity. Wasteful underuse can arise when ownership rights and regulatory controls are too fragmented. Making the tragedy of the anticommons visible upends our intuitions about private property. Private property no longer can be seen as the end point of ownership. Privatization can go too far, to the point where it destroys rather than creates wealth. Too many owners paralyze markets because everyone blocks everyone else. Well-​ functioning private property is a fragile balance poised between the extremes of overuse and underuse.

7.3.  The Magical Parking Lot So far, I’ve introduced the nutshell version of the commons and anticommons. To understand the concepts more fully, imagine you’ve discovered an empty paved lot near Leicester Square in London. At first, the parking paradise is free and open to all. No one tickets or tows. You park and go to the theater. No problem. Later, you tell friends, who park there too. No problem. But then others notice, and soon the lot is jammed. Cars are blocked in. Doors are dinged. Fights break out. The lot becomes a scary place. You pay to park elsewhere. This overused lot is an example of a tragedy of the commons. It’s a tragedy because every parker is acting reasonably, but their individual actions quickly sum to collective disaster. Similarly, if a single shepherd has access to a field, the result is well-​trimmed grass and fat sheep. But open the field to all shepherds, each of whom may add sheep without regard to the others, and soon there may be nothing left but bare dirt and hungry animals. Overuse tragedies are everywhere: species extinction, ozone depletion, and highway congestion. After Garrett Hardin popularized the “tragedy of the commons” metaphor in 1968, people gained a new language for a phenomenon that was widely experienced,

 

COMMONS AND ANTICOMMONS    183 but had been difficult to name. The concept helped people give voice to then-​emerging concerns about environmental degradation. Metaphors can be powerful. The tragedy of the commons concept revealed hidden links among innumerable resource dilemmas, large and small. Spotting this shared structure helped people identify shared solutions. For example, the International Association for the Study of the Commons brings together a global network of scholars, policymakers, and practitioners, while the Digital Library of the Commons hosts an online database that cites about fifty thousand articles related to the commons.6 How do we solve such tragedies? There are three distinct approaches: privatization and markets, cooperative engagement, and political advocacy and regulation. Bear in mind that each solution has an analogue on the anticommons side of the property spectrum. Private property and market transactions can solve overuse tragedy. Recall that in the parking example, you were the first to discover the empty lot. You might claim ownership for yourself based on your original discovery and first possession. Being first is a standard (but not necessarily fair or efficient) way to hand out rights in resources. Another path to private ownership passes through state control. The state might reject your claim of original discovery and instead appropriate the lot and auction it to the highest bidder or transfer it quietly to a crony. However the lot arrives in private hands, it will likely be managed better than if it had remained open to all. Owners can profit if they spruce up the lot, repave it perhaps, paint lines, and keep it clean, safe, and well used. As a parker in a private property regime, you lose the freedom of the commons but gain order and access. The moral justifications for private ownership are controversial for philosophers, but as a practical matter, moving to private property often does prevent overuse in a commons. Harold Demsetz, author of the leading economic theory of ownership, argues that this “conservation effect” is the main reason private property emerges in, and provides a benefit to, society (1967, 354–​359). Because of our private-​property focus, we tend to overlook cooperative solutions to overuse dilemmas. Cooperative solutions are often small-​scale, context specific, local, and not reliant on legal structures—​thus invisible. In the case of our magical parking lot, notes under windshields, gossip on the street, and other neighborly devices can coordinate the parkers. Parkers may figure out how to keep the parking lot running smoothly without state coercion or private ownership. In Governing the Commons, Ostrom demonstrated that close-​knit communities around the world have succeeded in managing group property without tragic outcomes.7 There are thousands of stories of successful cooperation that preserve contested resources and promote overall social welfare. 6 

On the IASC, see http://​www.iascp.org (visited 2 October 2012). On the Digital Library of the Commons, see note 5. 7  See generally E. Ostrom (1990, 1–​28), setting out the theoretical framework, and (1999) discussing solutions.

 

184   MICHAEL HELLER Finally, state coercion can solve overuse. Cooperative mechanisms may break down if there are too many newcomers coming and going, if people don’t really know each other, or if it is otherwise hard to discipline deviants. Then, parkers may move from polite notes under windshields to breaking antennae, purposely scratching cars, slashing tires, and fistfights. The state might assert ownership over the lot, put up a gate, and hand out or sell parking permits. But bureaucracy is costly and often capricious. Political pressure may lead to bizarre uses of the lot. States are rarely nimble or efficient parking lot operators. Public ownership and management can eliminate the tragedy of the parking lot commons, but they may create new costs and inconveniences for the parkers. Privatizing a commons may cure the tragedy of wasteful overuse and lead to orderly parking; but it also may inadvertently spark the opposite, a lot that no one can use. The tragedy of the anticommons describes this problem of wasteful underuse. Though the anticommons concept refers at its core to fragmented ownership, the idea extends to fragmented decision-​making more generally. Resource use often depends on the outcome of some regulatory process. If the regulatory drama involves too many uncoordinated actors—​neighbors and advocacy groups; local, state, and federal legislators; agencies and courts—​the sheer multiplicity of players may block use of the underlying resource. How could the parking lot become an anticommons? Recall that underuse in an anticommons is the mirror image of overuse in a commons. Much can go wrong when politicians privatize state-​owned resources, when resources are owned for the first time, or when owners divvy up property later on. For example, in privatizing the lot, politicians might not want to annoy parkers who are also voters. So they might give free parking spots based on every parker’s previous use of the lot. (This is approximately how US regulators have allocated ocean fishing quotas and tradable pollution permits.) If there were thousands of parkers, but say one hundred spots, dozens might have to share each spot. Assembling the fractional shares back into a usable parking lot would require too many deals. Even if each parker behaved reasonably, bargaining is costly. And many of us are not reasonable, especially at seven o’clock in Leicester Square when shows are about to start. So the “privatized” lot may sit empty and unused—​an anticommons. Now substitute sheep in a meadow for the parkers in the lot. If a common field were privatized down to the square inch, no shepherd would be able to graze a single sheep. The same might happen if innumerable heirs separately owned scattered strips of an ancestor’s farm. In an anticommons, the grass may be lush and tall and unused; in a commons, it may be picked bare. In both cases, the pasturage can be wasted and the sheep starve. The parking lot and shepherd’s field show that creating private property can solve the problem of overuse in a commons. But privatization can go too far. When it does, we can tip into an anticommons, and again everyone loses. Adding the concepts of underuse and anticommons makes visible a new frontier for private bargaining, political debate, and wealth creation. Our goal should be to find the sweet spot for property rights, between commons and anticommons.

 

COMMONS AND ANTICOMMONS    185

7.4.  The New World of Use My tales of the magical parking lot are a bit of a sleight of hand. They give a succinct overview of overuse and underuse, commons and anticommons. But underuse and anticommons are still squiggly—​until recently, my Word spellchecker rejected them by underlining each with red squiggles, and instead suggested undersea and anticommunist as replacements. These squiggles are a signal: the nonexistence of a word can be as telling as its presence. When we lack a term to describe some social condition, it is because the condition does not exist in most people’s minds. So, it should be no surprise that we have overlooked the hidden costs of anticommons ownership. We cannot easily fix the problem until we have created a shared lexicon to spot tragedies of the anticommons. Besides highlighting the language problem, the squiggles prompted me to look around the Internet at overuse and underuse. Googling overuse yielded about ten million hits in late 2012, while underuse generated under one million. (Commons had 800 million hits and anticommons had 25,000.) To me, the data immediately suggest two possibilities: either overuse is about ten times more important a social problem than underuse is, or we are only about 10% as aware of underuse as we should be. You will not be surprised that I believe the latter to be correct. To understand the Google results, start with overuse. According to the Oxford English Dictionary, overuse entered the language as a verb in the early 1600s. One of the earliest usages is as apt today as it was centuries ago: “When ever we overuse any lower good we abuse it.” By 1862, the noun form was well recognized: “The oyster beds are becoming impoverished, partly by over-​use.”8 Overuse continues to mean “to use too much” and “to injure by excessive force,” definitions that have been stable for hundreds of years. Many of Google’s top links for overuse come from medicine. Doctors diagnose “overuse syndrome” and dozens of “overuse injuries”—​injuries from too much tennis, running, violin playing, book reading, whatever. So what is the opposite of overuse? Ordinary use. The opposite of injuring yourself through too much use and excessive force is staying injury free by using an ordinary amount of force. Instead of abandoning an activity, do it in a reasonable, sustainable way. In medicine as in everyday language, the opposite of overuse is ordinary use (Figure 7.3). Since the 1600s, overuse and ordinary use have been an either–or proposition. Either you will feel pain in your elbow or you will be able to play happily, if not well. When you overuse a resource, bad things happen. It is much better to engage in ordinary use. How do we achieve ordinary use? Recall the problem of the magical parking lot. The usual solutions to tragedies of the commons are, as I’ve mentioned, privatization, cooperation, and regulation. These three solutions map onto the traditional view that 8 

OED, “overuse,” v., http://​www.oed.com/​view/​Entry/​135291 (visited 2 October 2012).

 

186   MICHAEL HELLER Overuse

Ordinary Use

Figure 7.3  Ordinary use as the end point.

State

Commons

Private

Figure 7.4  The trilogy of ownership.

ownership can be organized into three basic types of property: private, commons, and state (Figure 7.4).9 We all have strong intuitions about private property, but the term is surprisingly hard to pin down. A good starting point is William Blackstone, the foundational eighteenth-​ century British legal theorist. His oft-​quoted definition of private property is “that sole and despotic dominion which one man claims and exercises over the external things of the world, in total exclusion of the right of any other individual in the universe” (Blackstone 1765–​1769). In this view, private property is about an individual decision maker who directs resource use. Commons property refers to shared resources, resources for which there is no single decision maker. In turn, the commons can be divided into two distinct categories. The first is open access, a regime in which no one at all can be excluded, like anarchy in the parking lot or on the high seas. Mistakenly, the legal and the economics literature have long conflated the commons with open access, hence reinforcing the link between commons and tragedy. The second type of commons has many names, but for now let’s call it group access, a regime in which a limited number of commoners can exclude outsiders but not each other. If the ocean is open access, then a small pond surrounded by a handful of landowners may be group access (or consider the shared mews behind houses in Notting Hill, London’s keyholder-​only squares, and New York City’s Gramercy Park). Group access is often overlooked even though it is the predominant form of commons ownership and often not tragic at all.10 State property resembles private property in that there is a single decision maker, but it differs in that resource use is directed through some process that is, in principle, responsive to the needs of the public as a whole. In recent years, state property has 9 

On the property trilogy, see Heller (2001, 82–​92). On open access versus group access property, see Eggertsson (2003, 74–​85). I advocate that we use the term liberal commons to describe many forms of legally sanctioned group ownership. See generally Dagan and Heller (2001). 10 

 

COMMONS AND ANTICOMMONS    187 Commons

Private

Figure 7.5  The familiar split in ownership.

become less central as a theoretical category: the Cold War is over; most socialist states have disappeared; intense state regulation of resources has dropped from favor; and privatization has accelerated. Today, for many observers, the property trilogy can be reduced to an opposition of private and commons property, what one scholar calls simply “all and none” (Figure 7.5) (Barzel 1989, 71). I believe a substantial cause of our cultural blindness to the costs of fragmented ownership arises from the dominance of this too simple image of property. Note how the commons–​private opposition tracks the overuse–​ordinary use opposition. The former implies that there is nothing beyond private property; the latter suggests that we cannot overshoot ordinary use. Together, these oppositions reinforce the political and economic logic of the global push toward privatization. We assume, without reflection, that the solution to overuse in a commons is ordinary use in private ownership. This logic makes it difficult to imagine underuse dilemmas and impossible to see the uncharted world beyond private property. According to the Oxford English Dictionary (OED), underuse is a recent coinage. In its first recorded appearance, in 1960, the word was hedged about with an anxious hyphen and scare quotes: “There might, in some places, be a considerable ‘under-​use’ of [parking] meters.” By 1970, copy editors felt sufficiently comfortable to cast aside the quotes: “A country can never recover by persistently under-​using its resources, as Britain has done for too long.” The hyphen began to disappear around 1975.11 In the OED, this new word means “to use something below the optimum” and “insufficient use.” The reference to an “optimum” suggests to me how underuse entered English. It was, I think, an unintended consequence of the increasing role of cost–​benefit analysis in public policy debates. What happens when we slot underuse into the opposition in Figure 3? Although the result seems simple, it leads to conceptual turmoil (Figure 7.6). In the old world of overuse versus ordinary use, our choices were binary and clear-​ cut: injury or health, waste or efficiency, bad or good. In the new world, we are looking for something subtler—​an “optimum” along a continuum. Looking for an optimum level of use has a surprising twist: it requires a concept of underuse and surreptitiously changes the long-​standing meaning of overuse. Like Goldilocks, we are looking for something not too hot, not too cold, not too much or too little—​just right. Figure 7.7 suggests how underuse changes our quest. How can we know whether we are underusing, overusing, or optimally using resources? It’s not easy, and not just a matter for economic analysis. Searching for an optimum between overuse and underuse sets us on the contested path of modern regulation of risk, an inquiry that starts with economic analysis but quickly implicates our 11 

OED, “under-​use, n.,” http://​www.oed.com/​view/​Entry/​212195 (visited 2 October 2012).

 

188   MICHAEL HELLER Overuse

Underuse

Figure 7.6  The new spectrum of use.

Overuse

Optimal Use

Underuse

Figure 7.7  Goldilocks’ quest for the optimum.

core beliefs. We have to put dollar values on human lives and on the costs of overuse and underuse behaviors—​a process filled with moral and political dilemmas. I note this difficult topic to show that finding the optimum requires the idea of underuse and that this new word in turn transforms the meaning of overuse. Overuse no longer just means using a resource more than an ordinary amount. The possibility of underuse reorients policy making from relatively simple either–​or choices to the more contentious trade-​ offs that make up modern regulation of risk.

7.5.  The Tragedy of the Anticommons Adding the idea of “underuse” sets the stage for the anticommons. Looking back at Figures 7.3–​7.7, there is a gap in our labeling scheme. We have seen the complete spectrum of use, but not the analogous spectrum of ownership. What form of ownership typically coincides with squiggly underuse? The force of symmetry helped reveal a hidden property form. Figure 7.8 shows my path to the anticommons. I coined the term tragedy of the anticommons to help make visible the dilemma of too fragmented ownership beyond private property. Just as the idea of “underuse” transforms the continuum of resource use, “anticommons” transforms the continuum of ownership. It shows that the move from commons to private can overshoot the mark (Figure 7.9). When privatization goes too far, resources can end up wasted in an unfamiliar way. Seeing the full spectrum of ownership has another benefit. Our understanding of commons ownership may help inform solutions to anticommons tragedy. To start, consider the distinction between open access (anarchy open to all) and group access (property that is commons to insiders and private to outsiders). This distinction can do some work on the anticommons side of the spectrum as well. The conventional wisdom has often overlooked group access, but we don’t have to. Under the right conditions, groups of people succeed at conserving a commons resource without regulation or privatization (Ellickson 1991, 167–​183). Cooperation can get us to optimal use. Under what conditions does cooperation work, and what does that teach us about fixing underuse dilemmas?

 

COMMONS AND ANTICOMMONS    189 Overuse Commons

Optimal Use Private

Underuse ???

Private

Anticommons

Figure 7.8  An ownership puzzle.

Commons

Figure 7.9  The full spectrum of ownership.

At the extreme of open access, group norms don’t stick. For example, anyone can fish for tuna on the high seas. Tuna fleets work in relative isolation, and their catches can be sold anonymously to diverse buyers. Conservation norms, such as voluntary limits on fishing seasons, may gain little traction. Gossip and other low-​cost forms of policing don’t work for wide-​ranging international fleets. Unless states intervene, overuse is hard to avoid. Whales were saved from extinction more through naval powers enforcing international treaties than through gossip at the harbor bar. The state can sponsor hybrid solutions. What if the state asserted ownership over lobsters and fish, and then created private rights (such as licenses and tradable quotas) that complement cooperative solutions? Often, such hybrid regimes lead to fairer and higher-​yielding results than informal group access can achieve. For example, in Australia, the government issues licenses for a sustainable number of lobster traps and enforces strict harvesting limits. Lobstermen can wait to harvest until the lobsters mature, or they can sell their government-​created rights, secure in their markets and property. With far less fishing effort, this system yields more and bigger lobsters than US lobstermen catch either in coastal harbor gangs or on the open ocean (see, generally, Tierney 2000; Acheson 1988). Hybrid systems are the cutting edge of natural resource management:  examples include not only tradable fishing quotas but also carbon-​emission markets and transferable air-​pollution permits (Rose 1999). These solutions can work far beyond lobsters and tuna, even beyond natural resources generally. They may reach the edge of high-​ tech innovation. Solutions to commons property dilemmas give clues to solving anticommons tragedy. For open access, like the high seas, states must command resource use directly or create hybrid rights, such as fishing quotas. The anticommons parallel to open access is full exclusion in which an unlimited number of people may block each other. With full exclusion, states must expropriate fragmented rights or create hybrid property regimes so people can bundle their ownership. Otherwise, the resource will be wasted through underuse. There is, however, one important respect in which full exclusion differs from open access: an anticommons is often invisible. You have to spot the underused resource before you can respond to the dilemma. Group access in a commons also has an anticommons parallel: group exclusion in which a limited number of owners can block each other. Recall the multiple owners

 

190   MICHAEL HELLER Zone of Cooperative and Market-Based Solutions

Open Access

Group Access

Private Property

Group Exclusion

Full Exclusion

Figure 7.10  The full spectrum of property, revealed. I develop an early version of this spectrum in Heller (1999, 1194–​1198).

of our magical parking lot. For both group access and group exclusion, the full array of market-​ based, cooperative, and regulatory solutions is available. Although self-​ regulation may be more complex for anticommons resources (Depoorter and Vanneste 2007), close-​knit fragment owners can sometimes organize to overcome anticommons tragedy. For group exclusion resources, the regulatory focus should be support for markets to assemble ownership and removal of roadblocks to cooperation. Group property on the commons or anticommons side of private ownership is exponentially more important than the rare extremes of open access or full exclusion. Much of the modern economy—​corporations, partnerships, trusts, condominiums, and even marriages—​can be understood as legally structured group property forms for resolving access and exclusion dilemmas (see Dagan and Heller 552–​554; Dagan and Frantz 2004). We live or die depending on how we manage group ownership. Now, we can see the full spectrum of property, as shown in Figure 7.10.

7.6.  The Economics of the Anticommons After I proposed the possibility of anticommons tragedy, economist and Nobel laureate James Buchanan and his colleague Yong Yoon undertook to create a formal economic model. They wrote that the anticommons concept helps explain “how and why potential economic value may disappear into the ‘black hole’ of resource underutilization” (2000, 2). According to their model, society gets the highest total value from a resource—​say, the magical parking lot—​when a single decision maker controls its use. As more people can use the lot independently, the value goes down—​a tragedy of the commons. And as more people can block each other from the lot, the value also goes down symmetrically—​a tragedy of the anticommons. Figure 7.11 shows their graphic summarizing this finding. After developing their proof and showing how the anticommons construct may apply to a wide range of problems, Buchanan and Yoon concluded, “the anticommons construction offers an analytical means of isolating a central feature of sometimes disparate institutional structures…. [People] have perhaps concentrated too much attention

 

COMMONS AND ANTICOMMONS    191 Total Value

1

# of Excluders

# of Users

Figure 7.11  Value symmetry in an anticommons and a commons. See Heller (1999, 8). A HERE

B C

THERE

D HERE

E

F THERE

Figure 7.12  Substitutes versus complements.

on the commons side of the ledger to the relative neglect of the anticommons side” (2000, 12). In recent years, economic modeling of the anticommons, including game theory approaches has become quite sophisticated. See Parisi (2002); Parisi, Schulz, and Depoorter (2003); see, also, Fennell (2004), arguing that tragedies of the commons are best modeled as prisoner’s dilemma games and anticommons as chicken games. At the simplest level, anticommons theory can be understood as a legal twist on the economics of “complements” first described by Antoine-​Augustin Cournot in his 1838 Researches into the Mathematical Principles of the Theory of Wealth (and discovered independently by Charles Ellet in 1839 in his work on railway tariffs). Anticommons theory is a partial corrective for modern economic models that focus on “substitutes” and often neglect the role of “complements.”12 What’s the difference? In Figure 7.12, Railways A, B, and C are substitute ways to get from here to there. Say the fare is 9. If railway A finds a way to provide service for 8, it will win riders. B and C must become more efficient to keep up. In markets with robust substitutes, competitors have incentives to innovate, lower prices, and thereby indirectly benefit society as a whole. By contrast, Railways D, E, and F are complements. When inputs are complementary, generally you want all or none of them. Again, assume the fare from here to there is 9. D, E, and F each charge 3. Railway D knows that if you want to ride, you must buy its ticket. So why innovate? Instead, D may raise its 12  On the problem of complements in an information economy, see Varian, Farrell, and Shapiro (2005, 43–​45). On the interaction of substitutes and complements in the anticommons context, see Dari-​ Mattiacci and Parisi (2006).

 

192   MICHAEL HELLER fare to 5, hoping that E and F lower theirs to 2 each. But why would E and F go along? More likely, they too would raise fares, so the total exceeds 9, and ridership falls below the optimal level. With complementary competition, incentives to innovate are blunted: if D did lower fares, then E and F just might raise theirs. It’s the same problem if D, E, and F are complementary patents instead of railways. Then innovators face what economist Carl Shapiro calls a “patent thicket,” many phantom tollbooths on the route to commercializing new technology.13 Cournot proved that in markets dominated by complements—​whether railways or patents—​we can get higher overall social welfare if D, E, and F merge. Here, monopoly trumps competition. Anticommons theory, in turn, moves from railways and patents to ownership and regulation generally. All these concepts describe facets of the same dilemma: too many uncoordinated owners or regulators blocking optimal use of a single resource.

7.7.  Anticommons Puzzles Our Rhine boaters from a little earlier may seem an esoteric example, but there are a near-​infinity of everyday puzzles that share this common structure—​one whose solution could jump-​start innovation, release trillions in productivity, and help revive the global economy. Let me return to the drug patent example that opens this article. The Alzheimer’s drug that never came to market is not alone. Numerous potential drugs may be lost to anticommons ownership. In the past 30 years, drug R&D has been going steadily up, but discoveries of major new classes of drugs have been declining. Instead, drug companies focus on minor spinoffs of existing drugs for which they have already assembled the necessary property rights. How did this new drug discovery gap happen? Patent anticommons. Paradoxically, more biotech patent owners can mean fewer life-​saving innovations. Drugs that should exist, could exist, are not being created. To date, the most debated application of anticommons theory has built on the drug patent example. The field was sparked in part by my article with Rebecca Eisenberg (1998) on the anticommons in biomedical research, an article that has since been cited several thousand times.14 There has been a flurry of follow-​on papers, studies, and reports, many of which conclude that patents should be harder to obtain, in part to avoid potential anticommons tragedy effects.15 The 2011 reforms to the Patent Act, a string of

13  See Shapiro (2001, 119); cf. Burk and Lemley (2009, 75–​78, 86–​92), distinguishing anticommons and patent thickets in the patent context. 14  See, generally, Heller (2008, 49–​78), reviewing 10 years of scholarship and debate on anticommons effects in drug development. 15  ibid 65, discussing three influential policy-​oriented reports that argue for patent reform based on anticommons concerns.

 

COMMONS AND ANTICOMMONS    193 Supreme Court patent decisions, and a number of pending bills before Congress all aim to ameliorate potential anticommons tragedies.16 In their 2009 book, The Patent Crisis, Dan Burk and Mark Lemley review recent literature on patents and biotech innovation and conclude that, “the structure of the biotechnology industry seems likely to run high anticommons risks,” particularly when companies are attempting to bring products to market (see Burk and Lemley 2009, 89). However, the empirical basis for finding that anticommons effects are stifling innovation remains inconclusive. Scholars such as Jonathan Barnett are skeptical.17 This remains an important area for empirical work.18 In addition, the anticommons framework has been used to analyze ownership across the high-​tech frontier, ranging from Thomas Hazlett’s work on broadcast spectrum ownership, what he calls the “tragedy of the telecommons,” to Rosemarie Ham Ziedonis’s study of anticommons effects in technology patenting.19 In my Gridlock Economy book, I show that it is not just biotech that is susceptible to anticommons tragedy. Cutting-​ edge art and music are about mashing up and remixing many separately owned bits of culture.20 And as I earlier said, even with land the most socially important projects often require assembling multiple parcels.21 Anticommons theory is now well established, but empirical studies have yet to catch up. How hard is it to negotiate around ownership fragmentation? How much does ownership fragmentation slow down technological innovation? Does the effect vary by industry?22 It is difficult to measure discoveries that should have been made but weren’t, industries that could exist but don’t. We are just starting to examine these conundrums. In 2006, a team of law, economics, and psychology researchers reported experimental findings that rejected the presumed symmetry of commons and anticommons. They found that anticommons dilemmas “seem to elicit more individualistic behavior than commons dilemmas” and are “more prone to underuse than commons dilemmas are to overuse.” The researchers conclude that “if commons leads to ‘tragedy,’ anticommons may well lead to ‘disaster’ ” (Vanneste et al. 2006).23 16 

See Barnett (2015, 3), collecting sources. ibid. See also Burk and Lemley (“there is little concrete evidence that intensive levels of IP acquisition and enforcement restrain innovation or output”) (2009, 5). 18  Several survey-​based studies have questioned whether anticommons tragedy is blocking academic biomedical research (see, e.g., Walsh, Arora, and Cohen 2004, 285–​340, 324). 19   On the anticommons in the telecom context, see Heller (2008, 79–​106). See also Hazlett (2005); Ziedonis (2004). 20  On the anticommons in the copyright context, see Heller, (2008, 9–​16), discussing tragedy in filmmaking, art, history, and music; 27–​30, discussing the anticommons in online search, such as Google Books. See also Parisi and Depoorter (2003). 21  On the anticommons in land, see Heller (2008, 107–​142), detailing anticommons tragedies and solutions in land resources. 22  The best recent collection of empirical sources addressing these issues is Barnett (2015, 16–​57) detailing numerous mechanisms, such as standard setting organizations, patent pooling, and market-​ making intermediaries, that overcome potential anticommons tragedies in various industries. 23  Follow-​up studies looked at why negotiations fail when presented in anticommons form. They found more failure as the number and complementarity of fragment owners increases. Also, as 17 

 

194   MICHAEL HELLER These preliminary findings of bargaining failure around anticommons ownership may help provide some insight into otherwise puzzling economic phenomena. For example, some of the world’s biggest energy companies have for years failed to agree on joint management of oil and gas fields they own together. If one company pumps the oil too fast, it can wreck the pressure in the gas field; if the other extracts gas too fast, it traps the oil. American law has offered them an effective legal tool, called “unitization,” to overcome anticommons tragedy and smooth joint management of divided oil and gas interests. Yet firms block each other year after year.24 How can this be? Oil units are not a case of two spiteful neighbors arguing over a broken backyard fence. They involve arm’s-​length business negotiations between savvy corporations. Everyone has good information about the underlying geologic and technical issues. The gains from cooperation are in the billions of dollars. Why doesn’t one firm sell its interest to the other? Why don’t the firms merge? What’s going on? The experimental studies are beginning to give us explanations rooted in the psychology of the anticommons. Even the most sophisticated businesspeople can fail to reach agreement when a negotiation is framed in anticommons terms.

7.8.  Rounding out the Lexicon—​ Caveats and Cautions In rounding out the anticommons lexicon, there are some caveats:  first, my term focuses on one form of underuse, the tragedy that arises when ownership is too fragmented. Here, multiple owners block each other from using a scarce resource. Underuse can also arise in the monopoly context, when a single owner blocks access to a resource. This situation may be tragic, but it is not an anticommons tragedy in my sense of the term. In the old economy, many companies held monopolies: Ma Bell (the American telephone monopoly), railways, local water utilities, and others. Society gained the economic benefits of scale and scope from allowing these sectors to be monopolized. The state policed against abuse of monopoly power through complex rate regulation and oversight. Phone lines were cheaper and more available than in many other countries. The costs of these monopolies were often invisible, costs such as deferred and dampened innovation. In an information economy, any piece of intangible property, such as a patent, is also a monopoly. We award patents because monopoly profits create incentives to invent and uncertainty increases, losses become even more pronounced. See also Depoorter and Vanneste (2007, 21–​23). 24 

On the costs of “excessive anarchy” in the oil industry, see Libecap and Smith, (2002). The same tragedy affects excessive pumping of groundwater. See Thompson (2000, 241–​250).

 

COMMONS AND ANTICOMMONS    195 because patents give inventors incentives to disclose their discoveries (without patents people might prefer to invent things they could keep secret). On the other hand, drugs would be cheaper and lives could be saved if competitors could make generic copies at will. To balance the values of innovation, disclosure, and competition, the US Congress keeps shifting the bundle of rights that a patent confers. The dilemmas of any individual monopoly in the old or new economy are a great topic, but are not relevant here. For better or worse, these quandaries are familiar. However, we do not have much experience dealing with the interaction of ownership fragments or an array of blocking patents. The anticommons lexicon addresses not monopoly per se, but ownership multiplicity. Next, here’s a caution: when talking about an anticommons, stay away from absolutes. First, you shouldn’t assume that anticommons ownership is inevitably tragic.25 If we lived in a world where people had perfect information and could bargain with each other at no cost, they could avoid anticommons tragedy every time (just as, in a perfect world, there would be no commons tragedy, or for that matter, tragedy of any sort) (Coase 1960). In practice, however, bargaining is never free: people shirk duties and hold out for payoffs, and there are cognitive limits that shape owners’ decisions. In the real world, anticommons ownership is not necessarily tragic, but it does tend that way. Second, it’s theoretically possible that an anticommons may face overuse instead of underuse.26 For example, consider real estate development along the California coast. It’s a mess. Multiple community groups, environmentalists, neighbors, and government agencies may each prefer different versions of a project. Even in that regulatory morass, though, overbuilding may occur if it is sufficiently costly to exercise each right to veto development. Every opponent of development may prefer to go surfing and hope the others sit through the boring public hearings. If enough people opt for a free ride, a project might face too little opposition, not too much. It’s an empirical question whether the California coast tips toward over-​or underbuilding. That said, in most cases I’ve seen, anticommons regulation tends to be associated with too little economic development, not too much.27 Finally, there is a caveat that comes from the legal theorist Carol Rose. Certain resources, such as roads and waterways, are sometimes owned most efficiently in common. As Rose points out, creating and enforcing private-​property rights is itself costly; sometimes these costs exceed the gains, not just economically but also socially. Village greens and town halls may strengthen communities in ways that are socially valuable but hard to quantify in monetary terms. Rose (1986) coined the phrase “the comedy of the commons” to describe such social and economic benefits that can flow from group access (see also Ellickson 1993, 1336–​1338). 25 

These points are developed in Heller (1998, 676). Fennell develops this insight in her theory of common interest tragedies (see 2004, 934–​937). 27  Just as an anticommons theoretically may lead to overuse, it is possible for a commons to be associated with underuse (ibid). For a real-​world example, see Buzbee (2005). 26 

 

196   MICHAEL HELLER Rose’s insight is equally true on the anticommons side: there are both economic and social reasons that we may prefer group exclusion to sole ownership. For example, it’s possible that creating multiple vetoes may help preserve a treasured resource against transient political pressures for development: for instance, Central Park in New York City or Indian burial grounds in Arizona.28 Similarly, “conservation easements” intentionally use anticommons ownership to foster environmental goals Mahoney (2002, 739–​785). (With a conservation easement, the owner sells or gives away the right to develop land, gets a tax break, and retains the right to continue a current use such as farming.) The underuse created by split ownership may be justifiable if the environmental gains exceed fragmentation costs. On balance, though, I’m skeptical. What happens a generation from now when communities want to reduce sprawl but face a patchwork of easements that make “in-​fill” development prohibitively difficult? Many conservation easements look to me like potential anticommons tragedies.29 The “comedy of the anticommons” insight suggests that sometimes, for some resources, we should promote little or no use. Most of the time, for most resources, however, some positive level of use will be socially most valuable. Underuse is rarely the optimum.

7.9.  Toward a Nonsquiggly Language We have millennia of practice in spotting tragedies of the commons. When too many people fish, fisheries are depleted. When too many people pollute, we choke on dirty air. Then, we spring into action with market-​based, cooperative, and legislative solutions. Similarly, we have a lot of experience spotting underuse caused by a particular monopoly owner. We have created regulatory bodies that know (more or less) what to do with such dilemmas. But underuse caused by multiple owners is unfamiliar. The affected resource is hard to spot. Our language is new. A tragedy of the anticommons may be as costly to society as the more familiar forms of resource misuse, but we have never noticed, named, debated, or learned how to fix underuse. How do we stumble into the problem of too many owners? How do we get out? As a first step, underuse in a tragedy of the anticommons should be squiggly no more. Fixing anticommons tragedies is a key challenge for our time. Individual entrepreneurship and legal reform will be important. But, I want to stress that the first and most 28  On the potential use of an anticommons to preserve Central Park, see Bell and Parchomovsky (2003, 3–​4, 31–​36, 60–​61). 29  See Bell and Parchomovsky (2003, 785–​86) noting potential anticommons tragedy in conservation easements.

 

COMMONS AND ANTICOMMONS    197 important step to solving an anticommons is to name it and make it visible. With the right language, we can all spot links among ownership puzzles, and we can all come together to fix them. Nothing is inevitable about an anticommons. Every ownership puzzle results from choices we make (and thus can change) about how to control the resources we value most.

References Acheson, James. 1988. Lobster Gangs of Maine. Lebanon, NH: University Press New England. Aristotle. 1996. The Politics and the Constitution of Athens. Ed. S. Everson. Trans. B. Jowett. Cambridge: Cambridge University Press. Barnett, J. 2015. “The Anti-​Commons Revisited.” Harvard Journal of Law & Technology 1, 3, pp. 16–​57. Bell, A. and Parchomovsky, G. 2003. “Of Property and Antiproperty.” Michigan Law Review 102, pp. 1–​61. Blackstone, W. 1765–​1769. Commentaries on the Laws of England:  In Four Books. Bk. 2. Oxford: Clarendon Press. Buchanan, S. and Yoon, Y. J. 2000. “Symmetric Tragedies:  Commons and Anticommons.” Journal of Law and Economics 43, pp. 1–​12. Burk, D. L. and Lemley, M. A. 2009. The Patent Crisis and How the Courts Can Solve It Chicago: University of Chicago Press. Buzbee, W. W. 2005. “The Regulatory Fragmentation Continuum, Westway, and the Challenges of Regional Growth.” ’ Journal of Law and Politics 21, pp. 323–​364. Chamberlain, J. P. 1923. The Regime of the International Rivers:  Danube and Rhine. New York: Columbia UP. Coase, R. H. 1960. “The Problem of Social Cost.” Journal of Law and Economics 1, pp. 1–​44. Cournot, A. 1838 [1897]. Researches into the Mathematical Principles of the Theory of Wealth. Trans. N. T. Bacon. London: Macmillan. Dagan, H. and C. J. Frantz. 2004. “Properties of Marriage.” Columbia Law Review 104, pp. 75–​133. Dagan, H. and Heller, M. A. 2001. “The Liberal Commons.” Yale Law Journal 110, p. 549. Dari-​Mattiacci, G. and Parisi, F. 2006. “Substituting Complements.” Journal of Comparative Law and Economics 2, pp. 333–​347. Demsetz, H. 1967. “Toward a Theory of Property Rights.” American Economic Review, Papers and Proceedings 57, pp. 347–​359. Depoorter, B. and Vanneste, S. 2007. “Putting Humpty Dumpty Back Together:  Pricing in Anticommons Property Arrangements.” Journal of Law, Economics, and Policy 3, pp. 1–​28. The Digital Library of the Commons. http://​dlc.dlib.indiana.edu [Accessed 2 October 2012]. Eggertsson, Thráinn. 2003. “Open Access versus Common Property.” In:  T. Anderson and F. McChesney, eds., Property Rights: Cooperation, Conflict, and Law. New York: Princeton UP, 2003. pp. 74–​75. Ellet, C. Jr. 1839. An Essay on the Laws of Trade in Reference to the Works of Internal Improvement in the United States. Richmond: P.D. Bernard. Ellickson, R. C. 1993. “Property in Land.” Yale Law Journal 102, pp. 1315–​1338.

 

198   MICHAEL HELLER Ellickson, R. C. 1994. Order without Law: How Neighbors Settle Disputes. Cambridge: Harvard UP, 1991. Fennell, L. A. 2004. “Common Interest Tragedies.” Northwestern University Law Review 98, pp. 907–​937. Hardin, G. 1968. “The Tragedy of the Commons.” Science 162, pp. 1243–​1244. Hazlett, T. W. 2005. “Spectrum Tragedies.” Yale Journal of Regulation 22, p. 242. Heller, M. A. 1998. “The Tragedy of the Anticommons: Property in the Transition from Marx to Markets.” Harvard Law Review 111, pp. 621–​688. Heller, M. A. 1999. “The Boundaries of Private Property.” Yale Law Journal 108, pp. 1163–​1223. Heller, M. A. 2001. “The Dynamic Analytics of Property Law.” Theoretical Inquiries in Law 2, pp. 79–​92. Heller, M. A. 2008. The Gridlock Economy: How Too Much Ownership Wrecks Markets, Stops Innovation, and Costs Lives. New York: Basic Books. Heller, M. A., ed. 2011. Commons and Anticommons. Cheltenham: Elgar Publishing. Heller, M. A. and Eisenberg, R. S. 1998. “Can Patents Deter Innovation? The Anticommons in Biomedical Research.” Science 280, pp. 698–​701. IASC. On the Digital Library of the Commons. http://​www.iascp.org [Accessed 2 October 2012]. Libecap, G. D. and Smith, J. L. 2002. “The Economic Evolution of Petroleum Property Rights in the United States.” Journal of Legal Studies 31, p. S589. Mahoney, J. D. 2002. “Perpetual Restrictions on Land and the Problem of the Future.” Virginia Law Review 88, pp. 739–​786. Michelman, F. J. 1982. “Ethics, Economics and the Law of Property.” Nomos 24(3), pp. 667–​669. OED, “overuse,” v., http://​www.oed.com/​view/​Entry/​135291 [Accessed 2 October 2012]. OED, “under-​use, n.,” http://​www.oed.com/​view/​Entry/​212195 [Accessed 2 October 2012]. Ostrom, E. 1990. Governing the Commons: The Evolution of Institutions for Collective Action. Cambridge: Cambridge University Press. Ostrom, E. 1999. “Coping with Tragedies of the Commons.” Annual Review of Political Science 2, p. 493. Parisi, F. 2002. “Entropy in Property.” American Journal of Comparative Law 50, p. 595. Parisi, F., Schulz N., and Depoorter, B. 2003. “Fragmentation in Property: Towards a General Model.” Journal of Institutional and Theoretical Economics 159, p. 594. Parisi, F. and Depoorter, B. 2003. ‘Fair Use and Copyright Protection:  A  Price Theory Explanation.’ International Review of Law and Economics 21, p. 453. Rose, C. M. 1986. “The Comedy of the Commons: Custom, Commerce, and Inherently Public Property.” University of Chicago Law Review 53, p. 711. Rose, C. M. 1999. Expanding the Choices for the Global Commons: Comparing Newfangled Tradable Allowance Schemes to Old-​ Fashioned Common Property Regimes.” Duke Environmental Law and Policy Forum 10, p. 45. Shapiro, C. 2001. “Navigating the Patent Thicket: Cross Licenses, Patent Pools, and Standard Setting.” In: A. B. Jaffe. et al., eds, Innovation Policy and the Economy. Vol 1. Cambridge: MIT Press, 2001. 119–​150. Thompson, Barton H.  Jr. 2000. “Tragically Difficult:  The Obstacles to Governing the Commons.” Environmental Lawyer 30, pp. 241–​250. Thucydides. 1910. History of the Peloponnesian War. Bk. 1, sec. 141. Trans. R. Crawley. New York: E.P. Dutton. Tierney, J. 2000. “A Tale of Two Fisheries.” New York Times Magazine 27 August 2000.

 

COMMONS AND ANTICOMMONS    199 Vanneste, S. et  al. 2006. “From ‘Tragedy’ to ‘Disaster’:  Welfare Effects of Commons and Anticommons Dilemmas.” International Review of Law and Economics 26, p. 104. Varian, H. R., Farrell, J. V., and Shapiro, C. 2005. The Economics of Information Technology: An Introduction. Cambridge: Cambridge UP. Walsh, J. P., Arora, A., and Cohen, W. M. 2004. “Effects of Research Tool Patents and Licensing on Biomedical Innovation.” In: W. M. Cohen and S. A. Merrill, eds., Patents in the Knowledge-​ Based Economy. Washington, DC: National Academies Press. 285–​340. Yoram Barzel. 1989. Economic Analysis of Property Rights. Cambridge: Cambridge UP. Ziedonis, R. H. 2004. “Don’t Fence Me In: Fragmented Markets for Technology and the Patent Acquisition Strategies of Firms.” Management Science 50, p. 804.

 

Chapter 8

EC ONOMI C S OF INTELL E C T UA L PROPERT Y L AW Robert P. Merges

The economic study of intellectual property law is now well into its second wave.1 It first emerged in a series of ad hoc writings by pure economists stretching from the nineteenth century to the early 1960s. The first wave culminated in a series of classic articles done in the emerging methodology of Chicago school law and economics. But beginning in the 1980s, there began a new and muscular wave of research. This second wave is characterized by two primary features: (1) increasing methodological diversity and sophistication; and (2)  an emphasis on contextualization—​understanding how intellectual property (IP) law is embedded in larger social and economic systems, and how IP interacts with other aspects of those systems to foster innovative ideas and economic growth. This chapter assesses the state of this second wave. It outlines some chief points of contention in the literature, and points to some frontier issues just now appearing on the horizon. It also strives, in the conclusion, to push scholars to spend less time trying to answer the Big Question in the field—​whether IP law can be justified on economic grounds—​and more time understanding specific features and individual rules in the IP system. The Big Question has proven to be an elusive quarry. Much of the payoff in the field has come from asking a series of discrete Little Questions, so this enterprise should naturally attract most of the effort in the field.

8.1. Introduction It is useful, once in a while, to look at a complex thing in whole cloth, instead of thread by slender thread. When I do that with IP, here is what I see. I see that IP is beginning 1 

Robert P. Merges, is Wilson Sonsini Goodrich & Rosati Professor of Law and Technology, UC Berkeley Law School.

 

ECONOMICS OF INTELLECTUAL PROPERTY LAW    201 to live up to the aspirations of an earlier generation, which yearned to make law a much more expansive field of study, one that integrated the insights of sociology, history, psychology, and economics. All these threads are in place in the IP field now. Gone are the days when virtually every article on the subject began and ended with a discussion of case law and statutes. Now, though straight doctrinal work is far from dead, the pages of academic journals are alive with studies of all facets of IP law, informed and influenced by a wide range of academic disciplines. These disciplines supply a cornucopia of tools, which scholars put to work on all sorts of issues and problems. Scholars no longer stick to the close knitting of IP statutes and cases. What we have now is a very lively tapestry. Take for instance the outlines of a typical article from IP’s “classical” period. Starting from well-​accepted background assumptions—​that IP law is about incentives, for example; or that education is an important value and ought to be fostered by IP law—​a scholar would approach a recent line of cases, or an important decision by the Supreme Court.2 Such facts as entered into the discussion would be drawn from the legal record in the case. The influence of older cases would be plumbed and dissected. Implications of a legal holding would be fleshed out—​incentives are reduced, for example, by tightening patent standards; or educational values will be undermined, by constricting fair use. Based on these observations, the author would advise doctrinal course corrections to set things right. Contemporary work is very different. A book or article might be addressed to foundational assumptions. How does IP work in the overall context of creativity in an industry? What kind of growth has the industry undergone, and what if any role has IP been shown to play in that?3 Do firms and people in the industry rely on informal norms or other techniques—​apart from formal IP rights—​to protect investments in creative works?4 And, within the province of IP doctrine, how do courts actually apply the doctrines found in IP law?5 From a completely different perspective, a scholar might address the cost side of IP protection: what happens in the market when IP rights expire—​for a patented pharmaceutical, for instance, or a copyrighted book?6 To be sure, the literature continues to address topics of interest to first-​wave scholars. Does IP really operate as a viable incentive? In a particular industry or field of creativity? How does IP compare with other motivations and incentives that drive creativity, generally and in specific industries? Did a change in the law actually lead to greater or lesser investment or activity?7

2 

See, e.g., Kitch (1966); see, also, Adelman (1977). See, e.g., Allison, Dunn, and Mann (2007) (patents and the software industry); Mann and Sager (2007) (patents and startup firms in the software industry). 4  See, e.g., Raustiala and Sprigman (2012). 5  See, e.g., Beebe (2008) (empirical study of copyright’s fair use factors); Sag (2012) (analyzing copyright cases so as to predict the outcome of fair use disputes). 6  See Buccafusco and Heald (2013). See also Bechtold and Tucker (2013) (court ruling relaxing trademark enforcement against Google “keyword” ad sales in Europe had little effect on user visits to trademark owners’ websites). 7  Another type of study tests assumptions behind doctrine or doctrinal change. See, e.g., McKenna (2009) (testing trademark law’s assumption that consumers may be confused when a trademark owner’s mark is used in markets unrelated or ancillary to the owner’s primary markets). 3 

 

202   ROBERT P. MERGES These examples highlight two major differences between early or “first-​wave” scholarship and the work of more contemporary “second-​wave” researchers: (1) increased attention to IP rights in a broader economic context (contextualization); and (2) greater methodological diversity. This chapter is organized around these two themes. I first discuss the many ways that second-​wave scholarship seeks to show how IP rights are embedded in broader economic contexts, and thus diverges from first-​wave research, which tended to focus exclusively on IP rights as the central determinants of economic activity. Next, I consider the many different methodologies now being deployed to study issues in the economics of IP rights—​from large-​scale empirical work to surveys to interviews to experimental research.

8.2. Contextualization As a fine poet once wrote, “I have had to learn the simplest things last.”8 This seems true of IP economics. While the earliest forays into the field tended to see the legal rules concerning IP protection as part of a broader set of issues relating to investment in creative works (e.g., Levin et al. 1987), much of first-​wave scholarship conceived of formal IP rights as the sole determinant of various types of economic activity.9 (This is of course a feature of comparative statics, with its famous—​in some quarters, infamous—​ assumption of ceteris paribus.) As applied to IP in its early days, law and economics were concerned with the net welfare effects of tweaking specific rules and doctrines. A  crucial contribution of the second wave of IP economics has been to widen the field of view within which economic activity is studied.10 In this new approach, legal rules are but one determinant of activity in the creative sectors of the economy. Other aspects are just as important:  industry norms; technological protections; structural factors such as lead time (i.e., degree of difficulty to copy new products), the effectiveness of trade secrecy, and the ability to bundle innovative products with other products 8 

Charles Olson. Maximus, to Himself (1983). The full lines are “I have had to learn the simplest things/​last. Which made for difficulties.” 9  See, e.g., Pigou (1924); Clark (1927); and Plant (1934). For an excellent overview of this and other early work, see Menell and Scotchmer (2007) in A. Mitchell Polinsky and Steven Shavell (2007). 10  A body of work by the noted economist of technological change, Richard Nelson, stands either as a glaring exception to the standing assumptions of the first and second wave, or as the vanguard of the third wave; and perhaps both. See, e.g., Nelson and Rosenberg (1993, 3–​5) (“[T]‌he concept [of a system] is that of a set of institutional actors that, together, plays the major role in influencing innovative performance.”). Indeed, the highly influential original “Yale survey” on innovation embodied in 1987 the assumption that formal IP rights were but one among a series of “appropriability mechanisms” through which private firms sought to recoup the costs of their R&D investments. See Levin et al. (1987). Nelson deserves far more credit than he usually gets for thoroughly understanding and forcefully arguing for the “contextual” view of IP at what now seems a very early date. He is truly “Il miglior fabbro,” as T. S. Eliot said of Ezra Pound. For an appreciation, see Dosi (2006). Other early work by Edwin Mansfield also deserves to be mentioned here. See Mansfield (1985, 1986, 1986); Mansfield et al. (1977).

 

ECONOMICS OF INTELLECTUAL PROPERTY LAW    203 or services over which a firm has pricing power so as to recoup investments in innovation.11 Even alternative policy mechanisms such as R&D tax credits or firm relocation credits come into play. In second-​wave thinking, IP is just one of many ways firms make money by creating new things. Formal IP rules are seen as embedded in a larger economic context. This scholarship argues that only by understanding this larger context can we understand the true impact of formal legal rules on “the rate and direction of technological growth.” As mentioned, second-​wave thinking is in one sense simply a return to the roots of the field. One of the earliest survey-​based studies of IP rights asked R&D managers to compare patents with other, non-​legal “appropriability mechanisms”—​techniques for recouping investments in creative works. Perhaps not surprisingly, many respondents listed lead-​time advantage and trade secrecy as more important than formal IP rights such as patents and copyrights.12 Second-​wave research has now returned to this basic theme. In this new line of scholarship, contextualization is explored along a number of dimensions. In this chapter, I want to emphasize just three examples.13 Taken together, these give some idea of the richness that follows from understanding IP as but one aspect of a much broader innovative ecosystem in which creative firms and people operate. First, I will discuss what are known as IP’s “negative spaces”: trades and pursuits in which participants rely on informal norms instead of formal legal protection. These illustrate how IP rights are embedded in a larger system of social norms, which some scholars say render formal IP redundant. Second, I will turn to the function of IP rights in “platform technologies,” for example, cell phone standards and computer operating systems. Firms that own or dominate platforms often use IP selectively to control access to their platforms. For these firms, IP is embedded in a complex strategic game in which they seek profits through some combination of technology platform and complementary products. IP can be used in a number of ways in this game: to erect a toll for access to the platform; to selectively exclude competitors from the platform while sharing access with allies; or to enforce an open or nonproprietary platform model with other industry participants. Third, I will discuss the growing literature that links IP rights to economic organization. Scholars have come to see that IP plays an important role in the setting of firm boundaries, subtly affecting issues such as whether small, R&D-​intensive

11  This particular topic is very well presented in Teece (1986). Other survey-​based studies back this theory with empirical findings. See Cohen, Nelson, and Walsh (2000); Graham et al. (2010). 12  Levin et al. (1987, 794). 13  Three of what could have been main examples, I should add. So for instance, I could have added an entire section on contemporary trademark scholarship, which has shown that simplistic models of trademarks as a way to reduce consumer search costs are incomplete because they ignore the sophisticated techniques of branding experts and marketers. Thus, Jeremy Sheff shows that investments in branding can actually bias consumer’s perceptions about what they need or about the characteristics of a relevant product. See Sheff (2011); Lemley and McKenna (2012) (deploying marketing literature to explore the complexities of brand loyalty, and its impact on the actual economic effects of trademark law); and Lee, DeRosia, and Christensen (2009).

 

204   ROBERT P. MERGES firms can survive independently or instead are fated to be absorbed into large, vertically integrated companies. IP policy, from this perspective, helps determine not only aggregate investments in creativity, but also the locus of innovative activity: individuals, small firms, or large firms. It is a new and exciting aspect of the overall trend toward greater contextualization in the economic study of IP rights.

8.2.1. Case Study in Contextualization: Fields where IP Is “Unnecessary,” or “IP’s Negative Spaces” In recent years, IP scholars have described a number of fascinating trades and pursuits where people get along quite well without the protection of formal, enforceable IP rights. From French chefs (Fauchert and von Hippel 2008) to stand-​up comics (Sprigman and Oliar 2008), and from fashion designers (Sprigman and Raustiala 2006) to tattoo artists (Perzanowski 2013) and magicians (Loshin 2010), creative people working in various industries develop norms and practices that provide an adequate, and perhaps often superior, alternative to formal IP protection. Some have taken to calling these areas, collectively, IP’s “negative spaces” (Rosenblatt 2011). From these studies, a consensus theory has begun to emerge, which holds that IP law is far less necessary than many have traditionally supposed. From a descriptive or positive beginning, in other words, negative space theory often moves to a more normative point: the essential wrongheadedness of the traditional story that IP rights are always and everywhere necessary to call forth creative works. In terms of contextualization, then, negative space theory says at times that IP policy needs to be sensitive to one important aspect of context: the presence of norms that can act as “IP substitutes.”

8.2.1.1. How Big Is the Negative Space? One important point about negative space studies is that they tend to concentrate on “niche” fields, industries, and pursuits that are, in the aggregate, relatively small when compared with the “IP economy” as a whole, Merges (2013). So, for example, French restaurants, stand-​up comedy, and tattoo parlors are relatively modest economic enterprises.14 Compared with the estimated size of the overall IP-​based economy, they represent only a small percentage of activity. The International Intellectual Property Protection Alliance (IIPPA), for example, estimates that the “copyright industries” alone add $1 trillion to the US economy each year.15 It is very difficult to arrive at similar estimates for the “patent industries” because companies in so many industries obtain large numbers of patents every year. But we can say that the pharmaceutical industry is 14 

An exception might be Schultz (2006), who studies norms surrounding the copying of recorded music concerts by fans of “jam bands” such as the Grateful Dead. It is not clear how large this genre is as a percentage of all popular music, but it is clearly substantial. 15  Figures such as these should always be taken with a grain of salt because they are based on large aggregate data sources and prepared by interest groups with a string agenda. See L.A. Times (2013).

 

ECONOMICS OF INTELLECTUAL PROPERTY LAW    205 worth $ 340 billion per year alone,16 and that the chemical industry totals roughly $450 billion.17 Most of the fields that negative space theorists explore are, in reality, fairly small scale. They fit the description of property theorist Robert Ellickson (1993), who some time ago noted that in “close-​knit” groups, formal property rules were often less efficient than an effective set of social norms.18 But, as Ellickson also noted, once groups grow beyond a certain size, and the economic value of their activities passes some threshold, informal norms cease to work. I believe this is as true for IP as for conventional tangible property. So, while we may have some interesting things to learn about context from IP’s negative spaces, the most important lesson in the end may be that most industries, as they grow, end up recapitulating the history of IP protection generally. As the larger context changes, the negative spaces are filled in.

8.2.1.2. Changing Fashions? But there is one industry studied by negative space theorists that is not small scale: fashion. A conservative estimate is that this is a $200 billion per year field (Hemphill and Suk 2009).19 Clearly, the magnitude of this industry is much closer to the “copyright industries” plus pharma/​chemicals. So where does fashion fit into discussions of the “negative space”? Perhaps the most important point to note in this respect is that the fashion industry itself seems quite motivated to escape the negative space. According to Scott Hemphill and Jeanie Suk (2009): The adverse effects of copying explain why many [fashion] designers oppose copying, just as they oppose counterfeiting of handbags. ([The negative space] argument, if correct, ought to apply to fashion trade-​marks and copyrights as well.) [Raustiala and Sprigman] pitch their paradox as an explanation for the otherwise puzzling equanimity with which designers greet copyists. But that premise is faulty. In fact, many designers are vocal advocates against copying, and … make use of the currently limited legal tools available to curb copyists.20

In addition to enforcement using the established (though flawed) legal tools available, fashion designers have been active in lobbying for various new forms of design protection.21 This lobbying effort is perhaps the best indicator that innovative firms are indeed not satisfied with the current copying-​centric equilibrium in the fashion industry. As a number of scholars have noted, in fact, the speed and ease of copying have created a 16 

Wikipedia, Pharmaceutical Industry. www.Chemistryviews.org. 18  See Ellickson (1993, 1520). 19  See Hemphill and Suk (2009, 1148). US apparel sales reached $196 billion in 2007 (NPD Group 2008). Among fashion accessories, considering just one category, handbags, adds another $5 billion in sales. Krim (2007) reports US sales exceeding $5 billion in 2005. 20  See Hemphill and Suk (2009, 1183). 21  See, e.g., Eguchi (2011, 145): “A design protection bill [from 2010] garnered support from several well-​known designers and the Council of Fashion Designers of America (CFDA)—​the creative core of the fashion industry….” 17 

 

206   ROBERT P. MERGES significant threat to the livelihoods of creative fashion designers, making future lobbying more likely.22 Viewed from the perspective of the evolution of IP protection, the fashion industry begins to look less like an instructive outlier and more like a very conventional industry.23 As with many other industries, as the value of the creative products increases, the calls for strengthened IP protection increase as well. It may not be long until fashion crosses the great divide from negative to positive IP space.

8.2.1.3. Learning about Context from Negative Spaces From all this, I think there are two things we can learn. First, there is indeed room for IP-​ free zones—​negative spaces in the fabric of IP protection. IP need not permeate every crease and corner of the economy. Negative space theorists have shown real creativity and resourcefulness in searching out and documenting examples of fields that survive and even thrive outside the shadow of IP protection. As with all empirical research, each case study adds to our understanding of how our economy and society actually function, and the role of IP rights in different economic contexts. Yet, the other lesson from exploring negative spaces has to do with their limits. Careful attention to context reveals that, in most cases, norms are effective substitutes for IP rights primarily in niche industries where participants are relatively close-​knit socially. Out in the larger, more anonymous economy, it remains to be seen whether norms can effectively replace the formal protection provided by IP rights.

8.2.2. IP and Platform Technologies When scholars study IP in context, IP becomes not the sole fulcrum determining economic outcomes but instead one of many interconnected facets of economic activity. On one hand, this diminishes the significance of IP rights in the analysis; they are not the exclusive determinant of investment, innovation, or whatever dependent variable is affected by the existence or strength of IP protection. On the other hand, this more contextualized understanding helps to isolate the unique role that IP rights can play in a large, complex economic system. An example from the world of platform technologies will show what I mean. Platform technologies are technical systems such as computer hardware (e.g., the Apple iPhone or Samsung Android cell phone handsets) or software (the Microsoft Windows or LINUX operating systems). They provide a common starting point for further technological development: the creation of special chipsets for the iPhone: for instance, or a cell phone application designed for use with the Google Android operating system software. For some time, scholars have understood the economic/​technological forces 22  23 

See, e.g., Scafidi (2008). See Merges (2001).

 

ECONOMICS OF INTELLECTUAL PROPERTY LAW    207 at work in these “platform markets” (e.g., Shapiro and Varian 1998). What is of interest to IP scholars is how firms in platform markets use IP rights as strategic instruments. Some firms profit from proprietary platforms, which are covered by various IP rights. For them, exclusive IP rights exclude direct competitors, obviously. But in the case of “allies,” companies that develop complementary technologies or content that increase the overall value of the platform “ecosystem,” IP rights are often purposely waived (see Simcoe 2008; Merges 2008, 2011; Barnett 2011b). So IP law permits a pattern of selective enforcement and selective waiver that is used by sophisticated platform players to advance their interests in this complex area. Some platforms are “open,” rather than proprietary. IP plays a role in this context too. Access to an open platform can be made conditional on agreement to perpetuate the open access policy. This is achieved through licensing contracts granting open-​source participants access to a platform. Licensees agree that in exchange for obtaining access to an open-​source platform, their contributions will in turn be subject to an open licensing policy. Thus—​somewhat paradoxically—​IP rights are the foundation upon which open access is built. The IP rights that cover an open platform are the basis for the legal threat that maintains open access. The literature on IP and platforms treats platform access as a strategic variable that can be manipulated by for-​profit firms, those usually associated with the strong IP rights/​ proprietary innovation model side of the IP debate. So open innovation and profit maximization are no longer opposite ends of an ideological spectrum; the former becomes a strategy pursued by private firms when conditions warrant. This is true of for-​profit companies permitting access to their platforms, but in the case of private investment in open-​source projects. In each case, firms decide strategically to waive their property rights in the service of firm advantage. We see here an illustration of the growing contextualization of IP rights in the economic literature. Scholars have shown that they understand why the ability to waive IP rights selectively is an essential dimension of their usefulness in the context of platform competition. This is a far cry from a simple, unidimensional “incentive” story where an increase in IP protection is tested for its effect on total R&D investment. The effectiveness of IP rights in the platform environment depends not only on the attributes of the IP rights when granted but also on the various licensing and enforcement strategies employed by the holders of IP rights. By studying this very special setting, economists have learned some new important lessons about the effects of IP rights on firm strategy and overall welfare.

8.2.3. IP Rights, Firm Boundaries, and Economic Organization Traditional first-​wave scholarship followed the conventions of classical microeconomics in paying little attention to the organization of economic production. IP rights

 

208   ROBERT P. MERGES influenced the overall production level in society, but it was never mentioned which types of organizations did the producing (e.g., Nordhaus 1969). In these early models, IP rights stimulate creative output (while raising consumer prices), but the precise locus of that output is never mentioned: individuals, small firms, large firms, etc.—​this body of theory abstracts away from these issues. But as with economics generally, IP economics has in later years discovered the fascinating set of issues surrounding firm boundaries.24 This is one of the clearest and most rewarding aspects of the general trend toward the contextualization of IP rights. The basic insight from this literature is that IP rights can actually affect the location of firm boundaries (Teece 1986b). The key to this new understanding of IP is to see it not primarily as something that affects overall incentive levels, but instead as an instrument that affects transactions—​and hence the organization of production (Merges 2005; Arora and Merges 2004; Burk and McDonnell 2007; Barnett 2011a). Advocates of this view see IP as a way for small, specialized firms to protect against opportunism when contracting with larger firms. IP makes it easier for specialized firms to sell technology and know-​how via arm’s-​length contracts, which permits specialized producers to exist as independent firms. IP rights can then be said to affect industry structure; without these rights, specialized knowledge subject to opportunistic copying would have to be produced within large, vertically integrated firms. This in turn would mean a loss of the “high powered incentives” (to use Williamson’s term) available to independent firms who sell their output via contracts (Merges 2005). The upshot is that IP at the margin may enable more small and independent firms to remain viable even in industries where multicomponent products are assembled and sold by large, vertically integrated firms. This literature on IP and the boundaries of the firm thus shows in very clear relief the importance of studying IP rights in context. Context here is everything: IP rights and the nature of the firms that use them are inextricably interlinked. Another insight is that, by affecting the location of firm boundaries, IP may exert an indirect incentive effect that has not been appreciated in the past. As Barnett (2011a, 787) puts it, in the case of patents: “Organizational effects proxy for innovation effects:  where patents alter organizational behavior, they alter innovation behavior.” This is because the smaller firms enabled by stronger IP rights may be more innovative, as compared to their large, vertically integrated counterparts. Barnett (2004) and others make an additional point, which in some sense brings this literature back around to one of the pioneer studies in the contextualization of IP rights (Levin et al. 1987): because large firms have at their disposal numerous ways to recoup expensive R&D investments (e.g., by bundling technology-​intensive products with other products over which the firm has pricing power), while small companies do not (owing to the 24 

Good syntheses of the literature include Burk and McDonnell (2007); Barnett (2011b). See also Heald (2005); Bar-​Gill and Parchomovsky (2009). Perhaps the earliest reference to this theme in the legal literature is Merges (1996), arguing that weak formal IP rights forced the Japanese software industry into a vertically integrated production model that disfavored small, independent firms; however, see Mashima (1996), offering corrections to this thesis.

 

ECONOMICS OF INTELLECTUAL PROPERTY LAW    209 absence of complementary assets), IP rights are particularly important for small firms. The strength of IP rights, as a policy variable, affects small firms disproportionately.25

8.3.  Methodological Diversity The basic methodology of the first wave was comparative statics of one variety or another. The classic version was a formal comparative model, showing overall welfare with and without whichever feature of the IP system was being modeled (see, e.g., Nordhaus 1969). Even the early contributions, from what might be called the primordial era of the first wave, employed this approach. (They used analytic descriptions, or “word models,” instead of “formal” or mathematical models; see, e.g., Plant [1934].)

8.3.1.  The Growth of Empirical Studies The second wave is far more diverse. IP scholars now use all sorts of methods. Chief among them in recent years has been the use of empirical data. Beginning with some path-​breaking work in the late 1980s (e.g., Levin et al. 1987) and 1990s (e.g., Allison and Lemley 1998), and accelerating especially in recent years, empirical research has burgeoned so much that it now has its own specialty abstracting service.26 Rarely does a week go by without some new and interesting contribution looking at music sales, patent prosecution, or trademark searches. Scholars are beginning to specialize; familiarity with large datasets is now almost required in certain subfields of IP economics, especially patent law.27 Some studies rely on traditional data sources, such as patent citation studies. Others are based on novel, labor-​intensive datasets the researchers put together themselves. And more sophisticated statistical methods are brought to bear, compared with the relatively simple early work (see, e.g., Ziedonis 2004; Hall, Jaffe, and Trajtenberg 2005; Galasso and Schankerman 2010). Each study adds to the growing integration of IP empirics into the main body of contemporary social science. A heightened sense of context is evident in the newer generation of empirical work, following the general theme of recent IP scholarship. Early empirical studies were highly aggregated, comparing for example GDP growth with patent grant rates. Recent studies pursue a number of more sophisticated strategies, including a comparison of 25 

While in general a society interested in maximizing social welfare would be indifferent to whether large or small firms did most of the innovating, it has been suggested that the enhanced autonomy which accompanies smaller firms provides an independent social good (Merges, 2011a). Hence, in some cases at least slightly higher transaction costs ought to be tolerated for the sake of this independent social good. 26  See SSRN.com. 27  See, e.g., Alcacer and Gittelman (2006) on the use of patent-​examiner data to study patent examination patterns; Abrams, Akcigit, and Popadak (2013), on sophisticated study of patents through citation data.

 

210   ROBERT P. MERGES how related patents for a single invention fare at three of the major international patent offices, in Japan, Europe, and the US (see Webster, Palangkaraya, and Jensen 2007). Another set of studies identifies variation in patent-​examiner behavior, and the impact this has on patent outcomes such as validity determinations in litigation. See, e.g., Cockburn, Kortum, and Stern (2003). But empirical studies are just one of the methods second-​wave scholars put to use. Structured interviews, experimental methods, and comparative institutional approaches have also become popular in the past ten years or so (see Merges 2000; Vertinsky 2012). These, together with the continued deployment of economic modeling and analysis, represent important elements of the diversity of methods that now characterize the field. And beyond economics, all manner of alternative methodologies are in play as well (see, e.g., Merges 2011; Dinwoodie 2013). In the sections that follow, I describe what I see as the main thrust of several of the most important methodologies being used in the IP field today. As with everything in this chapter, though I try to be comprehensive, the result is no doubt a personal and idiosyncratic survey of the literature.

8.3.2. Models and Analytics Even while a host of newer methods has been brought to bear on IP-​related issues, scholars continue to use traditional approaches as well. For example, economic modeling still plays an important part in the discussion of IP-​related issues. One rich source of modeling-​based studies is the classic article, which a contemporary scholar will explore anew with updated insights. In this genre, consider for example Duffy (2004)—​a thorough reexamination of Edmund Kitch’s classic “prospect theory” of patents (1977). Other recent work considers different types of models, such as Yoo (2004), which describes a product differentiation view of copyright law. Still other work pushes classic insights a step further: for example, Lunney (1996), who argues that allocative efficiency, and not the classic “incentive-​access tradeoff,” ought to drive copyright policy. And an excellent unified synthesis of the economics of IP and innovation is Scotchmer (2006). At the same time, IP has caught the attention of scholars whose interests lie with theorizing about property rights generally. So, for example, several important articles by Henry Smith have pushed his highly analytic synthesis of property theory into the domain of IP law. See Smith (2007, 2009). So too with work by scholars like Fennell (2004, 2009), and Merrill (2009). In each case, IP provides examples for an analytic approach to important issues in property law. The authors integrate IP into the general fabric of property, using it for examples of a new way to conceive or categorize some longstanding bodies of property theory. Merrill, for example, argues that certain patent doctrines embody the property principle of “accession,” or extension of ownership to natural outgrowths of previously owned works. And Fennell (2009) points to interesting twists in IP inalienability rules in support of her argument for a more flexible conception of alienability generally. Still another area where general property theory

 

ECONOMICS OF INTELLECTUAL PROPERTY LAW    211 overlaps with treatment of IP issues is in the study of IP access regimes as a case study in governance of a common resource. See Samuelson (2003).

8.3.3. Experimental Studies A series of articles by Sprigman and Buccafusco describe experimental work on IP rights. One demonstrates that a special version of the well-​known endowment effect28 exists with respect to works covered by IP rights. The article reports that cognitive biases cause creators and owners of IP to set the price for IP considerably higher than what buyers, on average, are willing to pay (Sprigman and Buccafusco 2010). The research therefore suggests that IP transactions may occur at suboptimal levels. They report that cognitive and affective biases are likely to be more serious for transactions in works of relatively low commercial value, or for which no well-​established custom or pattern helps to inform valuation. The article explores the implications their findings have for the structuring of IP rights, IP formalities, IP licensing, and fair use. In a related article, Sprigman and Buccafusco (2011) try to move beyond the simple endowment effect story, to explore valuation issues when the seller of a work covered by IP rights is also the original creator of that work. Not surprisingly perhaps, they find an even stronger version of the endowment effect. They dub this “the creativity effect.” In a cleverly structured series of experiments, they not only confirm the existence of the endowment effect, they also show that when someone creates a work he or she sets an even higher valuation on it than when an owner/​possessor considers selling it. That is, they separate creators from mere owners. This is important in the IP field because IP rights are often justified by the fact that they permit original creators to market their works, which supports creative autonomy and can enhance creators’ incomes. Their conclusion is that markets for IP are systematically distorted by subjective bias. The authors make some interesting points regarding policy implications. An oft-​discussed feature of IP law centers on the fact that large companies own the rights to the creative works of their employees. In the usual discussion, this is seen as a problem, in that incentives to create are diluted by corporate ownership. Sprigman’s results suggest that corporate ownership in fact may not be such a bad thing. By removing the rights over IP-​protected works from the hands of individual creators, aggregating ownership in the hands of large companies may in fact facilitate more efficiency in IP markets. The final paper tests the value of attribution (having one’s name attached to one’s creative works) and publication (reaching an actual audience) in IP markets. The findings here are based again on experiments involving subjects and carefully constructed scenarios. The authors show that artists (in this case, photographers, and painters) place a very high value on having their names attached to their works. This is established by a series of experimental trials in which some artists are given the option of retaining

28 

See “Endowment Effect,” Wikipedia, http://​en.wikipedia.org/​wiki/​Endowment_​effect.

 

212   ROBERT P. MERGES attribution; the value of this right is determined by looking at the price demanded by artists when they retain attribution and the price when they do not. In other words, the authors of the study have a methodology that allows them to infer the value artists attach to attribution. The same is true of the value they place on publication, which again is determined by looking at the price differential under two conditions. Despite its sophistication, this body of work remains somewhat ambiguous as regards its policy implications. The most important issue here is implicit baselines concerning the optimal rate of transactions. On the “creativity effect,” Sprigman and Buccafusco (2011) argue that the valuations placed on creative works by potential buyers and by noncreator “owners” (those who were simply given a work created by someone else) are more reliable guides to the actual market value of those creative works. This sets up the policy conclusion that markets for IP-​protected works are often inefficient because creators themselves overvalue their works. But if intrinsic value is treated neutrally, then a different conclusion might follow: many artists choose to retain sole ownership of their works because, having created them, they are the highest-​value owners on the scene. From this perspective the issue is not “failed transactions” but instead the importance of intrinsic valuation when it comes to creative works. In a related vein, it might be that the studies here show aggregate tendencies. But this does not drive a strong policy conclusion, because all it takes is for some outliers—​some “creative professionals” whose subjective valuation is closer to that of buyers—​to make a market. Under this interpretation, high-​valuation creators will remain outside the market, but low-​valuing creators will enter it. This could in turn drive a sort of “reverse market for lemons” dynamic. “High valuers” will be driven out of the market, because they can’t sell their works at market-​ clearing prices. And “low valuers” will come to predominate. If this became a trend, it would undermine the notion that disparate valuation makes IP markets too “thin.” Another line of research, conducted primarily by experimental psychologists, eschews large-​bore policy prescriptions while testing what might be called the development of intuitions regarding IP rights. The chief characteristic of these studies is that they take insights from cognitive psychology and test their relevance to IP-​specific scenarios. Some study the development of children’s intuitions about ownership (Fasig 2000; Friedman and Neary 2008; Neary and Friedman 2013). Others show that children have an intuitive appreciation for the value of labor in justifying ownership (Kangiesser and Hood 2013). A fascinating branch of this field with special relevance to IP studies developmental cognition, testing whether children have any sense of idea ownership, and if so, at what age. So, for example, cognitive scientists have shown that children as young as 6 years old “apply simple rules of ownership to ideas as well physical objects (Shaw, Li, and Olson 2012, 1389). Further, they also note that young children sense that some ideas are inherently common property, and they will not apply ownership concepts and rules to a common word, for example: “[C]‌hildren and adults used first possession for determining ownership of ideas, but not words” (2012, 1398). It is also worth noting that, according to another set of studies, children begin to recognize and censure plagiarism between the ages of 4 and 6. See Olson and Shaw (2011).

 

ECONOMICS OF INTELLECTUAL PROPERTY LAW    213 There is room here for much more nuance. A critical issue, for example, is whether children’s intuitions about idea ownership are, as some researchers suggest, culture-​ specific. If so, the studies are noteworthy in showing cultural influences over moral intuitions at a very early age. If not, they support a more universalistic claim: that IP rights are rooted in moral intuitions that develop early in the human mind. In any event, it is safe to say that there are a good number of fascinating questions left to explore in this area, and experimental studies of IP law may continue to enlighten the field for some time to come.

8.4. Conclusion I have emphasized two major trends in the economics of IP rights: An increased sense of IP in context, and the growth of methodological diversity. Taken together, they have the field poised for deeper insights and greater policy relevance than ever before. By studying how IP works in various contexts, scholars have progressed significantly from the stylized, simplistic models of the past. Each study is distinct, taking as its starting point a specific economic setting. But taken together, they reveal the many diverse ways IP rights exert an influence on economic behavior. The contrast with older scholarship is striking: contextual studies begin not with simple models but with real-​world industries and practices, and ask how IP rights play into various features of the setting under study. The payoff is a much better understanding of the distinct contribution IP rights make to economic life. Though a unified view of these rights (e.g., as simple incentives or as economic monopolies) is of course sacrificed, a multifaceted reality is revealed that is infinitely more varied and interesting than the simple worldview of the old models. In one sense, the benefits of methodological diversity are even more self-​evident. Each approach, indeed each individual study, has something to teach us about IP rights. An empirical study of variability among patent examiners, for example, might lead to a call for better Patent Office quality control or it might simply push private actors to diversify their patent filings among different examining groups. Better ways of grouping and organizing doctrines might stimulate new thoughts on core features of IP systems. The advantages are as diverse as the methods and studies themselves. But a more significant payoff comes when multiple methodologies reach the same conclusion. Convergence of this sort sends a powerful message. We can more confidently argue for policy prescriptions when multiple scholars, using different tools, arrive at the same conclusion. So, for example, when economic modeling, ethnographic interviews, and large-​scale event studies all indicate that extending the term of copyright protection adds nothing to creators’ incentives, we can be confident in advocating against further increases in the length of copyright. At the very least, consensus on this scale can help reveal the naked power of lobbying groups; a policy proposal that is

 

214   ROBERT P. MERGES opposed by all serious scholars, but which gains political traction nonetheless, teaches just how completely legislative muscle can trump objective policy analysis. Methodological diversity is actually essential if we are to have real confidence in our policy positions. As with the natural sciences, a successful strategy for understanding complex issues such as the place of IP rights requires two distinct phases. The first is highly reductionist: a small feature of the complex system is isolated and studied in great depth. Then, when many such reductionist studies are available, the disparate results are sifted and collected into a new synthesis. In this second phase, the disparate pieces are reassembled into a comprehensive new understanding. This new synthesis can be robust precisely because it is built on many solid, discrete studies. Real confidence comes when we can see the overall pattern that emerges from this diversity of building block studies. To reach this point in the study of IP rights, we will need many more diversified studies of discrete phenomena. But as these pile up, we will arrive at new syntheses that provide a firmer foundation for policy than the field has ever known.

References Abrams, David, Akcigit, Ufuk, and Popadak, Jillian A. 2013. “Patent Value and Citations:  Creative Destruction or Strategic Disruption?” University of Pennsylvania, Institute for Law & Economics Research Paper, No. 13-​23. http://​papers.ssrn.com/​sol3/​ papers.cfm?abstract_​id=2351809. [Accessed 20 August 2016]. Adelman, Martin J. 1977. “Property Rights Theory and Patent-​ Antitrust:  The Role of Compulsory Licensing.” New York University Law Review 52, pp. 77–​121. Alcacer, Juan and Gittelman, Michelle. 2006. “Patent Citations as a Measure of Knowledge Flows:  The Influence of Examiner Citations.” Review of Economics and Statistics 88(4) (Nov.), pp. 774–​779. Allison, John R. and Lemley, Mark A. 1998. “Empirical Evidence on the Validity of Litigated Patents.” AIPLA Quarterly Journal 26, pp. 185–​274. Allison, John R., Lemley, Mark A., Moore, Kimberly A., and Trunkey, Derek R. 2004. “Valuable Patents.” Georgetown Law Journal 92, pp. 435–​483. Allison, John R., Dunn, Abe, and Mann, Ronald J. 2007. “Software Patents, Incumbents and Entry.” Texas Law Review 85, pp. 1579–​625. Arora, Ashish and Merges, Robert P. 2004. “Specialized Supply Firms, Property Rights and Firm Boundaries.” Industrial and Corporate Change 13, pp. 451–​472. Bar-​Gill, Oren and Parchomovsky, Gideon. 2009. “Law and the Boundaries of Technology-​ Intensive Firms.” University of Pennsylvania Law Review 157, pp. 1649–​1675. Barnett, Jonathan M. 2004. “Private Protection of Patentable Goods.” Cardozo Law Review 25, pp. 1251–​1313. Barnett, Jonathan M. 2011a. “Intellectual Property as a Law of Organization.” Southern California Law Review 84, pp. 785–​857. Barnett, Jonathan M. 2011b. “The Host’s Dilemma: Strategic Forfeiture in Platform Markets for Informational Goods.” Harvard Law Review 124, pp. 1861–​1936. Bechtold, Stefan and Tucker, Catherine. Forthcoming. [2013]. “Trademarks, Triggers and Online Search.” Journal of Empirical Legal Studies. https://​papers.ssrn.com/​sol3/​papers. cfm?abstract_​id=2266945 [Accessed 8 August 2016].

 

ECONOMICS OF INTELLECTUAL PROPERTY LAW    215 Beebe, Barton. 2008. “An Empirical Study of U.S. Copyright Fair Use Opinions, 1978–​2005.” University of Pennsylvania Law Review 156, pp. 549–​638. Buccafusco, Christopher and Heald, Paul. 2013. “Do Bad Things Happen When Works Enter the Public Domain? Empirical Tests of Copyright Term Extension.” Berkeley Tech. Law Journal 28, pp. 1–​44. Burk, Dan L. and McDonnell, Brett H. 2007. “The Goldilocks Hypothesis:  Balancing Intellectual Property Rights at the Boundary of the Firm.” University of Illinois Law Review 2007, pp. 575–​595. Clark, John Bates. 1927. Essentials of Economic Theory. New York: Macmillan. Cockburn, Ian, Kortum, Samuel, and Stern, Scott. 2003. “Are All Patent Examiners Equal? Examiners, Patent Characteristics and Litigation Outcomes.” In:  Stephen A. Merrill, ed., Patents in the Knowledge-​Based Economy. Washington DC: National Academies Press. 19–​53. Cohen, Wesley, Nelson, Richard R., and Walsh, John P. 2000. “Protecting Their Intellectual Assets: Appropriability Conditions and Why U.S. Manufacturing Firms Patent (or Not).” Working Paper No. 7552, National Bureau of Economic Research. Dinwoodie, Graeme B. 2013. Methods and Perspectives in Intellectual Property. Cheltenham: Edward Elgar. Dosi, Giovanni, Malerba, Franco, and Silva, Francesco. 2006. “Information, Appropriability, and the Generation of Innovative Knowledge Four Decades After Arrow and Nelson: An Introduction.” Industrial and Corporate Change 15, pp. 891–​919. Duffy, John. 2004. “Rethinking the Prospect Theory of Patents.” University of Chicago Law Review 71, pp. 439–​499. Eguchi, Aya. 2011. “Note, Curtailing Copycat Couture: The Merits of the Innovative Design Protection and Piracy Prevention Act and A Licensing Scheme for the Fashion Industry.” Cornell Law Review 97, pp. 131–​156. Eisenmann, Thomas, Parker, Geoffrey, and Van Alstyne, Marshall. 2009. “Opening Platforms:  How, When and Why?” In:  Annabelle Gawer, ed., Platforms, Markets and Innovation. Cheltenham: Edward Elgar. 131–​162. Ellickson, Robert. 1993. “Property in Land.” Yale Law Journal 102, pp. 1513–​616. Fasig, Lauren G. 2000. “Toddlers’ Understanding of Ownership: Implications for Self-​Concept Development.” Social Development 9, pp. 370–​382. Fauchart, Emmanuelle and von Hippel, Eric. 2008. “Norms-​Based Intellectual Property Systems: The Case of French Chefs.” Organization Science 19, pp. 187–​207. Fennell, Lee Anne. 2004. “Common Interest Tragedies.” Northwestern Law Review 98, pp. 907–​990. Fennell, Lee Anne. 2009. “Adjusting Alienability.” Harvard Law Review 122, pp. 1403–​1465. Friedman, Ori and Neary, Karen R. 2008. “Determining Who Owns What: Do Children Infer Ownership from First Possession?” Cognition 107, pp. 829–​849. Galasso, Alberto and Schankerman, Mark. 2010. “Patent Thickets, Courts, and the Market for Innovation.” Rand Journal of Economics 41(3, Autumn), pp. 472–​503. Graham, Stuart J. H., Merges, Robert P., Samuelson, Pamela, and Sichelman, Ted. 2010. “High Technology Entrepreneurs and the Patent System:  Results of the 2008 Berkeley Patent Survey” Berkeley Technology Law Journal 24, pp. 1256–​1327. Hall, Bronwyn H., Jaffe, Adam, and Trajtenberg, Manuel. 2005. “Market Value and Patent Citations.” Rand Journal of Economics 36(1, Spring), pp. 16–​38. Heald, Paul J. 2005. “A Transaction Costs Theory of Patent Law.” Ohio State Law Journal 66, pp. 473–​498.

 

216   ROBERT P. MERGES Hemphill, C. Scott and Suk, Jeannie. 2009. “The Law, Culture, and Economics of Fashion.” Stanford Law Review 61, pp. 1147–​1155. Kanngiesser, Patricia and Hood, Bruce. 2013. “Not by Labor Alone: Considerations for Value Influence Use of the Labor Rule in Ownership Transfers.” Cognitive Science 37, p. 8. Kitch, Edmund. 1966. “Graham v. John Deere Co.: New Standards for Patents.” Supreme Court Review 1966, pp. 293–​346. Kitch, Edmund. 1977. “The Nature and Function of the Patent System.” Journal of Law and Economics 20, pp. 265–​291. Krim, Tanya. 2007. “There’s nothing ‘Trivial’ About the Pursuit of the Perfect Bag.” Brandweek March 29. Lee, Thomas R., DeRosia, Eric D., and Christensen, Glenn L. 2009. “An Empirical and Consumer Psychology Analysis of Trademark Distinctiveness.” Arizona State Law Journal 41, pp. 1033–​1109. Lemley, Mark A. and McKenna, Mark P. 2012. “Is Pepsi Really a Substitute for Coke?” Georgetown Law Review 100, pp. 2055–​2117. Levin, Richard C., Klevorick, Alvin, Nelson, Richard R., and Winter, Sidney. 1987. “Appropriating the Returns from Industrial Research and Development.” Brookings Papers on Economic Activity 3, pp. 783–​831. Los Angeles Times. 2013. “U.S. Copyright Industries Add $1 Trillion to GDP.” Nov. 19, 2013. http://​articles.latimes.com/​2013/​nov/​19/​business/​la-​fi-​ct-​intellectual-​property-​20131119 [Accessed 20 August 2016]. Loshin, Jacob. 2010. “Secrets Revealed: Protecting Magicians’ Intellectual Property Without Law.” In:  Christine A. Coros, ed., Law and Magic:  A  Collection of Essays. Durham, NC: Carolina Academic Press. 123–​148. Lunney, Glynn S., Jr. 1996. “Reexamining Copyright’s Incentive-​Access Paradigm.” Vanderbilt Law Review 49, pp. 483–​655. Mann, Ronald J. and Sager, Thomas. 2007. “Patents, Venture Capital and Software Startups.” Research Policy 36, pp. 193–​208. Mansfield, Edwin. 1985. “How Rapidly Does New Industrial Technology Leak Out?” Journal of Industrial Economics 34, pp. 217–​223. Mansfield, Edwin. 1986a. “Patents and Innovation: An Empirical Study.” Management Science 32, pp. 173–​181. Mansfield, Edwin. 1986b. “The R&D; Tax Credit and Other Technology Policy Issues.” American Economic Review, Papers and Proceedings 76, pp. 191–​197. Mansfield, Edwin, Rapoport, John, Romeo, Anthony, Wagner, Samuel, and Beardsley, George. 1977. “Social and Private Rates of Return from Industrial Innovations.” Quarterly Journal of Economics 71, pp. 221–​240. Mashima, Rieko. 1996. “The Turning Point for Japanese Software Companies:  Can They Compete in the Prepackaged Software Market?” Berkeley Technology Law Journal 11, pp. 429–​459. McKenna, Mark P. 2009. “Testing Modern Trademark Law’s Theory of Harm.” Iowa Law Review 95, pp. 63–​116. Merges, Robert P. 1996. “A Comparative Look at Property Rights and the Software Industry.” In: David Mowery, ed., The International Computer Software Industry: A Comparative Study of Industry Evolution and Structure. Oxford: Oxford University Press. 272–​303. Merges, Robert P. 2000. “Intellectual Property Rights and the New Institutional Economics.” Vanderbilt Law Review 53, pp. 1857–​1876.

 

ECONOMICS OF INTELLECTUAL PROPERTY LAW    217 Merges, Robert P. 2001. “One Hundred Years of Solicitude: Intellectual Property Law 1900-​ 2000.” California Law Review 88, pp. 2187–​2233. Merges, Robert P. 2005. “A Transactional View of Property Rights.” Berkeley Technology Law Journal 20, pp. 1477–​1520. Merges, Robert P. 2008. Intellectual Property Rights and Technological Platforms. Working Paper, University of Berkeley School of Law. http://​papers.ssrn.com/​sol3/​papers. cfm?abstract_​id=1315522 [Accessed 20 August 2016]. Merges, Robert P. 2011. Justifying Intellectual Property. Cambridge: Harvard University Press. Merges, Robert P. 2011a. “Autonomy and Independence: The Normative Face of Transaction Costs.” Arizona Law Review 53, pp. 145–​163. Merges, Robert P. 2013. “What Can We Learn from IP’s Negative Spaces?” Media Institute—​ Intellectual Property Issues. http://​www.mediainstitute.org/​IPI/​2013/​120913.php [Accessed 20 August 2016]. Menell, Peter S. 1987. “Tailoring Legal Protection for Computer Software.” Stanford Law Review 39, pp. 1329–​1372. Menell, Peter S. 1989. “An Analysis of the Scope of Copyright Protection for Application Programs.” Stanford Law Review 41, pp. 1045–​1104. Menell, Peter and Scotchmer, Suzanne. 2007. “Intellectual Property.” In: A. Mitchell Polinsky and Steven Shavell, eds., Handbook of Law and Economics. Elsevier. Merrill, Thomas W. 2009. “Accession and Original Ownership.” Journal of Legal Analysis 1, pp. 459–​510. Neary, Karen R. and Friedman, Ori. 2013. “Young Children Give Priority to Ownership When Judging Who Should Use an Object.” Child Development 84, p. 6 Nelson, Richard R., ed. 1962. The Rate and Direction of Inventive Activity. Princeton: Princeton University Press. Nelson, Richard R. and Rosenberg, Nathan. 1993. National Innovation Systems: A Comparative Analysis. New York: Oxford University Press. Nordhaus, William D. 1969. Invention, Growth and Welfare:  A  Theoretical Treatment of Technological Change. Cambridge, MA: MIT Press. North, Douglass. 1990. Institutions, Institutional Change and Economic Performance. Cambridge: Cambridge University Press. NPD Group. 2008. “The U.S. Apparel Market 2007 Dresses Up … Way Up.” Business Wire. Mar. 18. http://​www.businesswire.com/​news/​home/​20080318005099/​en/​U.S.-​Apparel-​Market-​ 2007-​Dresses-​Up...Way [Accessed 21 August 2016.] Olson, Kristina R., and Shaw, Alex, 2011. “‘No Fair, Copycat!’:  What Children’s Response to Plagiarism Tells Us About Their Understanding of Ideas,” Developmental Science 14 , pp. 431–​439. Perzanowski, Aaron. 2013. “Tattoos and IP Norms.” Minnesota Law Review 98, pp. 511–​567. Pigou, A.C. 1924. The Economics of Welfare. 2nd ed. London: Macmillan. Plant, Arnold. 1974 [1934]. “The Economic Theory Concerning Patents for Inventions.” In: A. Plant, Selected Economic Essays and Addresses of Arnold Plant. London:  Routledge & Kegan Paul. 35–​56. Priest, George L. 1986. “What Economists Can Tell Lawyers about Intellectual Property: Comment on Cheung.” Research in Law and Economics 8, pp. 19–​24. Radin, Margaret J. 1982. “Property and Personhood.” Stanford Law Review 34, pp. 957–​1015. Radin, Margaret J. 1987. “Market Inalienability.” Harvard Law Review 100, pp. 1849–​1937. Raustiala, Kal and Sprigman, Christopher. 2012. The Knockoff Economy. New York: Oxford University Press.

 

218   ROBERT P. MERGES Rawls, John. 1971. A Theory of Justice. Cambridge, MA: Harvard University Press. Rawls, John. 1993. Political Liberalism. Cambridge, MA: Harvard University Press. Rosenblatt, Elizabeth L. 2011. “A Theory of IP’s Negative Space.” Columbia Journal of Law and the Arts 34, pp. 317–​341. Sag, Matthew. 2012. “Predicting Fair Use.” Ohio State Law Journal 73, pp. 47–​114. Samuelson, Pamela. 2003. “Mapping the Digital Public Domain.” Law and Contemporary Problems 66, pp. 147–​171. Scafidi, Susan. 2008. “F.I.T.:  Fashion as Information Technology.” Syracuse Law Review 59, pp. 69–​81. Schultz, Mark F. 2006. “Fear and Norms and Rock & Roll: What Jambands Can Teach Us About Persuading People to Obey Copyright Law.” Berkeley Technology Law Journal 21, pp. 651–​681. Scotchmer, Suzanne. 2006. Innovation and Incentives. Cambridge, MA: MIT Press. Shaw, Alex, Li, Vivian, and Olson, Kristina R. 2012. “Children Apply Principles of Physical Ownership to Ideas,” Cognitive Science, 36, pp. 1383–​1403. Sheff, Jeremy. 2011. “Biasing Brands.” Cardozo Law Review 32, pp. 1245–​1314. Simcoe, Timothy S. 2008. “Open Standards and Intellectual Property Rights.” In:  Henry Chesbrough, Wim Vanhaverbeke, and Joel West, eds., Open Innovation: Researching a New Paradigm. New York: Oxford University Press. 161–​183. Smith, Henry. 2007. “Intellectual Property as Property:  Delineating Entitlements in Information.” Yale Law Journal 116, pp. 1742–​1827. Smith, Henry. 2009. “Institutions and Indirectness in Intellectual Property.” University of Pennsylvania Law Review 157, pp. 2083–​2133. Social Science Research Network. N.d. Intellectual Property: Empirical Studies eJournal. Ed. Christopher Buccafusco and David Schwartz. http://​papers.ssrn.com/​sol3/​Jeljour_​results. cfm?nxtres=121&form_​name=journalbrowse&journal_​id=1649836&Network=no&SortOr der=ab_​approval_​date&stype=desc&lim=false [Accessed 22 August 2016]. Sprigman, Christopher, and Buccafusco, Christopher J. 2010. “Valuing Intellectual Property: An Experiment.” Cornell Law Review 96, pp. 1–​45. Sprigman, Christopher, and Buccafusco, Christopher J. 2011. “The Creativity Effect.” University of Chicago Law Review 78, pp. 31–​52. Sprigman, Christopher and Oliar, Dotan. 2008. “There’s No Free Laugh (Anymore):  The Emergence of Intellectual Property Norms and the Transformation of Stand-​Up Comedy.” Virginia Law Review 94, pp. 1787–​1856. Sprigman, Christopher and Raustiala, Cal. 2006. “The Piracy Paradox:  Innovation and Intellectual Property in Fashion Design.” Virginia Law Review 92, pp. 1687–​1746. Sprigman, Christopher J., Buccafusco, Christopher J., and Burns, Zachary C. 2013. “Valuing Attribution and Publication in Intellectual Property.” Boston University Law Review 93, pp. 1389–​1435. Teece, David. 1986a. “Profiting from Technological Innovation: Implications for Integration, Collaboration, Licensing and Public Policy.” Research Policy 15, pp. 285–​305. Teece, David. 1986b. “Firm Organization, Industrial Structure and Technological Innovation.” Journal of Economic Behavior and Organization 31, pp. 193–​211. Varian, Hal, and Shapiro, Carl. 1998. Information Rules. Boston: Harvard Business School Press. Vertinsky, Liza. 2010. “Comparing Alternative Institutional Paths to Patent Reform.” Alabama Law Review 61, pp. 501–​550. Vertinsky, Liza. 2012. “An Organizational Approach to the Design of Patent Law.” Minnesota Journal of Law, Science and Technology 13, pp. 211–​262.

 

ECONOMICS OF INTELLECTUAL PROPERTY LAW    219 Webster, Elizabeth, Palangkaraya, Afons, and Jensen, Paul H. 2007. “Characteristics of International Patent Application Outcomes.” Economics Letters 95, pp. 362–​368. Yoo, Christopher. 2004. “Copyright and Product Differentiation.” New York University Law Review 79, pp. 212–​280. Ziedonis, Rosemarie. 2004. “Don’t Fence Me In: Fragmented Markets for Technology and the Patent Acquisition Strategies of Firms.” Management Science 50(6), pp. 804–​820.

 

Chapter 9

T R ADEM ARKS A ND U NFA I R C OM PETI T I ON Clarisa Long

9.1.  Introduction The patent and copyright forms of intellectual property protection both turn on the same underlying dilemma: how to solve the public goods problem. Inventions and artistic creations, if unprotected, become public goods once revealed. Legal rules allow individuals to maintain some elements of control over information that would otherwise be lost to the public domain. By granting exclusive rights to the creative work, legal rules make it possible for creators to be compensated for that revelation.1 Without protection, not enough information will be produced, but with protection, information will be suboptimally distributed. This is an old and familiar story. But that motif does not squarely fit with the traditional justification of trademark law, which has long been classified as a subpart of the law of unfair competition.2 While the stated purpose of the patent and copyright regimes has long been to encourage the creation of inventions and expressive goods—​to “promote the progress of science and the useful arts”3—​trademark law’s historic goal was not to encourage the creation of branded goods per se. As a part of unfair competition law, trademark protection, and its

1  See, e.g., Graham v John Deere Co., 383 U.S. 1, 9(1965) (describing the patent system as “a reward, an inducement, to bring forth new knowledge”); Easterbrook (1984, 21–​22) (observing that inventors would not make information public without the promise of compensation); Machlup (1958) (discussing the “exchange for secrets” theory of the patent system). 2  Two of the classic papers establishing the law and economics approach to trademark law are Landes and Posner (1987, 265) and Economides (1988, 523). 3  See US Const. Art. I, § 8, cl. 8 (“[Congress shall have the power] [t]‌o promote the Progress of Science and useful Arts, by securing for limited Times to Authors and Inventors the exclusive Right to their respective Writings and Discoveries”).

 

TRADEMARKS AND UNFAIR COMPETITION    221 cousin, false advertising law, was a strand of the common law’s general attempt to implement standards of commercial morality and fair dealing in economic markets. The forerunner to unfair competition law was concerned with protecting established merchants’ entitlement to custom.4 The common law tort of passing off, in which merchant A would represent its own (usually inferior) goods as those of merchant B, was originally made actionable in order to prevent competitors from improperly diverting business from producers in the marketplace. (The common law tort of reverse passing off, in which merchant A buys merchant B’s goods and then falsely represents those goods as being merchant A’s own goods, is relatively rare for obvious reasons. In cases of reverse passing off, merchant A is most likely taking an economic loss and selling merchant B’s goods at or below their purchase price in order to attract customers.) As markets expanded in their geographical scope and market structure became more complex, the relationships between producers and consumers became more attenuated. Consumers no longer necessarily knew the producer of a set of goods personally or as a member of a local community. Trademarks thus began to be important in advertising and product recognition. One commentator has argued that ownership of a trademark was “assigned to the person who adopted the mark for her trade, not because she had created it or its favourable associations, but because such person was conveniently placed and strongly motivated to vindicate the broader public interest in a mark’s ability to identify accurately the source of the goods to which it was attached.”5 By being able to differentiate the products of competing sellers, trademarks allowed consumers to associate a product’s qualities with a single source. One goal of trademark law became, therefore, to police the marketplace to protect consumers from confusion as to the source of a good arising from confusingly similar trademarks being applied to competing goods. On this theory, trademark law served as one way of policing competitive markets and thereby as providing an indirect form of consumer protection. Over the past century, however, the rise of the regulatory state means that trademark law in particular, and unfair competition law more generally, are no longer the dominant modes of regulating competitive behavior in the market. As trademark law in the United States has evolved, one strand of trademark theory—​dilution law—​has gained traction. Trademark dilution law treats trademarks as being valuable for the goodwill they embody. Producers who build up positive associations in the consumers’ minds with the mark or who otherwise amass goodwill from their marked products can reap the benefit, so long as the producer can continue to control the way in which the mark is used. Conferring on trademark holders the power to exclude third parties from nonconfusing uses of the marks allows mark owners to reap the benefits of their investment in product quality. Under dilution theory, the emphasis is not on protecting consumers, but in incentivizing producers to create goodwill.

4  Restatement (3d) of Unfair Competition § 9, cmt. c (1995) (discussing concerns with “the unfairness inherent in a diversion of trade from the owner of the mark through a fraudulent use of the trademark.”). 5  Lunney (1999, 367, 417).

 

222   CLARISA LONG Across the various fields and subfields of modern law, the law and economics mode of analysis, with its emphasis on economic efficiency, has been a more powerful explanatory tool in some fields than it is in others. Trademark law has been one of the fields for which law and economics theories have had explanatory traction. (Unfair competition law generally, with its emphasis on equity, has been more resistant to traditional law and economics analysis.) Commentators have recognized that the law and economics model of trademark law has been well accepted and indeed is the dominant model in theoretical explanations for trademark law today.6 I posit that there are two main reasons for this. The first is to trademark law’s common law foundation. Although Congress comprehensively codified US trademark law in 1946 in the Lanham Act, the statute made few changes to the common law. As a result, the field’s underlying strata of judge-​made rules, many of which can be justified in efficiency terms, remained largely unchanged. The second is the relative balance of power between the courts and Congress. Unlike copyright law, in which Congress has been the dominant evolutionary force in recent decades, Congress delegated interpretive authority over the Lanham Act to the courts.7 Over the past few decades, courts have increasingly used the federal Lanhan Act as a remedy for matters that had long been considered subsumed under state unfair competition law, such as false advertising and copying of trade dress. Courts have maintained an active presence even after trademark law’s codification, which has provided a counterbalancing force to the pressures of interest groups on the legislative process.8 This chapter will summarize the dominant law and economic approach to some of the doctrines comprising the two major strands of modern trademark law: what I call classic trademark law, in which trademark protection is justified as reducing consumer search costs and preventing consumer confusion, and dilution law, which is explained by the more traditional intellectual property justifications of encouraging investment in creating valuable marks and protecting producer goodwill. It does not purport to be a comprehensive description of the field, but rather provides a light overview.

9.2.  The Law and Economics of Classic Trademark Law Trademark law, at least in its classical form, is directed at preventing harm to consumers through the misleading or confusing use of trademarks. Trademarks indicate the source or origin of goods.9 Because many things can serve as source identifiers, trademarks are not confined to words, phrases, pictures, logos, symbols, and other merely literary 6 

See Barton (2004, 621). See Leval (2004, 187, 198). 8  For a discussion of this, see Long (2006, 1029). 9  See 15 U.S.C. § 1127 (2012) (defining trademark as “any word, name, symbol, or device, or any combination thereof [used] to identify and distinguish [a person’s] goods”). 7 

 

TRADEMARKS AND UNFAIR COMPETITION    223 elements. The look and feel of a product, a product configuration, a color, sound, or scent may serve as a source identifier.10 Thus, in this chapter, I refer to all product identifiers protected under the Lanham Act as “marks” and “trademarks,” even though some of these may actually be trade dress or service marks. I shall also follow the convention of indicating a word or logo used as a trademark by displaying it in all capital letters.

9.2.1. Some Benefits of Trademarks Law and economics has posited two economizing functions of modern day trademarks. The first of these is to create symbols that are shorthand for a product’s source. The second is that trademarks communicate something about the trademark holder itself as opposed to a particular product. Balanced against the economizing functions of trademarks, however, must be considered the social costs of trademark protection. The traditional justification for trademarks is that they reduce consumer search costs. On this view, trademarks are a compact and efficient means to communicate information to consumers. Consumers use the mark as source identifier, expecting that goods branded with the mark will be of the quality they have come to associate with past purchases bearing the mark. Use of confusingly similar marks on competing goods may cause consumers mistakenly to purchase goods from another source whose quality may differ from the goods the consumer intended to purchase. In the absence of trademark law, unscrupulous competitors could take advantage of the association between mark and quality in order to “pass off ” their own goods as those of another goods producer. Trademark law thus benefits consumers by allowing them to engage in heuristics that, on balance, save them time and mental processing, as well as protecting consumers from some kinds of deception and unfair competition in the marketplace. Classical trademark theory is directed at protecting this source of trademark value as a reducer of consumer search costs. Producers of competing goods differentiate their goods in various ways, often by varying the attributes of the good. For example, Coke has orange-​flavored undertones, whereas Pepsi tends toward lemon-​based notes. Coke has a higher degree of carbonation, whereas Pepsi is sweeter.11 Some of the attributes of goods are readily observable; others cannot be assessed simply by inspection of the good. On this view, the less readily observable the product attributes, the greater the social value of trademarks. Search costs will vary depending upon the nature of the good. Because consumers can use a trademark as a proxy for a set of hard-​to-​measure product attributes, the 10 

See, e.g., Qualitex Co. v Jacobson Prods. Co., 514 U.S. 159, 162 (1995) (including “color within the universe of things that can qualify as a trademark”); Two Pesos, Inc. v Taco Cabana, Inc., 505 U.S. 763, 767 (1992) (finding restaurant decor to be protected trade dress); In re Clarke, 17 U.S.P.Q.2d (BNA) 1238, 1238 (T.T.A.B. 1990) (scent); In re Gen. Elec. Broad. Co., 199 U.S.P.Q. (BNA) 560, 563 (T.T.A.B. 1978) (sound). 11  For a discussion of the differences between Coke and Pepsi drinks and an example of how a consumer taste test is performed, see Thumin (1962, 358, 358–​59).

 

224   CLARISA LONG identification function of trademarks is especially important for experience goods, or goods that yield information about their qualities over time.12 For example, a particular model of car is an experience good because latent defects or performance advantages generally become apparent only after purchasers have used the car for some time. Reliance goods are particularly hard for consumers to judge on their own. Consumers either lack access to information about the good or lack the expertise to evaluate the available information. Instead, consumers must rely on the assessment of some trusted third party, usually an expert, whose expertise allows the expert to assess the good on the consumers’ behalf. Examples of this include electrical equipment, medical advice, or cryptographic software. Underwriters’ Laboratories can probably find a defect in a lamp faster than the average consumer can. (Most consumers would probably prefer that Underwriter’s Laboratories identify a problem in an electrical device, rather than consumers stumbling on it themselves.) Similarly, consumers often don’t know if medical advice is any good without using some proxy for quality such as asking for a second opinion. Sometimes a consumer can make an assessment on her own, but most of the time she can’t. Trademarks and certification marks are helpful for consumers to assess reliance goods, because the reputation of the producer or the certifier stands as a proxy for the quality of the good. In essence, for both experience goods and reliance goods, the mark encapsulates the experience of the consumer or other consumers, becoming a sort of “expert” signal on which the consumer relies. By contrast, inspection goods such as produce readily yield information about their attributes, often by casual observation.13 Trademarks are less valuable in reducing search costs for inspection goods. Consumer benefit, and by extension social benefit, is enhanced by making sure that when consumers rely on source identifiers, such marks are not used in a confusing manner. The social cost is that competitors of a trademark holder incur avoidance costs; they must create their own nonconfusing trademark. Trademarks and advertising send another kind of signal as well, and that is a self-​ referential message. A consumer can learn something about the company, even if they don’t learn much about a product, from the advertising surrounding the product. One message might be “we advertise, and therefore we must sell a good of sufficiently high quality that we can afford this high-​cost expenditure.” Another message might be “through our advertising you can detect a certain kind of corporate culture or vision which will appeal to you and therefore you should buy our products.” Marks thus are a shorthand way for consumers to associate a lot of information with a single product.

9.2.2. The Boundaries of Trademark Protection As with any form of intellectual property protection, the boundaries of trademark law are important. Trademark’s boundaries tend to be more context-​dependent than other 12 

For a discussion of experience goods and inspection goods, see Landes and Posner (2003, 117, n. 51).

13 See ibid.

 

TRADEMARKS AND UNFAIR COMPETITION    225 forms of intellectual property protection such as patent law. Some of the ways in which the boundaries of trademark law are circumscribed include the strength of the mark, functionality, and allowing recovery for infringement only if the defendant’s use of the same or similar mark results in a likelihood of consumers being confused. Not all source identifiers are created equally, however. Some present higher avoidance costs than others; some are more euphonious or aesthetically appealing than others; some are more memorable.

9.2.2.1. The Continuum of Protection for Word Marks Word marks are arrayed along a continuum of strength. The strongest marks are fanciful marks, which are made up words. Suppose, for example, that a manufacturer was producing a brand new version of a portable device for carrying hot liquids—​otherwise known as a coffee mug—​and wanted to create a mark under which to sell it. The producer could mark it GRODROK, which would be fanciful although not terribly euphonious.14 Slightly less strong but still protected are arbitrary marks, which are pre-​existing words applied in a new concept. If the producer decided to use the mark PIANO to sell the mug, that would be an example of an arbitrary mark. Suggestive marks are weaker on the continuum, and convey to the user some features of the product. In this case, using the mark OASIS would be suggestive of the qualities the producer wants the user to associate with the mug. Marks that are fanciful, arbitrary, or suggestive receive protection from their first use on goods in commerce, and it is easy to see why: they allow consumers to identify the source of the good without removing from the public domain words that other producers need to describe their own products. Words that are deemed descriptive—​those that convey the product’s qualities directly to the consumer—​are unprotected, at least ab initio. For example, attempting to use the mark COFFEE-​PORTER for the mug would likely fail for being descriptive. (Although, to be sure, the line between suggestive marks, which are protected, and merely descriptive words, which are not, is highly context-​dependent and variable.) This makes sense because a descriptive word, without more, does not distinguish the manufacturer’s product from those of other manufacturers. Giving protection to a merely descriptive word ab initio would impose avoidance costs on competitors and consumers alike that would be particularly high because the descriptive word would require competitors to use alternative, non-​protected, words to describe their own products. That could in turn inhibit the communication of useful information to consumers by other producers. Trademark protection is dependent on thick context, however, so over time a mark can move across the line of protection as consumers come to associate it in a particular context with a particular producer. If consumers came to associate a descriptive word such as “coffee porter” with a single source of a product, the descriptive word could receive protection. It would be declared to have “secondary meaning,” which means that 14  Professor Carter cited the fictional trademark GRODROK in his article (1990, 759, 770) as an example of an unlyrical mark. Be that as it may, I find it quite charming, myself. But de gustibus non est disputandum.

 

226   CLARISA LONG in addition to its descriptive qualities, it was recognized by consumers as a source identifier when applied to that product. Generic words do not receive protection because they are the name of the genus of goods for which the would-​be-​marked product is the species. For example, our producer of a portable device for carrying hot liquids could not use the words coffee mug as a source identifier for its product because that would be generic. Once again, the same reasoning applies to the non-​protection of generic words as is applied to descriptive words: allowing a producer to use a generic word to mark its product would require all other sellers of the same product in the marketplace to create a new word to refer to the general class of goods, and would necessitate consumers learning the new word for the genus of goods. If arbitrary and fanciful marks are stronger and provide protection ab initio, why don’t producers always seek to use arbitrary or fanciful marks? Why would they use suggestive marks? Why come up with a mark that is close to the line and therefore risks being deemed descriptive? Imagine the following scenario. You have just moved to a new town in a region of the country unfamiliar to you and are looking for a grocery store. As you drive down the street, you encounter the GRODROK grocery store. That doesn’t sound too appetizing, so you drive on. You next encounter a grocery store with the arbitrary mark ALPHA BETA. (This was the real trademark of a now-​defunct chain of grocery stores in the United States.) At least it’s not stomach churning, but the name doesn’t tell you much about the qualities of the store. As you drive on, you come across the LUCKY grocery store (also the name of a real chain of supermarkets in the United States). This name sounds more promising, but it still doesn’t tell the users much about the qualities of the store. How are consumers going to get lucky? Will they meet the partner of their dreams in the produce aisle? Will they win an in-​store lottery? A few things are left to the imagination. Finally, you encounter the SUPERFRESH grocery store. This name, which most likely comes awfully close to the distinctive/​descriptive line, directly conveys the qualities that the producer would like consumers to associate with the good. The closer a mark comes to the distinctiveness line, the less mental processing consumers need to do. If producers want to communicate information to consumers with the least amount of effort, marks that come close to being purely descriptive accomplish that objective.

9.2.2.2. Trade Dress and Functionality It doesn’t make much sense to try to fit trade dress—​which comprises all non-​word forms of source identifiers, such as product shape—​into the categories of fanciful, arbitrary, etc., that are used for word marks. Nevertheless, trade dress, like word marks, has rules about the boundaries of protection. If denying protection for generic words sets a limitation on the boundaries of protection for word marks, the doctrine of functionality serves as one boundary for trade dress. The doctrine of functionality should be thought primarily as a policy tool intended to exclude trade dress from protection in circumstances where granting protection could have anticompetitive effects. As the US Supreme Court has stated, “The functionality

 

TRADEMARKS AND UNFAIR COMPETITION    227 doctrine prevents trademark law, which seeks to promote competition by protecting a firm’s reputation, from instead inhibiting legitimate competition by allowing a producer to control a useful product feature…. If a product’s functional features could be used as trademarks … a monopoly over such features could be obtained without regard to whether they qualify as patents and could be extended forever (because trademarks may be renewed in perpetuity).”15 There are several formulations for determining if a feature of a product should be denied trade dress protection on the grounds that it is legally functional. One touchstone for determining that an element is functional is if it is “essential to the use or purpose of the article or if it affects the cost or quality of the article.”16 The first of these two types of functionality, which can be thought of as the core of the doctrine of functionality, excludes from protection physical features that have a primarily utilitarian function or that could potentially be eligible for patent protection. For this reason it is often referred to as “mechanical functionality.” For example, the dumbbell shape of a baby bottle could serve as the trade dress for the producer of the baby bottle. But the shape also has a utilitarian function in that it allows the baby to grip the bottle more easily. As a result, the shape of the baby bottle would be denied trade dress protection.17 Usually excluded from the legal definition of functionality, it should be noted, is product packaging. Of course the packaging of a product (think, e.g., of a glass Coke bottle) serves a functional purpose—​to hold the product and keep it from being damaged or dirtied before the consumer can consume it. But although product packaging has a utilitarian purpose, it can also serve as a powerful source identifier. In most such cases, if a court deems that consumers use the product packaging as a source identifier—​and again, the Coke bottle is a good example of strong trade dress—​then its functional characteristics as product packaging do not serve to eliminate trade dress protection. A slightly different set of circumstances under which an aspect of a product will be deemed functional and therefore beyond the realm of trademark protection is where granting protection to the trade dress would have an anticompetitive effect. Under these circumstances, a product’s feature need not be eligible for patent protection or mechanically useful. But if granting protection for that feature could raise manufacturing costs for competitors because competitors would be required to design around it and there are a limited number of alternative designs possible, then the feature will be deemed de jure functional. Such a doctrine, however, appears to be more theoretical than actual. In most cases, courts hesitate to deny trade dress protection on the grounds that competitors will have a hard time designing around it.18

15 

Qualitex Co. v Jacobson Products Co., 514 U.S. 159, 164–​65 (1995). Inwood Lab. v Ives Lab., 456 U.S. 844, 850 n.10 (1982). 17  See, e.g., In re Babies Beat, Inc., 13 U.S.P.Q.2d (BNA) 1729 (T.T.A.B. 1990). 18  For a case denying trade dress protection as on competitiveness grounds, see Valu Engineering, Inc. v Rexnord Corp., 278 F.3d 1268 (Fed. Cir. 2002) (upholding the Board’s determination that the designs were functional as used in the context at issue and thus had a “competitively-​significant application”). 16 

 

228   CLARISA LONG A subspecies of the doctrine of functionality, aesthetic functionality, fits this latter justification. The notion of aesthetic functionality may sound like a contradiction in terms but it is actually consistent with the policy goals of unfair competition law. It attempts to make sure that a popular design is not removed from the public domain, even if that design has no mechanical function.19 A design is considered aesthetically functional if its “aesthetic value lies in its ability to confer a significant benefit that cannot practically be duplicated by the use of alternative designs.”20 For example, a manufacturer selling individual china plates that could be purchased to replace damaged pieces in china patterns already owned by consumers was deemed not to be liable for trademark infringement because the ability to purchase individual plates was beneficial to consumers. Consumers found the design on the replacement pieces of china valuable, not as a source identifier, but because it allowed them to complete sets of china that had had individual pieces broken, rather than having to buy a full set of plates in a new pattern.21

9.2.2.3. Likelihood of Confusion as the Touchstone for Protection under Classic Trademark Law The boundaries of trademark protection, as mentioned previously, are highly contextual. One very important limitation on the boundaries of protection for trademarks under classic trademark law is that use of the mark by other producers in the marketplace by a defendant without the trademark holder’s permission gives rise to liability only to the extent that consumers are confused by the defendant’s use as to the source of the goods.22 This is why, for example, use of the identical word mark DELTA by Delta Airlines for airline services, Delta Dental for dental insurance, Delta Power Equipment Corporation for power tools, Delta Moving and Storage for moving services, and Delta Faucet for bathroom fixtures does not give rise to liability by any of these users against any of the others: in each context, consumers are not confused as to the source of the particular good at issue. Such a doctrine raises both producers’ and consumers’ costs along some margins and lowers them along others. It makes classic trademark protection much more nuanced than it otherwise would be if the rule were to assign strong exclusionary rights on a non-​ contextual basis. As a result, producers who want to use a mark similar or identical to

19   See, e.g., Publ’ns Int’l, Ltd. v Landoll, Inc., 164 F.3d 337, 342 (7th Cir. 1998) (“Gold is a natural color to use on a fancy cookbook.”); Pagliero v Wallace China Co., 198 F.2d 339, 343–​44 (9th Cir. 1952) (stripping away trademark protection when the aesthetically pleasing nature of a trademark conferred market power over the good itself). For a slightly unusual take on aesthetic functionality, see Plasticolor Molded Prods. v Ford Motor Co., 713 F. Supp. 1329, 1335 (C.D. Cal. 1989) (stating in dictum that use of FORD on automobile floor mats by a maker of replacement floor mats is “functional” because it is “designed to help the mats contribute to a harmonious ensemble of accessories and decorate the interior of a car”), vacated after settlement and consent decree, 767 F. Supp. 1036 (C.D. Cal. 1991). 20  Qualitex Co. v Jacobson Products Co., 514 U.S. 159, 170 (1995). 21  Pagliero v Wallace China Co., 198 F.2d 339 (9th Cir. 1952). 22  See Lanham Act §§ 32(1), 43(a), 15 U.S.C. §§ 1114(1), 1125(a) (establishing liability for infringement of registered and unregistered marks, respectively).

 

TRADEMARKS AND UNFAIR COMPETITION    229 one already being used must assess whether consumers would be likely to be confused. This increases producers’ avoidance costs in that they have to make a determination about the consumer perception of various marks. If a new entrant into the market for making residential water pipes wanted to use the mark DELTA, would consumers be likely to confuse it with Delta Faucet? After all, both companies are offering a product through which water is conveyed in a residence. One product is placed inside the walls of the house, whereas the other is external to the walls. The new entrant must identify the appropriate consumer market—​would it be homeowners? contractors? plumbers? all of the above?—​and anticipate whether its use of the mark would be likely to cause confusion as to the source of its product in the relevant market, whatever that might be. On the other hand, a focus on consumer confusion lowers the costs to producers of creating trademarks since they do not need to create marks that are absolutely novel. Using potential consumer confusion as the touchstone for liability preserves classic trademark law’s focus on consumers. If consumers are not harmed, then producers cannot exclude other sellers in the marketplace from using the same mark, much as they may desire exclusive rights. As I will demonstrate, in recent years, trademark law has added a new cause of action—​dilution law—​that allows trademark holders to recover for infringement of their marks under certain circumstances even when consumers are not confused. This represents a fundamental shift away from focusing on consumers as boundary for trademark protection.

9.2.3. Some Costs of Trademarks Commentators have not left unchallenged the economic justifications that have been posited to support a robust system of trademark protection. Some commentators have argued that strong trademark protection can act as a barrier to entry for potential competitors while giving existent market entrants an advantage by virtue of being first.23 Strong trademark protection can raise the costs to sellers of communicating useful information to consumers. As the Restatement (Third) of Unfair Competition notes, “The extension of trademark protection to packaging and product features can undermine the freedom to copy successful products unprotected by patent or copyright. Such encroachments on the public domain can result in higher prices and decreased access to goods and services. Although these criticisms have not undermined the basic commitment to the protection of trademarks, their influence is evident in the continuing effort to delineate the appropriate scope of protection.”24 Trademark protection, whether of word marks or trade dress, imposes avoidance costs on other producers in the marketplace, who must design their own trademarks and trade dress to avoid infringement. Third parties trying to fulfill their duties of avoidance must not just figure out the contexts in which they cannot use a protected mark; 23 

24 

See, e.g., Carter (1990). Restatement (3d) of Unfair Competition § 9, cmt. c (1995).

 

230   CLARISA LONG they must also figure out the contours of what they must avoid. Put another way, other sellers in the marketplace need to define protected marks in order to avoid infringement. The higher the definition costs of the mark, the more costly it becomes to avoid the mark. Concerns about the costs of trademarks can help explain why marks are not allowed to be “warehoused,” and instead must be used on goods in commerce in order to be valid.25 If trademark holders were allowed to stockpile trademarks they were not using in commerce and claim exclusionary rights to them, the avoidance costs to competitors would be positive but the benefits to consumers would be zero. Unlike a patent, a trademark does not have to be registered with the Patent and Trademark Office in order to receive protection. But registration confers several advantages on a trademark registrant, such as putting third parties nationwide on constructive notice of ownership of the mark.26 Registration has the advantage to other sellers of lowering their avoidance costs, since they can search a single source to determine which trademarks not to use for their own products. Trade dress generally presents higher definition costs to trademark holders and third parties alike than do word marks. In recent years, the types of product features that can qualify as protectable trade dress has expanded to include source identifiers, such as product configuration, product design, and product ambience, that often have higher definition costs than traditional pictorial or literary marks. Trademarks composed of words, logos, and pictures can be defined fairly readily by a simple representation on paper. Product design, configuration, and ambience, by contrast, often require descriptions that are more detailed to convey the concept of the trademark. Trade dress is also more likely to remain unregistered than word marks. Why is this? Consider the following example. Pepperidge Farm uses the word GOLDFISH as a trademark for its fish-​shaped cracker.27 It also uses the appearance of the cracker as trade dress. The word GOLDFISH is registered as a trademark with the US Trademark Office, whereas the appearance of the fish-​shaped cracker is not. Registration is one reason that makes the existence and boundaries of the trademark easier for third parties to define than the trade dress of the cracker. Even if the word GOLDFISH were not registered, however, the boundaries of the trademark GOLDFISH would be easy to figure out just by looking at the mark: they are the letters G-​O-​L-​D-​F-​I-​S-​H. Depending on the circumstances, consumers might be confused by misspellings like “Goldfisch,” but even so, there is not much room to expand or contract the zone of protection surrounding the mark. There are several reasons that the boundaries of trademark protection for the cracker’s appearance are more malleable and harder for third parties to define than the contours

25 

Lanham Act § 45, 15 U.S.C. § 1127 (2012). See 15 U.S.C. § 1072 (2012) (stating that federal registration of mark provides constructive notice of ownership). 27  See Reg. No. 0739118 (1962) (GOLDFISH). 26 

 

TRADEMARKS AND UNFAIR COMPETITION    231 of protection for the mark itself. It’s not clear which features of the cracker consumers are using as a source identifier, or whether consumers are using the appearance of the cracker as a source identifier at all. The source-​identifying features could be shape, size, texture, color, smell, flavor, or some combination of these, but competitors in the marketplace cannot be sure, at least not before litigation. Pepperidge Farm’s refusal to register the elements of the cracker that it considers its trade dress means that it has chosen not to commit itself to a definition in advance of a lawsuit. Pepperidge Farm itself may not know what elements of the appearance of the cracker it considers the protected trade dress until a competitor comes along with a similar product. At that point Pepperidge Farm has the incentive to claim that the definition of its protected trade dress maps onto whatever the competitor is doing.28 This helps explain why the owners of trade dress have incentives not to register the trade dress for which they seek protection. The disadvantages of failing to register are often outweighed by the benefits of non-​registration. By refraining from registering trade dress such as product appearance, trademark owners retain the ability to expand and contract the definition of their mark.

9.3.  The Rise and Rise of Trademark Dilution Law If one long-​standing justification for the legal protection of trademarks is that they serve as a source identifier that benefits consumers, another justification for trademark law has arisen more recently. This justification states that producers invest time, energy, and effort into creating positive associations in consumers’ minds about their products. Providing legal protection for trademarks allows producers to reap the benefits of creating goodwill among consumers.29 Over time, consumers can be expected to form opinions about the products they consume or encounter and by extension about the trademarks associated with these products. Such opinions can influence their future purchasing decisions about products associated with the trademark. If consumers’ opinions are positive, trademark holders stand to reap the benefit, but this requires trademark holders being able to control the way in which the mark is used. Conferring on trademark holders the power to exclude third parties from nonconfusing uses of the marks allows mark owners to reap the benefits of their investment in the attributes of the product. 28 

See, e.g., Nabisco, Inc. v PF Brands, Inc., 191 F.3d 208, 218 (2d Cir. 1999), in which Pepperidge Farm argued successfully that the similarities of color, shape, size, and taste between the appearance of its cracker and the appearance of its competitor’s product (a fish-​shaped cracker that appeared as part of a mix of crackers labeled CATDOG) constituted the elements of its protected trade dress whereas the differences (the baker’s markings on each cracker and mix of crackers in the package) did not. 29  The existence of dilution law, in both the United States and the European Union, can be traced back to an article written by Professor Frank I. Schechter (1927, 813).

 

232   CLARISA LONG Before the inclusion of dilution in the trademark statute, the owner of a mark could recover for trademark infringement under the Lanham Act only if the commercial use of its mark by another seller in the marketplace caused consumer confusion. Under its dilution provisions, the Lanham Act protects “[t]‌he owner of a famous mark … against another person who, at any time after the owner’s mark has become famous, commences use of a mark or trade name in commerce that is likely to cause dilution by blurring or tarnishment of the famous mark.”30 Dilution law is a departure from the traditional justification of trademark law, and one that has sparked no small amount of commentary. Commentators have described dilution law as “a fundamental shift in the nature of trademark protection.”31 Judge Richard Posner has worried about dilution’s “seductive appeal.”32

9.3.1. Trademarks and Good Will Dilution law is producer-​focused rather than consumer-​focused, even though some commentators have argued that dilution law is geared toward protecting consumers because the use of a mark by more than one producer increases consumers’ search costs.33 Dilution law seeks to prevent diminution in the value of a famous mark if that mark is used by someone other than the trademark holder.34 The underlying assumption of a dilution theory of trademark law is that even when consumers are not confused by the use of the mark, the unauthorized use of a famous mark by other producers in the marketplace can diminish the mark’s value because it is no longer associated with a single source. Dilution is a more exclusionary variant of the trademark entitlement than the classic likelihood-​of-​confusion form. Because classic trademark law requires consumers to be confused about the source of goods before a defendant could be held liable, it is more nuanced and contextual than dilution law is, which does not ask about consumer confusion. But that does not mean that dilution law is easier to define. There are two variants of trademark dilution, which is an indication that the existence of trademark dilution can be hard to identify. These two variants are dilution by blurring and dilution by tarnishment.35

9.3.1.1. The Definitions of Dilution: Blurring and Tarnishment Blurring is the use of a famous mark by another producer in the marketplace to identify a different product in a way that weakens the connection in consumers’ minds between 30 

15 U.S.C. § 1125(c)(1) (2012). Lemley (1999, 1687, 1698). 32  Posner (2003, 621, 623). 33  See, e.g., Dogan and Lemley (2005, 461, 493) (“[P]‌roperly understood, dilution is targeted at reducing consumer search costs, just as traditional trademark law is.”). 34  What the literature customarily refers to as “dilution law” is technically antidilution law. 35  See 15 U.S.C. § 1125(c)(1) (2012) (setting forth the two forms of dilution). 31 

 

TRADEMARKS AND UNFAIR COMPETITION    233 the mark and the original product, even when the defendant’s use does not create confusion among consumers as to source.36 Dilution by blurring is statutorily defined as “association arising from the similarity between a mark or trade name and a famous mark that impairs the distinctiveness of the famous mark.”37 Blurring can occur when a third party uses a famous mark to identify its own product in a nonderogatory way. One example is the use of the mark TIFFANY as the name of a restaurant.38 Let us assume that the restaurant is upscale, so that tarnishment by association with low-​quality goods is not a problem. Consumers are not likely to confuse the restaurant with the jewelry store, but nonetheless the same mark is now being used on two very different consumption items: restaurant services and jewelry. This arguably increases the cognitive costs to consumers of keeping the products separate. Tarnishment, by contrast, is defined as “association arising from the similarity between a mark or trade name and a famous mark that harms the reputation of the famous mark.”39 Tarnishment can “occur where the effect of the defendant’s unauthorized use is to dilute by tarnishing or degrading positive associations of the mark and thus, to harm the reputation of the mark,” and often involves using the mark to cast aspersions on the mark holder.40 A tarnishment theory of dilution law is used to justify protection of goodwill against association with an unwholesome product, such as pornography or illegal drugs.41 Among trademark plaintiffs, tarnishment as a theory of dilution has tended to be less popular than blurring has.42 Similarly, when courts have granted recovery on dilution claims, they usually have done so on grounds of blurring rather than tarnishment, although it is not always easy to distinguish between the two definitions of dilution.43 Because the actions that give rise to tarnishment claims, such as critical or derogatory uses of a trademark, often have free speech implications, it is important for courts to take this into consideration when adjudicating a dilution claim.

9.3.1.2. Some Boundaries on Dilution Law Dilution law represents a fundamentally different approach to trademark law than the more traditional likelihood of confusion variant. Like other theories of trademark protection, however, dilution law has various boundaries that limit its scope.

36 

See McCarthy (2005), (defining blurring). 15 U.S.C. § 1125(c)(2)(B) (2012). 38  See, e.g., Ty Inc. v Perryman, 306 F.3d 509, 511 (7th Cir. 2002). 39  15 U.S.C. § 1125(c)(2)(C) (2012). 40  See McCarthy (2005) (defining tarnishment). 41  See, e.g., Anheuser-​Busch Inc. v Andy’s Sportswear Inc., 40 U.S.P.Q.2d 1542, 1543 (N.D. Cal. 1996) (holding that “Buttwiser” T-​shirts tarnished BUDWEISER mark); Coca-​Cola Co. v Gemini Rising, Inc., 346 F. Supp. 1183, 1188 (E.D.N.Y. 1972) (finding that “Enjoy Cocaine” posters tarnished ENJOY COCA-​COLA mark). 42  See Long (2006, 1029). 43 See Ty Inc., 306 F.3d at 511 (“Analytically [tarnishment] is a subset of blurring.”). 37 

 

234   CLARISA LONG One boundary placed on dilution law is that the mark must be famous.44 This fits with the justification of dilution law being a means of protecting the goodwill with which a trademark holder has attempted to imbue its mark. Although the definition of “fame” is itself someone malleable, fame can serve as proxy for goodwill. Moreover, the remedy is limited only to injunctive relief unless the defendant has behaved willfully, in which case it can include damages.45 Another limitation exempts several types of uses by defendants from liability. Generally, these uses have speech implications. If trademark holders were allowed to enjoin truthful criticism or discussion of their trademarks by others, this could create social costs by allowing famous mark holders to control truthful speech. Thus federal dilution law creates exceptions from liability for comparative advertising by competitors using the trademark, for descriptive use of the mark that does not constitute a source identifier, for media use of the trademark, and for noncommercial use of the mark.46 Like a likelihood-​of-​confusion theory of trademark infringement, dilution law only requires that plaintiffs prove a likelihood of dilution.47 Since dilution of a mark can take more than one form, asking plaintiffs to prove a mere likelihood of dilution reduces the burden of proof on plaintiffs, but it is not clear that having such a loose proof requirement makes the entitlement more clear. Even with these boundaries, dilution law has expanded beyond its original narrow origins.48 The loosening and expansion of dilution law by, for example, shifting from a standard that required proof of actual dilution to one that required plaintiffs to prove a mere likelihood of dilution has caused one commentator to describe the dilution form of protection as ‘bloated.’49

9.3.2. The Costs and Benefits of Trademark Dilution Law Justification of dilution law on the ground that it allows holders of famous marks to recover for their investment in the good will surrounding the mark is consistent with traditional explanations for intellectual property protection in other areas, such as patent and copyright law. Viewed this way, famous marks with positive consumer associations are a classic public good. The cost of creating a public good is high, but once the 44 

See 15 U.S.C. § 1125(c)(2)(A) (2012) (listing some considerations for determining the fame of a mark for dilution purposes). 45  See 15 U.S.C. § 1125(c)(5) (2012). 46  See 15 U.S.C. § 1125(c)(3) (2012). 47  See 15 U.S.C. §1125 (c)(1) (2012). 48  See Schechter (1927, 813, 828–​31), (proposing a form of dilution law limited to conflicts between identical marks, where the plaintiff ’s mark was not only famous but also arbitrary and where the defendant’s use of the mark was on noncompeting and nonsimilar goods). 49  McCarthy (2005) (stating, “It is my belief that the present state of antidilution law has been bloated far out of proportion to its original purpose and intent.”).

 

TRADEMARKS AND UNFAIR COMPETITION    235 public good is created, the cost to third parties of using it is low.50 Without some form of protection, the reasoning goes, creators of famous marks will not be able adequately to internalize the benefits of the goodwill they have created and will therefore underproduce such marks. Legal protection allows creators of famous marks a greater opportunity to police the use of the mark, prevent third parties from free riding, and thereby enjoy the profits associated with their marks.51 In order for this type of protection to be efficient, however, sellers would need to be incentivized to produce marks with positive goodwill and would not invest sufficiently in creating famous trademarks without protection. Dilution law, if improperly applied, presents the possibility of social cost, so the net benefits of allowing trademark holders the kind of control over their marks that dilution law does must be weighed against the net social costs. Dilution law may create a net social cost, for example, if the net avoidance costs incurred by third parties attempting to avoid the mark exceed the value of the goodwill reaped by the mark holder. Let us consider the costs and benefits to trademark holders when third parties use a mark on their own goods without authorization. The costs to the trademark holder could be positive, zero, or even negative if the trademark holder is benefited by the third-​party use, such as if the third-​party use made the legitimate trademark holder’s goods more desirable. An example of a positive cost to a trademark holder occurs when unauthorized third-​party use of a trademark diverts sales away from the trademark holder toward a cheaper product bearing a counterfeit mark. If the counterfeit product is of lower quality than the authorized product, the mark holder may suffer a cost because consumer perceptions of the trademark are diminished. By contrast, a famous trademark holder may be benefited by another producer’s unauthorized use of the mark if the use gives the mark more airtime in front of consumers.52 Although statistically unlikely, it is possible if the factor driving a mark’s value is the quantity rather than the quality of publicity the mark receives. Producer goodwill is most likely to be damaged in the case of tarnishment of the mark, but this is not always the case. The social welfare calculus surrounding allegedly tarnishing uses is nuanced. Third-​party uses of a mark that trademark holders believe are tarnishing to the mark may in fact have positive social value. Nonconfusing use of a mark by unauthorized third parties for purposes of criticism, social commentary, parody, or other speech-​related purposes can have net positive social benefits that outweigh the harm to the trademark holder. As a result, dilution law explicitly excludes comparative commercial advertising, news reporting, and news commentary from liability of actionable 50 

See Arrow (1962, 609, 614–​16), (describing the public goods phenomenon as related to patented goods) and Lemley (1997, 989, 995–​96), (describing the intellectual property and public goods problem). 51  See Franklyn (2004, 117), (arguing that US dilution law “really is about preventing free-​riding on famous marks.”). 52  See, e.g., Dreamwerks Prod. Group, Inc. v SKG Studio, 142 F.3d 1127, 1129 (9th Cir. 1998) (discussing possibility, in likelihood of confusion context, that harm to a plaintiff trademark holder might be “somehow offset by any extra goodwill plaintiff may inadvertently reap as a result of the confusion between its mark and that of the defendant”).

 

236   CLARISA LONG harm.53 Nonetheless, it is important for courts to police the boundary carefully to assure that claims of harm by tarnishment do not shut down free speech. If trademark holders were allowed to silence parodic or critical speech involving their mark under the guise of tarnishment, the result could be a net social loss. The case for producer goodwill being diminished when blurring occurs is harder to make. What is not clear is whether the benefits to consumers of dilution law translate into benefits for trademark holders. Under a likelihood-​of-​confusion standard, for instance, trademark holders and consumers alike benefit from consumers not being confused by the source of a good. But dilution can occur in the absence of consumer confusion. How is a trademark holder benefited by limitations on the use of the mark by third parties when consumers are not confused? Or put another way, if there is no consumer confusion or tarnishment of the mark, has a trademark holder been harmed when consumers associate its mark with more than one source or product? One could say that when consumers must share their attention between two sources or products associated with the same mark, such as a restaurant named TIFFANY and a famous jewelry store of the same name, they will thus devote less attention to the jewelry store. But should producers have an entitlement to consumers’ attention span? And if so, are producers injured when consumers don’t pay them the attention the producer feels they deserve? As a result, some scholars have argued that the only appropriate justification for dilution law is preventing harm to consumers.54 On this view, dilution by blurring should be enjoined only if the blurring imposes higher search costs on consumers.55 Blurring may harm consumers, so protecting them from blurring may be beneficial. Commentators have argued that blurring can harm consumers by increasing search costs because it requires consumers to devote more mental energy to evaluating the context in which the mark is being used.56 Indeed, there is empirical evidence that blurring of a famous mark can increase consumer search costs.57 One example of harm to consumers from blurring occurs when the use of TIFFANY for a restaurant increases consumers’ cognitive costs in distinguishing the use of TIFFANY for the jewelry store. If it really is the case that the use of the same mark by two different parties on unrelated goods in the marketplace increases consumers’ search costs, then dilution law can be justified on consumer protection grounds. Otherwise, the case for protection against blurring is ambiguous from an efficiency perspective.

53

  See 15 U.S.C. § 1125(c)(3)(A)–​(C) (2012). See Dogan and Lemley (2004, 777, 790–​91) (“[D]‌ilution—​at least as properly understood—​turns on injury to the informative value of a mark.” [footnote omitted]). 55  See, e.g., Klerman (2006, 1759, 1766–​67), (arguing that if the use of the same trademark on two different products does not increase search costs, there is no harm). 56  See O’Rourke (1998, 277, 306–​07, and n.114) (“Dilution by blurring is concerned with preventing the erosion of the distinctiveness of the mark because of its use on non-​related products. The ‘noise’ that this creates around the mark may increase consumer search costs.”). 57  See, e.g., Morrin and Jacoby (2000, 265, 274), (showing that even when consumers correctly matched trademarks with products, they took longer to do so if a mark was associated with more than one product). 54 

 

TRADEMARKS AND UNFAIR COMPETITION    237

9.4. Conclusion The culmination of trademark law in the form of protection again dilution represents completion of a full circle of development of unfair competition law. From its start as a means of protecting merchants’ entitlement to custom, unfair competition law and its subsidiary, trademark law, have evolved in several stages: first as a primitive common law means of regulating markets in the absence of antitrust law, then as a means of protecting consumers in an era when consumer protection law was not as robust as it later became, and finally to encompass a cause of action that allows protection for reasons similar to those contained in the original justification.

References 15 U.S.C. § 1072 (2012). 15 U.S.C. § 1125(c)(1) (2012). 15 U.S.C. § 1125(c)(2)(A) (2012). 15 U.S.C. § 1125(c)(5) (2012). 15 U.S.C. § 1125(c)(2)(B) (2012). 15 U.S.C. § 1125(c)(2)(C) (2012). 15 U.S.C. § 1125(c)(3) (2012). 15 U.S.C. § 1125(c)(3)(A)–​(C) (2012). Anheuser-​Busch Inc. v Andy’s Sportswear Inc., 40 U.S.P.Q.2d 1542, 1543 (N.D. Cal. 1996). Arrow, Kenneth J. 1962. “Economic Welfare and the Allocation of Resources for Invention.” In: Richard Nelson, ed., Nat’l Bureau of Econ. Research, The Rate and Direction of Inventive Activity: Economic and Social Factors. pp. 609, 614–​16. Beebe, Barton. 2004. “The Semiotic Analysis of Trademark Law.” UCLA Law Review 51, p. 621. Carter, Stephen L. 1990. The Trouble with Trademark.” Yale Law Review 99, pp. 759, 770. Clarke, 17 U.S.P.Q.2d (BNA) 1238, 1238 (T.T.A.B. 1990). Coca-​Cola Co. v Gemini Rising, Inc., 346 F. Supp. 1183, 1188 (E.D.N.Y. 1972). Dogan, Stacey L. and Lemley, Mark A. 2004. Trademarks and Consumer Search Costs on the Internet. Houston Law Review 41, pp. 777, 790–​791. Dogan, Stacey L. and Lemley, Mark A. 2005. “The Merchandising Right: Fragile Theory or Fait Accompli?” Emory Law Journal 54, pp. 461, 493. Dreamwerks Prod. Group, Inc. v SKG Studio, 142 F.3d 1127, 1129 (9th Cir. 1998). Easterbrook, Frank H. “Foreword: The Court and the Economic System.” Harvard Law Review 98(4), pp. 21–​22. Economides, Nicholas S. 1988. “The Economics of Trademarks.” Trademark Rep 78, p. 523. Franklyn, David J. 2004. “Debunking Dilution Doctrine: Toward A Coherent Theory of the Anti-​Free-​Rider Principle in American Trademark Law.” Hastings Law Journal 56, p. 117. Gen. Elec. Broad. Co., 199 U.S.P.Q. (BNA) 560, 563 (T.T.A.B. 1978). Graham v John Deere Co., 383 U.S. 1, 9 (1965). In re Babies Beat, Inc., 13 U.S.P.Q.2d (BNA) 1729 (T.T.A.B. 1990). In re Clarke, 17 U.S.P.Q. (BNA) 1238 (T.T.A.B. 1990).

 

238   CLARISA LONG In re General Electric Broadcasting Co., 199 U.S.P.Q. (BNA) 560, 563 (T.T.A.B. 1978). Inwood Lab. v Ives Lab., 456 U.S. 844, 850 n.10 (1982). Klerman, Daniel. 2006. “Trademark Dilution, Search Costs, and Naked Licensing.” Fordham Law Review 74, 1759, 1766–​1767. Landes, William M. and Posner, Richard A. 1987. “Trademark Law: An Economic Perspective.” JL & Econ 30, p. 265. Landes, William M. and Posner, Richard A. 2003. The Economic Structure of Intellectual Property Law. New York: Nicholas Thompson. 117, n. 51. Lemley, Mark A. 1997. “The Economics of Improvement in Intellectual Property Law.” Texas Law Review 75, pp. 989, 995–​996. Lemley, Mark A. 1999. “The Modern Lanham Act and the Death of Common Sense.” Yale Law Journal 108, pp. 1687, 1698. Leval, Pierre N. 2004. “Trademark: Champion of Free Speech.” Columbia Journal of Law & Arts 27, pp. 187, 198. Long, Clarisa. 2006. “Dilution.” Columbia Law Review 106, p. 1029 15. Lunney, Glynn S. Jr., 1999. “Trademark Monopolies.” Emory Law Journal 48, pp. 367, 417. Machlup, Fritz. 1958. An Economic Review of the Patent System, Study No 15, Subcommittee on Patents, Trademarks, and Copyrights of the Senate Committee on the Judiciary, 85th Cong., 2d Sess. 1, 21. McCarthy, J. Thomas. 2005. McCarthy on Trademarks and Unfair Competition § 24:115. 4th ed. Thomas Reuters Westlaw. Morrin, Maureen and Jacoby, Jacob. 2000. “Trademark Dilution: Empirical Measures for a Concept.” J Pub Pol’y & Marketing 19, pp. 265, 274. Nabisco, Inc. v PF Brands, Inc., 191 F.3d 208, 218 (2d Cir. 1999). O’Rourke, Maureen A. 1998. “Defining the Limits of Free-​Riding in Cyberspace: Trademark Liability for Metatagging.” Gonzaga Law Review 33, 277, 306–​307, n.114. Pagliero v Wallace China Co., 198 F.2d 339, 343–​44 (9th Cir. 1952). Plasticolor Molded Prods. v Ford Motor Co., 713 F. Supp. 1329, 1335 (C.D. Cal. 1989). Posner, Richard A. 2003. “Misappropriation: A Dirge.” Houston Law Review 40, pp. 621, 623. Publ’ns Int’l, Ltd. v Landoll, Inc., 164 F.3d 337, 342 (7th Cir. 1998). Qualitex Co. v Jacobson Prods. Co., 514 U.S. 159, 162, 164–​65 (1995). Reg. No. 0739118 (1962). Restatement (3d) of Unfair Competition § 9, cmt. c (1995). Schechter, Frank I. 1927. “The Rational Basis of Trademark Protection.” Harvard Law Review 40, p. 813. Thumin, Frederick J. 1962. “Identification of Cola Beverages.” Journal of Applied Psychology 46, pp. 358, 358–​359. Two Pesos, Inc. v Taco Cabana, Inc., 505 U.S. 763, 767 (1992). Ty Inc. v Perryman, 306 F.3d 509, 511 (7th Cir. 2002). U.S. Const. Art. I, § 8, cl. 8. Valu Engineering, Inc. v Rexnord Corp., 278 F.3d 1268 (Fed. Cir. 2002).

 

Chapter 10

L AW AND EC ONOMI C S OF INFORM AT I ON Tim Wu

Information is an extremely complex phenomenon not fully understood by any branch of learning, yet one of enormous importance to contemporary economics, science, and technology (Gleick 2011; Pierce 1980). Beginning in the 1970s, economists and legal scholars, relying on a simplified “public good” model of information, have constructed an impressively extensive body of scholarship devoted to the relationship between law and information. The public good model tends to justify law, such as the intellectual property laws or various forms of securities regulation that seek to incentivize the production of information or its broader dissemination. A review of the last several decades of scholarship based on the public good model suggests the following two trends. First, scholars have extended the public good model of information to an ever-​increasing number of fields where law and information intersect. An incomplete list of fields covered includes intellectual property, securities regulation, financial regulation, contract theory, financial regulation, consumer protection, communications, and the study of free speech. While scholars in all of these fields are interested in information, they tend to focus on different market failures and different properties of information. Generally, scholars of intellectual property have focused on problems of underproduction—​the concern that, absent government intervention, less than optimal amounts of information will be produced. In contrast, scholars in other fields, like securities regulation or consumer protection, analyze the dissemination of information, or “information asymmetries”—​failures to distribute information in an optimal fashion. Second, over the last decade, scholars have sharply questioned the simplified model, and asked whether, in practice, information actually has the characteristics of a public good. The public good model of information relies on two purported qualities: (1) that information tends to be difficult or impossible to exclude others from, and (2) that its consumption does not eliminate its value for others. The first assumption, in particular, has undergone considerable attack; a closer look suggests that context, subject matter,

 

240   TIM WU and industry structure tend to yield great variation in how much intervention really is required to ensure adequate production or dissemination. This tends to support the existence of dynamic legal regimes attuned to differences in subject matter or perhaps industry structure. The review closes by asking if the public good model, while well established and relatively easy to understand, ought really be the exclusive focus of the economic and legal understanding of information. The article closes by considering other, less investigated, but potentially important, properties of information that have not received as much scholarly attention.

10.1.  Information’s Peculiar Characteristics Information is a complex abstraction that has been the subject of intense study by scientists and philosophers for more than a century. It remains incompletely understood: some physicists, for example, believe that every particle and force in the universe might actually be best understood as a form of information (Wheeler 1990). In the sciences, a minimal, though not-​uncontested definition of information defines it “as one or more statements or facts that are received by a human and that have some form of worth to the recipient” (Losee 1998). What are the economic properties of this abstraction? Economists and legal scholars have generally been uninterested in the scientist’s concept of information and instead more captivated by the premise that information is a “public good.” Stated otherwise, the economic and legal scholarship has sought to analyze information as a member of a category of goods first described in modern times by Mill (1848) as those that require public intervention to ensure an adequate supply. If Mill (1848) did not invent the model, he certainly popularized it. His most famous example of a public good was the lighthouse: something from which all benefited, but might be unwilling to pay for privately. Other classic examples of public goods include a strong national defense, clean air, and so on, and Mill may have seeded the current treatment of information by describing knowledge as follows. “The cultivation of speculative knowledge” wrote Mill, “though one of the most useful of all employments, is a service rendered to a community collectively, not individually, and one consequently for which it is, primâ facie, reasonable that the community collectively should pay.” In the 20th century, Paul Samuelson (1954) stated Mill’s idea more precisely by describing a public good as what he called a “collective consumption good.” According to Samuelson, the category included those goods which” … all enjoy in common in the sense that each individual’s consumption of such a good leads to no subtractions from any other individual’s consumption of that good …” (Samuelson 1954, 1955). Kenneth Arrow (1962) provided one of the first linkages between Samuelson’s concept and

 

LAW AND ECONOMICS OF INFORMATION    241 intellectual property regimes, like patents, by which government intervenes in the market for information.1 By the 1980s, it had become commonplace to link the public good model to government regimes that concerned themselves with information. Meanwhile, from the 1970s onward, the study of “information asymmetries” popularized by George Akerlof and others (notably A. Michael Spence and Joseph E. Stiglitz) has influenced and served as an important complement to the study of information production (Akerlof 1970). The study of asymmetries is essentially concerned with the distribution, as opposed to the creation of information. The basic observation that a suboptimal distribution of information may yield a variety of problems (such as adverse selection, moral hazard, or worse) has influenced most of the writing described here.

10.2.  The Spread of the Public Good Model Today, some version of the public good model of information now dominates analysis of the economics of information production. Among other fields, scholars have applied information-​as-​public-​good arguments to fields as diverse as the regulation of securities, contract, consumer protection laws, communication laws, and constitutional law, among others. However, as we shall see, there are important variations in how the arguments appear in different fields. The use of public good arguments to justify grants of intellectual property has perhaps the longest lineage—​one probably older than the public good model itself. Consider Lord Macaulay’s (1841) famous argument that copyright is a “tax on readers for the purpose of giving a bounty to writers,” justified because it is “desirable that we should have a supply of good books: we cannot have such a supply unless men of letters are liberally remunerated.” But the theory has spread far from its origins in intellectual property. Since the 1980s, a public good theory of information has been used to justify mandating the disclosure of information for consumer or investor protection. “[B]‌ecause information has many characteristics of a public good,” wrote Coffee (1984), “securities research tends to be underprovided.” The related concept of “informational” or “transparency” regulation such as hazard warnings, medical disclosure, and certain forms of campaign finance regulation has been justified using similar concepts.2 Farber (1991) has relied on public good arguments to explain or justify the American First Amendment’s protection of speech. “[I]nformation is likely not only to be underproduced in the private market,” he writes, “but also to be insufficiently protected by the political system.” 1 

“In the absence of special legal protection, the owner cannot … simply sell information on the open market. Any one purchaser can destroy the monopoly, since he can reproduce the information at little or no cost” (Arrow 1962, 609). 2  See, e.g., Fung et al. (2007); Hasen (1996) (media loopholes in campaign finance regulations); Lyndon (1989) (chemical toxicity disclosure); Magat and Viscusi (1992).

 

242   TIM WU In each area, the key theory is that there is a market failure: without state action, important information will either be underproduced or too much will be kept secret from the public. On closer examination, there are actually two different concerns here: underproduction and suboptimal distribution. These, as we shall see, can be more generally tied to two properties of information: nonexclusion and nonrivalry, respectively.

10.3. Underproduction Mill’s (1848) original theory focused on nonexclusion. As he wrote, “it is impossible that the ships at sea which are benefited by a lighthouse, should be made to pay a toll on the occasion of its use.”3 The idea is that if it is difficult or impossible to exclude nonpayers from consuming the good in question, no one will have an incentive to provide the good, justifying public provisioning of the good. Jefferson (1813), writing before Mill, opined similarly that an idea, once divulged, “forces itself into the possession of everyone, and the receiver cannot dispossess himself of it.” In contemporary times, it is commonplace to describe information as “impossible” or “very difficult” from which to exclude anyone.4 It makes sense that a concern about the underproduction of information ought to depend on a concern about nonexcludability. The idea is that if the producer of information cannot exclude nonpayers, he will lack the means to recoup his initial investment, hence eliminating any desire to create information in the first place. If an author invests heavily in writing a book, and then lacks any mechanism to exclude nonpayers, he will be unable to reap the proceeds of his investment later and, therefore, will have no direct financial incentive to write books in the first place (though he might have indirect or personal incentives). More realistically, we might say that if there were no mechanism for excluding nonpayers, then publishers would be unlikely to invest in an author’s work, therefore, making a career as an author difficult to support. The key question then is whether there is something about information that makes it impossible to exclude nonpayers from consuming. A moment’s reflection makes it obvious that in most contexts this premise is cannot be right, at least in its strong form. 3   See also Samuelson (1954). This point was famously challenged by Coase (1974), who established that private lighthouses were funded by port fees. “We may conclude that economists should not use the lighthouse as an example of a service which could only be provided by the government.” 4  See, e.g., Boyle (2003) explaining that “[ideas] are also assumed to be non-​excludable (it is impossible, or at least hard, to stop one unit of the good from satisfying an infinite number of users at zero marginal cost).”; Krawiec (2001) stating that “by labeling informat ion a collective good in this Article, I do not mean to imply that it is literally impossible to exclude nonpurchasers, but rather the slightly weaker condition that such exclusion is extremely difficult.”; Menell (1988) arguing that “as [the National Commission on New Technological Uses of Copyrighted Works] well recognized, the information comprising innovation in application programs is a prime example of a public good. Given the ease and low cost of copying application programs, it is often impossible to exclude nonpurchasers from an application program’s benefits once it is commercially available.”

 

LAW AND ECONOMICS OF INFORMATION    243 Consider the text of a book locked in a vault for which the key is lost: we are all excluded from it. If you don’t have a ticket, you won’t see that movie. The information contained in an engraving written in a lost language, like hieroglyphs before the discovery of the Rosetta stone, is inaccessible to everyone. Two basic ideas from the basic science of information make it clear why the nonexcludability assumption is hard to support. First, information consists of patterns, which must subsist in some physical or electronic form: ink on paper, stored magnetic charges, or whatever else. Second, for a human to process information, that information must reach the brain (unlike, say, national defense, which can be consumed unknowingly), and be in a form that the brain can process. These necessities combine to suggest one can exclude others from information. Why, then, have so many thinkers insisted that information has the “property” of nonexcludability? What writers like Jefferson (1813) or Mill (1848) seem to have meant by nonexcludability seems to be something meant at a high level of abstraction, really a property of knowledge or wisdom more than information. For example, one might be the beneficiary of the Christian injunction to “love thy neighbor as thyself ” without ever hearing the phrase. Similarly, you might enjoy the comforts of air-​conditioning without having read any of the original patents. But this is different from being a property of information. Alternatively, sometimes what is meant by the “nonexcludability” of information is really the idea that it is cheap to copy information. Since the invention of the printing press, and especially since digitalization, copying information is usually far cheaper than creating valuable information. This point can also be expressed by saying that information goods have a high initial and low marginal cost of production. The fact that nonexcludability is not some intrinsic quality of information, but a technological contingency, is a challenge for the intellectual property laws. Breyer (1970) noticed as much when he argued that copyright is hard to justify given the existence of alternative means of exclusion, such as the “lead time” enjoyed by the first publisher of a book.5 Breyer’s argument has been criticized for its technological naiveté (the piece presumed, for example, that software would be hard to copy), but the central insight seems correct; namely, when copying is expensive, the case for government intervention weakens. That’s why, for example, the later Picasso never suffered from a lack of financial incentives to paint, because only he could create a Picasso. The prospect of non-​legal mechanisms of exclusion is what Sprigman and Rastiala (2012) rely on in their study of creative industries, such as fashion, cooking, and stand-​up comedy, which seem to prosper without intellectual property. In each, the industry devises its own means of exclusion, which seem good enough to incentivize production. Moglen (2003) argues that since non-​market mechanisms yield sufficient information production, the actual effect 5  “A copying publisher, faced with the problems of ‘lead time’ and ‘retaliation’ is unlikely to see much profit in copying low-​volume titles. It seems unlikely, for example, that a publisher thinking of copying the type of tradebook that now sells about 4000 copies, would count on selling the 2000 or more copies needed to earn a profit” (Breyer 1970, 301).

 

244   TIM WU of the creation of exclusionary rights in information is merely a giant wealth transfer from the proletariat to the bourgeoisie.6 Nowadays most scholars hedge their bets by describing exclusion as partially nonexcludable, “hard” to exclude people from, or by drawing a line between private and public information.7 Some scholars, such as Christopher Yoo, Amy Kapczynski, and Talha Syed argue that nonexcludability shouldn’t be considered a defining feature of information at all. Yoo (2007) suggested that the costs of exclusion depend on the technological context of consumption, rather than any inherent characteristic of information.8 Hence, it may be very expensive to exclude the ships that benefit from a lighthouse, but that is irrelevant to whether people without tickets may be kept out of a movie theater on opening night. Consequently, Yoo argues that information should be understood an “impure” public good, yielding policy outcomes different from the pure public goods assumption.9 Kapczynski and Syed (2013) argue that excludability is: highly variable across information goods, and is affected not only by formal legal entitlements, but also by existing technologies for detecting or tracing such uses (and their costs); existing social norms regarding ‘acceptable’ or ‘reasonable’ enforcement efforts (in light of concerns about privacy, freedom of thought and speech, and so forth); and the existing institutions—​or social roles, relations, and organizational forms—​within which the predominant uses of the good will be made.

The weakness of the nonexcludability assumption cannot be said to have destroyed the case for intellectual property or other forms of government action. Defenders of the intellectual property regimes have attempted to justify intellectual property by relying on several alternative theories. First, one might rely not on the assumption of nonexcludability, but, as stated previously, the empirical observation that it is, in today’s technological context, usually 6  “Creators of knowledge, technology, and culture discover that they no longer require the structure of production based on ownership and the structure of distribution based on coercion of payment. Association, and its anarchist model of propertyless production, makes possible the creation of free software, through which creators gain control of the technology of further production” (34). See also Breyer (1970, 289), explaining, “we do not ordinarily create or modify property rights, nor even award compensation, solely on the basis of labor expended.” 7  Benkler (2000), in a typical example, writes, “Information is generally understood to be perfectly nonrival and partially nonexcludable.” 8  “Indeed, it has long been recognized that exclusion is typically possible, with the costs of exclusion depending on the state of technology” (Yoo 2007, 659). Yoo attributes this point to Bator (1958) (describing how nonappropriability can cause market failure). 9  Yoo’s (2007) argument is more complex than captured by this sentence. More specifically, he believes that Samuelson’s theory of public goods does not depend on nonrivalry and nonexcludability, but rather, a condition whereby consumers consume the same quantity of the good and signal their preferences by prices, the inverse of the situation with private goods. Yoo believes that this incentivizes users of a public good to understate their true willingness to pay, in the hopes that others will bear the costs of creating that good (670).

 

LAW AND ECONOMICS OF INFORMATION    245 cheaper to copy information than create it in the first place. Whether low marginal costs of production are intrinsic to information or simply a matter of technological context is an interesting one and not subject to easy answer. In the days when monks copied bibles by hand, copying costs were perhaps comparable to the costs of creating the work in the first place. However, ever since the invention of the printing press, copying information has tended to be cheaper than producing it, which is an explanation for the appearance of the earliest copyright laws in that era. In any event, the low marginal cost of production for informational goods creates a free-​riding argument. Represented in accounts by Palmer (1989), among others,10 the argument asserted that the production of information would naturally create, within groups, either problems of collective action or a “tragedy of the commons.” Given an incentive to copy information and thereby free ride on the production efforts of others, none will be incentivized to produce information, therefore yielding less production than might be ideal. This argument depends not on the impossibility of excluding consumers but rather the low costs of copying in certain contexts, as just discussed (Palmer 1989). Second, some scholars justify the existence of government enforcement of intellectual property rights by stressing the costs of the alternatives. The creators of information tend to regard it as theirs, and want to protect it. When private parties rely on private remedies, those remedies may themselves be quite socially expensive. Consider that real property can be defended by its owners using fences and armed guards, yet government grants exclusion rights in land, whether to encourage investment or to facilitate the development of markets. In the Hobbesian sense, the legal system may be a less wasteful alternative to private exclusion schemes: for example, if it displaces expensive warfare between information producers and their copiers (Kovarsky 2006). This scholarship tends to allude to the costs of “races” of various kinds, including “arms races” between copiers and creators. “The existence of a cost-​effective self-​help remedy,” argues Lichtman (2004), should not always preclude “government regulation as a means to accomplish similar ends.” In another example in the scholarship, Hemphill and Suk (2009) argue that in the fashion industry an inability to exclude copiers creates a reliance on “logoification.” That, they argue, “pull[s]‌fashion toward a status-​conferring function and away from the communication of diverse messages.” Third, intellectual property has sometimes been separately justified by what might be described as a Demsetzian theory. The idea is that group use of a given resource, like information, will create externalities of various sorts that should be internalized by property rights. For example, Kitch (1977), in a famous paper, argued that patent ought to give out broad “prospects”—​that is, a patent covering the initial invention and subsequent inventions as well. Otherwise, Kitch argued the owner might lack incentives to make further investments in research beyond the initial invention in research for fear that the benefits will be appropriated by others.

10 

See also Kitch (1977); Dam (1999); Kieff (2001).

 

246   TIM WU Kitch’s idea has faced criticism too voluminous to summarize. Lemley’s (2004) criticism is typical, and it echoes earlier papers by Rose (2003), Gordon (1992), and others.11 Lemley (2004), as in his other work, relies on the nature of information; it “cannot be depleted,” he wrote, and therefore is not subject to a tragedy of the commons or the negative externalities that justify real property rights. Rather, by its very nature, the tendency was for the creation of information to throw off positive externalities, such as the example of the multiple beneficiaries of the invention of the steam engine. Compensation for positive externalities, or spillovers, Lemley argued in this and other works,12 should almost never be the subject of government intervention (Frischmann and Lemley 2007). “If ‘free riding’ means merely obtaining a benefit from another’s investment, the law does not, cannot, and should not prohibit it” (Lemley 2005, 1049).13 Fourth and finally, where the public good model doesn’t strongly justify intervention, it is certainly possible government may have entirely different goals in mind unrelated to the economics of information. For example, a strong copyright or patent regime may be understood as a subsidy for the entertainment or pharmaceutical industries. As such, the law might really draw little justification from the economics of information, but rather be better described as such by a form of industrial policy, or perhaps a component of a “strategic trade policy.” Such laws might also be understood as a form of political patronage. If nothing more, a generation of scholarship on the existence of private exclusion mechanisms should force any contemporary policy maker to consider the comparative efficiency of private and public reward schemes for the production of information products, along with due consideration of the choice among public alternatives. That point is central to Raustiala and Sprigman’s (2012) work on alternatives, and Benkler’s description of alternative means of information production. It has also lead to a flourishing body of literature, including contributions by Steven Shavell and Tangay van Ypersele (2001) among others, examining the potential of reward systems as an alternative to patent for incentivizing product research.14

10.4.  Information Distribution Problems Problems of information distribution, or “underdissemination,” form a problem distinct from underproduction. Here, the problem can be phrased as follows. Markets and 11  Rose (2003, 90) wrote that the case for property rights is inherently weaker in what she labeled “intellectual space” because “there is no physical resource to be ruined by overuse. Books, tapes, and words may be copied, inventions may be imitated, pictures may be reproduced all without the slightest damage to the original.” 12  “Such intervention may be unnecessary and in fact may lead to welfare-​reducing distortions.” Frischmann and Lemley (2007, 299–​300). 13  See also Gordon (1992, 167), noting, “a culture could not exist if all free riding were prohibited within it.” 14  See also Abramowicz (2003). Benkler (2006), similarly, suggests that there may be alternative private models of producing valuable information that depend on what he terms “peer production” models.

 

LAW AND ECONOMICS OF INFORMATION    247 other systems involving human decisions, such as elections, require a certain amount of information to function well and may require that the information be distributed symmetrically among buyers and sellers (or their equivalents). Where information is scarce, or if distributed asymmetrically, systemic failure can be expected. By metaphor, information may act like oil in an engine, and if insufficient, or poorly distributed, the engine may seize. While some quantity of information is disseminated naturally, so to speak, the idea that natural sources will be inadequate tends to support some forms of public or private intervention to ensure enough information is disseminated to keep things running smoothly. As described previously, the problem of information distribution was a focus of the information asymmetry literature that originated in the 1970s. Among that literature’s first prominent area of legal relevance was the analysis of capital markets. In 1984, law professors Gilson and Kraakman (1984) theorized that the distribution of information was a key determinant of capital market efficiency. Hence, institutions (like investment banks) capable of reducing the costs of obtaining information, and thereby making information more widespread, increased the efficiency of the capital markets. Writing in the same year, Coffee (1984) argued that the need for investors to have adequate information to make investment decisions justified the existence of regulations requiring disclosure of financial information. The argument that a market needs information in order to function efficiently has subsequently been used to justify the existence of quasi-​ private institutions like credit rating organizations.15 But capital markets are not the only systems that require adequate and well-​distributed information to function well. Markets for consumer goods and services do as well, yielding theoretical support for interventions like consumer warnings, the publication of nutritional information,16 or the consumer protection requirement that information on labels be generally accurate (Fung et al. 2007; Magat and Viscusi 1992; Sage 1999; Sunstein 1999). Meanwhile, in the regulation of banks, the markets for money claims has been said to depend on the maintenance of symmetric ignorance (Judge 2016). Non-​market systems, like the political system or elections, also require information to function well, which may justify various measures, such as the prohibition of political censorship in the First Amendment (Farber 1991). The extent to which the need for better distribution of information justifies government intervention depends, of course, on whether the private mechanisms of distribution are adequate. This question is very hard to answer in the abstract. Gilson and Kraakman (1984) theorized that the “distribution of a particular piece of information is a function of its cost,” by which they meant the cost of acquiring it or perhaps of producing it. They suggested that private mechanisms, such as investment banks, might aid the distribution of information but they did not make clear when such mechanisms might 15  See, e.g., White (2002, 43–​44), but see Fitzpatrick, IV and Sagers (2009), who question whether CROs can be justified by information-​production or disseminating role. 16  See, e.g., Valuck (2004), discussing the need for government intervention in the provision of information about dietary supplements. More broadly, see Magat and Viscusi (1992)

 

248   TIM WU be adequate. It is also true that markets themselves are, as Fitzpatrick and Sagers (2009) put it, “machines for generating information,” and many financial economists have long supposed that markets generate enough information to generate accurate prices without much intervention. On the other hand, it is obvious that there is much important information that one cannot expect to be widely distributed without any public intervention. Some might be simply too expensive to be worth producing (such as census information), creating the underproduction problem described herein. Alternatively, there are forms of information where the parties who hold it have good reason to hide it. Companies might want to hide their true revenues and profits, or the fact that a product is defective or causes disease. In these sorts of situations, scholars and government have suggested public action is necessary. While the literatures centered on information asymmetries and the previous discussion of underproduction are authored by different groups of scholars, a moment’s analysis reveals that they rely on the same conceptual framework, and ultimately the same observations about information itself. Asymmetries are typically assumed to result from some initial allocation of information that is expensive to overcome. Another way to express the same point is to suggest that high information prices cause problems (Frischmann 2009). But what might make problems of distribution of information different than any other good or commodity? The differences lie in the key concept of “nonrivalry,” which is the link between the information asymmetry literature and the public good scholarship. The concept is simply that consumption of information does not reduce its utility for others. Your reading of my book doesn’t “use it up” the same way that eating my sandwich does. This concept is expressed by economists in various ways, including the idea that information is “infinite in supply,” or experiences “jointness of consumption.” In mathematic models, nonrivalry is captured by the assumption of zero marginal cost of production. Perhaps reflecting its slightly mysterious quality, legal writers have often employed parables or analogies to capture the concept of nonrivalry. Lemley (1997) writes, “if I give you a fish, I no longer have it, but if I teach you to fish, you or I can teach a hundred others the same skill without appreciably reducing its value.”17 Perhaps the most famous parable in the intellectual property tradition is Jefferson’s (1813) analogy to fire and air. “He who receives an idea from me, receives instruction himself without lessening mine; as he who lights his taper at mine, receives light without darkening me.” Therefore, ideas are “like fire, expansible over all space, without lessening their density in any point, and like the air in which we breathe, move, and have our physical being, incapable of confinement or exclusive appropriation.” Jefferson took this to mean that private rights in

17  See also Goldstein (1994, 12) “A loaf of bread, once eaten, is gone. But ‘Oh, Pretty Woman,’ once sung and heard, is still available for someone else to sing and to hear. Countless fans can listen to the song, indeed copy it, without diminishing its availability to anyone else who wants to sing or listen to or copy it.”

 

LAW AND ECONOMICS OF INFORMATION    249 ideas cannot be justified “[i]‌f nature has made any one thing less susceptible than all others of exclusive property, it is the action of the thinking power called an idea.” Copyright’s fair use doctrine, while sometimes said to overcome failures of bargaining, has sometimes been justified by what is, on closer inspection, the solution to a distributional problem. Provided the producer is unaffected, or not unduly affected, the nonrivalrous nature of information suggests the public should want others to use the information at a price approaching zero. The famous Sony Corp. of America v. Universal City Studio, Inc. (1984) decision, which held legal the home taping of TV shows for later viewing, is easily viewed as the solution to an underuse problem. If the home taping did not actually deplete or affect the value of the television shows, there was no good reason not to allow it. In that case, might say the court set the price at zero. The analogies to fish and fire notwithstanding, it is worth asking, as we did with nonexcludability, does information actually have the characteristics of nonrivalry? In its purest form, this concept would presume that one’s usage of information would have no effect on another’s, that “a unit of the good can be consumed by one individual without detracting, in the slightest, from the consumption opportunities still available to others from that same unit” (Cornes and Sandler 1996). To use a common example, it seems implausible that one person’s use of a stop sign would change that experience for someone else. Some scholars believe that information’s nonrivalry is incontestable, as a matter of physics rather than law or economics. “The degree to which a good is or is not rivalrous is a fact of nature,” writes Benkler (2000), “a thing either does, or does not have this unusual attribute that, once produced, many can enjoy it without added cost.” But others argue that information, like private goods, might in fact sometimes be “overgrazed” or suffer from congestion, like a parcel of land or a highway (Magliocca 2001). The trademark laws, which employ concepts like blurring, tarnishing, and dilution, seems to contemplate this possibility, and Posner and Landes (2003) so asserted explicitly. They argued that excessive use of an image of Humphrey Bogart, for example, over time might cause “confusion, the tarnishing of the image, or sheer boredom … [e]‌ventually the image might become worthless.” Turning to the well-​known character of Mickey Mouse, Posner and Landes argued that the character might similarly become overgrazed without proper management: “not only would the public rapidly tire of Mickey Mouse, but his image would be blurred, as some authors portrayed him as a Casanova, others as catmeat, others as an animal-​rights advocate, still others as the henpecked husband of Minnie.” Posner and Landes’s (2003) assertion has gained a few adherents,18 but it has also been subject to sharp criticism (Yoo 2007; Lemley 2004). Such congestion arguments, argued Mark Lemley (2004), “misunderstand[] the nature of information,” which “cannot be depleted.” Christopher Yoo (2007) argues that the Posner and Landes argument is hard to sustain beyond the particular examples of celebrity images or characters, which have an unusual economics all of their own. Yoo accepts the possibility of congestion, but 18  Barnes (2011, 553) seeks to extend the concept as follows: “analytically, it is useful to distinguish between the simple ‘over-​reproduction’ type of overuse that causes the public to tire of the image and ‘transformative’ overuse that blurs the image.”

 

250   TIM WU notes that usage of information will not predictably decrease its value; moreover, if there are congestion externalities, says Yoo, they might be either technological or pecuniary; and if the latter, the policy implications become highly ambiguous. Examples make clear that congestion problems or overgrazing concerns are certainly not present for all forms of information. (This is a point Landes and Posner (2003) concede by alluding to Shakespeare’s works, which “seem undiminished by the proliferation of performances and derivative works, some of them kitsch[.]‌”) Strictly speaking, Landes and Posner were writing about copyrighted works, which are a subset of information; but consider a useful piece of information such as the circumference of the earth. While clearly a form of valuable information, how it might be used up or tarnished is unclear. Even if made to serve as a central plot element in a bad romantic comedy, the number (40,075 km) would retain its value and remain useful to others. What does seem to be potentially subject to dilution or tarnishing is not the information itself, but something like the reputation attached to the information, which seems an analytically distinct category. The idea that underusage is a problem also depends on information actually being nonrivalrous, or infinite in supply, not just as an ideal object, but as a market reality—​ how often is the information itself actually the relevant good? Nachbar (2007) argues that for many purposes, it is other economic factors with which policy should concern itself. “Intellectual works” he states, “do not exist except as the product of human labor, which is itself the subject of considerable rivalry.” For example, in the case International New Service v. Associated Press (1918), as Nachbar (2007) explains, it may have been true that the information taken from one news service by another was a public good. However, the court, according to Nachbar, was not concerned with the unusual properties of information, but with competition for profit as between the two news services. His point is that esoteric questions about the nature of information, such as those just considered at length here, can often be irrelevant for many markets. * * * We can here summarize the current state of understanding. Information is widely agreed to have unusual properties as compared with physical resources, chief among which are some measure of nonrivalry, nonexclusivity, and a low marginal cost of production. These properties have been used both to justify specialized legal regimes designed either to overcome underproduction and asymmetry problems. The extent of all of these properties is contested, both as a matter of theory and of market reality, or is said to be context dependent. This tends to suggest that the case for laws governing information is difficult to state generally, and may vary widely by industry, context, or the means of production involved. While describing information as a public good provides a simple justification for government intervention, as Coase (1974) famously pointed out with respect to lighthouses, the reality is often far more complex (there were, in fact, private lighthouses). This might lend support to the idea, for example, that the subject matter of copyright and patent might be better served by rules that are adjusted by an ongoing judicial process, rather than being subject to blanket rules. In areas outside of intellectual property,

 

LAW AND ECONOMICS OF INFORMATION    251 it similarly suggests that regulators ought to pay careful attention to determine whether, in fact, the information they wish to see disclosed requires intervention, and how effectively consumers will really make use of the information. Meanwhile, a continuing problem for regulators lies in dealing with informational asymmetries as a class. For while most government interventions seek to get more information “out there,” an asymmetry is not necessarily cured by interventions designed to increase the production of information, for unless the information gets into the right hands, it may in fact worsen the problem. It is also curious that, given the myriad properties of information, nonexcludability and nonrivalry have received so much attention. One may be suspicious that the attention is prompted by how neatly this model fits into the pre-​existing concept of a public good more than the underlying realities of what properties information holds. In any event, it is worth suggesting that lawyers’ or economists’ understanding of information’s properties might be broader and might begin to draw less on just anecdotal examples but more on the study of the science of information. It may turn out that information’s other less-​studied properties will be equally important for public policy. Consider, for example, the question of what effect usage has on the market value of information. While the question has not been well studied, two views are implied by the literature: namely the “overgrazing” view that information necessarily degrades or is tarnished by usage, and its rejoinder that usage cannot have any effect on the value of information. Neither seems to be correct, and the relationship between usage and value may in fact be more complicated than first appears. It is interesting to notice that many suppliers in the information economy devote themselves to trying to convince the public to consume information, and will even sometimes pay them to do so. The idea of the “attention economy” refers to concentrated efforts by suppliers not to avoid the usage of their information but to encourage it to the broadest extent possible—​captured by the desire to see information “go viral” and reach millions of users. This is not a new phenomenon: record labels have long wanted their music to play on the radio, and have long been willing to pay radio stations to, in effect, give away their product for free. If information were predictably subject to overgrazing, this would be a terrible idea; similarly, if information’s value could not possibly be affected by usage, why try so hard to get people to consume it? Something is missing. There exist several potential explanations for the phenomenon. Some theorists describe a certain class of informational goods that gets more valuable with usage. They are what Sunstein and Ullmann-​Margalit (2001) call “solidarity goods”:  goods that increase in value with joint consumption. As the authors write, “Solidarity goods have more value to the extent that other people are enjoying them.” While solidarity goods are not exclusively informational, many of Sunstein and Ullmann-​Margalit’s examples are forms of information. “The value of a magazine or television program focusing on a current topic (genetic engineering of food, e.g.) may increase significantly if many other people watch or read them.” This suggests that, rather than trying to discourage usage of some information, the owner of the information has some incentive to increase consumption of the information in question and to maximize the value of the good.

 

252   TIM WU The theory of solidarity goods also seems an incomplete explanation. It does not explain, for example, the intuition that a song may become more valuable by repeated play to even a single consumer, regardless of any group effects. Another and perhaps more obvious explanation is that one seeks to push information out for free as a form of advertising. Advertising, at least in Galbraith’s (1958) account, is a tool for creating demand, and, by this theory, the song is played to create demand for itself. But that doesn’t tell us enough: the usual definition of the word advertising is something that “calls attention to goods for sale.” It seems a different matter when information generates demand for itself and, therefore, by usage, increases its own value. Watching the exploits of a copyrighted character like Iron Man on the screen or page may generate an insatiable appetite for more Iron Man. By establishing itself in the minds of consumers in this way, information can become incredibly valuable, in part, because it can then be used to draw consumers to still other objects by an attention merchant.19 A different theory suggests that we are approaching the problem incorrectly, for someone listening to a “free” song or watching a “free” video is actually paying for it, not in units of money, but in human attention, a different scarce resource. By this measure, deciding to spend time with any information should be seen as an expenditure, dissolving the apparent paradox. In the law, trademark is best acquainted with this dynamic. With its doctrines of blurring, tarnishment, and dilution, trademark seems to recognize that certain uses of information might reduce the value of information, or its reputation. On the other hand, firms pay millions to have their brands announced during sporting events, or placed in Times Square to be consumed by millions without payment, or even to have the brand mocked or portrayed in strange circumstances, so long as there is exposure (consider the prominent role played by FedEx in the film Cast Away). The result of such mass exposure is rarely degradation, but rather, the creation of brands, which are easily the most valuable form of intellectual property. In any event, understanding these mechanisms is one of many ways we might better understand the information economy and its regulation. It might entail trying to understand better the competition for cognitive space and attention that is so central to the information economy.

References Abramowicz, Michael B. 2003. “Perfecting Patent Prizes.” Vanderbilt Law Review 56, pp. 115–​145. Akerlof, George A. 1970. “The Market for ‘Lemons’:  Quality Uncertainty and the Market Mechanism.” Quarterly Journal of Economics 84, pp. 488–​500. 19 

Cf. Evans (2013).

 

LAW AND ECONOMICS OF INFORMATION    253 Arrow, Kenneth J. 1962. “Economic Welfare and the Allocation of Resources for Invention.” In:  National Bureau of Economic Research, The Rate and Direction of Inventive Activity: Economic and Social Factors. Princeton NJ: Princeton University Press. 609–​626. Barnes, W. David. 2011. “Congestible Intellectual Property and Impure Public Goods.” Northwestern Journal of Technology and Intellectual Property 9, pp. 533–​563. Bator, Francis M. 1958. “The Anatomy of Market Failure.” Quarterly Journal of Economics 72, pp. 351–​379. Benkler, Yochai. 2000. “An Unhurried View of Private Ordering in Informational Transactions.” Vanderbilt Law Review 53, pp. 2063–​2080. Benkler, Yochai. 2006. The Wealth of Networks. New Haven, CT: Yale University Press. Boyle, James. 2003. “The Second Enclosure Movement and the Construction of the Public Domain.” Law and Contemporary Problems 66, pp. 33–​74. Breyer, Stephen. 1970. “The Uneasy Case for Copyright:  A  Study of Copyright in Books, Photocopies, and Computer Programs.” Harvard Law Review 84, pp. 281–​351. Coase, Ronald. 1974. “The Lighthouse in Economics.” Journal of Law and Economics 17, pp. 357–​376. Coffee, John. C., Jr. 1984. “Market Failure and the Economic Case for a Mandatory Disclosure System.” Virginia Law Review 70, pp. 717–​753. Cornes, Richard and Sandler, Todd. 1996. The Theory of Externalities, Public Goods, and Club Goods. Cambridge, MA: Cambridge University Press. Dam, Kenneth W. 1999. “Self-​ Help in the Digital Jungle.” Journal of Legal Studies 28, pp. 393–​412. Evans, David S. 2013. Attention Rivalry among Online Platforms. University of Chicago Institute for Law & Economics Online Research Paper No. 627. Farber, Daniel A. 1991. “Free Speech without Romance:  Public Choice and the First Amendment.” Harvard Law Review 105, pp. 554–​583. Fitzpatrick, Thomas J. IV, and Sagers, Chris. 2009. “Faith Based Financial Regulation: A Primer on Oversight of Credit Rating Organizations.” Administrative Law Review 61, pp. 557–​610. Frischmann, Brett M. 2009. “The Pull of Patents.” Fordham Law Review 77, pp. 2143–​2167. Frischmann, Brett M. and Lemley, Mark A. 2007. “Spillovers.” Columbia Law Review 107, pp. 257–​302. Fung, Archon, Graham, Mary, and Weil, David. 2007. Full Disclosure: The Perils and Promise of Transparency. New York: Cambridge University Press. Galbraith, John K. 1958. The Affluent Society. Boston, MA: Houghton Mifflin Harcourt. Gilson, Ronald J. and Kraakman, Reinier H. 1984. “The Mechanisms of Market Efficiency.” Virginia Law Review 70, pp. 549–​644. Gleick, James, 2011. The Information: A History, a Theory, a Flood. New York: Pantheon Books. Goldstein, Paul. 1994. Copyright’s Highway:  from Gutenberg to the Celestial Jukebox. New York: Hill and Wang. Gordon, Wendy. 1992. “On Owning Information: Intellectual Property and the Restitutionary Impulse.” Virginia Law Review 78, pp. 149–​282. Hasen, Richard L. 1996. “Clipping Coupons for Democracy:  An Egalitarian/​Public Choice Defense of Campaign Finance Vouchers.” California Law Review 84, pp. 1–​59. Hemphill, C. Scott and Suk, Jeannie. 2009. “The Law, Culture, and Economics of Fashion.” Stanford Law Review 61, pp. 1147–​1200. International New Service v. Associated Press, 248 U.S. 215 (1918).

 

254   TIM WU Jefferson, Thomas. 1813. “Letter to Isaac McPherson, Monticello, August 13 1813.” In: Andrew Adgate Lipscomb, ed., The Writings of Thomas Jefferson. Washington, DC: Thomas Jefferson Memorial Association of the United States. Judge, Kathryn. 2016. “Information Gaps and Shadow Banking.” Draft, 2016, on file with author. Kapczynski, Amy and Syed, Talha. 2013. “The Continuum of Excludability and the Limits of Patents.” Yale Law Journal 122, pp. 1900–​1963. Kieff, F. Scott. 2001. “Property Rights and Property Rules for Commercializing Inventions.” Minnesota Law Review 85, pp. 697–​754. Kitch, Edmund. 1977. “The Nature and Function of the Patent System.” Journal of Law and Economics 20, pp. 265–​290. Kovarsky, Lee. 2006. “A Technological Theory of the Arms Race.” Indiana Law Journal 81, pp. 917–​970. Krawiec, Kimberly D. 2001. “Fairness Efficiency, and Insider Trading: Deconstructing the Coin of the Realm in the Information Age.” Northwestern University Law Review 95, pp. 443–​503. Landes, William M. and Posner, Richard A. 2003. “Indefinitely Renewable Copyright.” University of Chicago Law Review 70, pp. 471–​518. Lemley, Mark A. 1997. “The Economics of Improvement in Intellectual Property Law.” Texas Law Review 75, pp. 989–​1084. Lemley, Mark A. 2004. “Ex Ante versus Ex Post Justifications for Intellectual Property.” University of Chicago Law Review 71, pp. 129–​150. Lemley, Mark A. 2005. “Property, Intellectual Property, and Free Riding.” Texas Law Review 83, pp. 1031–​1076. Lichtman, Douglas. 2004. How the Law Responds to Self-​Help. John M. Olin Law & Economics Working Paper No. 232, University of Chicago Law School. Losee, Robert. 1998. “A Discipline Independent Definition of Information.” Journal of American Social Information Science 48, pp. 254–​269. Lyndon, Mary L. 1989. “Information Economics and Chemical Toxicity: Designing Laws to Produce and Use Data.” Michigan Law Review 87, pp. 1795–​1861. Macaulay, Lord Thomas Babington. 1841. “Speech Delivered in the House of Commons.” Magat, Wesley A. and Viscusi, W. Kip. 1992. Informational Approaches to Regulation. Cambridge, MA: The MIT Press. Magliocca, Gerard N. 2001. “One and Inseparable: Dilution and Infringement in Trademark Law.” Minnesota Law Review 85, pp. 949–​1036. Menell, Peter S. 1988. “An Analysis of the Scope of Copyright Protection for Application Programs.” Stanford Law Review 41, pp. 1045–​1104. Mill, John Stuart. 1848. Principles of Political Economy with Some of Their Applications to Social Philosophy. London: John W. Parker, West Strand. Moglen, Eben. 2003. The dotCommunist Manifesto. http://​emoglen.law.columbia.edu/​my_​ pubs/​dcm.pdf [Accessed 21 August 2016]. Nachbar, Thomas B. 2007. “The Comedy of the Market.” Columbia Journal of Law & the Arts 30, pp. 453–​466. Palmer, Tom G. 1989. “Intellectual Property: A Non-​Posnerian Law and Economics Approach.” Hamline Law Review 12, pp. 261–​304. Pierce, John, 1980. An Introduction to Information Theory:  Symbols, Signals and Noise. New York: Dover Publications.

 

LAW AND ECONOMICS OF INFORMATION    255 Raustiala, Kal and Sprigman, Christopher. 2012. The Knockoff Economy: How Imitation Sparks Innovation. New York: Oxford University Press. Rose, Carol M. 2003. “Romans, Roads, and Romantic Creators: Traditions of Public Property in the Information Age.” Law and Contemporary Problems 66, pp. 89–​110. Sage, William M. 1999. “Regulating Through Information:  Disclosure Laws and American Health Care.” Columbia Law Review 99, pp. 1701–​1829. Samuelson, Paul A. 1954. “The Pure Theory of Public Expenditure.” Review of Economics and Statistics 36, pp. 387–​389. Samuelson, Paul A. 1955. “Diagrammatic Exposition of a Theory of Public Expenditure.” Review of Economics and Statistics 37, pp. 350–​356. Shavell, Steven and van Ypersele, Tangay. 2001. “Rewards versus Intellectual Property Rights.” Journal of Law & Economics 44, pp. 525–​547. Sony Corp. of America v Universal City Studio, Inc., 464 U.S. 417 (1984). Sunstein, Cass R. 1999. “Informational Regulation and Informational Standing:  Akins and Beyond.” University of Pennsylvania Law Review 147, pp. 613–​674. Sunstein, Cass R. and Ullmann-​Margalit, Edna. 2001. “Political Goods.” Journal Political Philosophy 9, pp. 129–​149. Valuck, Tom. 2004. “Keeping Dietary Supplement Regulations Slim and Fit: Finding a Healthy Balance between Paternalism and Consumer Choice.” Georgetown Journal of Law & Public Policy 2, pp. 285–​314. Wheeler, John Archibald. 1990. “Information, Physics, Quantum:  The Search for Links.” In: Wojciech H. Zurek, ed., Complexity, Entropy, and the Physics of Information. Boulder, CO: Westview Press. White, Lawrence J. 2002. “The Credit Rating Industry: An Industrial Organization Analysis.” In: Richard M. Levich et al., eds., Ratings, Rating Agencies, and the Global Financial System. New York: Springer Publishing Company. 41–​63. Yoo, Christopher S. 2007. “Copyright and Public Good Economics:  A  Misunderstood Relation.” University of Pennsylvania Law Review 155, pp. 635–​7 15.

 

Chapter 11

OPEN-​A CCESS AND INFORMATION COMMONS Yochai Benkler

Open-​access commons are a family of institutional arrangements that are far more pervasive in modern complex economies than usually recognized in the economic literature. Core resources necessary for communications, innovation, trade, transportation, and energy are managed based on symmetric use privileges open to all, deploying nondiscriminatory allocation based on queuing where congestion occurs, rather than on exclusive proprietary control used to achieve price clearance. Highways, roads, sidewalks, navigable waterways, and open squares are central to intercity and urban commerce. The public domain in knowledge, data, information, and culture forms the foundation of innovation, markets, and creativity. Open standards, the core Internet protocols, unlicensed wireless spectrum, and the privately created open-​access commons in Free and Open Source Software (FOSS) runs most Internet infrastructure. More ambiguously, common carriage and public utilities that characterize basic infrastructures for energy and wired communications exhibit full or partial characteristics of open-​access commons. The defining characteristic of open-​access commons, by comparison to property, whether individual or collective, is their utilization of (a) symmetric use privileges for (b) an open, undefined set of users in the general public, rather than (a′) asymmetric exclusive control rights located in the hands of (b′) an individual legal entity or defined group (club) use, and (c) their primary reliance on queuing and some form of governance-​based allocation, rather than (c′) price-​cleared models, for congestion-​clearance and management. Far from being necessarily less efficient or unsustainable, for substantial classes of resources, we have observed many critical resources migrating from provisioning based on a reasonably well-​developed market in private exclusive rights to open-​access commons. These migrations have not occurred through regulatory intervention, but rather through private actions of users and producers. Carol Rose documented this transition in roads and squares in the nineteenth century;1 and it was 1 

See Rose (1986).

 

OPEN-ACCESS AND INFORMATION COMMONS    257 the case of open-​access Internet, which displaced proprietary services like CompuServe and Prodigy, in Wi-​Fi and similar open-​access wireless commons, which now carry the majority of wireless data communications despite early dominance of proprietary wireless carriage; it has been the case in software, where FOSS, in competition with proprietary substitutes, now accounts for much of the basic software for using the Internet; and it is developing in diverse models of open-​access publication, both in “net native” sites like Wikipedia and many hundreds of thousands of lesser forums, and in more traditional publication sectors, primarily in the scientific and educational fields. Despite their ubiquity, an apparently critical role in the growth-​critical domains of innovation, transportation, and infrastructures, and competitive successes in modern economies, little economic theory addresses itself to the emergence and resilience of open-​access commons. The baseline assumption is, instead, that open-​access commons are tragic:2 rational actors pursuing their self-​interest under an open-​access regime will suffer congestion and disinvestment. Public goods theory explains state or public ownership of some classes of resources—​like lighthouses. These resources are still owned as property, usually by the government or a publicly created body, and access to them may be open, where they are strictly nonrivalrous, as with lighthouses or national security, or may be allocated by some other mechanism of public decision making, whether democratic or authoritarian, rational or corrupt. We have theories for group ownership of certain resources, initially club goods theory,3 and more recently prominent, the Ostrom School of commons theory.4 Both theoretical approaches, however, require exclusion of nonmembers or noncommoners from the resource set owned as a club or a participants in the common property regime. A defining feature of both theories is that, by their own terms, they would converge with standard economic theory in predicting that, for example, open-​access wireless spectrum would fail, and certainly underperform a market in spectrum clearance rights. Yet, over the past 15 years, Wi-​Fi and similar open-​ spectrum approaches have outpaced property-​like wireless carriage and now carry the majority of data communications. They would similarly predict that a protocol capable of auctioning or otherwise prioritizing clearance slots for congestible data-​carriage networks such as IBM’s Token Ring for local area networks or some network protocols that competed with the Internet protocol would be more efficient and would outperform protocols based on a simple first-​come, first-​served protocol, with no limitations on who can use the system, such as Ethernet or the Internet Protocol, TCP/​IP. Yet the history of network technology in the past 30 years has seen protocols based on an open-​ access commons model of management (Ethernet and TCP/​IP) outcompete protocols able to achieve price clearance, and gain dominance as the core standards of network communications.

2 

See Hardin (1968, 1244–​1245). See Buchanan (1965, 1–​14); Olson (1965); Berglas (1976, 116–​121). See, generally, Cornes and Sandler (1996). 4  See Ostrom (1990). 3 

 

258   YOCHAI BENKLER Section 11.1 of this chapter offers a brief overview of the commons literature, emphasizing the distinction between Ostrom commons and open access. Section 11.2 offers thumbnail case studies of the emergence of open-​access commons in the digitally networked environment. The emergence of unlicensed wireless in physical infrastructure space; the rise of TCP/​IP and Ethernet in networking protocol space; the emergence of FOSS; the emergence of commons-​based production of content on the Net, and of open-​ access publishing in scientific and educational materials. Section 11.3 suggests that the emergence of open-​access commons reflects the combined effect of innovation economics under conditions of high uncertainty and the diversity of human motivations (generally studied in economics as a branch of behavioural economics). It concludes with a brief typology of open-​access commons and their proprietary parallels, organized along the dimensions of the models of provisioning the resource and models of governing the resource once provisioned. For all its richness, the work stimulated by the emergence of the Internet over the past quarter century has not yet caused a revision of property theory. This chapter is intended, in part, to help stimulate a reimportation of theory from what we have been learning at the unstable edge, to the core of traditional theory.

11.1.  Two Tales of all Commons: An Intellectual History Garrett Hardin’s parable of the Tragedy of the Commons set the terms of the debate over the commons for the following two generations. Following closely on the heels of Mancur Olson’s Logic of Collective Action, it laid the intellectual foundations for an abiding scepticism about the feasibility of open-​access commons. Resources to which anyone had a right of access would be overused and underinvested. Every extracted unit provided its full benefits to the extractor, while sharing its disinvestment costs with all other potential and future extractors; by contrast, every invested unit imposed its full costs on the investor, but the benefits it produced were shared with all other and future extractors. To avoid overextraction and underinvestment under these conditions, the resource had to be owned and managed—​regulated either by the state or by private owners. Demsetz’s theory of property rights followed these basic insights, arguing that property rights would be introduced as soon as the marginal increase in the value of the resource gained by converting it from commons to property became larger than the transactions costs of creating and maintaining a property regime in the commons.

11.1.1. The Ostrom School Two schools of work on the commons developed in response to this baseline story. The first, anchored in the work of Elinor and Vincent Ostrom at the Workshop in Political

 

OPEN-ACCESS AND INFORMATION COMMONS    259 Theory and Policy Analysis at Indiana University was primarily a response to Olson. Over decades of painstaking field research, the Ostrom School showed that groups could solve the problems of collective action without relying on the state for either of the two then-​dominant models: directly regulating behaviour or defining and enforcing private property rights. The work emphasized detailed studies of a carefully delineated set of institutions—​limited common property regimes (CPRs)—​applicable to a carefully defined class of physical resources:  common-​pool resources.5 Using highly context-​specific, detail-​rich case studies of these settings, under the Institutional Analysis and Development (IAD) framework Elinor Ostrom developed,6 and abstracting from them to the mainstream game theory and public choice theory, Ostrom was able to carve out a distinct and robust field that had enormous real-​world implications for development policy. It played a critical role as a major intellectual critique of the dominant model that privileged property rights as the core solution to collective action problems.7 CPRs range from the lobster gangs of Maine,8 through Spanish irrigation districts,9 to Japanese fisheries.10 CPRs are not open-​access commons. Indeed, Ostrom insisted on “the difference between property regimes that are open-​access, where no one has the legal right to exclude anyone from using a resource, and common property, where members of a clearly defined group have a bundle of legal rights including the right to exclude nonmembers from using that resource.”11 This definition would exclude congestible open commons like roads and highways, Wi-​Fi, or the Internet. The Ostrom School focused not on open access, but on the fact that groups could solve their collective actions problems without the state or exclusive property as among the members of the group. Nowhere is this clearer than in Ostrom’s description of the irrigation districts in Alicante, which used a CPR-​specific system of tradeable, divisible scrip-​ denoted rights in fractions of minutes of water. Such a fluid market was “a commons” in Ostrom’s framework, because it derived from a collectively created, non-​state, non-​ state-​defined-​property system. The critical policy claim of the Ostrom School was that these systems embodied local knowledge, and were superior both to state regula­ tion and standardized property rights systems, in that both of the latter abstracted too greatly from the diverse and distinct features of the resource set governed by the CPR. The primary policy implication was that rationalized modernization programmes, whether implemented as regulatory interventions or state-​sponsored public works, or as “privatization” through parcelling the CPR into individualized property rights, were likely to cause greater disruption and loss of local knowledge

5 

See E. Ostrom (1990). See Elinor Ostrom (2011, 9–​11). 7  See E. Ostrom (1990). 8  See Schlager and E. Ostrom (1992, 249, 257–​259). 9  See E. Ostrom (1990, 69–​82). 10  See Satria, Matsuda, and Sano (2006, 226, 233–​34). 11  See E. Ostrom (1990, 121). 6 

 

260   YOCHAI BENKLER about proper management of the common-​pool resource than leaving the existing CPR in place.12

11.1.2. Open Commons The majority of the work on open commons in the past quarter century has revolved around information and the Internet. But the first work to challenge the idea that open-​access commons were systematically tragic was Carol Rose, The Comedy of the Commons.13 Rose examined common law doctrines under which private property came to be declared open to the public as a whole: where the set of individuals who have the right is open and undefined. She discussed, in particular, roads and navigable waterways, as well as open squares or fields where gatherings (both social and market) were traditional. Rose’s primary explanation was an early version of network economics—​ demand-​side increasing returns to scale made it so that opening these resources to public use substantially increased usage, which, in turn, produced substantial enough positive externalities to dominate congestion costs caused by making the road or navigable waterway an open-​access commons. Most of the work on open commons has occurred in the relatively “easy” case of noncongestible information goods and the public domain, as well as in the “harder” case of congestible network goods: Internet and wireless communications. Information is nonrivalous and partially nonexcludable, which means that to the extent that proprietary mechanisms succeed in reducing the nonexcludability, they systematically lead to underutilization of information relative to optimal (the nonrivalry means the marginal cost of information, once produced, is zero, and any positive price leads to some deadweight loss),14 and because existing information is a core input into the production of new information (e.g., established scientific facts as a basis for new investigation, past innovations leading to new ones, and news reports input into analyses), treating information as property also increases the cost of new information production.15 So much is well established and thoroughly modeled, and it is commonplace that intellectual property sets up a tradeoff between the incentives it creates for innovators and creators through enabling them to extract rents from the products of their investment, and the underutilization of the information by users and second-​ generation information goods producers that result from the nonrivalrous nature of information.16

12 

See Hess and Ostrom (2003) p. 123. See Rose (1986, 778–​780). 14  This has been a standard argument since Nelson (1959). For another discussion of the perverse effect on incentives of open rights to information, see Arrow (1962, 616–​617). 15  See Scotchmer (1991). 16  See Boyle (2008, 17–​41); Lerner (2002, 221–​222). 13 

 

OPEN-ACCESS AND INFORMATION COMMONS    261 In the early 1990s, Litman published an early analysis of how copyright systematically preserved a substantial public domain, by design, as a fundamental resource set for works subject to copyright,17 and Pamela Samuelson investigated the critical role of open access to the incremental development process at the heart of software development.18 Boyle then began to expand this view and locate it in a political economy, framing the tension over property in information, knowledge, and culture as one between producers who depended on access to existing information, such as software developers, journalists, or rap artists, and producers who depended on control over stocks of cultural goods.19 Lemley criticized efforts to strengthen intellectual property rights based on endemic market failures in information production, as well as the “on-​the-​ shoulders-​of-​giants” effect and the significance of positive externalities in innovation,20 and Cohen underscored the unique economics of information to negate the growing use of the analogy of physical property to support stronger intellectual property rights.21 My own work built on these insights and extended them in two ways. First, by analyzing open-​spectrum commons, I expanded the analysis from the domain of strictly nonrival public goods (information, knowledge, and culture) to rival or congestible goods where open innovation effects dominated congestion-​clearance efficiency effects.22 Second, I explained the shared institutional form of the public domain and other open commons in terms of a shift in the institutional foundation of the industrial organization: decentralization of innovation.23 A shift from asymmetric exclusive rights to symmetric use privileges underwrote a decentralization of innovation, creativity, production, and exchange, and thereby permitted greater experimentation and diversity.24 Lessig combined these insights into a broad argument for preserving commons at every layer of the Internet: at the level of physical and logical infrastructures, and the level of the creative content people exchange over it.25 Frischmann then tied the work on Internet, wireless, and information to the work Rose did by developing a more general theory of commons in infrastructure.26 He developed an approach based on demand-​ side failures. Defining “infrastructure” as a broad range of goods that are important in the downstream creation of other goods, particularly in a broad range of public goods and “social goods” like rule of law or basic capabilities that a society’s moral commitments require that everyone have, and which “may be consumed non-​rivalrously for some appreciable range of demand,” Frischmann argued that failures in demand to express the full social value of these infrastructure goods will be systematic, and will 17 

Litman (1990, 965). Samuelson, (1990a, 1025; 1990b, 23). Her work clearly influenced the thinking of leaders in the software developer community: see Garfinkel, Stallman, and Kapor (1991, 50–​53). 19  See Boyle (1996, 174–​184). 20  Lemley (1997, 1049–​1057). 21  See Cohen, (1998, 462–​466). 22  See Benkler (1998, 287–​359; 2002). 23  See Benkler (1998, 1999, 354–​424). 24  See Benkler (2001, 84–​88). 25  See Lessig (Random House 2001, 147–​233). See, also, Benkler (2000, 561). 26  See Frischmann (2005, 917). 18 

 

262   YOCHAI BENKLER systematically lead to underprovisioning of the infrastructure good if left as a private property model.27

11.2.  Open-​Access Emerging in the Presence of Proprietary Alternatives: Cases from the Networked Environment Over the past 25 years, open-​access models repeatedly have emerged and become stable, sometimes dominant, models for provisioning and managing critical segments of all layers of the information environment. The repeated success of these practices suggests that, at a minimum, open-​access commons are more stable and sustainable than the standard tragedy of the commons model permits. The fact that they were chosen by agents in market and social settings rather than through regulatory fiat, and succeeded in terms of market adoption by consumers or firms in the presence of well-​developed proprietary alternatives, suggests further that they provide some affirmative advantages over proprietary or closed-​commons alternatives with which they compete.

11.2.1. Unlicensed Spectrum No core resource in contemporary society better reflects the intellectual and institutional history of thinking about rationalized regulation of public goods, the shift to privatization and price-​cleared markets, and, finally, the emergence of commons, than spectrum regulation.28 Ubiquitous connected computing would simply be impossible without extensive use of wireless communications. From the now-​mundane smartphone to the exotic driverless car, through heart monitors, smart grids, or inventory management and shipping, the major innovations in early twenty-​first century information technology, built on tiny computers embedded in everything, could not develop if they were limited to communicating through wires. From the early 1910s until the 1990s, “spectrum,” the range of frequencies usable for wireless radio communications, was treated as a regulated public good, subject to administrative rationing and regulation. It was the subject of public ownership and funding in most of the world, and the subject of licensing and close federal regulation in the public interest in the United States. For half that period, spectrum regulation was also the subject of a sustained intellectual effort to explain that clear definition of exclusive property rights, initial allocation through 27 

28 

See Frischmann (2012). For a more complete intellectual history with citations, see Benkler (2012b, 76–​100).

 

OPEN-ACCESS AND INFORMATION COMMONS    263 auction, and subsequent allocation through flexible secondary markets would be a far superior alternative to the command and control regulation that was the exclusive model used throughout the world to regulate spectrum. Indeed, it was in developing his argument for spectrum auctions in a 1959 paper entitled the Federal Communications Commission that Ronald Coase first developed the argument that became The Problem of Social Cost a year later. By the early 1990s, the superiority of a property-​in-​spectrum market had become the orthodoxy among economists working on wireless regulation, and the private property approach gained partial acceptance in institutional practice. Over the course of the 1990s, spectrum auctions became the norm, and by the early 2000s, secondary markets, particularly in the United States and Australia, were made substantially more flexible. By the late 1990s, however, a technological alternative had developed that utilized “junk bands” that had been set aside as open-​access commons for the emissions of industrial, scientific, and medical radio signals. In 1998, the precursor to the Wi-​Fi standard was adopted, and in 1999, the first Wi-​Fi standard 802.11b was adopted. Initially treated as a regulatory backwater, and intellectually rejected by mainstream economic thinking on the subject, the unlicensed wireless spectrum drew innovation from a wider range of companies and amateurs than those involved in the proprietary spectrum. Investing in computation-​intensive cooperative architectures to manage congestion, rather than on clearing competing bids for exclusive pricing, the wireless capacity of the unlicensed spectrum began to roughly double every 20 months, roughly parallel to Moore’s Law for computation capacity. By 2012, entire industries that had been projected as major areas of growth for proprietary wireless carriers—​mobile health applications, smart-​grid communications, mobile payments, and inventory management—​had come to be dominated by a range of unlicensed wireless technologies, leaving proprietary spectrum-​ based applications niche markets. Even the most obvious success of cellular proprietary architecture, mobile Internet access data, had come to mix Wi-​Fi and cellular, and, over time, the majority of data, even to smartphones, came to be carried over Wi-​Fi.29 Wi-​Fi and similar bands that have become the predominate infrastructure for wireless data communications are open-​access commons with precisely the characteristics that the tragedy of the commons predicted would lead to tragedy, under conditions that attracted substantial orthodox scholarship that predicted precisely this failure. Reality has gravitated in favour of a true open-​access commons. This outcome was the product of two competing dynamics. One dynamic was that at the margin, an owner of any given slice of spectrum could clear competing uses and deliver more reliable communications capacity in that band. The other dynamic was that an open-​access band permitted any device, manufactured by any manufacturer and deployed by any consumer for any purpose, to be deployed, tried, and adopted or abandoned based on its effectiveness for the task required. In the open-​access bands no entity has exclusive control to regulate who deploys what device where. That openness created a sufficient level of innovation by diverse producers, innovation in using computation or coordination to solve 29 

See Benkler (2012a), for detailed analysis of these sectors.

 

264   YOCHAI BENKLER “noise” and innovation in services and architectures, that it has resulted in greater carrying capacity for the commons than is available in proprietary bands. As a practical matter, the open, decentralized innovation permitted in open-​access spectrum dominated whatever higher efficiency was obtainable from centralizing decisions about allocation and assignment in proprietary bands in most classes of communication. Proprietary bands retain their advantage in certain important types of communications, particularly communications that move very fast and need continuous coverage, but have been overtaken by commons-​based models in most sectors that require wireless data communications capacity.

11.2.2.  TCP/​IP vs. ATM In 1985, TCP/​IP was adopted by the Internet Advisory Board (an informal collection of engineers working on the Internet problem) as the Internet protocol. TCP/​IP is an open-​access commons, in the sense that it treats all packets identically, and requires the developers of end applications to design their applications so they are robust to any likely delay that results from queuing being the congestion management protocol. It developed and was adopted as an informal protocol, by volunteers who claimed no proprietary interest in it, even though there were both proprietary firm-​developed alternatives from IBM and DEC (Digital Equipment Corp.), and formal standard body alternatives, the International Organization for Standardization (ISO) and the International Telecommunications Union (ITU). Most pertinent from our perspective here was the effort, primarily in the 1990s, to develop a protocol that would allow telecommunications carriers to differentiate between packets and provide priority to some packets, which, in turn, would permit price clearance over queuing. The Asynchronous Transmission Protocol (ATM) was touted by telephone companies as the next generation of IP, but in fact failed to outcompete its open-​access competitor. The rate of innovation enabled by the fully open standard, with its commitment to end-​to-​end open innovation outcompeted those applications developed to take advantage of a more controlled set of network resources.30

11.2.3. Free and Open-​Source Software Moving from physical infrastructure and standards to software, the well-​known story of FOSS has served as a poster child for the success of commons-​based strategies for over a decade.31 FOSS reflects a licensing practice that voluntarily “contracts out” of a proprietary regime and instead adopts an open-​access commons. All FOSS licenses create an open-​access regime for the software developed. Anyone can copy the code, modify 30  31 

See van Schewick (2010). See (Benkler 2002); Kelty (2008); Schweick and English (2012).

 

OPEN-ACCESS AND INFORMATION COMMONS    265 it, use it, and redistribute modifications if they so choose. The major division among the various licenses is that some licenses, most prominently the BSD (Berklely Software Distribution) licenses create a simple open-​access regime. The other class, most prominently the GPL (General Public License), imposes a reciprocity condition on the rights of any user who modifies and distributes the software. That is, any modified software under the GPL must be released under the same open-​access terms that the modifier received it. But even the GPL is open access (a) with regard to use, redistribution, and modification for own use; and (b) imposes a nondiscriminatory reciprocity requirement on all, essentially a requirement to reseed the commons as a condition for making certain, more intensive uses of it. It is decidedly not a common property regime that is based on exclusion from the shared resource of all but a defined set of commoners. By the end of the first decade of the twenty-​first century, FOSS had come to account for between 65% and 70% of the web-​server software market; about 80% of server-​side scripting languages; an undocumented but large portion of embedded computing running Linux, as well as forming the kernel of the handheld Android operating system; and accounting for about one third of the web browser market, in the form of Firefox. By one account, about 40% of firms engaged in software development engaged at least in some of their work in FOSS development.32 FOSS has certainly not driven proprietary software development out of the market, but considering that it is based on a development model in which no one exerts exclusive rights over the final project, and about half the people involved in any significant extent are not paid for their contributions, the growth, success, and sheer technical excellence of FOSS defies explanation under traditional models that privilege proprietary and firm-​based development. There are aspects of FOSS that go directly to organization theory, but are beyond the scope of this chapter.33 For our purposes here, it is enough to note that one of the most dynamic, growing areas of innovation and production has seen a wide-​scale, sustained, and effective adoption of an open-​access commons in its core resources, both inputs and outputs, even on the background of a well-​developed property system and a well-​developed set of firms that were developing software on a proprietary model before FOSS burst on the scene as a major organizational and institutional alternative.

11.2.4. Wikipedia and Commons-​Based Peer Production of Content More Generally Paralleling FOSS, Wikipedia is one of the handful of most visited sites on the Internet, and it has established itself as one of the most important general knowledge utilities on the Web. It is edited by thousands of volunteers who manage their affairs internally without contracts, property, or state fiat, and without payment. Its inputs and outputs are all open-​access commons, licensed under a Creative Commons Attribution ShareAlike 32  33 

Lerner and Schankerman (2010). For a review chapter, see Benkler, (2016).

 

266   YOCHAI BENKLER license, which shares its core features with the GPL described previously, applied to cultural creations as opposed to software. Comparative studies over the years have mostly found Wikipedia to be of reasonable quality: imperfect, but not more so than other encyclopedias, including the standard-​setter, Britannica. Studies oriented in particular towards scientific entries found Wikipedia to be reliable. The National Cancer Institute study in 2010 was a particularly powerful example, where Wikipedia articles on various common cancers were found to be of equivalent accuracy, though less user-​friendly and readable, than the National Cancer Institute’s professionally produced explanations for patients.34 In terms of formal institutional framework, Wikipedia is an open-​access commons. Moreover, organizationally and technically, it is designed to permit anyone to edit it, whether they log in as a user or not, although no one is paid to do so. Lacking contract, exclusive property, or fiat, Wikipedia is the most complex and successful instance of large-​scale sustained self-​governance that we have observed on the Net, and quite possibly anywhere. This self-​governance process is best understood through the prism of the IAD and common property regime literature, since while formally there are no constraints on membership, and institutionally there are no exclusive rights, the social practice revolves around a well-​articulated social-​norms structure governing this vast community. For close to a decade, as of this writing, the number of editors who contributed more than five edits per month to Wikipedia in all languages has floated between 75 and 85 thousand, and the number of editors who contributed more than 100 edits per month has floated between 10.5 and 11.5 thousand; the English-​language Wikipedia has about 33% to 40% of those numbers, respectively. By any account, that is a very large number of active contributors who are managed in a complex, vague system of overlapping elements, none of which quite fit any crisp, well-​defined model of governance. As Wales put it, “Wikipedia is not an anarchy, though it has anarchistic features. Wikipedia is not a democracy, though it has democratic features. Wikipedia is not an aristocracy, though it has aristocratic features. Wikipedia is not a monarchy, though it has monarchical features.”35 A particularly insightful analysis of this set of overlapping features is developed in three chapters of Joseph Reagle, Good Faith Collaboration,36 although the work on Wikipedia governance is extensive, and a substantial portion of it is more criti­ cal of one or many aspects of the community’s governance processes and practices.37 While Wikipedia’s position as the leading instance of commons-​based peer production is clear, smaller-​scale projects that rely on open-​access commons for cultural production are legion. Stack Overflow or Quora are likely best known, but Wikia, the company founded by Jimmy Wales to host Wikis, hosts several hundreds of thousands of Wikis. The P2PValue project, funded by the EU, is building a large database of 34 

See Rajagopalan et al. (2010). See Wales (2007). 36  See Reagle (2010, Chap. 4–​6). 37  An excellent bibliography is found in Morell (2011); the volume generally collects a substantial amount of recent work that develops a critique of the more optimistic interpretations of Wikipedia. 35 

 

OPEN-ACCESS AND INFORMATION COMMONS    267 hundreds of case studies of peer production projects.38 Rather than an exception or a quirk, building knowledge bases using commons-​based, rather than proprietary models has become a standard approach in the menu of possible models. Wikipedia and FOSS are the most visible and successful examples of an alternative production model that has developed online: commons-​based peer production.39 The focus of this chapter is on commons, and largely excludes discussion of the economics of distributed innovation or peer production specifically. Briefly, the simplest model of peer production focuses on transactions costs. Social exchange is a transactional framework parallel to price-​cleared, managerial, and government transactional frameworks. The Coasean transactions costs explanation of the firm can then be extended to social transactional frameworks. In particular, where human capabilities and motivations are diverse and therefore hard to specify and contract, where tasks are complex and require diverse forms of human capital and insight, and where resources that could go into an information production task are similarly diverse, and may be possessed by different people, the transactions costs associated with a proprietary information production project—​including contracting both for the necessary information inputs and the necessary human resources, can be very high. Social production allows people to self-​assign, explore a large opportunity space of information resources in the commons and potential collaborators, and get together without the associated contracting costs, and without the monitoring and compensation system costs associated with firm-​based, proprietary production. Where information inputs that are nonrivalrous can be combined with practices that require little capital or capital already in service in households (e.g., computers and communications capacity), and human time that would otherwise be focused on consumption can be refocused on production that treats the tasks as play, social production can emerge as the most efficient model.40 Similarly, collaborative user innovation is particularly effective where communication costs are low and the design task can be rendered modular but would be very expensive if borne by a central actor.41

11.2.5. Open-​Access Publication and Creative Commons Beyond peer production, we are observing substantial efforts to shift practices that, in the past, have emphasized proprietary control to commons-​based models. Academic publication in particular has seen a significant shift towards open-​access publication.42 In some disciplines, most prominently physics and computer science, academic publication is almost exclusively open access. That is, the research outputs published are available under an open-​access license. In other disciplines with more well-​established, 38 See http://​directory.p2pvalue.eu/​home. 39 

See Benkler (2002, 369). See Benkler (2002). 41  See Baldwin and von Hippel (2010). 42  For an extensive review of the history, types, and progression of open access, see Suber (2012). 40 

 

268   YOCHAI BENKLER dominant publishers, most importantly biology and medicine, where the dominance of proprietary publications is difficult to surmount because of the importance of publication in those venues for authors’ professional advancement, open-​access publishing has developed more slowly. Nonetheless, even in these areas, the emergence of the PLoS (Public Library of Science) journals has created a significant venue for high-​quality scientific papers even in the presence of these extremely high-​impact proprietary journals. At the time of this writing, PLoSOne is the largest scientific publisher in the world. Open-​access scholarly communication has its roots in several efforts beginning in a 2000 petition called for by Harold Varmus, Patrick O. Brown, and Michael Eisen, calling on senior scientists to commit to publish only in open-​access journals. In late 2001 early 2002, many of the aspirations of the movement were set in the Budapest Open Access Initiative. The core idea was that scientific publication has never been driven by royalty payments to authors or reviewers, and while the cost of professional production could be significant in some journals, the pricing of journals by publishers reflected monopoly power over access to knowledge that was generated by scientists funded through public and philanthropic funding, who had interest in the widest possible dissemination of their work, rather than high royalties, which they themselves did not enjoy. Over the 15 years of its development, open-​access publication has increased, several major universities have undertaken to have at least the prepublication version of their faculty’s work available in open-​access repositories (often called Green Open Access repositories), and several major government and philanthropic funding agencies have mandated, or at least supported, open-​access publication fees, which go to publishers in lieu of royalties to fund the professional editing aspect of the work while keeping the works available for downloading by anyone, anywhere, free of charge. Of all other forms of successful open-​access practices, open-​access publishing of academic work is the least puzzling. Basic science is a public good that enjoys public and philanthropic funding, and so separates the production price and the consumption price in a way that overcomes some of the standard problems with private provisioning of public goods using proprietary exclusion. Academic scientists likely self-​select because of a preference profile that is happy to trade off money income for a range of nonmonetized desiderata, from status to freedom to be creative. Nonetheless, it offers a valuable example of a practice that is a critical element of growth in market societies that has moved away from an exclusively proprietary model, towards an open-​access model over the past 15 years.

11.2.6. Open-​Access Case Studies: Conclusion In all, the diverse case studies, from spectrum and standards to software, general knowledge, cultural production, and academic publication are intended to underscore the extent to which in the digitally networked information environment we have seen, repeatedly and significantly, the voluntary adoption of open-​access commons-​based practices in the presence of pre-​existing proprietary models. This has happened not

 

OPEN-ACCESS AND INFORMATION COMMONS    269 as holdovers from preindustrialization cultural practices in the global market periphery, but at the heart of the most advanced economic sectors in the most advanced economies.

11.3.  Theories of Commons Open-​access commons are a family of institutional solutions that respond to three practical problems under certain resource conditions. The three practical problems are (a) high persistent positive externalities, of which nonrivalry in information goods is an extreme case; (b) uncertainty, under which exploration trumps appropriation and has its primary impact in innovation; and (c) social disembeddedness, or the risk that markets will drive resource utilization in ways that will lead to social instability or political intervention. This latter set of problems, following from Karl Polanyi’s definition rather than the more recent usage, must await treatment in a different publication, owing to space constraints and placing greater demands on law and economics literature as it stands. Pursuing the former two in a narrower, new institutional economics framework will have to suffice for this chapter. In terms of characteristics of the resource set, open access is most feasible in the case of resources that are either nonrivalrous (information, knowledge, and standards) or partially congestible, with variable loads over time such that for substantial ranges of their operation their use is not congested—​roads, electricity, and spectrum have this characteristic. The extent to which the resource is depleted by use or renewable also contributes to its amenability to open-​access management. The less congestible the resource is, the less benefit is gained from introducing asymmetric excludability except as a solution to initial provisioning. The more prominent the periods of noncongestion in the total utilization range, the less benefit there is for instituting an asymmetric exclusivity regime to clear peak demand periods. The more a partially congestible resource is renewable, like spectrum, the less significant the problem of disinvestment, or cumulative congestion over time, is. The more the resource requires continuous reinvestment, such as with roads or the electricity grid, the more we see members of the open-​access family that are integrated with public provisioning or some other form of payment for use that nonetheless retains the symmetric use privileges but attaches it to use payments. Open access can be required in a society even where all these conditions are absent, where the social implications of exclusion dominate the efficiency concerns of nonexclusion. Emergency room care is an obvious example. As Kapczynski and Syed have shown, there are classes of rules of intellectual property best explained by this form of deep nonexcludability.43

43 

See Kapzcynski and Syed (2013, 1900).

 

270   YOCHAI BENKLER

11.3.1. Positive Externalities Beginning with Rose’s Comedy of the Commons, which explored the emergence of open-​access commons in roads and navigable waterways as a function of the positive returns to scale that travel provided a growing continental commercial system,44 a core explanation of the emergence and success of open-​access commons has been their utility in providing the resources necessary to support high positive externality activities. Frischmann then emphasized these “spillovers” as central to a wider range of “infrastructure” goods.45 Information goods are a particular subset of this problem, in that (a) the nonrivalry means that efforts to internalize the positive externalities by creating enough exclusivity to support appropriation through charging a price will necessarily result in deadweight loss;46 and (b) high exclusion leads to higher costs for downstream innovators or creators, because of the “shoulders of giants” effect (innovation is both input and output of its production; proprietary exclusion therefore increases both costs and expected benefits). As a result, the property-​like solutions represented by patents and copyrights suffer from the well-​known limitations and imperfections, while the public domain plays a critical role in seeding innovation and creative expression. Open access to academic publication is an example. The understanding of the role academic science plays in early-​stage research and investigation as producing high positive spillovers that cannot be captured through intellectual property is longstanding. The basic idea is that investigation that tries to internalize all its social value will necessarily focus on appropriable innovation, and, therefore, will necessarily be more narrowly focused. Given that early innovation is critical to later innovation, that it is a cumulative process, and that it is critical to growth, a proper growth-​oriented policy will seek to assure that there is some level of public funding for “basic” research—​that is, research usable as input in a wide range of research projects. Open access to that basic science as input for follow-​on innovation and investigation maximizes its spillovers. Debates over patenting of university innovation occur precisely along the lines of whether some degree of exclusivity will lead to greater effort to convert the basic science into usable technology, as compared with the risk that this exclusivity will lead to a narrowing of focus and a gain, in terms of appropriable investment, that is outweighed by the loss in focus on basic, high-​positive-​spillover science.47 The open-​access publication movement of the past 15 years takes this basic logic and applies it to the area where the argument in favour of appropriation is even weaker than the patent, because the investment necessary to convert a finished research product into a published paper is much smaller than the investment necessary to bring a science innovation to a product market, and the majority of the labour is done in peer review, as professional activity attendant to 44 

See Rose (1986, 768). See Frischmann, supra (2005, 2012). 46  For the original statement of the trade-​off, see Arrow (1962), supra; for an overview of the problem, see Bracha and Syed (2014, 1841). 47  Mowery, Nelson, Sampat, and Ziedonis (2004); Owen-​Smith and Powell (2001, 109). 45 

 

OPEN-ACCESS AND INFORMATION COMMONS    271 publication. Similarly, because science is incremental, the on-​the-​shoulders-​of-​giants effect is pronounced, further supporting the open-​access models. This also explains in large measure why software, where innovation is widely seen as incremental, is a field where we see open-​access commons in the form of FOSS licensing being chosen by both market and nonmarket actors. Roads, in the nineteenth and twentieth centuries, and the Internet and spectrum, in the twenty-​first, are examples of congestible goods associated with very high positive externalities. The more people use roads to travel to a city or between homes and workplaces; or use the Internet or spectrum to connect to services and social practices, the more congested the shared resource becomes, but also the more valuable the dependent activities. The city centre becomes more valuable as a trade centre the more potential trading partners there are; the application development market grows and becomes more valuable to all users when there are more users who use the Internet or wireless communications more often. Efforts to perfectly calibrate the price of using the infrastructure so as to maximize both (a) the efficient utilization of the infrastructure at any point, including congestion peaks, and (b)  the positive externalities associated with high usage require enormous amounts of information, about all users, and about all possible combinations of users and uses that might benefit from meeting each other. Given imperfect information and transactions costs, it is practically impossible to maxi­ mize on both dimensions. Where we have seen infrastructures with very large positive externalities, like roads and navigable waters or the Internet, we have seen that tension resolved in favour of growth in the system as a whole, through symmetric universal open access to the infrastructure resource, rather than in favour of efficiency of utilization of that resource. Depending on the cost of provisioning and the risk of disinvestment, we have seen these symmetrically managed resources range from simple open access and minimal-​use rules, like navigable waters, unlicensed spectrum, or the Internet protocol, to owned and priced symmetric use models like common carriage in telephony or public utilities in electricity distribution.

11.3.2. Uncertainty, Freedom to Operate, and Exploration Open-​access commons and property can also be interpreted as institutional mechanisms that represent significantly different information and motivation models. Property centralizes the point at which information and incentives necessary to determine the access, use, management, and disposition of a given resource in a single entity by giving that entity asymmetric power to determine who will get to access or use the resource, at what time, and for what purposes. The defining feature of commons is that there is no such asymmetric power. Instead, the resource is subject to a set of symmetric rules concerning access, use, extraction, and management. The absence of asymmetry removes the owner as a focal point for transactions and as the coordinating mechanism

 

272   YOCHAI BENKLER for competing claims on the resource. The symmetry allows diverse users the freedom to operate without transacting, within the symmetric constraints and subject to the congestion characteristics of the resource. As in the case of property and unlike regulatory decisions, information is gathered and processed by decentralized actors. Unlike the case of property, information gathered by these decentralized actors is not collated in a single decision point. Rather, diverse actors act upon information they have or exchange without the need to translate it into a universally understood expression (currency, e.g.) that compares competing uses and clears them. Where the level of uncertainty is such that freedom of action (to adapt to changed circumstances) is an important desideratum, in some cases, more than security in holdings (whose value and utility are part of the uncertainty) and power to appropriate outputs directly through exclusion (whose coming into being is part of the uncertainty)—​we need, and find ubiquitously around us, both commons and property. On this analysis, with perfectly frictionless markets under perfect information, we wouldn’t need commons. But this is no more relevant than saying that with perfectly selfless individuals under perfect information and frictionless social exchange we wouldn’t need property. Given imperfect markets, imperfect information, diversely motivated individuals, and imperfect systems of social cooperation and exchange, some mix of property and commons is necessary for reasonable planning and pursuit of goals. This is from the private-​returns perspective, setting aside collective goals such as efficiency and growth, and explaining the widespread adoption of commons-​based practices (like FOSS) even in the presence of property-​based alternatives. From an individual agent’s perspective, having a mix of resources—​ some commons, some property—​will increase his or her utility over time, given imperfect markets, persistent uncertainty, and change. The histories of spectrum commons and the Internet protocol offer nice illustrations. The combination of doubling of computation capacity every 18 months for decades, coupled with the global reach of the innovation system and its escape from the confines of a few well-​known labs, like Bell Labs, created a rate and range of change that led to true uncertainty (as opposed to risk, where we know the range of outcomes and distribution of probabilities) in innovation practices that depended on computation (whose innovation rate caused uncertainty, but was not itself uncertain, and has not been managed in a commons) and communication as core resources. TCP/​IP, the core Internet protocol, implemented an “end-​to-​end” design principle that effectively refused to optimize for any particular function within the network, and required all applications to take care of themselves—​that is, solve whatever higher-​level functions and optimization they required without making any demands on the network design itself. This design choice sacrificed optimization and efficiency of a known set of applications (e.g., voice or real-​time streaming) in exchange for high flexibility and decentralization of the capacity to innovate. It meant that when four clever Israeli programmers figured out instant messaging, or four Estonian programmers figured out a better voice codec, they did not need to ask permission from a network operator, they did not require the change

 

OPEN-ACCESS AND INFORMATION COMMONS    273 of network design that would still have to accommodate hundreds of other applications developed by others, but instead could simply design ICQ, the grandfather of instant messaging, or Skype, respectively, to do all of its work on the end-​user devices, and send through the network only the minimal simple packets the network was designed to receive and route. ICQ launched the Israeli high-​tech startup culture, but at the time, came out of nowhere. As for Skype, while the idea of video telephony had been around for decades, Skype’s solution used a modified peer-​to-​peer network that was built initially on an architecture that its creators had originally developed for the KaZaa peer-​to-​peer file-​sharing network. It used end-​user nodes to relay packets that had no quality of service assurance in the network. The approach would have been treated as a pipe dream within the telephony engineering system before it was successfully introduced. Indeed, Internet telephony was the most prominent driver of efforts to replace the Internet protocol so that carriers could distinguish packets and sell priority (quality of service commitments) to voice packets. Instead, the end-​to-​end, open-​access architecture of the Internet allowed for hundreds, or thousands, of low-​cost experiments in this and related fields to be run, implemented, and fail in a fully decentralized form until one or a few of them resulted in a superior solution. The same can be said of all the major innovations on the Internet—​from Berners Lee’s innovation of the World Wide Web, through the browser, the search engine, to the social network and the cloud storage company—​these were all the results of extensive experimentation that depended on the open-​access commons model of the Internet and came out as the winning solution among many parallel efforts of exploration in the face of a rapidly changing and highly uncertain innovation challenge. Similarly, in open-​access spectrum, we saw diverse companies develop diverse products to take advantage of open-​access spectrum bands, mostly Wi-​Fi but also other standards in diverse unlicensed bands, in a way that dramatically outpaced innovation by those few carriers who owned spectrum. In areas as diverse as smart-​grid communications systems, medical device wireless communications, or inventory management, the more cumbersome carriers that depended on proprietary spectrum allocations failed to keep up with the diverse range of innovative companies that relied on the commons.48 Ultimately, even the carriers themselves ended up turning to the commons to carry a majority of their wireless data requirements. When presented with major spikes in its network after introduction of the iPhone, AT&T had major congestion problems with its mobile data network. It could have gone to the secondary spectrum markets set up by the Federal Communications Commission (FCC) a few years earlier, where it could have leased the additional capacity in a spot market. It did not. Instead, it combined a long-​term proprietary strategy—​seeking to purchase licenses from Qualcomm—​with a short-​term, more dynamic solution that was based on the commons. AT&T invested in Wi-​Fi hotspots and encouraged users to off-​load traffic to their home and public Wi-​Fi

48 

See Benkler (2012b).

 

274   YOCHAI BENKLER spots. SFR in France, the second-​largest mobile provider and third-​largest home broadband provider, went one further and harnessed all of its home broadband subscribers—​ about 22% of the French market—​to become Wi-​Fi load-​balancing points for all their mobile data subscribers. Wi-​Fi off-​loading by carriers has become the norm, carrying anywhere from 35% to 65% of mobile data.49 More generally, we can say that the more diverse, uncertain, and rapidly changing the environment, the harder it is to codify the value of resources, uses, and outcomes, and the more attractive the freedom of action associated with having a resource in the commons is to these users. The symmetric constraints coupled with a general privilege to use the resource under these constraints mean that the need for transactions at the margin is eliminated, and with it transaction-​cost barriers: strategic behaviour of platform or essential-​facilities owners, imperfect information with its widespread risk of unmatched offer–​ask differences as a user seeks to obtain a sufficient flow of the resource, and so forth. The commons can be said to have a private option value to users whose price is (a) the reduced certainty of availability of a stated quantity of the resource as is available in markets, itself a function of how perfect or imperfect the relevant market is; (b) the lost appropriation opportunity from not having the resource controlled in a proprietary form, relative to nonexclusion-​based forms of appropriation that remain available without exclusion from the resource; and (c) the cost differential between the desired use in the market, given its imperfections and the cost of using the commons. The greater the background uncertainty as to the required quantity or quality of the resource and the market imperfections, the higher the option value—​that is, the more of the benefits of property an agent would be willing to forgo in exchange for the greater flexibility offered by commons, within its known constraints. Uncertainty is connected to one additional dimension of the economic advantage of open-​access commons. Different tasks are more or less amenable to the diverse motivations individuals bring to their actions. The better defined, more routine a task, the easier it is to specify the desired level and quality of effort, to monitor the outcome and connect it to performance and, therefore, to subject the behaviour to reward and punishment on a standard incentives model. The more a desired outcome depends on initia­ tive, tacit knowledge, insight, or creativity, the more uncertain a task environment, and the more the worker needs to continuously examine, explore, innovate, adapt, and apply diverse forms of effort to the task, the harder it is to subject performance to monitoring, or to price effort accurately. As a result, intrinsic motivations become more important, and price-​driven performance is harder to apply well. This difference helps explain why Mertonian science takes over from managerial models as the explored knowledge space becomes less known; why industrial labs that are oriented towards generating big innovation steps create organizational bubbles relatively protected from managerial control (e.g., Bell Labs and Xerox Labs), and why entrepreneurial firms tend to cluster around 49 

See ibid (103). The scale and scope of use, rather than the precise numbers, are what is important for purposes of this review.

 

OPEN-ACCESS AND INFORMATION COMMONS    275 universities, which are excellent institutions for harnessing a range of motivations, from the pleasure of inquiry, through status, to freedom and flexibility over time and area of application. Where resources are subject to open-​access commons, they more readily lend themselves to models that do not tie use directly to payment. Agents who seek to operate on non-​priced motivations can access the resources without concern for the ways in which their use could be translated into enough revenue to secure continued access to the resource. The success of commons-​based strategies in areas like software development, consumer reviews, video creativity, or factual writing reflects precisely the freedom to operate on nonmonetary motivations permitting the development of highly diverse creative and innovative practices.

11.4.  Types of Open-​Access Commons The family of institutional arrangements that fall under the category of open-​access commons is defined by its use of symmetric use privileges, rather than asymmetric exclusive rights, as the core allocation mechanism. The primary branches in the family tree depend on the provisioning of the resource and the governance of the symmetric use privileges. Table 11.1 offers an overview of the members of the open commons family and their exclusive property parallels. The table reflects four major provisioning systems (government, market, social, and natural) and four major governance approaches (state, property and contract, social norms, and no constraint). Each cell is divided into two: a light-​grey shaded subcell, where access to the resource is available to an open class on nondiscriminatory terms, and a black-​shaded subcell, where asymmetric exclusion is the organizing principle. The traditional antipodes (market and state) are represented by the categories of market provisioned, property and contract-​governed, asymmetric exclusivity subcell (hot dogs, homes, and so on), and the state-​provisioned, state-​ regulated, asymmetric exclusivity subcell (military bases and food stamps). Classic public goods are represented in the state-​provisioning, no constraint cell (lighthouses). The dominant modes of commons that serve as the foundation of commercial, industrial economies fall in the nonexclusivity subcells of the state-​and market-​provisioning cells. These can be subject to state regulation (highways, public utilities, mass transit when state provisioned, or common carriers, privately held utilities, or unlicensed wireless bands when market provisioned), or no constraint (open government data, or formerly IP protected materials now in the public domain). The more exotic phenomena that have developed in networked society—​free software, both commercial and purely socially produced—​occupy the subcells of social provisioning, mostly with no asymmetric exclusivity, while many of the CPRs studied by the Ostrom School occupy both the symmetric nonexclusivity (for uses within the CPR) and asymmetric exclusivity (for the relations between insiders and outsiders in CPRs) subcells of social norms-​ organized, socially produced goods.

 

Table 11.1  Types of Open Access Commons by Governance and Provisioning Model Governance/ Provisioning

State regulation

State: tax, bonds, Highways, and fees streets; public utilities; water; mass transit

Military bases; food stamps Markets: direct Common payment, indirect carriers; “private” appropriation public utilities; unlicensed wireless bands

Property and contract

Social-​cultural norms

No constraint

Null (if universal symmetric access right, then law, not contract, allocates)

Peer review for publicly funded science not patented; parks; city squares; sidewalks Publicly funded science that results in patents

Lighthouses; government data: weather, labour/​GDP measurements

Street performers; online musicians; voluntary compliance systems

Cultural materials and innovation originally commercial now in the public domain

Government contracts Broadcast reception (provision in market, but equal privilege to use); GPL/​BSD software by firms (for example, Android)

Automobile Hot dogs; Effort in high-​ safety standards; homes; personal commitment zoning computers; IP organizations goods in coverage Social: labor and Solid organ goods; donations donations

Nature

Contractually reconstructed commons; BSD, GPL? CC-​BY; CC-​ SA?; CC-​NC

Null

Null

CPRs inside, if need provisioning: for example, a dam; von Hippel innovation; Wikipedia editing; much CBPP; GalaxyZoo; Foldit; culturally constructed commons

TCP/​IP; the web; Wi-​Fi standards; much CBPP outputs; Wikipedia reading

Health regulation Enrollment applied to church in socially day care provisioned services

CPRs on the outside; Alicante irrigation system

Null

Pollution Privately created controls; national open nature parks; fisheries preserves

CPRs that require Air inhalation; open allocation: for ocean transit example, pastures

Tradeable permits Private recreation CPRs from the parks: for outside example, hunting lodges

Null

 

OPEN-ACCESS AND INFORMATION COMMONS    277

11.5. Conclusion This chapter was intended to familiarize readers with the literature on open-​access commons and the factual of prevalence open-​access commons in contemporary advanced economies. Contrary to the tragedy of the commons fable, open-​access commons are in fact ubiquitous in modern, complex economies, and play a critical role in making market economies function. Open-​access commons is not a single institution, but a family of institutional arrangements. The core-​defining feature of the family is that its subtypes all apply an institutional model that provides symmetric use privileges to an open general class of users, rather than assigning an asymmetric exclusion right to an individual or known class of individuals as do individual private property, club goods, or CPRs. Open-​access commons have emerged through choice in markets, social arrangements, or public policy, even where earlier property-​based institutions already existed. This has largely happened where use of the resource involved high positive externalities that cannot be internalized without substantial loss of total welfare; where innovation and exploration using the resource as input is particularly valuable, so that the innovation effects of permitting everyone to explore with productive uses of the resource dominate the efficiency effects of maintaining more controlled use for congestion avoidance; and where the resource is useful for a range of uses, including socially motivated use, which is unlikely to be able fully to express its social value if forced to be monetized. The economic theory of open-​access commons as a general theoretical problem is still in its infancy. Substantial work has been done on CPRs, on information commons, or the public domain versus patents and copyrights, on the Internet protocol and end-​ to-​end innovation, and on wireless spectrum regulation. There nonetheless remains substantial work to be done to synthesize these diverse forms of open-​access commons and explain at a more general level how these diverse commons interact with property to offer a more comprehensive theory of market economies and societies’ use of these two families of institutions. Even if the outline of the theoretical explanations available to date is unpersuasive to the reader, the fact of the large role of open-​access commons demands that we offer better theories. Economists cannot ignore the commons and continue to imagine that property is the interesting core, and commons a negligible periphery, in the face of large-​scale, critical elements of growth that adopt commons by choice and sustain them over time in preference to available proprietary alternatives.

References Baldwin, Carliss, and von Hippel, Eric. 2010. Modeling a Paradigm Shift:  From Producer Innovation to User and Open Collaborative Innovation. Working Paper, Harvard Business School. Benkler, Yochai. 1998a. “Overcoming Agoraphobia: Building the Commons of the Digitally Networked Environment.” Harvard Journal of Law and Technology 11, pp. 287–​400.

 

278   YOCHAI BENKLER Benkler, Yochai. 1998b. “The Commons as a Neglected Factor of Information Policy.” Presentation at the 26th Annual Telecommunications Research Conference. Alexandria, VA (October 3–​5). Benkler, Yochai. 1999. “Free as the Air to Common Use: First Amendment Constraints on Enclosure of the Public Domain.” New York University Law Review 74, pp. 354–​446. Benkler, Yochai. 2000. “From Consumers to Users:  Shifting the Deeper Structures of Regulation towards Sustainable Commons and User Access.” Federal Communications Law Journal 52, pp. 561–​579. Benkler, Yochai. 2002. “Coase’s Penguin, or Linux and the Nature of the Firm.” Yale Law Journal 112, pp. 369–​446. Benkler, Yochai. 2012. “Open Wireless vs. Licensed Spectrum:  Evidence from Market Adoption.” Harvard Journal of Law and Technology 26(Fall), pp. 69–​163. Benkler, Yochai. 2016. “Peer Production and Cooperation.” In: J. M. Bauer and M. Latzer, eds., Handbook on the Economics of the Internet. Cheltenham: Edward Elgar. 91–​119. Berglas, Eitan. 1976. “On the Theory of Clubs.” American Economic Review 66, pp. 116–​121. Boyle, James. 1996. Shamans, Software, and Spleens:  Law and the Construction of the Information Society. Cambridge, MA: Harvard University Press. Boyle, James. 2008. The Public Domain:  Enclosing the Commons of the Mind. New Haven, CT: Yale University Press. Bracha, Oren and Syed, Talha. 2014. “Beyond the Incentive-​ Access Paradigm:  Product Differentiation and Copyright Revisited.” Texas Law Review 92, pp. 1841–​1920. Buchanan, James. 1965. “An Economic Theory of Clubs.” Economica 32, pp. 1–​14. Cohen, Julie.E. 1998. “Lochner in Cyberspace:  The New Economic Orthodoxy of ‘Rights Management.’” Michigan Law Review 97, pp. 462–​563. Cornes, Richard and Sandler, Todd. 1996. The Theory of Externalities, Public Goods, and Club Goods. New York: Cambridge University Press. Frischmann, Brett, M. 2005. “An Economic Theory of Infrastructure and Commons Management.” Minnesota Law Review 89, pp. 917–​1030. Frischmann, Brett, M. 2012. Infrastructure:  The Social Value of Shared Resources. New York: Oxford University Press. Garfinkel, Simson L., Stallman, Richard M., and Kapor, Mitchell. 1991. “Why Patents Are Bad for Software” Issues in Science and Technology 8(Fall), pp. 50–​55. Hardin, Garrett. 1968. “The Tragedy of the Commons.” Science 162, pp. 1243–​1248. Hess, Charlotte, and Ostrom Elinor. 2003. “Ideas, Artifacts, and Facilities: Information as a Common-​Pool Resource the Public Domain.” Law and Contemporary Problems 66(Winter/​ Spring), pp. 111–​145. Kapzcynski, A. and Syed, Talha. 2013. “The Continuum of Excludability and the Limits of Patents.” Yale Law Journal 122, pp. 1900–​1963. Kelty, Christopher. 2008. Two Bits:  The Cultural Significance of Free Software. Durham, NC: Duke University Press. Lemley, Mark A. 1997. “The Economics of Improvement in Intellectual Property Law.” Texas Law Review 75, pp. 989–​1084. Lerner, Josh. 2002. “150 Years of Patent Protection.” American Economic Review 92(May), pp. 221–​225. Lessig, Lawrence. 2001. The Future of Ideas: The Fate of the Commons in a Connected World. New York: Random House. Litman, Jessica. 1990. “The Public Domain.” Emory Law Journal 39, pp. 965–​1023.

 

OPEN-ACCESS AND INFORMATION COMMONS    279 Moody, Glenn. 2002. Rebel Code: Linux and the Open Source Revolution. New York: Basic Books. Morell, Mayo F. 2011. “The Wikimedia Foundation and the Governance of Wikipedia’s Infrastructure, Historical Trajectories and its Hybrid Culture.” In:  Geert Lovnik and Nathaniel Tkacz, eds., Critical Point of View. Amsterdam: Institute of Networked Cultures. Mowery, David C., Nelson, Richard R., Sampat, Bhaven N., and Ziedonis, Ardvis. 2004. Ivory Tower and Industrial Innovation, University-​Industry Technology Transfer Before and After the Bayh-​Dole Act. Palo Alto, CA: Stanford University Press. Nelson, Richard R. 1959. “The Simple Economics of Basic Scientific Research.” Journal of Political Economy 67, pp. 297–​306. Olson, Mancur. 1965. The Theory of Collective Action. Cambridge, MA:  Harvard University Press. Ostrom, Elinor. 1990. Governing the Commons:  The Evolution of Institutions for Collective Action. New York: Cambridge University Press. Ostrom, Elinor. 2011. “Background on the Institutional Analysis and Development Framework.” Policy Studies Journal 39(February), pp. 7–​27. Owen-​ Smith, Jason, and Powell, Walter W. 2001. “Career and Contradictions:  Faculty Responses to the Transformation of Knowledge and its Uses in the Life Sciences.” Research in the Sociology of Work 10, pp. 109–​140. Rajagopalan, M.S., Khanna, V., Stott, M., Leiter, Y., Showalter, T., Dicker, A., and Lawrence, Y. 2010. “Accuracy of Cancer Information on the Internet: A Comparison of a Wiki with a Professionally Maintained Database.” Journal of Clinical Oncology 28, p. 6058. Reagle, Joseph, M., Jr. 2010. Good Faith Collaboration: The Culture of Wikipedia. Cambridge, MA: MIT Press. Rose, Carol. 1986. “The Comedy of the Commons: Custom, Commerce, and Inherently Public Property.” University of Chicago Law Review 53, pp. 711–​781. Samuelson, Pamela. 1990a. “Benson Revisited:  The Case against Patent Protection for Algorithms and Other Computer Program-​Related Inventions.” Emory Law Journal 39, pp. 1025–​1154. Samuelson, Pamela. 1990b. “Legally Speaking:  Should Program Algorithms Be Patented?” Communications of the ACM 33, pp. 23–​27. Satria, Arif, Matsuda, Yoshiaki, and Sano, Masaaki. 2006. “Contractual Solution to the Tragedy of Property Right in Coastal Fisheries.” Marine Policy 30, pp. 226–​236. Schankerman, M., and Lerner, Josh. 2010. The Comingled Code: Open Source and Economic Development. Cambridge, MA: MIT Press. Schlager, Edella, and Ostrom, Elinor. 1992. “Property-​ Rights Regimes and Natural Resources: A Conceptual Analysis.” Land Economics 68, pp. 249–​262. Schweick, Carles M., and English, Robert C. 2012. Internet Success: A Study of Open-​Source Software Commons. Cambridge, MA: MIT Press. Scotchmer, Suzanne. 1991. “Standing on the Shoulders of Giants: Cumulative Research and the Patent Law.” Journal of Economic Perspectives 5(Winter), pp. 29–​41. Suber, Peter. 2012. Open Access. Cambridge, MA: MIT Press. van Schewick, Barbara. 2010. Internet Architecture and Innovation. Cambridge, MA: MIT Press. Wales, Jimmy. “Talk: Benevolent Dictator.” Wikimedia. March 16, 2007. http://​meta.wikimedia. org/​?oldid=544462 [Accessed 21 August 2016].

 

Chapter 12

FAM ILY AND H OU SE H OL D EC ONOMI C S Amy Wax

The past decade has seen a dramatic surge in women into the labor force, accompanied by significant changes in gender roles within the family, the workplace, and society as a whole. These developments have elicited growing academic interest and prompted proposals for legal and policy reforms designed to improve outcomes for women, families, and everyone else (Blau, Ferber, and Winkler 2011). The purpose of this chapter is to explore some aspects of recent work in this area, with emphasis on the economic analysis of the interplay between labor markets and family roles.

12.1. Household Decision-​M aking Models Research by economists and others has revealed that family well-​being, and the economic fate of men and women within families, is a complex, interactive function of social and economic conditions, relationships within households, and individual choices. Gary Becker, in Treatise on the Family (1981), starting from the analysis of the family as a single decision-​making unit, pioneered the insight that models for household behavior needed to take into account how family members with differing preferences make decisions under the constraints imposed by others in the family and the economy as a whole. In the wake of Gary Becker’s groundbreaking work, the analysis of household decision making has developed from a simple unitary model toward more complex and nuanced understandings of the diversity of interests and the dynamic interactions that take place within families (Himmelweit et al. 2013). The unitary model of household functioning, which regards the family as a single decision-​making unit, assumes that household members agree by consensus to

 

FAMILY AND HOUSEHOLD ECONOMICS    281 maximize a single utility function. That function is subject to a pooled budget constraint, composed of the sum of all family members’ income (Vermeulen 2002). If total income is held constant, the unitary model predicts that individual contributions to the pooled household budget, or changes in those contributions, are independent of each person’s influence over how the budget is allocated (Vermeulen 2002). Although the unitary model assumes that individual preferences and resources within the household could be combined through consensus between the household members, it does not indicate how such a consensus could be reached (Samuelson 1956; Alderman et al. 1995). Becker (1974) sought to improve on the model by positing a benevolent head of the family who took into account the preferences of all household members, but also adjusted allocations in response to family members’ behavior (Becker 1981). Thus, if a family member—​for example, a “rotten kid”—​took actions to raise his or her own consumption, thereby lowering the consumption of others, the head of the household could reduce transfers to try to induce the “rotten kid” to behave less selfishly. Becker predicted that this sort of response by the head of household would cause individual preferences and behavior within the household to converge. The unitary model’s approach to family resources has come under fire repeatedly as overly simplistic about the distributive effects of a diversity of preferences and interests among family members (Doss 2011). For example, further analysis of the “rotten kid” theorem, which echoes criticisms leveled at the unitary model generally, has found that the theorem only holds under restrictive circumstances (Alderman et  al. 1995; Bergstrom 2008). Other models that seek to incorporate a more dynamic approach to how family decision making affects household allocations have had more predictive success. For example, nonunitary models predict that reallocating income from fathers to mothers tends to increase children’s consumption, nutrition, and well-​being. These are outcomes unitary models do not predict (Himmelweit et al. 2013; Alderman et al. 1995; Vermeulen 2002). The validity of these predictions has received empirical support. A 1997 study of UK government policy found that redistributing child benefits from husbands to wives resulted in greater expenditures on women’s and children’s clothing (Lundberg et al. 1997) Likewise, a higher income share for wives in rural Mexico was associated with greater expenditures on food and children’s needs, and lower expenditures on services and alcohol, a result not predicted by a simple income-​pooling method (Lechene and Attansio 2002). Bargaining models offer the most significant advantage over unitary models in explaining intrahousehold interactions among individuals with disparate preferences (Himmelweit et  al. 2013). These models represent the application of game-​theoretic approaches, developed and refined by economists working in various fields, which analyze how individuals make decisions when their fate depends on what other individuals do. The models recognize that the income and non-​monetary resources available to families are not independent of the interactive dynamics of the unit, which are in turn crucially affected by the individual, distinct utility functions of each member. An important insight of bargaining models is that interactions among family members determine both the total resources generated by the unit and the allocation of those

 

282   AMY WAX resources. Thus, on some assumptions, household members might make decisions that generate suboptimal well-​being for individuals within the unit or the family as a whole. Household bargaining models have been subdivided into cooperative and non-​ cooperative models. In cooperative models, household members are assumed to act together in order to reach a Pareto-​efficient result, defined as one where no person can achieve greater utility without the utility of the others being reduced (Doss 2011). Non-​ cooperative models drop the assumption that household members enter into binding, costlessly enforceable agreements. This allows for possible inefficient, non-​Pareto optimal situations (dubbed “burnt toast” options) as each individual household member acts to maximize his or her own utility at others’ expense in light of a budget constraint and other household members’ decisions (Lundberg and Pollack 1996). Evidence of inefficient forms of behavior in actual marriages and families, such as child abuse, domestic violence, and economically suboptimal geographic relocations and labor market choices, are most consistent with non-​cooperative models (Lundberg and Pollack 1996; Doss 2011; McPeak and Doss 2006; Udry 1996). According to bargaining models, the bargaining power of each family member is determined by his or her “threat point”—​the utility level that each individual would achieve if cooperation broke down (Mahony 1995; Wax 1998). Threat-​point analysis has generally focused on married couples by analyzing the interactions between husbands and wives. In the divorce-​threat model, the threat point is determined by the situation each member of the couple would face if the marriage dissolved (Himmelweit et al. 2013; Lundberg and Pollack 1993; Chiappori et al. 2002). This model predicts that the bargaining power of each spouse within the marriage, as well as the likelihood that a spouse will seek or resist divorce under different legal regimes, is a function of the resources each spouse can generate and retain for him or herself as a single person. Thus, an increase in the wife’s share of the total household income will give her greater bargaining power within the relationship, and vice versa (Blundell et al. 2007; Wax 1998). This prediction is consistent with the finding (discussed further later in the chapter) that women decrease their share of the housework as their income increases up to a 50–​50 split (Bittman et al. 2003). Using a divorce-​threat model, Wax has posited that, because a number of key factors combine to give husbands an average extramarital threat-​point advantage over their wives, men will tend to receive the lion’s share of marital resources and egalitarian marriage will be rare (Wax 1998; see, also, Mahony 1994). According to Wax the most critical male advantages are a longer reproductive life and a greater potential to remarry someone younger and have more children. These factors greatly enhance men’s utility relative to women’s in the event of a divorce. In addition, it can be shown through game-​theoretic modeling that women’s stronger preferences for domestic order and investments in “quality” children (because of limited reproductive potential), along with men’s greater earning power, tend to boost men’s bargaining advantage within marriage. This enables men to shift domestic responsibilities to women even in the face of women’s preference for equal sharing of domestic and childcare tasks (Wax 1998). That men tend to enjoy a large share of the marital gains, or “surplus,” is confirmed by the observation

 

FAMILY AND HOUSEHOLD ECONOMICS    283 that married men enjoy better health and well-​being than single men do, with the gains for married over single women being much smaller (Wax 1998). Positing a strong, average intramarital bargaining advantage for men also helps explain the paradox that most divorces are initiated by women, despite women having poorer extramarital prospects. In accordance with a threat-​point model, Margaret Brinig observes that a woman will leave a marriage if its value approaches or falls below the value of her next-​best alterna­ tive. It is suggested that this can happen when the husband appropriates most of the marital gains for himself, which may result in his “overreaching” and making his wife worse off than she would be without him (Brinig and Allen 2000; Brinig and Crafton 1994; Wax 1998). Under the alternative, non-​cooperative bargaining model of marital interaction, sometimes referred to as the “separate spheres model” (Lundberg and Pollack 1993), the option of marital dissolution is considered unavailable, and the threat point is represented by the couple staying together even after cooperation between them breaks down (Bennett 2013). Under this model, bargaining power is increased for the spouse with more resources that can be withheld from the other within the marriage (Bittman et al. 2003). For example, a breadwinner husband can withhold income from a nonworking wife. Or a homemaker wife can reduce her level of household services (Lundberg and Pollack 1993). With the advent of no-​fault divorce and the substantial likelihood that spouses will terminate non-​cooperative relationships, this model’s ambit may be limited. However, it retains vitality for couples concerned about the high costs of divorce, including well-​documented harms to children—​a concern that some economists have observed tends to predominate among the upper middle class (Wax 2014). Because spouses’ circumstances and threat points are so individualized within this model, it’s validity has been difficult to prove (Himmelweit et al. 2013).

12.2.  Division of Labor Within Households Housework can be defined as “the set of unpaid tasks performed to satisfy the needs of family members or to maintain the home and the family’s possessions” (Lachance-​ Grzela and Bouchard 2010). Because it is customarily unpaid and exists outside the market, housework is difficult to observe, measure, and quantify (Shelton and John 1996). This presents an empirical challenge, and researchers have often responded by ignoring or excluding less visible or overlapping types of effort (such as child minding or passive “babysitting,” household management, and various kinds of emotional labor) from direct examination (Shelton and John 1996). Despite these limitations, research on time-​use patterns within families reveals a recurring constant. As stated by two prominent investigators of family time use, “[t]‌he most notable characteristic of

 

284   AMY WAX the current division of household labor is that, whether employed or not, women continue to do the majority of housework” (Shelton and John 1996). Although women’s and men’s time spent doing housework has gradually converged (Bianchi et al. 2000), the most recent data from an Organization for Economic Cooperation and Development (OECD 2013) study show that men spend 30% more time on paid work and 50% less time on unpaid work than women do. Also, men and women tend to spend their time on different sorts of tasks. Women do more time-​consuming, routine housework, including meal preparation and cooking, housecleaning, shopping for groceries and household goods, washing dishes and cleaning up after meals, and laundry. Men’s efforts are directed at household tasks that require attention only intermittently, such as household repairs, yard care, driving other people, managing finances, and paying bills (Lachance-​Grzela and Bouchard 2010; Coltrane 2000). Bargaining theory predicts that increasing female involvement in the workplace will lead to greater bargaining power at home and reduced time spent on housework. Consistent with this theory, women’s on housework time has been observed to decline over time as their workplace experience has increased. Nonetheless, complete convergence has not been achieved. Female labor market gains have not translated into equality at home (Lachance-​Grzela and Bouchard 2010), with women continuing to perform a larger share of unpaid home labor (Pinto and Coltrane 2009). Several models have been developed to explain these patterns and explore the factors thought to influence how much housework each spouse performs. Each of the major models makes the simplifying assumption that housework is undesirable and therefore each spouse will tend to avoid it—​an assumption that might be questionable in some cases (Greenstein 2000).

12.2.1. The Relative-​Resources Model The relative-​resources approach, sometimes also referred to as the economic-​exchange hypothesis or the economic-​dependence model, builds on broader insights of bargaining models that predict the allocations of benefits and burdens within families. Like other assignments of value generated within a marriage, who does the housework as opposed to who enjoys its fruits is a function of the alternatives available outside the relationship. Divorce-​threat and internal allocation bargaining models are thus relevant to understanding the division of housework responsibility. A key insight of these models is that a partner’s individual resources and endowment, such as income, education, and the value of extramarital options (including economic, marital, and reproductive alternatives) (Wax 1998), provide decision-​making power within the relationship, which in turn affects the allocation of household labor (Mannino and Deutsch 2007). Since housework is presumed to be undesirable and requires the expenditure of energy for the benefit of others, family members with more resources and, therefore, more power, will tend to bargain their way out of performing housework (Knudson and Wærness 2008). These assumptions predict that wives’ relative paucity of individual resources compared

 

FAMILY AND HOUSEHOLD ECONOMICS    285 with husbands’ puts women in a weaker negotiating position and explains the chronic housework “gap” between men and women (Greenstein 2000). In general, the research literature provides some support for the relative-​resources model as an explanation for observed patterns. There are data showing that the secular trend in increased earnings for women, and rising income relative to men, helped cause the growth over a 31-​year period in men’s participation in housework (Cunningham 2007). Other studies have found that wives’ contribution to outside household income is inversely related to the allocation of household labor, even when other individual characteristics are held constant (Knudson and Wærness 2008; Parkman 2004). The relative education of spouses also correlates with the division of household labor, with data indicating that couples in which wives’ education exceeds their husbands’ have smaller gender gaps in the quantity of household labor performed (Bianchi et al. 2000). Finally, couples who share the breadwinner role appear to have the closest to an equal distribution of housework, although not to the point of strict equality (Davis and Greenstein 2004). One notable shortcoming of the relative-​resources approach has been its inability to explain the persistent gap in housework despite women’s steadily improving currency in the workforce and even in couples where the wife’s earnings equal or exceed the husband’s. The relative-​resources model would suggest that parity in the workplace would produce equality at home, but the evidence instead indicates that women who have comparable resources to their partners still do most of the housework (Evertsson and Nermo 2007). A number of scholars have noted the data’s failure to confirm the linear relationship between the partners’ earning differentials and the allocation of housework as predicted by the relative-​resources model. They propose instead that the actual relationship is curvilinear, with more traditional labor allocations seen in households where women’s earnings exceed or fall short of their spouses’, and more equal divisions in those where the spouses’ incomes are more closely matched (Bittman et al. 2003; Greenstein 2000). This curvilinear relationship is arguably in tension with the suggestion that women’s share of housework decreases when their economic independence increases (Knudsen and Wærness 2008; Fuwa 2004), as well as with other research suggesting that only a woman’s absolute earnings, not her relative earnings, matter in predicting her time spent doing housework (Gupta 2006, 2007). The observation that women who earn more compared with their husbands do more housework than in households with equal earnings remains a puzzle that straightforward relative-​resources models fail to explain.

12.2.2. The Time-​Availability Model The time-​availability model is based on the premise that the amount of time each partner spends on paid labor affects the amount of time that he or she has avail­ able to perform housework (Artis and Pavalko 2003). Because men are traditionally more involved in the labor force, and husbands invest more hours in paid labor even

 

286   AMY WAX when both spouses work either part time or full time, this model predicts that men will have less time available to perform housework and women will do the greater share (Lachance-​Grzela and Bouchard 2010). The model also predicts that rising female labor-​force participation will result in a reallocation of the housework burden toward men because of women’s decreased availability to perform housework (Robinson and Hunter 2008), but that parity will be hard to achieve without gender equality in hours of paid work. The time-​availability model has received some support from the data on the relationship between the partners’ paid work patterns and their work at home. As a general matter, both full-​time and part-​time employed men and women perform less housework than the unemployed of either sex do (Bianchi et al. 2000). Studies also confirm a fairly robust correlation between the household members’ hours of employment and the distribution of housework within the family (Cunningham 2007), with a woman’s share of housework decreasing as her hours of paid employment increase (Knudson and Wærness 2008; Artis and Pavalko 2003). Qualitative elements arguably related to the intensity and demands of paid work that could distract attention from household tasks (including length of employment history and earnings trajectory) also have been shown to correlate with both partners’ involvement in housework, although women’s employment status has a greater effect than does men’s (Cunningham 2007). Some studies of the relationship of housework and time availability have revealed that women’s employment seems to lead to a more equal distribution of housework through the mechanism of exposing women to workplace environments that are supportive of gender equality (Bolzendahl and Myers 2004). It is also possible, however, that this link is due to selection, with employed women tending to have more egalitarian gender views from the start (Fan and Marini 2000). Like the resource-​allocation hypothesis, the time-​availability model falls short in explaining some aspects of the gender gap in household responsibility, including the finding that working women sometimes do more housework than their employed husbands do, even when they work the same hours at jobs with similar status (Bartley et al. 2005). One study showed that, in a sample of full-​time male and female workers with approximately equal working hours, women spent the equivalent of 80% as much of their paid-​work time doing housework, whereas men spent the equivalent of only 40% (Lincoln 2008), with the result that the women worked more hours overall. Another study suggests, however, that when time spent in both paid employment and unpaid domestic work are combined, there is an equal split of hours invested between partners (Gershuny and Sullivan 2003). The variations in the findings of the relationship between spouses’ paid​and domestic work hours counsels caution. Accurate evidence on working hours is essential to any proper empirical test of the time-​availability model, but it is hard to come by. As discussed more fully later in the chapter, the designations of full time and part time are imprecise and may fail to capture actual hours worked, which can differ by gender even within the same job categories (Hymowitz 2011a, 2011b, 2012). Another shortcoming of the time-​availability model is that it does not answer the question of whether women’s greater responsibility for housework is a cause or an

 

FAMILY AND HOUSEHOLD ECONOMICS    287 effect of devoting fewer hours to paid work. Implicit in some discussions of the time-​ availability model is the assumption that each spouse chooses how much time to work for pay according to individual preferences, with the housework responsibility falling into place in response to those choices. The problem is that choices about paid work may be dependent on household expectations. Some women may feel forced to reduce their work hours because of weighty household responsibilities imposed by other factors—​ such as the type of power imbalances identified by the resource constraint model, or the pressures of gender ideology as discussed in Section 12.2.3. Thus, the mere fact that the hours of paid and unpaid work are inversely related for each member of the couple does not necessarily settle the question of why a particular balance of paid and unpaid work is observed and, thus, whether the time-​availability model offers a complete causal explanation of household divisions of labor.

12.2.3. Gender-​Ideology Model Gender ideology is defined as “how a person identifies [himself or herself] with regard to marital and family roles traditionally linked to gender” (Greenstein 2000). Notions about proper male–​female roles in society and within relationships are positioned on a spectrum ranging from traditional ideas (favoring distinct spheres and a strict male-​ breadwinner-​and-​female-​homemaker structure) to more egalitarian norms (which embrace a vision of both partners participating more equally at home and at work) (Lachance-​Grzela and Bouchard 2010). According to the gender-​ideology model, ideas matter. People’s views about gender roles, and their desire to represent and fulfill those roles, are as important as economic concerns in their decisions about how to divide responsibility for earning income and running the household. On this assumption, couples with views closer to the egalitarian end of the spectrum would be predicted to more equally divide housework, whereas those with more traditional attitudes would show a lopsided assignment of paid and unpaid labor between spouses, with the performance of more traditional female tasks by wives and male tasks by husbands (Davis et al. 2007). Although gender-​related perceptions and social norms are elusive and difficult to measure directly (De Henau and Himmelweit 2013), some studies suggest that gender ideology has an independent influence on men’s and women’s behavior. Compared with more traditional women, women with strong egalitarian gender attitudes are likely to report doing less household labor, whereas more egalitarian men tend to perform more hours of housework compared with men subscribing to more traditional norms (Davis et al. 2007; Fuwa 2004; Parkman 2004). Some empirical observations suggest that women’s behavior is more sensitive to the couple’s gender attitudes than is men’s. For example, men’s egalitarian attitudes tend to correlate with less labor time for their female partners without producing an absolute increase in their own housework hours. Also, although women with attitudes that are more egalitarian do less work than traditional women do, egalitarian women’s partners do not show a significant increase in household labor relative to the partners of more conventional women (Bianchi et al. 2000).

 

288   AMY WAX Other research indicates, however, that men with higher levels of education do a greater share of the housework than do men with less education (Gershuny and Sullivan 2003), an observation that may be explained by the strong correlation between high economic status and egalitarian gender ideologies (Cha and Thébaud 2009). Indeed, because men who are more educated would tend to command greater resources and higher earnings that would relieve them of household duties, their willingness to do a greater share of housework, if robust, supports the hypothesis that gender attitudes do have some effect on men’s domestic behavior. As a general matter, there is evidence that gender ideology evolves in the direction of egalitarian viewpoints, with individuals increasing their support of egalitarian perspectives over time (Fan and Marini 2000). In light of these observations, the gender-​ideology model predicts that the societal trend toward widespread adoption of egalitarian attitudes “should translate into a more equal division of household labor between men and women,” even controlling for other factors (Lachance-​Grzela and Bouchard 2010; Davis and Greenstein 2004; Artis and Pavalko 2003). The overall direction has indeed been toward divisions of labor that are equal, but it remains to be seen whether actual convergence is ultimately achieved or whether women continue to do the lion’s share of unpaid work.

12.2.4. The Economic-​Dependency Model and the Gender-​Construction Model As a variant on the relative-​resources model, Julia Brines (1994) initially posited that economic dependency would be a decisive influence in the allocation of household labor. The model predicts that, as economically dependent wives—​that is, those who rely on their husbands for basic financial support—​increase their share of the household earnings, their portion of domestic labor will decrease. This prediction follows from a conventional notion of economic exchange, which posits that a wife swaps her unpaid labor for a share of the household income, with her need to gain access to her husband’s income decreasing as her earnings rise. Based on empirical time-​use studies, however, Brines observed that the economic-​exchange model could not be so straightforwardly applied to couples where husbands are economically dependent on wives. Although exchange theory predicts that the husband would assume responsibility for most of the domestic work under those circumstances, Brines discovered that economically depen­ dent men’s share of the housework actually declines. Brines’ conclusion is that men and women appear to respond differently to economic dependence on their spouses. Brines hypothesized that couples who violate the traditional structure of a breadwinner husband with a dependent wife feel constrained to respond to this “role reversal” by enacting a form of gender display. Reallocating housework back to the woman counterbalances the man’s compromised breadwinner role and mitigates the resulting threat to the husband’s masculine identity. By having the husband

 

FAMILY AND HOUSEHOLD ECONOMICS    289 do less housework, the couple “does gender” and achieves “gender accountability” in the eyes of themselves, their spouses, and their friends. Brines’ findings led to the elaboration of the gender-​construction hypothesis, which builds on the gender-​ideology model to identify gender as not only a distinct aspect of couple’s self-​definition but also something that is dynamically created and recreated in interactions with others (Potuchek 1992). According to Coltrane (2000), “gender construction theories suggest that women and men perform different tasks because such practices affirm and reproduce gendered selves, thus reproducing a gendered interaction order.” Brines’ application argues that housework is a key activity through which men and women display or produce gender (Bianchi et al. 2000; Brines 1994). Since household labor is widely perceived as expressing the caring and nurturing aspects of femininity (Badr and Acitelli 2008), women may perform more housework to reinforce their feminine gender identities, and men may resist performing housework to protect their male gender identities (Knudson and Wærness 2008; Erickson 2005). In keeping with this observation, there is evidence that American men who face challenges to their masculinity in the workplace tend to avoid housework (Arrighi and Maume Jr. 2000). Likewise, Greenstein (2000) found that, regardless of gender ideology, couples with non-​normative earner roles engage in “gender deviance neutralization” by distorting the amount of housework they report for each spouse in a direction that appears consistent with gender expectations. In sum, the focus on the function of norm and identity reinforcement, and the inclusion of gender as a “product” of housework, helps make sense of some seemingly irrational, or at least economically perplexing, aspects of the household division of labor (Shelton and John 1996).

12.2.5. Other Aspects of the Division of Household Labor In addition to those mentioned already, other factors have been found to predict the division of household labor. These include the couple’s marital status and race, and the contributions of children.

12.2.5.1. Marital Status Studies have indicated that married women spend significantly more time on housework than cohabitating women do, even after controlling for other demographic characteristics (South and Spitze 1994). Research also consistently shows that cohabitating couples have a closer to equal distribution of household labor than do married couples (Baxter 2005; Batalova and Cohen 2002). This is mostly a function of each partner spending less time on traditionally gender-​specific tasks, resulting in a more dramatic reduction of housework time for women and a convergence of unpaid work time between partners (Domínguez-​Folgueras 2013). While some studies report no significant difference in housework performed by married and cohabiting men, a few have found that marriage

 

290   AMY WAX decreases men’s share of housework, which is consistent with married men spending longer hours at paid work (Gupta 1999; Shelton and John 1993).

12.2.5.2. Race The few studies that have focused on race’s role in the division of household labor have yielded mixed results. Although research indicates that African-​American males, net of other predictors, do a larger proportion of the housework than white men do, African-​ American women still do twice as much as their partners (Shelton and John 1996; Kamo and Cohen 1998). Also, white husbands were significantly less likely than black husbands to perform traditionally feminine household tasks (Orbuch and Eyster 1997). In contrast, wives in dual-​earner Mexican-​American families expect their husbands to be involved in all aspects of household labor (Herrera and DelCampo 1995). Overall, white husbands and wives perform significantly fewer hours of household labor compared with non-​white husbands and wives (Bianchi et al. 2000).

12.2.5.3. Contributions of Children The findings are equivocal on whether children perform a significant amount of household labor (Bianchi et al. 2000), but researchers have found that children’s housework is divided by gender, with boys often performing traditionally “male” tasks, and girls performing traditionally “female” tasks (Coltrane 2000). An area worth investigating is whether, given lower birth rates, smaller families, and increased emphasis on developing children’s human capital, children’s participation in household maintenance has undergone a secular decline in recent years. This could increase the relative burden that adults in the home are required to bear.

12.2.6. Fairness in Labor Distribution, the Meaning of Full Time, and the Leisure Gap This section addresses a number of topics important to both the division of household labor and the issue of work–life balance. It reviews research into how fairness in the distribution of household labor is viewed by spouses. It then considers the “gender leisure gap,” which refers to data indicating that women’s average combined hours of paid and unpaid work exceed men’s. Finally, it elaborates on the problem that assertions concerning the gender division of labor, as well as gaps in pay, may be distorted by a lack of precision in measurements of “full-​time” and “part-​time” work, with those terms having different meanings across genders.

12.2.6.1. Fairness One interesting consequence of the division of household labor is that even though women perform the majority of household labor, only 20% to 30% of women perceive this distribution as unfair (Mikulka 1998). Research has found that there is a high

 

FAMILY AND HOUSEHOLD ECONOMICS    291 degree of agreement between married men and women on what is equitable and that most couples do not use a 50–​50 split of housework as an “equality point.” Men believe the division of labor is fair when they contribute 36% of the time devoted to household tasks, whereas women designate their performance of 66% of the work as fair (Lennon and Rosenfeld 1994). Nonetheless, not all men and women are satisfied with the specific allocation within their own marriage. There is evidence that women who view the distribution of household labor as unfair are more likely to experience depression (Bird 1999; Lennon and Rosenfeld 1994) and that a perception of inequity decreases marital satisfaction for both men and women and increases women’s (but not men’s) likelihood of divorce (Frisco and Williams 2003). Yet other research has found that a woman is more likely to perceive an unequal housework division as inequitable when she encounters low time availability (usually because of working long hours in the labor market), is not highly dependent on the partner’s income, and adheres to a non-​traditional gender ideology (Braun et al. 2008). Another strand of research has focused on the existence of a “gender leisure gap,” a term referring to the finding that women’s hours of paid plus unpaid work generally exceed those of men, thus allowing men more leisure time (Bianchi and Milkie 2010). Estimates of this gap vary from men having more than 30 more minutes per day of leisure (Mattingly and Sayer 2006) to 3 hours per week (Wang 2013). Research has also found that while leisure time is more meaningful to women, it is also more stressful and less refreshing, possibly because of more frequent interruptions and attempts at multitasking during nonworking periods (Wang 2013; Bianchi et al. 2006; Offer and Schneider 2011). Alison Wolf (2013) has disputed the existence of a gender leisure gap, claiming that recent evidence shows that the gap is closing, albeit with some variation between countries and over the life course. She also questions the importance of the leisure gap in women’s quest to achieve work–life balance, asserting that its role has been exaggerated relative to other factors more critical to women’s status and well-​being. One important aspect of research on the gendered leisure gap has to do with methodology. Are tallies of working hours, whether at home or in the paid workplace, accurate? Leisure gap reports are sometimes based on time-​diary studies, which are subject to individual distortions in reporting. But data about working hours, especially in official government reports, are especially prone to imprecision. Kay Hymowitz (2013) has confronted this problem in analyzing the purported “gender pay gap,” revealed by statistical reports that full-​time working women earn between 75 and 81 cents for every dollar men earn. She has questioned those numbers by arguing that full time does not designate the same duration of work for men and women. Hymowitz points out that the definition of full time often includes anyone who works more than 35 hours per week. Within that category, the evidence indicates that men as a group work significantly longer hours than women do. In 2007, 27% of male full-​time workers worked 41 hours or more per week, compared with only 15% of female full-​time workers. At the opposite end of the spectrum, 12% of women working full time put in between 35 and 39 hours, whereas only 4% of men did (Hymowitz 2011a, 2011b). Based on these statistics, the gender pay gap could well be a function of men working longer hours. This would not betoken pay discrimination

 

292   AMY WAX but rather lower compensation for women based on fewer hours at work. Hymowitz also attacks the frequent claim that men make more money than women who perform the same job. For example, it has been reported that female physicians and surgeons earn 64.2% as much as male physicians and surgeons do. But Hymowitz argues that these numbers ignore important differences. Males are disproportionately more likely to choose medical specialties that require longer hours and that offer higher compensa­ tion than those chosen by women (Hymowitz 2011a, 2012a, 2012b, 2013). All in all, once factors such as education and hours are controlled for, the pay gap between genders shrinks to single digits.

12.3.  Work–​Family Balance The following section discusses several theoretical approaches to explaining why women differ, on average, from men in their choices for balancing paid work and unpaid domestic labor. The landscape can be roughly divided between those who believe that structural and institutional factors are the most important causes of gender divergence, as opposed to those who cite women’s (and men’s) personal preferences, interests, and outlook as driving differences in outcomes. The contribution of externally imposed versus internal preference-​based elements to women’s “choices” is a critical fault line in the debate about gender divergence in labor market participation and time use. One prolific and leading proponent of the external, structural view is Joan Williams (2001). Williams believes that women’s labor market prospects and success are routinely circumscribed by strong institutional and workplace expectations that favor men. She argues that employers envision an ideal worker who has numerous “masculine” traits—​including the ability to work full time, move when required, take little time off, be continuously available on the job, and give paid work absolute priority over domestic responsibilities. The ideal worker is quintessentially “masculine,” Williams asserts, because functioning in that role requires a worker to have a stay-​at-​home spouse to handle childcare and domestic duties. Because women bear the brunt of household responsibilities, they are less able than men to function as ideal workers, and thus reap fewer rewards in the workplace. This pattern is self-​reinforcing, with women often withdrawing from paid work and shouldering greater domestic responsibility because their failure to function as ideal workers results in lower pay and diminished market opportunities. Williams also makes clear her belief that the “ideal worker” structure is contingent and dispensable. Although the “ideal worker” expectation is deeply embedded in the current culture and structure of the workplace, it incorporates arbitrary features that can and must be reformed before women can achieve true equality. Recent work by the economist Claudia Goldin (2014) carries forward some of Joan Williams’s themes. According to Goldin, the “last chapter” of change that is needed to achieve true economic equality between men and women is a radical revision in how jobs are structured and remunerated. Goldin’s focus is primarily on workplace reforms

 

FAMILY AND HOUSEHOLD ECONOMICS    293 and compensation rather than on trying to alter the household division of labor. She emphasizes the goal of allowing women to combine domestic responsibilities with high workplace status and pay. Like Williams, Goldin wants employers to reject the paradigm of the “ideal worker” who is constantly available to fulfill employer demands. She seeks to create a more temporally flexible workplace that de-​emphasizes the need for long, continuous, fixed hours or “face time,” thus leaving more time for the necessary tasks of domestic life. According to her research, family-​friendly workplace restructuring has gained momentum in various sectors, such as technology, science, and health, with beneficial results for women’s workplace status and gender pay parity. Other economic sectors, such as the corporate, financial, and legal worlds, have proved more resistant. Nonetheless, she argues, these areas might be amenable to the type of team approach that gives individual workers more leeway to attend to responsibilities outside the workplace while still maintaining efficiency and continuity within it (Goldin 2014). In contrast to Williams and Goldin, other theorists stress women’s preferences, tastes, and interests as the principal driving forces underlying distinct life roles for men and women. Kingsley Browne (2002) adopts the perspective of evolutionary psychology to explain why women are less successful than men at work and why they tend to be more focused on the home front. According to Browne, existing structures and institutions are neither arbitrary nor pivotal in holding women back, and the existence of a “glass ceiling” is a chimera. Rather, because women are more invested in their offspring, they are naturally inclined to give greater attention to domestic pursuits, and have less of a drive to reach the pinnacles of the workplace. Thus, existing arrangements primarily reflect the average preferences, leanings, and ambitions of men and women, which in turn are based on significant innate differences. A study by Steven and Christopher Rhoads (2012) of gender roles and childcare also urges consideration of “biological and evolutionary” explanations, and the role of deep-​ seated gender-​specific preferences, to explain gender differences in workplace status and success. The Rhoads conducted a study of male and female tenure-​track professors at an elite university to determine what factors were associated with more equal or unequal distributions of childcare duties among similarly educated and ranked junior academics. The Rhoads study found that even male professors who took parental leave and purported to embrace non-​traditional gender roles did not generally share childcare duties equally with their wives, even if both members of the couple were full-​time working academics. In attempting to get at why wives and husbands with similar professorial status and jobs nevertheless performed different amounts of childcare, the authors administered a questionnaire that asked the subjects to report their enjoyment of child-​related tasks on a numerical scale. The women almost uniformly reported a significantly higher average enjoyment of even burdensome tasks associated with their children (like taking care of sick child). The authors conclude that an important reason for the unequal distribution of childcare is that the average female gets more positive utility from childcare than does the average male—​a preference that persists despite otherwise similar professional interests and abilities. The authors also suggest that, because neither gender-​role attitudes nor leave-​taking status were strongly correlated with enjoyment of childcare,

 

294   AMY WAX efforts to change external work–family policies or gender-​role ideology would probably not have a significant effect on the study subjects’ behavior, beyond what had already been achieved. Like Brown, the authors argue that “biological and evolutionary factors” are important drivers of residual gender differentials in domestic behavior, even among sophisticated and highly educated populations, and that these tendencies might be beyond the reach of proposed “family-​friendly” policy changes. Catherine Hakim’s preference theory (2000) also argues that attitudinal factors such as women’s work–lifestyle preferences, motivations, and aspirations help explain observed work and family patterns. Unlike Browne, however, Hakim does not argue from an evolutionary perspective, and she assigns an important role to structural and institutional factors, especially in influencing the behavior of women whose preferences are mixed. According to Hakim, the female population is highly diverse, in that it spans a wide variety of interests and orientations. Strongly work-​centered women anchor one end of the population. These women are characterized by their determination to develop a career and a willingness to subordinate domestic pursuits to achieving workplace goals. The other end of the continuum is occupied by home-​centered women, who place family and domestic priorities at the fore and are relatively indifferent to work and career. Finally, there are women in the middle, whom Hakim refers to as “drifters.” These women are happiest when achieving some mix of workplace and domestic goals. Hakim posits that, at present and historically, women at the extremes of the spectrum are more robust in their preferences, with hard-​core work-​oriented women generally rejecting domesticity, and strongly domestic women opting out of the workforce if possible. Regardless of talent, education, and social expectations, the women in these categories do their best to make good on their preferences, which may be easier or harder to achieve in different eras. In contrast, the behavior and work-​family choices of the “drifters” are more amenable to influence by situational factors and contextual variables. These can include cultural attitudes, social norms, and workplace conventions (such as the existence of part-​time positions and flexible hours), as well as official policies such as the availability of generous parental leave and financial subsidies for families. Hakim estimates that between 40% and 80% of the female population occupies the middle category, while the opposite ends of the spectrum each comprise between 10% and 30% of women. Some researchers have critically analyzed Hakim’s preference theory. One study examined work–​life opportunities across seven European countries to determine whether Hakim’s theory could be substantiated or whether more emphasis should be placed on structural factors (Kangas and Rostgaard 2007). These authors found a significant correlation between home-​centered attitudes and labor-​market status, which was consistent with Hakim’s framework, at least regarding “home-​centered” women. They also found that generous leave schemes and other family support policies, as well as husbands’ attitudes and familial orientation, played a large role in women’s work–​life choices, including hours worked. Although none of these observations would appear inconsistent with Hakim’s theory (because she concedes that the majority of women are influenced by such variables), the authors nonetheless criticize Hakim for giving too

 

FAMILY AND HOUSEHOLD ECONOMICS    295 much weight to “intrinsic preferences” and de-​emphasizing the extent to which con­ textual variables, especially in some societies, can drastically shape both what women prefer and their ability to make good on those preferences.

12.4.  Protections for Caregivers: Mandated Benefits and Antidiscrimination Laws Although, as the discussion so far demonstrates, the reasons for this social fact are in dispute, it is well established that women on average bear more responsibility than men for maintaining the household and caring for children. Necessarily, women’s economic status and prospects in the workplace are intimately related to that caregiving function. This section discusses two important legal protections for American women with caregiving responsibilities—​the Family and Medical Leave Act (FMLA) and Title VII of the Civil Rights Act. It looks at the work of Christine Jolls and Jonathan Gruber, who adopt a neoclassical economic framework to examine the effects of mandatory protections for caregivers, such as parental leave laws, on women’s employment and compensation. It then examines this author’s own game-​theoretic analysis, which argues that some family-​unfriendly workplace conventions might represent a dysfunctional or suboptimal equilibrium that could benefit from selective reforms. Finally, the section reviews some empirical research on the effects of mandated parental and maternity leave worldwide.

12.4.1. The FMLA The FMLA, passed in 1993, guarantees up to 12 weeks of unpaid leave annually to employees who meet designated statutory conditions related to illness or family responsibilities. Individual workers are entitled to leave owing to a serious personal illness or to care for a newborn baby, newly adopted child, or a family member with a serious health condition. See 29 U.S.C. § 2612(a)(1) (2013). An employee returning from an FMLA-​covered leave must be restored to the same job or one with equivalent pay, benefits, and other terms and conditions of employment. See 29 U.S.C. § 2614(a)(1). In protecting the rights of pregnant workers and mothers, the FMLA has several limitations. Only employers with at least 50 workers within a 75-​mile radius of the leave-​taking employee’s workplace are obligated to provide leave, and the employee must have worked for the same employer for 1,250 hours in the previous year, which effectively excludes many part-​time workers (29 U.S.C. § 2611(2)(A)(ii)). A 2000 US Department of Labor study found that only 10.8% of employers in the United States meet the FMLA definition of a “covered” employer, and only 58.3% of employees work for a covered employer (Waldfogel 2001). In addition, many employees eligible to take leave under

 

296   AMY WAX the FMLA fail to do so because they cannot afford a long period without pay (Suk 2010). Paid maternity-​related leave in the United States is fully private and rare, with only 16% of employers offering this benefit as of 2008 (Families and Work Institute 2008). Thus, de facto job and legal protection for pregnant women and new mothers who take time off from work is still exceptional in the United States.

12.4.2. Employment Discrimination Litigation: Title VII and the Pregnancy Discrimination Act of 1978 Given the limitations of the FMLA, female employees with caregiving responsibilities have increasingly turned to employment discrimination litigation under Title VII of the Civil Rights Act of 1964 (42 U.S.C. 2000 et. seq.) to secure job protections in the workplace. The language of Title VII forbids discrimination based on sex, not on maternity or family status. In light of this, most Title VII suits brought by caregivers rely on a “sameness” theory, which holds that Title VII is violated when employers fail to treat pregnant women or mothers in the same manner as other employees (Suk 2010). This theory has been used mainly by women who claim that their work status has been compromised by becoming mothers. The Act was also amended in 1978, by adding the Pregnancy Discrimination Act (PDA), which provides specified types of protections for pregnant women (42 U.S.C. § 2000e(k) [2013]). The general turn to statutory antidiscrimination law and the enactment of the PDA followed the failure of litigants to persuade the courts to extend protection to pregnant women under the US Constitution. The plaintiffs in Geduling v Aiello had alleged that California’s disability insurance system violated the Equal Protection Clause of the United States Constitution by excluding pregnancy-​related disabilities from the system’s coverage. The US Supreme Court disagreed, 417 U.S. 484, 492 (1974), holding that there was no violation because California’s exclusion on the basis of pregnancy could not be equated with an exclusion on the basis of gender. Two years later, the Court rejected a Title VII-​based argument on similar grounds, holding that the exclusion of pregnancy coverage from an employer’s disability plan was a rational, nondiscriminatory decision based on cost considerations (General Electric Co. v Gilbert, 429 U.S. 125, 138–​140 [1976]). In response to these cases, Congress passed the PDA of 1978, which amended Title VII to clarify that discrimination “because of sex” included discrimination “on the basis of pregnancy” (42 U.S.C. § 2000e(k) [2013]). Although the statute does not guarantee job leave for pregnancy or childbirth and does not protect mothers in general, its main effect has been to force employers to treat pregnancy like any other physical disability that requires nonworking time and is entitled to health insurance coverage. The law has also helped to mitigate more general workplace discrimination (such as termination or demotion) based on pregnancy.

 

FAMILY AND HOUSEHOLD ECONOMICS    297 Recent data have indicated a surge in pregnancy discrimination litigation, with pregnancy discrimination lawsuits increasing by 65% between 1992 and 2007 (National Partnership for Women and Families 2008). Increasing numbers of cases go beyond invoking the limited protections of the PDA to assert Title VII claims based on unfair treatment of employees with family responsibilities, a phenomenon referred to as family responsibility discrimination (FRD) (Suk 2010; EEOC 2007). These types of cases increased four-​fold between 1996 and 2005 (Still 2006), and recent data indicate that plaintiffs win more than 50% of the cases—​a much higher rate than the 30% for other types of employment discrimination cases (Clermont and Schwab 2009; Parker 2006). The theory behind FRD cases traces its legal origins to the work of Joan Williams who, as already noted, asserts the quintessentially masculine nature of the “ideal worker” favored in the employment setting. On this view, the time demands and inflexibility institutionalized in workplace “ideal worker” structures routinely operate to the disadvantage of women with children and thus amount to a form of sex discrimination that Title VII is designed to correct. As an extension of this framework, many litigants identify gender stereotypes as a potent source of discrimination, arguing that employers routinely erect a “maternal wall” that blocks women’s advancement based on assumptions that female caregivers cannot be good workers and that men are not caregivers who need parental leave (Suk 2010). In keeping with these arguments, the EEOC’s 2007 Guidance identified gender stereotyping as the most important factor in FRD cases. In practice, the “maternal wall” theory has proved to be a successful strategy in FRD litigation. Some FRD cases involve “loose lips,” where supervisors make blatantly discriminatory remarks that reveal their assumptions about mothers. Others rest on allegations of discriminatory requirements, such as last-​minute travel and long continuous hours, that parents without a stay-​ at-​home spouse (who are mostly mothers) have difficulty meeting (Suk 2010). Victories for plaintiffs in such cases represent a significant extension of Title VII to individuals who claim they are disadvantaged in the workplace for their caretaking responsibilities.

12.4.3. Academic Approaches to Mandatory Benefits Several scholars, including Jonathan Gruber and Christine Jolls, have applied an economic perspective to analyze the pros and cons of so-​called mandated benefits, such as those extended through laws like the FMLA, which requires employers to offer affir­ mative, and often costly, protections to employees. Gruber’s work has focused on the effects of mandatory coverage of childbirth expenses in employer-​provided insurance policies. Gruber has hypothesized that requiring employers to include these benefits would lead to lower wages for women or women of childbearing age because employers would try to recapture the increased cost of employing women by paying them less (Gruber 1994). Empirically, Gruber has claimed that his predictions are validated. According to his data, mandated health coverage on the state and federal level caused the wages of women of childbearing age (between 20 and 40) to fall by at least the cost

 

298   AMY WAX of the mandate, while these women’s levels of employment were essentially unchanged. Gruber argues that these findings “consistently suggest shifting of the costs of the mandates [to the beneficiaries of the mandates] on the order of 100%, with little effect on net labor input.” In other words, the women who were supposed to benefit from coverage were bearing the cost through reductions in their compensation levels without any mea­ surable increase in the labor market participation of this group. According to Gruber, these requirements do not actually make women better off, at least in the short term. Christine Jolls (2000) has also analyzed the possible economic consequences of so-​ called accommodation mandates, such as the requirement of protected maternity leave under the FMLA. She concludes that the answer to the question of who bears the costs of such mandates—​whether the beneficiary class, workers as a whole, or some combination of employer and workers—​is: “it depends.” When occupations are significantly segregated by gender, it is easy for employers selectively to pass on a mandate’s costs by lowering wages only for the beneficiary population or by employing fewer people in that category. That is because antidiscrimination laws only require equal treatment by sex for people in the same job and allow differential pay and hiring rates for men and women in distinct jobs. Under these conditions, Jolls predicts that the FMLA’s accommodation mandates will either depress women’s wages or decrease women’s employment, depending on whether the value of the mandate to the workers exceeds its cost. Where cost is less than value, employees will accept the wage reduction and remain employed. If not, the number of workers in the benefited category will decline as employees voluntarily leave the workforce. When significant occupational segregation is not present, in contrast, the constraints of antidiscrimination laws might limit the employer’s ability to pass on costs selectively to the beneficiary population. If “wage differentials” by gender are restricted (through, e.g., aggressive enforcement of equal-​pay laws), employers might try to effect “employment differentials” by gender—​that is, hiring fewer women. If those steps are restricted (because of strong or easily enforced antidiscrimination laws), employers will have no choice but to spread the costs of mandates to the workforce as a whole, which will result in less compensation for everyone, and also in cross-​subsidization of beneficiary groups by others. Whether or not the costs of the mandates exceed the benefits will also influence its effects, as overly expensive mandates will tend to produce lower employment levels and job losses. In sum, the expected consequences of mandates depend on a number of contingent parameters, including the characteristics of the particular workplace, the legal environment, and the relative value of the mandate. All of those factors are important in determining whether a particular mandate’s effects are positive or negative overall. Amy Wax (2004) questions whether some of the assumptions of the neoclassical approach utilized by Gruber and Jolls are valid in general and true for accommodation mandates in particular. She constructs a game-​theoretic model to demonstrate that long hours and excessive workplace demands can sometimes result in a costly “arms race” that gives a competitive edge to individual strivers but hurts some workers (including those with caregiving responsibilities) and is inefficient for the workplace as a whole. In

 

FAMILY AND HOUSEHOLD ECONOMICS    299 other words, certain “ideal worker” practices might be good for some workers relative to others, but bad for everyone because all workers might be happier with a more family-​ friendly set of workplace conventions (Frank 1995). Because existing job arrangements may not always be optimal, family-​friendly reforms might sometimes have the effect of “jolt[ing] management-​employee interactions out of stable, inefficient equilibria and toward more efficient alternatives.” These changes could come from an external source, such as governmental mandates for parental leave and workplace flexibility. Or they might arise informally and internally, through management directives or the evolution of spontaneous order. Wax cautions that although such reforms have the potential to be self-​sustaining, most will require either “firm direction by management or self-​selection by employees with similar preferences” to avoid erosion through turnover or disruption by employees with less family-​friendly priorities. One problem with implementing family-​friendly reforms, however, is that it is not always easy to know when existing workplace conditions represent an inefficient equilibrium or are necessary to the optimal functioning of a particular enterprise. Recently, as Claudia Goldin has demonstrated and has already been discussed (Goldin 2014), a diversity of practices among different types of workplaces has emerged, with some becoming more flexible and less demanding of long hours and continuous availability. Business, law, and finance have generally held out against these trends. Whether those sectors can be effectively restructured without harming productivity or efficiency remains to be seen. Aside from the issue of whether specific workplace mandates actually help or hurt women, some commentators, including economists, have argued that various steps should be taken to share the burden and cost of raising children more widely throughout society. Those proponents observe that society as a whole reaps economic benefits when its adult members are healthy, productive, and well socialized, and that the type of childrearing that achieves this result requires considerable parental investments (Folbre 1994; Burggraf 1997; Wax 1999; Olsaretti 2013). By taking on a large share of the expensive and time-​consuming burden of childcare, parents, and especially mothers, benefit not only themselves and their children, but everyone. Because the costs of raising children are imposed disproportionately on some members of society, we should adopt policies that reward caretakers by offering extra resources, such as job training and government subsidies. Amy Wax adds to this observation by noting that, because beneficiaries of parental care fail fully to compensate those who bear the burdens and provide the benefits, caregiving activities tend to generate uncompensated positive externalities. Classic economic theory predicts that care giving will therefore tend to be undersupplied relative to its optimal levels. In other words, “hands-​on caring and other ‘women’s work’ is vital to a good society and … potentially in short supply in a free one.” That prediction provides some justification, grounded in economic principles, for “a mix of policies and strategies, formal and informal, that returns to caregivers the rewards they would enjoy if the beneficiaries of their efforts really paid their own way” (Wax 1999). The main drawback of such an approach, however, is that not all parents are equally skilled and effective in socializing their children. Quality controls on parenting are notoriously poor, especially for the growing number of children born to single

 

300   AMY WAX mothers who are ill equipped to support or care for them (Wax 2000, 2003). It is thus unclear whether subsidies for all parents will result in a net benefit to society, especially if targeted to those with a paucity of resources.

12.4.4. Effects of the FMLA and Mandatory Paid Leave on Wages and Employment: Empirical Perspectives Researchers interested in the real-​world consequences of family-​friendly mandates have examined the empirical effects on wages and employment of the FMLA’s unpaid leave requirements, as well as paid leave guarantees, which tend to prevail in Europe. Work by Charles Baum and Jane Waldfogel indicates that, while the FMLA has induced more women to take maternity leaves in the United States, effects on wages and employment have not been significant. Christopher Ruhm’s studies of European family-​leave benefits suggest that extended periods of paid maternity leave can have positive effects on employment but negative effects on wages.

12.4.4.1. Effect of the FMLA on Wages and Employment In his empirical examination of the FMLA’s effects on labor markets in the United States, Charles Baum (2003) failed to find any statistically significant change in employment levels or wages. This result was observed even among the subset of women who worked for employers covered by the FMLA and when various controls for selection bias were introduced. Baum proposed several reasons for the FMLA’s minimal effect. First, he noted that many larger companies, which were most affected by the FMLA, had voluntarily provided some kind of maternity leave benefits before the law was passed. In addition, twelve states and the District of Columbia had already enacted and implemented legislation guaranteeing unpaid maternity leave for in-​state workers before the FMLA was enacted. Finally, Baum hypothesized that the short duration of the FMLA’s leave (less than 3 months) minimized its economic impact, especially since the unpaid status might induce many women to take shorter durations of leave or none at all. Based on these results, Baum argues that strengthening the law—​by, for example, extending coverage to small employers, lowering employee eligibility requirements, or lengthening the leave period somewhat—​would likely have modest to minimal effects on employment and wages. However, his data tell us nothing about whether significantly longer European-​style periods of paid leave would have positive or detrimental effects. Baum’s findings largely coincide with an earlier analysis by Jane Waldfogel (1999), who observed that the FMLA led to increased coverage and usage by eligible women without imposing significant costs in the form of lower employment or wages for women overall. Waldfogel used her study to dispel two “myths” about the FMLA. First, she argued that, despite the FMLA’s purported limitations (in covering a minority of workers, many of whom already had benefits, and in securing only unpaid leave), its effects on coverage

 

FAMILY AND HOUSEHOLD ECONOMICS    301 and leave utilization were not negligible. Although acknowledging that the law’s impact has been limited by some of these factors, she reports a “notable increase in coverage and an even bigger increase in leave-​taking” in the wake of its enactment. Waldfogel points toward two possible reasons for this effect: the modest but measurable increase in the population entitled to leave through the FMLA’s expanded coverage, and more utilization of other available leave benefits resulting from a greater acceptance by employees and employers of leave-​taking for family needs in the wake of the FMLA’s enactment. Waldfogel also disputes the FMLA’s presumed negative effects on worker compensation and participation by observing slightly higher employment levels (in part because of greater retention and lower quit rates among women) and minimal net wage effects post-​FMLA, at least in the short run.

12.4.4.2. Effects of Mandated Parental Leave in Europe Unlike the United States, all Western European countries offer at least 3 months of paid benefits to women after the birth of a child. Christopher Ruhm’s study (1998) argues that these mandates lead to more women going to work but lower women’s relative wages, especially for leaves of longer duration. While short periods of up to 3 months of paid parental leave increase the employment-​to-​population ratios of women by 3% to 4% with little effect on wages, longer periods of up to 9 months raise employment-​to-​ population ratios by 4% but cause hourly earnings to drop by around 3%. Ruhm hypothesized that the rise in women’s employment ratios can be explained by multiple factors, including increased workforce participation and duration, as well as changes in the ways the government and managers categorize women workers. According to Ruhm, approximately 25% to 50% of the change is likely due to more women being designated as “employed but absent from work” while on leave. The remaining change in employment ratios is probably due to increased labor-​force participation by females who would not otherwise be working, including those drawn into jobs by the availability of generous maternity benefits and those returning to work after becoming mothers rather than leaving the workforce entirely. As for wages, Ruhm’s results showed that brief leave entitlements had little effect on women’s earnings, but that longer leave periods resulted in 2% to 3% reductions in relative wages—​a difference he termed “substantial.” Ruhm hypothesized that shorter leave periods impose fewer costs on employers, especially where (as in some countries) the benefits are paid by the government. Longer leave periods, on the other hand, could decrease relative wages through several mechanisms. If more women are induced by the lure of maternity benefits to seek and take employment, the increases in the labor supply alone could lower female earnings. Extended periods of leave could also impose substantial nonwage costs on firms who must find and train temporary replacements. Finally, the pronatalist effect of maternity benefits might encourage women to have more children in a short time frame, resulting in absences stretching into years. Taking such extensive periods away from work could cause a “substantial depreciation of human capital.” However, Ruhm cautioned that his results and their interpretation should be viewed as speculative because of the small sample sizes and incomplete data. Thus, the

 

302   AMY WAX jury is still out on the long-​term effects of generous maternity benefits, and whether their impact on women’s pay, employment, and work status is positive or negative.

References Alderman, Harold, Chiappori, Pierre-​ André, Haddad, Lawrence, Hoddinott, John, and Kanbur, Ravi. 1995. “Unitary versus Collective Models: Is it Time to Shift the Burden of Proof?” World Bank Research Observer 10, pp. 1–​19. Arrighi, Barbara A. and Maume, David J., Jr. 2000. “Workplace Subordination and Men’s Avoidance of Housework.” Journal of Family Issues 21, pp. 464–​487. Artis, Julie E. and Pavako, Eliza K. 2003. “Explaining the Decline in Women’s Household Labor:  Individual Change and Cohort Differences.” Journal of Marriage and Family 65, pp. 746–​761. Badr, Hoda and Acitelli, Linda K. 2008. “Attachment Insecurity and Perceptions of Housework:  Associations with Marital Well-​ Being.” Journal of Family Psychology 22, pp. 313–​319. Bartley, Sharon J., Blanton, Priscilla W., and Gilliard, Jennifer L. 2005. “Husbands and Wives in Dual-​Earner Marriages: Decision-​Making, Gender Role Attitudes, Division of Household Labor, and Equity.” Marriage and Family Review 37, pp. 69–​74. Batalova, Jeanne A. and Cohen, Philip N. 2002. “Premarital Cohabitation and Housework:Couples in Cross-​National Perspective.” Journal of Marriage and Family 64, pp. 743–​755. Baum II, Charles L. 2003. “The Effect of State Maternity Leave Legislation and the 1993 Family and Medical Leave Act on Employment and Wages.” Labour Economics 10, pp. 573–​596. Baxter, Janeen. 2005. “To Marry or Not to Marry: Marital Status and the Division of Household Labor.” Journal of Family Issues 26, pp. 300–​321. Becker, Gary. 1974. “A Theory of Social Interactions.” Journal of Political Economy 82, pp. 1063–​1093. Becker, Gary. 1981. A Treatise on the Family. Cambridge, MA: Harvard University Press. Bennett, Fran. 2013. “Researching Within-​Household Distribution: Overview, Developments, Debates, and Methodological Challenges.” Journal of Marriage and Family 75, pp. 582–​597. Bergstrom, Theodore C. 2008. “Rotten Kid Theorem.” The New Palgrave Dictionary of Economics. 2nd ed. http://​www.econ.ucsb.edu/​~tedb/​archive/​palgrave.pdf [Accessed 21 August 2016]. Bianchi, Suzanne M. and Milkie, Melissa A. 2010. “Work and Family Research in the First Decade of the 21st Century.” Journal of Marriage and Family 72, pp. 705–​725. Bianchi, Suzanne M., Milkie, Melissa A., Sayer, Liana C., and Robinson, John P. 2000. “Is Anyone Doing the Housework? Trends in the Gender Division of Household Labor.” Social Forces 79, pp. 191–​228. Bianchi, Suzanne M., Robinson, John P., and Milkie, Melissa A. 2006. Changing Rhythms of American Family Life. New York: Russell Sage Foundation. Bird, Chloe E. 1999. “Gender, Household Labor, and Psychological Distress: The Impact of the Amount and Division of Housework.” Journal of Health and Social Behavior 40, pp. 32–​45. Bittman, Michael, England, Paula, Sayer, Liana, Folbre, Nancy, and Matheson, George. 2003. “When Does Gender Trump Money? Bargaining and Time in Household Work.” American Journal of Sociology 109, pp. 186–​214.

 

FAMILY AND HOUSEHOLD ECONOMICS    303 Blau, Francine, Ferber, Marianne, and Winkler, Anne. 2014. The Economics of Women, Men, and Work 7th edition. Boston, MA: Pearson. Blundell, Richard, Chiappori, Pierre-​André, Magnac, Thierry, and Meghir, Costas. 2007. “Collective Labor Supply:  Heterogeneity and Non-​Participation.” Review of Economic Studies 74, pp. 417–​445. Bolzendahl, Catherine and Myers, Daniel J. 2004. “Feminist Attitudes and Support for Gender Equality: Opinion Change in Women and Men, 1974–1998.” Social Forces 83, pp. 759–​790. Braun, Michael, Lewin-​ Epstein, Noah, Stier, Haya, and Baumgärtner Miriam K. 2008. “Perceived Equity in the Division of Household Labor.” Journal of Marriage & Family 70, pp. 1145–​1156. Brines, Julie. 1994. “Economic Dependency, Gender, and the Division of Labor at Home.” American Journal of Sociology 100, pp. 652–​688. Brinig, Margaret and Allen, Douglas W. 2000. “‘These Boots Are Made for Walking’: Why Most Divorce Filers Are Women.” American Law and Economics Review 2, pp. 126–​169. Brinig, Margaret and Crafton, Steven M. 1994. “Marriage and Opportunism.” Journal of Legal Studies 23, pp. 869–​894. Browne, Kingsley. 2002. Biology at Work:  Rethinking Sexual Equality. New Brunswick, NJ: Rutgers University Press. Burggraf, Shirley. 1997. The Feminine Economy and Economic Man. Cambridge, MA: Perseus. Cha, Youngjoo and Thébaud, Sarah. 2009. “Labor Markets, Breadwinning, and Beliefs: How Economic Context Shapes Men’s Gender Ideology.” Gender & Society 23, pp. 215–​243. Chiappori, Pierre-​André, Fortin, Bernard, and Lacroix, Guy. 2002. “Marriage Market, Divorce Legislation, and the Household Labor Supply.” Journal of Political Economy 110, pp. 37–​72. Clermont, Kevin M. and Schwab, Stewart J. 2009. “Employment Discrimination Plaintiffs in Federal Court: From Bad to Worse?” Harvard Law & Policy Review 103, pp. 128–​129. Coltrane, Scott. 2000. “Research on Household Labor: Modeling and Measuring the Social Embededness of Routine Family Work.” Journal of Marriage and Family 62, pp. 1208–​1233. Cunningham, Mick. 2007. “Influences of Women’s Employment on the Gendered Division of Household Labor over the Life Course: Evidence from a 31-​Year Panel Study.” Journal of Family Issues 28, pp. 422–​444. Davis, Shannon N. and Greenstein, Theodore N. 2004. “Cross-​National Variations in the Division of Household Labor.” Journal of Marriage and Family 66, pp. 1260–​1271. Davis, Shannon N., Greenstein, Theodore N., and Gerteisen, Jennifer P. 2007. “Effects of Union Type on Division of Household Labor: Do Cohabitating Men Really Perform More Housework?” Journal of Family Issues 28, pp. 1246–​1272. De Henau, Jerome and Himmelweit, Susan. 2013. “Unpacking Within-​Household Gender Differences in Partners’ Subjective Benefits from Household Income.” Journal of Marriage and Family 75, pp. 611–​624. Domínguez-​Folgueras, Marta. 2013. “Is Cohabitation More Egalitarian? The Division of Household Labor in Five European Countries.” Journal of Family Issues 34, pp. 1623–​1646. Doss, Cheryl. 2011. “Intrahousehold Bargaining and Resource Allocation in Developing Countries.” World Bank Research Observer 28, pp. 52–​78. EEOC. 2007. “Enforcement Guidance:  Unlawful Disparate Treatment of Workers with Caregiving Responsibilities.” http://​www.eeoc.gov/​policy/​docs/​caregiving.pdf [Accessed 21 August 2016]. Erickson, Rebecca J. 2005. “Why Emotion Work Matters: Sex, Gender, and the Division of Household Labor.” Journal of Marriage and Family 67, pp. 337–​351.

 

304   AMY WAX Evertsson, Marie and Nermo, Magnus. 2007. “Changing Resources and the Division of Housework:  A  Longitudinal Study of Swedish Couples.” European Sociology Review 23, pp. 455–​470. Families and Work Institute. 2008. “Employer Study from Families and Work Institute Shows Significant Changes for U.S. Workers Since 1998.” https://​benefitslink.com/​pr/​detail. php?id=41967 [Accessed 21 August 2016]. Fan, Pi-​Ling and Marini, Margaret M. 2000. “Influences on Gender-​Role Attitudes during the Transition to Adulthood.” Social Science Research 29, pp. 258–​283. Folbre, Nancy. 1994. “Children as Public Goods.” American Economic Review 84, pp. 86–​90. Frank, Robert. 1995. The Winner Take-​All Society. New York: Penguin Books. Frisco, Michelle L. and Williams, Kristi. 2003. “Perceived Housework Equity, Marital Happiness, and Divorce in Dual-​Earner Households.” Journal of Family Issues 24, pp. 51–​73. Fuwa, Makiko. 2004. “Macro-​Level Gender Inequality and the Division of Household Labor in 22 Countries.” American Sociology Review 69, pp. 751–​767. Geduling v Aiello, 417 U.S. 484, 492 (1974). General Electric Co. v Gilbert, 429 U.S. 125, 138–​140 (1976). Gershuny, Jonathan and Sullivan, Oriel. 2003. “Time Use, Gender, and Public Policy Regimes.” Social Politics 10, pp. 205–​228. Goldin, Claudia. 2014. “A Grand Gender Convergence: Its Last Chapter.” American Economic Review 104, pp. 1091–​1119. Greenstein, Theodore N. 2000. “Economic Dependence, Gender, and the Division of Labor in the Home: A Replication and Extension.” Journal of Marriage and Family 62, pp. 322–​335. Gruber, Jonathan. 1994. “The Incidence of Mandated Maternity Benefits.” American Economic Review 84, pp. 622–​641. Gupta, Sanjiv. 1999. “The Effects of Transitions in Marital Status on Men’s Performance of Housework.” Journal of Marriage and Family 61, pp. 700–​7 11. Gupta, Sanjiv. 2006. “Her Money, Her Time: Women’s Earning and their Housework Hours.” Social Science Research 35, pp. 975–​999. Gupta, Sanjiv. 2007. “Autonomy, Dependence, or Display? The Relationship between Married Women’s Earnings and Housework.” Journal of Marriage and Family 69, pp. 399–​417. Hakim, Catherine. 2000. Work-​ Lifestyle Choices in the 21st Century:  Preference Theory. Oxford: Oxford University Press. Herrera, Ruth S. and DelCampo, Robert L. 1995. “Beyond the Superwoman Syndrome: Work Satisfaction and Family Function among Working-​Class  Mexican-​American Women.” Hispanic Journal of Behavioral Science 17, pp. 49–​60. Himmelweit, Susan, Santos, Cristina, Sevilla, Almudena, and Sofer, Catherine. 2013. “Sharing of Resources within the Family and the Economics of Household Decision Making.” Journal of Marriage and Family 75, pp. 625–​639. H.R. Rep. No. 95-​948, at 2-​3 (1978), reprinted in 1978 U.S.C.C.A.N. 4749, 4750–​4751. Hymowitz, Kay S. 2011a. “Sorry, But the Gender Wage Gap is Here to Stay.” Pittsburgh Post-​ Gazette. http://​old.postgazette.com/​pg/​11317/​11892601090.stm?cmpid=newspanel#ixzz1dm ahftzw [Accessed 21 August 2016]. Hymowitz, Kay S. 2011b. “Why the Gender Gap Won’t Go Away. Ever.” City Life Journal 21. http://​www.city-​journal.org/​2011/​21_​3_​gender-​gap.html. [Accessed 21 August 2016]. Hymowitz, Kay. 2012a. “Why Women Make Less Than Men” Wall Street Journal (April 26, 2012). http://​online.wsj.com/​news/​articles/​SB100014240527023035924045773618830194142 96 [Accessed 21 August 2016].

 

FAMILY AND HOUSEHOLD ECONOMICS    305 Hymowitz, Kay. 2012b. “Rachel Maddow Sparks Gender Wage Fight.” POLITICO (May 4, 2012). http://​www.politico.com/​news/​stories/​0512/​75921.html. [Accessed 21 August 2016]. Hymowitz, Kay. 2013. “Do Women Really Want Equality?” TIME (Sept. 4, 2013). http://​ideas.time.com/​2013/​09/​04/​do-​women-​really-​want-​equality/​ [Accessed 21 August 2016]. Jolls, Christine. 2000. “Accommodation Mandates.” Stanford Law Review 53, pp. 223–​306. Kamo, Yoshinori and Cohen, Ellen L. 1998. “Division of Household Work between Partners:  A  Comparison of Black and White Couples.” Journal of Comparative Family Studies 29, pp. 131–​146. Kangas, Olli and Rostgaard, Tine. 2007. “Preferences or Institutions? Work-​Family Life Opportunities in Seven European Countries.” Journal of European Social Policy 17, pp. 240–​256. Knudson, Knud and Wærness, Kari. 2008. “National Context and Spouses’ Housework in 34 Countries.” European Sociology Review 24, pp. 97–​113. Lachance-​Grzela, Mylène and Bouchard, Geneviève. 2010. “Why do Women do the Lion’s Share of Housework? A Decade of Research.” Sex Roles 63, pp. 767–​780. Lechene, Valerie and Attansio, Orazio. 2002. “Tests of Income Pooling in Household Decisions.” Review of Economic Dynamics 5, pp. 720–​748. Lennon, Mary-​Clare and Rosenfeld, Sarah. 1994. “Relative Fairness and the Division of Housework: The Importance of Options.” American Journal of Sociology 100, pp. 506–​531. Lincoln, Anne E. 2008. “Gender, Productivity, and the Marital Wage Premium.” Journal of Marriage and Family 70, pp. 806–​814. Lundberg, Shelly and Pollack, Robert A. 1993. “Separate Spheres Bargaining and the Marriage Market.” Journal of Political Economy 101, pp. 988–​1010. Lundberg, Shelly and Pollack, Robert A. 1996. “Bargaining and Distribution in Marriage.” Journal of Economic Perspectives 10, pp. 139–​158. Lundberg, Shelly, Pollack, Robert A., and Wales, Terence J. 1997. “Do Husbands and Wives Pool their Resources? Evidence from the UK Child Benefit.” Journal of Human Resources 32, pp. 463–​480. Mahony, Rhona. 1995. Kidding Ourselves:  Breadwinning, Babies, Bargaining Power. New York: Basic Books. Mannino, Clelia A. and Deutsch, Francine M. 2007. “Changing the Division of Household Labor: A Negotiated Process between Partners.” Sex Roles 56, pp. 309–​324. Mattingly, Marybeth J. and Sayer, Liana C. 2006. “Under Pressure: Gender Differences in the Relationship between Free Time and Feeling Rushed.” Journal of Marriage and Family 68, pp. 205–​221. McPeak, John and Doss, Cheryl. 2006. “Are Household Production Decisions Cooperative? Evidence on Pastoral Migration and Milk Sales from Northern Kenya.” American Journal of Agricultural Economics 88, pp. 525–​541. Mikulka, Gerold. 1998. “Division of Household Labor and Perceived Justice: A Growing Field of Research.” Social Justice Research 11, pp. 215–​241. National Partnership for Women & Families. 2008. “The Pregnancy Discrimination Act: Where We Stand 30 Years Later.” http://​www.nationalpartnership.org/​site/​DocServer/​ Pregnancy_​Discrimination_​Act_​-​_​Where_​We_​Stand_​30_​Years_​L.pdf. Organization for Economic Cooperation and Development (OECD). 2013. “How’s Life? 2013:  Measuring Well-​Being.” http://​www.oecd.org/​statistics/​howslife.htm [Accessed 21 August 2016].

 

306   AMY WAX Offer, Shira and Schneider, Barbara. 2011. “Revisiting the Gender Gap in Time-​ Use Patterns:  Multitasking and Well-​ Being among Mothers and Fathers in Dual-​ Earner Families.” American Sociology Review 76, pp. 809–​833. Olsaretti, Serena. 2013. “Children as Public Goods?” Philosophy and Public Affairs 41, pp. 226–​258. Orbuch, Terri L. and Eyster, Sandra L. 1997. “Division of Household Labor among Black Couples and White Couples.” Social Forces 76, pp. 301–​332. Parker, Wendy. 2006. “Lessons in Losing: Race Discrimination in Employment.” Notre Dame Law Review 81, pp. 889–​954. Parkman, Allen M. 2004. “Bargaining Over Housework:  The Frustrating Situation of Secondary Wage Earners.” The American Journal of Economics and Sociology 63, pp. 765–​794. Pinto, Katy M. and Coltrane, Scott. 2009. “Divisions of Labor in Mexican Origin and Anglo Families: Structure and Culture.” Sex Roles 60, pp. 482–​495. Potuchek, Jean L. 1992. “Employed Wives’ Orientation to Breadwinning: A Gender Theory Analysis.” Journal of Marriage and Family 54, pp. 548–​558. Rhoads, Steven E. and Rhoads, Christopher H. 2012. “Gender Roles and Infant/​Toddler Care: Male and Female Professors on the Tenure Track.” Journal of Social, Evolutionary, and Cultural Psychology 6, pp. 13–​31. Robinson, Bryan K. and Hunter, Erica. 2008. “Is Mom Still Doing It All? Reexamining Depictions of Family Work in Popular Advertising.” Journal of Family Issues 29, pp. 465–​486. Ruhm, Christopher J. 1998. “The Economic Consequences of Parental Leave Mandates: Lessons from Europe.” Quarterly Journal of Economics 113, pp. 285–​317. Shelton, Beth Anne and John, Daphne. 1993. “Does Marital Status Make a Difference? Housework among Married and Cohabitating Men and Women.” Journal of Family Issues 14, pp. 401–​420. Shelton, Beth Anne and John, Daphne. 1996. “The Division of Household Labor.” Annual Review of Sociology 22, pp. 299–​322. Samuelson, Paul A. 1956. “Social Indifference Curves.” Quarterly Journal of Economics 70, pp. 1–​22. South, Scott J. and Spitze, Glenna D. 1994. “Housework in Marital and Nonmarital Households.” American Sociology Review 59, pp. 327–​347. Still, Mary C. 2006. “Litigating the Maternal Wall:  U.S. Lawsuits Charging Discrimination against Workers with Family Responsibilities.” http://​www.worklifelaw.org/​pubs/​ FRDreport.pdf [Accessed 21 August 2016]. Suk, Julie C. 2010. “Are Gender Stereotypes Bad For Women? Rethinking Antidiscrimination Law and Work-​Family Conflict.” Columbia Law Review 110, pp. 1–​69. Sunstein, Cass. 2001. “Human Behavior and the Law of Work.” Virginia Law Review 87, pp. 205–​276. Udry, Christopher. 1996. “Gender, Agricultural Production, and the Theory of the Household.” Journal of Political Economics 104, pp. 1010–​1046. Vermeulen, Frederic. 2002. “Collective Household Models:  Principles and Main Results.” Journal of Economic Surveys 16, pp. 533–​564. Waldfogel, Jane. 1999. “The Impact of the Family and Medical Leave Act.” Journal of Policy Analysis and Management 18, pp. 281–​302. Waldfogel, Jane. 2001. “Family and Medical Leave: Evidence from the 2000 Surveys.” Monthly Labor Review 17–​23. http://​academiccommons.columbia.edu/​download/​fedora_​content/​ download/​ac:152428/​CONTENT/​mlr.2001.09.art2.pdf [Accessed 21 August 2016].

 

FAMILY AND HOUSEHOLD ECONOMICS    307 Walker v Fred Nesbit Distributing Co., 2004 U.S. Dist. LEXIS 15969 (S.D. Iowa 2004). Wang, Wendy. 2013. “The ‘Leisure Gap’ Between Mothers and Fathers.” Pew Research Center (Oct. 17, 2013). http://​www.pewresearch.org/​fact-​tank/​2013/​10/​17/​the-​leisure-​gap-​between-​ mothers-​and-​fathers/​ [Accessed 21 August 2016]. Wax, Amy. 1998. “Bargaining in the Shadow of the Market: Is There a Future for Egalitarian Marriage?” Virginia Law Review 84(May), p. 509. Wax, Amy. 1999. “Caring Enough:  Sex Roles, Work, and Taxing Women.” Villanova Law Review 44, pp. 495–​524. Wax, Amy. 2000. “Rethinking Welfare Rights: Reciprocity Norms, Reactive Attitudes and the Political Economy of Welfare Reform.” Law & Contemporary Problems 63(Winter/​Spring), p. 257. Wax, Amy. 2003. “Something for Nothing: Liberal Justice and Welfare Work Requirements” Emory Law Journal 52(Winter), p. 1. Wax, Amy. 2004. “Family-​Friendly Workplace Reform: Prospects for Change.” Annals of the Academy of Political and Social Science 596, pp. 36–​61. Wax, Amy. 2014. “Diverging Destinies Redux.” Michigan Law Review 112(April), p. 925. Williams, Joan. 2001. Unbending Gender: Why Family and Work Conflict and What to Do About It. Oxford: Oxford University Press. Wolf, Alison. 2013. The XX Factor: How the Rise of Working Women Has Created a Far Less Equal World. New York: Crown Publishing.

 

Chapter 13

E C ONOMICS OF RE ME DI E S Ariel Porat

13.1. Introduction Analyzing the substantive law without its remedial part is almost meaningless.1 If we knew that the law imposes liability for negligence or for breach of a contract, but knew nothing about the remedies to which the victim is entitled in case his rights are infringed, we would know very little about the impact of the law on the real world. Therefore, in all legal systems, the remedies are interlinked with the substantive law. In civil legal systems, the same code that allocates entitlements among the parties also sets the remedies for protecting those entitlements. In both civil and common law legal systems, it is hard to imagine a court deciding a substantive law dispute without taking into account, explicitly or implicitly, the remedies that are available to victims. Indeed, it is hard to imagine the creation of the substantive law, by either legislatures or courts, without careful consideration of the remedial consequences of its breach. The claim that substantive law and remedies are interlinked with one another might imply that each legal field must have its own unique remedies. If that were the case, remedies would not be an independent topic, but rather a subtopic in each and every substantive field of the law.2 This implication, however, is wrong. Remedies in different legal fields have much in common, and the study of remedies as a topic can teach us a lot, especially when the goals of the substantive legal fields are similar (Cooter 1985). Consider tort law and contract law. Under its efficiency rationale, tort law should minimize social costs, thereby enhancing social welfare. In order to achieve this goal, tort law should provide incentives for both the injurer and the victim to take efficient 1  Ariel Porat is Alain Poher Professor of Law at Tel Aviv University and Fischel-​Neil Distinguished Visiting Professor of Law at the University of Chicago Law School. I thank Hed Kovetz for excellent research assistance. 2  In most law schools in common law jurisdictions, an important part of the Contract Law or the Tort Law class is remedies (for breaching a contract or for wrongfully inflicting harm, respectively), and only a few law schools offer a Remedies class.

 

ECONOMICS OF REMEDIES    309 precautions. Similarly, contract law should also provide the parties with efficient incentives, in order to enable them to maximize the contractual surplus. In both torts and contracts, providing the injurer/​promisor and the victim/​promisee with efficient incentives is done through a combination of substantive and remedial law. It should therefore come as no surprise that the remedies in both legal fields share much in common. This chapter emphasizes the common denominators of the remedies in torts and contracts. Some remedies that are more typical of either contracts or torts are also discussed. While the remedies in both fields are similar, they are not identical, and often are adapted to the legal context to which they apply.

13.2.  Property Rules and Liability Rules 13.2.1. General Framework In a seminal article, Calabresi and Melamed (1972) distinguished between the allocation of entitlements and the remedies for protecting them, as two distinct stages in promoting efficiency. In particular, they argued that once entitlements are allocated, they can be protected by either property or liability rules. Under a property rule, no one is allowed to deprive the owner of his entitlement without his consent; under a liability rule, other people are allowed to do so, but must compensate the owner for his losses. Thus, suppose Polluter inflicts harm on Resident. The law should allocate an entitlement, either to Resident to live without the pollution or to Polluter to pollute without interference. Assume first that the law made the former choice, so that the entitlement is allocated to Resident. Now a second choice must be made: to protect the entitlement with either a property rule or a liability rule. Under a property rule, Resident can sue Polluter in court and get an injunction, prohibiting further pollution (rule 1, in Calabresi and Melamed’s terms); under a liability rule, Resident is entitled to compensation only, so it is Polluter’s choice whether to stop polluting or, instead, pollute and compensate Resident for his losses (rule 2, in Calabresi and Melamed’s terms). Assume next that the law allocated the entitlement to Polluter rather than to Resident. Here, too, Polluter’s entitlement can be protected with either a property or a liability rule. Under a property rule, no one can stop Polluter from polluting without his consent (rule 3); under a liability rule, Resident can stop Polluter from polluting even without his consent, but if she chooses to do so she should compensate Polluter for the harm he suffers by ceasing the pollution (rule 4). The four rules are summarized in Table 13.1. Calabresi and Melamed analyzed the efficiency considerations in making the choices with regard to allocating and protecting the entitlements. First, they argued that when transaction costs are low it does not really matter, from an efficiency (rather than a distributional) perspective, whether the entitlement is allocated to Polluter or Resident: as

 

310   ARIEL PORAT Table 13.1 Allocating and Protecting Entitlements Entitlement

Protection

Rule 1

Resident

Property rule

Rule 2

Resident

Liability rule

Rule 3

Polluter

Property rule

Rule 4

Polluter

Liability rule

the Coase Theorem (1960) teaches us, with low transaction costs the parties would reach the efficient solution regardless of the initial allocation of entitlements. Thus, if Polluter is the cheapest cost avoider and rule 1 applies, he would take measures to prevent the harm (otherwise Resident would get an injunction in court, prohibiting pollution); if rule 3 applies, instead, Resident would offer Polluter a payment to stop polluting and Polluter would accept the offer. The same reasoning applies to the reverse case when Resident is the cheapest cost avoider: Resident would take measures either to avoid the harm (under rule 3) or be paid by Polluter to do the same thing (under rule 1). Things become more complicated when transaction costs are high, which makes contracting between the parties either hard or impossible. Here it is necessary to distinguish between two scenarios: first, when the cheapest cost avoider can be identified, and second, when it is unknown who the cheapest cost avoider is or if there is one at all. In the first scenario, the entitlement should be allocated to the party who is not the cheapest cost avoider. Such an allocation would provide incentives to the cheapest cost avoider to take measures to prevent the harm. Thus, if Polluter is the cheapest cost avoider, allocating the entitlement to Resident and protecting her entitlement with a property rule (rule 1) would incentivize Polluter to prevent the harm. The same logic applies to the reverse case when Resident is the cheapest cost avoider: here, rule 3 would do the work. The second scenario is the more interesting because it calls for liability rules. Assume that it is impossible for either courts or legislatures to know whether stopping pollution is efficient or not. Under those circumstances a property rule would not be an adequate solution, since once such a rule is applied, the parties might be stuck in an inefficient situation. For example, if rule 1 is applied, pollution is prohibited, so even if pollution is efficient, it will be prevented. Indeed, if the court or legislature knew that pollution is efficient, they would apply rule 3 and restore efficiency. But once courts or legislatures cannot know whether pollution is efficient or not, rule 2 could solve the problem. With rule 2, Polluter must decide whether to pollute and bear the resulting harm or stop polluting. Polluter will do whatever is cheaper for him, and that would be cheaper for society. Thus, if the harm is 100 and prevention costs are 50, Polluter will stop polluting, whereas if prevention costs are 150, Polluter will continue to pollute. In both cases Polluter’s interest and the societal interest align. Rule 2 is not the only alternative for solving the problem; rule 4 could be equally effective. Under rule 4, it is Resident rather than Polluter who decides whether to stop the

 

ECONOMICS OF REMEDIES    311 pollution or not. If she decides in the affirmative, she will order Polluter to stop polluting and reimburse him for prevention costs, but if she decides in the negative, she will do nothing and bear the harm. Thus, if the harm is 100 and prevention costs are 50, Resident will order Polluter to stop polluting, but if the costs are 150, Resident will do nothing and bear the harm. It is easy to see that the comparisons between harm and prevention costs under rules 2 and 4 are the same, the only difference being the identity of the party conducting the comparison and making the decision that follows. As with Polluter under rule 2, so too with Resident under rule 4, the interest of the party making the decision aligns with the societal interest. The choice between rules 2 and 4 has redistributive consequences (exactly like the choice between rules 1 and 3): rule 2 (and 1) favours victims (Residents), while rule 4 (and 3) favours injurers (Polluters). But, even if efficiency was the only consideration that matters for the law, the choice between the two rules could make a difference. The main efficiency consideration for choosing between the two rules is the availability of information to the courts applying the rules and the risk of errors that follow. Thus, if courts have better information about the harms to victims than about prevention costs, rule 2 will be more efficient than rule 4 is, and if the reverse is true, rule 4 will be the more efficient rule. To understand why, consider the preceding numerical example when harm is 100 and prevention costs are 50, and assume that rule 2 applies. Efficiency wise, the harm should be prevented because it is greater than prevention costs. But suppose now that the courts underestimate Resident’s harm—​say, because they are unaware of the high value Resident ascribes to her property—​and set damages at 40 instead of 100. Under those circumstances, Polluter will inefficiently pollute, because prevention costs are greater than damages. Applying rule 4 could solve the problem if, but only if, the courts estimate Polluter’s prevention costs accurately enough. Thus, if under rule 4, ordering Polluter to stop polluting would trigger Resident’s liability of 50, Resident will compare his harm of 100 (the realistic assumption here is that Resident accurately estimates her own harm) with expected liability of 50, and decide to stop the pollution. In this way, efficiency would be restored. In our last example, rule 2 leads to underdeterrence; with different numbers and overestimation of Resident’s harm by courts, rule 2 might lead to overdeterrence. In both cases, rule 4 might sometimes—​but not always—​solve the problem. In other cases the reverse might be true: under-​or overestimation of prevention costs by courts may result in inefficiencies under rule 4, which rule 2 might sometimes—​but not always—​ameliorate. In cases where courts’ errors are not solvable under either rule 2 or 4 in a satisfactory manner, the case for a liability rule becomes weaker, and the case for a property rule becomes stronger. With high transaction costs, uncertainty as to whether prevention by Polluter is efficient or not, and a high risk of courts’ errors with respect to both harm and prevention costs, it is hard to choose between a property or liability rule. In the real world, rule 4 is very rarely applied (Chang 2014).3 The main reason seems to be that in cases when transaction costs are high, typically, victims are numerous, 3  For a case where it was applied, see Spur Industries, Inc. v Dell E. Webb Development Co., 494 P. 2d 700 (Ariz. 1972).

 

312   ARIEL PORAT and cooperation between them—​a prerequisite for rule 4’s implementation—​is often implausible because of a free-​riding problem. Thus, when there are many Residents who should decide whether to stop Polluter from polluting in return for monetary payment—​as rule 4 requires—​each Resident might refuse to share the costs, knowing that if other Residents pay, pollution will stop anyway and she will be able to reap the benefit for free. The same problem typically does not arise with rule 2: the injurer needs no one’s cooperation in order to inflict harm on the victim and afterwards compensate him for it (Porat 2009).4

13.2.2. Refinements Calabresi and Melamed’s article has inspired many commentators who developed its original arguments, offered new applications, and criticized some aspects of them. Owing to limitations of space, only a few contributions to the literature will be mentioned in this chapter. Lucian Bebchuk (2001) pointed out that Calabresi and Melamed offered an ex post analysis that takes as a given the costs and benefits that would be generated for the parties with and without externality-​producing actions. This analysis, claimed Bebchuk, does not capture the entire picture. The allocation of entitlements and the way in which they are protected divide values between the parties differently, and this ex post division has a considerable impact on the parties’ ex ante decisions. Ex ante decisions take place before the decisions are made whether to undertake externality-​producing actions, and they influence the parties’ potential payoffs with or without these externality-​producing actions. A full account of the efficiency of any given allocation of entitlements and how they are protected—​Bebchuk’s argument goes—​must consider not only the ex post analysis but also the ex ante analysis. A couple of other articles, one authored by Ronen Avraham and the other by Ian Ayres, have dealt with the situation discussed in subsection 13.2.1, in which the court applying a liability rule lacks information about the values the parties ascribe to their entitlements. Avraham and Ayres suggested sets of rules, constructed on a combination of Calabresi and Melamed’s rules and some additional rules that can potentially encourage parties to reveal their true valuations of their entitlements, thereby facilitating outcomes that are more efficient (Avraham 2004; Ayres 1996). Lastly, Barbara Luppi and Francesco Parisi (2011) addressed the question regarding how remedies should be chosen when there are asymmetric transaction costs. They defined asymmetric transaction costs as situations in which different alternatives for reallocating resources entail different costs, such as when it is less costly to transfer an

4  Sometimes, however, injurers have an interest in directing their injurious activities towards the same victim/​s because marginal harm decreases when more injurers join the existing ones. In such cases, cooperation problems among injurers would emerge (Dillbary 2013).

 

ECONOMICS OF REMEDIES    313 entitlement from one use to another than it is in the reverse direction. Luppi and Parisi consider the possibility of using mixed remedies in such cases; for example, applying a property rule when A is the owner of the entitlement and B is the potential infringer, and a liability rule if the positions of A and B are reversed. Calabresi and Melamed’s framework is very useful in many contexts. In contract law, for example, specific performance could be characterized as a property rule, while damages are described as a liability rule. The next section further elaborates on this point.5

13.3.  Specific Performance, Damages, and Efficient Breach 13.3.1. General Framework There has been extensive debate in the legal literature over which remedy—​specific performance or damages—​should be the primary remedy and which the exception. For many, this debate represents a much broader dispute between law and economics and deontological scholars over the nature and goals of the law (Shiffrin 2009; Shavell 2009; Posner 2009). In the beginning, law and economics scholars argued that damages should be the primary remedy because specific performance, but not damages, discourages efficient and, therefore, desirable, breaches of contract. Later, law and economics scholars developed more nuanced arguments, showing that specific performance is often the more efficient remedy. The following two examples illustrate scenarios of efficient breach, and the discussion that follows clarifies the conditions under which a damages remedy would encourage such breaches and, at the same time, discourage inefficient breaches. Example  1. Gain-​Seeking Breach. Seller undertakes to manufacture a machine for Buyer. Expected costs of production are 80, the price, which is paid up front, is 90, and the value of the machine to Buyer is 100. After the contract is concluded, a second buyer shows up offering Seller 110 for the same machine. Seller can produce only one machine at a time, so he breaches the contract with Buyer and sells the machine to the second buyer. The only loss Buyer suffers is the machine’s lost value. Example 2. Loss-​Avoiding Breach. Seller undertakes to manufacture a machine for Buyer. Expected costs of production are 80, the price, which is paid up front, is 90, and the value of the machine to Buyer is 100. After the contract is concluded, owing 5 

The typical application of Calabresi and Melamed’s framework is intentional infliction of harm, as in nuisance (pollution) cases. Is that framework suitable for accidental harms? Should a negligence rule be characterized as a property or a liability rule? For the argument that negligence law should be characterized as liability rules, see Porat (2009, 199–​200) for the counterargument; also, see Coleman and Kraus (1986); Zipursky (1998, 55–​70).

 

314   ARIEL PORAT to a shortage of labour and materials, costs of production rise to 110. Seller breaches the contract. The only loss Buyer suffers is the machine’s lost value.

In both Examples 1 and 2, the breach of the contract is efficient. Under the assumption of perfect compensation with expectation damages as the measure of recovery, Seller pays damages of 100 to Buyer and reaps a benefit of 10 from the breach. An implicit assumption in the argument that the breach in Examples 1 and 2 is efficient is that, owing to high transaction costs, renegotiation between Seller and Buyer after contracting is too costly, and with Example 1, high transaction costs also preclude the sale of the machine (or assigning the right to the machine) by Buyer to the second buyer. Indeed, damages remedy, like any liability rule (see Section 13.2), could be justified only with high transaction costs because, otherwise, specific performance, like any property rule, would be preferable. With a remedy of specific performance, and with the assumption of high transaction costs, in both examples Seller will perform the contract inefficiently. Similarly, with disgorgement damages (in the amount of the profits made by Seller in Example 1 or in the amount of the savings made by Seller in Example 2), Seller will lack any motivation to breach efficiently because he gains nothing from a breach. Note that if compensation is lower than expectation damages, Seller, in both examples, might breach the contract even if performance is efficient. Thus, if damages are 90 (reliance damages), Seller will breach even if the second buyer in Example 1 offers him 95, or if costs of production in Example 2 rise to 95. In both cases the breach is inefficient, but it creates a benefit of 5 to Seller, which might motivate him to breach. The notion of efficient breach is justified by not only the general notion of promoting social welfare but also the more specific idea of incomplete contracts (Shavell 2009; Markovits and Schwartz 2011, 2012). According to the incomplete contracts idea, contract law provides the parties with default rules that apply to their contracts unless they opt out of those rules. The default rules save the parties transaction costs (in terms of negotiation and drafting costs), which they would have incurred but for the default rules. In order to achieve their goal, the default rules should be compatible with most parties’ interests because, otherwise, most parties would opt out of the defaults, and transaction costs would increase rather than decrease. The default rules will be compatible with most parties’ interests if they are efficient, viz., if they allocate risks and provide the parties with incentives in ways that increase the contractual surplus. The efficient breach idea—​so the incomplete contracts argument goes—​increases the contractual surplus, and is, therefore, compatible with most contractual parties’ interests. Therefore, it should be considered a desirable default rule. Note that increasing the contractual surplus serves not only the promisor’s interest, but also the promisee’s because the expected benefit of the option to breach efficiently will be shared by the two parties through price adjustment when making their contract. Therefore, basing the efficient breach idea on the theory of incomplete contracts might be more effective in defending this idea from deontological (or other nonutilitarian) attacks compared with basing it on the general notion of promoting social welfare.

 

ECONOMICS OF REMEDIES    315

13.3.2. Refinements The efficient breach argument is valid, as long as damages are fully compensatory. When damages are undercompensatory, specific performance often becomes the most efficient remedy. A notable category of cases where specific performance is the primary remedy because damages are typically undercompensatory is the sale of unique goods. In a now classical article, Anthony Kronman proposed a rationale for the willingness of courts to allow specific performance when the contract’s subject matter is a unique, rather than fungible, good. He suggested that with unique goods, far more so than with fungible goods, there is a substantial risk of undercompensation of buyers for two reasons: first, buyers often attach a subjective value to the unique good, and that value is not compensated for in the event of breach. Second, subsequent to breach of unique good contracts by sellers, buyers incur search costs in finding a substitute good, on top of the search costs they incurred when finding the original good, and those additional search costs are not compensated for (Kronman 1978). Since, according to Kronman, the parties would prefer specific performance over damages if the benefit to the seller from having the option to breach and pay damages is less than the costs of the breach to the buyer, and because the costs of the breach to the buyer with unique goods are typically higher than with fungible goods, Kronman concluded that specific performance would more often be preferred by the parties with unique goods than those with fungible goods. Another assumption underlying the efficient breach argument relates to transaction costs. In gain-​seeking breaches, such as in Example  1, the efficient breach argument assumes that transaction costs make it hard, even impossible, for Buyer to find the second buyer who values the subject matter of the contract more, and sell it to her. When this assumption is relaxed, specific performance might be the most efficient remedy. In an article published shortly after Kronman’s article, Alan Schwartz argued for a much broader application of specific performance than that suggested by Kronman. Schwartz claimed that the risk of undercompensation is substantial not only with unique goods, as Kronman suggested, but also with many fungible goods (Schwartz 1979). Furthermore, in contrast to Kronman, Schwartz claimed that the benefit to the seller of having the option to breach and pay damages is often less with fungible goods than with unique goods. Therefore, with fungible goods, the parties’ ex ante preferences would not necessarily warrant damages, rather than specific performance, as their preferred remedy. Further, for Schwartz, the central consideration in the choice between damages and specific performance as a remedy is which one entails lower post-​breach negotiation costs. This depends, according to Schwartz, mostly on whether (in our Example 1) it is easier for the seller or for the first buyer to find the second buyer who values the good more than the first buyer: only if it is the seller could damages be preferable to specific performance. Thomas Ulen took the argument for specific performance a step further, suggesting that specific performance should be the routine remedy for breach. Ulen, like Schwartz, regarded the post-​breach negotiation costs as a central factor in the efficiency of specific performance (Ulen 1984).

 

316   ARIEL PORAT While the first-​generation writings on efficient breach focused on the promisor’s decision to perform or breach, Richard Craswell (1988) analyzed the effects of the contractual remedies on various decisions made by the promisor and promisee. Craswell considered the effect of the remedies on the decisions as to whether to enter into the contract in the first place and what level of precautions to take in order to reduce the probability of breach. Craswell explained that even if post-​breach negotiation costs were zero, the prevailing remedy would still affect the parties’ decisions made before the promisor’s decision whether to perform or breach. In the economic analysis of contract law, it is implicitly assumed that the efficient breach argument is equally valid with respect to both loss-​avoiding and gain-​seeking breaches (Posner 2009). By contrast, lay people’s intuition is different:  experimental studies have indicated that people react more tolerantly to loss-​avoiding breaches than to gain-​seeking breaches (Baron and Wilkinson-​Ryan 2009). Moral philosophers have also distinguished between the two types of breaches, arguing that a breach to pursue a gain is more reprehensible than a breach to avoid a loss (Zamir and Medina 2010, 265). Lastly, behavioural law and economics could explain people’s different reactions to the two types of breach as a reflection of people’s different attitudes to losses as opposed to gains (Cohen and Knetsch 1992; cf. Zamir and Ritov 2010). Recently, Maria Bigoni, Stefania Bortolotti, Francesco Parisi, and Ariel Porat (2014) suggested, also from an economics perspective, that the case for allowing the promisor an option to breach is typically more vital in loss-​avoiding breaches than in gain-​seeking breaches.

13.3.3. The Promisee’s Incentives While the effects of remedies on the incentives of the promisor have been thoroughly analyzed in the literature, only a few scholars have analyzed their effects on the promisee’s incentives.6 A notable exception is Robert Cooter (1985), who pointed out that with full compensation, the victim’s incentives are eroded and overreliance might result. For example, if the promisee knows that there is a high likelihood of a breach, he might rely as if the likelihood of a breach is zero, knowing that he can reap the benefits of reliance if performance takes place but externalizes its costs to the promisee—​who will reimburse him for those costs—​if a breach occurs. Cooter suggested that when damages to the promisee are awarded at a fixed amount, the promisee relies efficiently because he fully internalizes both the costs and benefits of his reliance. Thus, liquidated damages, if they are set at the level of expected harm and remain invariant with respect to actual harm, might solve the overreliance problem.7

6  For an analysis of the mitigation of damages defence, see Goetz and Scott (1983). But the mitigation of damages defence applies only after a breach (or an anticipatory breach) occurs, and it is effective only when the victim’s behaviour is verifiable. 7  But it may create other inefficiencies: see infra Section 13.5.2.

 

ECONOMICS OF REMEDIES    317 Cooter and Porat (2002) discussed the erosion of the victim’s incentives not only with respect to overreliance, but also with respect to noncooperation. They noted that with fully compensatory damages, the promisee who could cooperate with the promisor and reduce the probability of a breach might be unwilling to do so, especially if noncoop­ eration is nonverifiable (otherwise a duty of cooperation or a comparative fault defence could solve the problem [Porat 2009]). Cooter and Porat suggested a novel theoretical solution, which they called “anti-​insurance.” According to their solution, the promisee and promisor make a contract with a third party (“anti-​insurer”), according to which the promisee assigns his right to damages to the anti-​insurer, so that in case of a breach the promisee receives no compensation and the promisor pays fully compensatory damages to the anti-​insurer. Before a breach occurs, the anti-​insurer pays to the promi­ sor and promisee for the valuable right to collect damages in case of a breach. Both the promisor and promisee now have efficient incentives because they fully internalize the costs and benefits of the precautions they will take to reduce the probability of a breach or reduce overreliance. Another more practical solution for improving the promisee’s incentives is to undercompensate her. With the risk of undercompensation, the promisee will be more willing to cooperate or avoid overreliance than with fully compensatory damages. Indeed, with imperfect compensation, the promisor’s incentives are deficient, but once the promisee’s incentives are taken into account, allowing some level of undercompensation could be optimal (Cooter and Porat 2004).

13.4.  Scope of Liability Not all harms are compensable, and not all victims can recover. First, in both torts and contracts, only foreseeable harms are recoverable. Second, in tort law, proximate cause is a limit to liability: if the harm is too remote, liability will not be imposed. Third, in negligence law, liability will be imposed only if the negligent injurer owed a duty of care to the victim. Both proximate cause and duty of care are used by courts as a means to limit liability for policy considerations (Restatement 3d Torts: Liability for Physical and Emotional Harm, § 7, cmt. a, 2010; Dobbs 2000, 448). Section 13.4 discusses several topics, all of which raise questions as to the appropriate scope of liability.

13.4.1. Foreseeability In both torts and contracts, foreseeability of losses is a precondition for the imposition of liability (Dobbs 2000, 443–​470; Farnsworth 2004, 792–​799). There are several efficiency justifications for this requirement. First, if the losses are not foreseeable there is no sense in imposing liability for their materialization, since such liability will not affect the behaviour of the injurer or the

 

318   ARIEL PORAT promisor. That is because if losses are unforeseeable, the costs of taking them into account when deciding which precautions to take are prohibitively high, so the injurer and the promisor will ignore them anyway, with or without liability (Landes and Posner 1987, 246–​247). Second, in negligence, when injury of any kind is unforeseeable—​namely, the probability of an injury is very low—​expected harm is low, and the costs of precautions typically exceed expected harm. When the costs of precautions are higher than expected harm, the injurer is non-​negligent, and liability should not be imposed. According to this justification for the foreseeability requirement, the unforeseeability of any injury is an indicator that there is no negligence on the injurer’s part to begin with. Note that this justification applies to a narrow set of cases where any injury—​as opposed to only the one that resulted in the litigated losses—​is unforeseeable. Third, on many occasions, the unforeseeable losses are foreseeable for the promisee and victim. In such cases the promisee and victim are typically the cheapest cost avoiders of the unforeseeable losses, whereas the injurer and promisor are not. Leaving the unforeseeable losses on the victim’s and promisee’s shoulders motivates them to take measures to avoid those losses, either before or after the wrong or breach takes place. This is especially essential when those measures are nonverifiable, and neither the comparative fault nor the mitigation of damages defence is applicable. Fourth and last, in the contractual context, the foreseeability requirement incentivizes the promisee to disclose private information to the promisor regarding the promisee’s unforeseeable losses, which is vital to the promisor for deciding whether to breach or to perform, as well as what level of precautions to take to avoid a breach. This justification for the foreseeability requirement is relevant, when conveying information from one party to another is possible and when the unforeseeable losses are foreseeable for the promisee (Ayres and Gertner 1989). To better understand this justification, assume that the promisee’s expected losses from a breach are 100 and foreseeable for her, but only a loss of 10 is foreseeable for the promisor. Without the foreseeability requirement, the promisee would not convey to the promisor any information regarding his high potential losses, because if the promisor knew about that, he would charge a higher price for his undertakings. As a result, with only partial information, the promisor would have deficient incentives to perform. In contrast, with the foreseeability requirement, the promisee would convey the information regarding her high potential losses to the promisor in order to make those losses foreseeable for him and, therefore, recoverable by the promisee. Conveying the information would secure the promisor’s efficient incentives, thereby increasing the contractual surplus.

13.4.2. Pure Economic Loss Victims often suffer pure economic losses. In contracts, losses are often purely economic in the sense that the promisee lost profits and nothing else, but as has already

 

ECONOMICS OF REMEDIES    319 been explained, liability is routinely imposed and for good economic reasons.8 In torts, in contrast, courts are commonly reluctant to impose liability for pure economic losses. There are several efficiency considerations that could justify this reluctance. The first consideration is that pure economic losses are often a private rather than a social cost9 (Bishop 1982). Imagine an injurer who created a nuisance to restaurateur 1, who shut down his restaurant for a week and lost profits of 100. Assume that the restaurant’s patrons fully mitigated their losses by dining during that week at another restaurant owned by restaurateur 2. Further, assume that the profits gained by restaurateur 2 from restaurateur 1’s patrons are 100, in addition to the regular profits made from his own regular patrons. From a social perspective—​so goes the economic argument—​no social harm has been done: profits were just transferred from one person to another. With no social harm, injurers should not take any costly precautions, so liability should be nil. This argument, however, has limits. To start with, if restaurateur 1 is a recurring loser and restaurateur 2 a recurring winner, no liability might inefficiently suppress restaurateur 1’s activity (Bishop 1982). Second, with no liability, restaurateur 1 might take costly precautions to avoid the harm; if the injurer, rather than restaurateur 1, is the cheapest cost avoider, liability would save the costs of precautions of restaurateur 1, even if at a cost to the injurer who—​given his expected liability—​would take precautions and prevent the harm (Dari-​Mattiacci and Schäfer 2007). Third, loss of profits—​as in our example—​are not necessarily just transfers of value from one person to another: the destruction of input might increase the marginal cost of production, leading to production decrease and higher prices (Rizzo, 1982a and 1982b). The second consideration against liability for pure economic losses is, that with such losses causation is often hard to prove. Thus, even if we think that those losses should be prevented, there is a risk that with liability the injurer will pay more than what he actually caused and be overdeterred. The reason why causation is harder to prove in pure economic loss cases than in physical injury cases is that those losses are of a type that occurs regularly with no wrongdoing, and, therefore, it is hard to distinguish them from non-​wrongful losses. To illustrate, assume that in the nuisance example, restaurant 1 was not shut down, but restaurateur 1 nonetheless lost profits owing to low attendance by patrons. Low attendance, however, could be caused by many other causes—​most of them non-​wrongful—​and it could be hard, sometimes even impossible, to isolate the effects of the wrongful from those of the non-​wrongful causes. If courts tend to resolve uncertainties in favour of victims—​and they are often so inclined—​they might impose too heavy a liability burden on the injurer (Abraham 2011, 1781–​1783). 8 

Supra Section 13.3. This, however, is not always so. Sometimes pure economic losses are a substitute for physical losses that are not compensated for practical reasons. For example, if the defendant polluted a river with chemical effluents and destroyed the wildlife in a certain area, since no one in particular owns the wildlife, recovery for the physical harm is impossible. Anglers, however, might be able to recover for lost profits even if those are pure economic losses: the lost profits could serve as a proxy for the social value of the lost wildlife. Cf. Pruitt v Allied Chemical Corp., 523 F. supp. 975 (1981). 9 

 

320   ARIEL PORAT The third and last consideration against awarding damages for pure economic losses is that those losses often have two characteristics that make the victim an especially effective cost avoider. The first characteristic is that those losses are accumulated over time, and, therefore, victims have a relatively long period to mitigate them. The second characteristic—​which has already been mentioned—​is that economic losses are of a type that occurs regularly with no wrongdoing; as a result, victims have expertise in handling and reducing them. The nuisance case is not the best example to illustrate these two characteristics, especially if the restaurant is shut down because of the nuisance. So let’s take another example: suppose a person is wrongfully injured in a road accident and has not showed up at work for 2 months. The injured person’s employer argues that he has suffered pure economic losses in terms of lost profits because of his employee’s absence from work. Obviously, the employer will not be able to recover. A  possible justification for this result is that the employer is a very effective cost avoider: her lost profits accumulate from day to day, so she has time to consider how to mitigate them (e.g. by hiring a substitute employee, reducing her activity level, or postponing the performance of some of the work for a few months). Furthermore, the employer is accustomed to handle such losses on a daily basis—​employees are often absent from work for non-​wrongful causes—​and must have expertise in minimizing them. Indeed, even if the employer had been entitled to compensation for her lost profits, she would have been required to mitigate those losses as a precondition for any recovery; however, most of the employer’s failures to mitigate losses are nonverifiable, and, therefore, the most effective way to encourage her to mitigate losses efficiently is just to let her bear them.

13.4.3. Nonpecuniary Loss Nonpecuniary losses are losses that have no economic impact on the victim, such as emotional distress, agony, disappointment, and pain and suffering. In contracts, nonpecuniary losses are compensated almost only when the main interest protected by the contract is nonpecuniary in nature (Farnsworth 2004, 810). Typical examples are tour package contracts,10 contracts to perform cosmetic surgeries,11 and contracts for providing services for weddings or funerals.12 In torts, nonpecuniary losses are typically compensated when accompanied by physical injury, mainly bodily injury, and only rarely when those are standalone losses (Dobbs 2000, 1050–​1053). One objection to the imposition of liability for nonpecuniary losses is that those losses are subjective, and, therefore, the risk of plaintiffs’ bringing frivolous claims is high. If those claims were to be allowed—​so the argument goes—​injurers might pay 10 

Jarvis v Swan Tours EWCA Civ 8 (1972). Sullivan v O’Connor 296 N.E. 2d 183 (1973). 12  Lewis v Holmes 109 La. 1030 (1903). 11 

 

ECONOMICS OF REMEDIES    321 excessive damages and be overdeterred. This objection, however, is less persuasive when there is some objective evidence for the existence of the nonpecuniary losses and their magnitude, such as when the victim suffers pain and suffering accompanied by bodily injury (Bovbjerg et al. 1989). Another objection is that inflicting nonpecuniary losses on the victim typically does not decrease her marginal utility of money, so transferring payments from the injurer to the victim typically does not improve social welfare (cf. Shavell 1987, 228–​231; Danzon 1984, 517, 521). This objection is attenuated, and it even disappears when deterrence is considered, and because nonpecuniary losses are social losses, injurers and promisors should internalize them and be adequately deterred (Rea 1982). The marginal utility of money argument coupled with the countervailing deterrence argument provides a compelling explanation of courts’ willingness to allow compensation for nonpecuniary losses (almost) only when the interest protected by the contract is mainly nonpecuniary. Imagine a construction contract between Builder and Owner, where the parties anticipate when making their contract that with a certain probability performance will be delayed and, as a result, Owner will be disappointed. Would they agree that in case of a breach Owner would be compensated for his disappointment? If yes, liability should be the contractual default rule; otherwise, no liability should be the default rule. Let’s start with the marginal utility of money argument. According to this argument, Owner will never insure against disappointment loss, because suffering this loss will not affect her marginal utility of money. Similarly, Owner will not “insure” against such a loss through the contract (because he will be required to pay a premium through the contract price). The deterrence argument does not change this conclusion, as long as we assume that most of the losses resulting from the breach are pecuniary. Because pecuni­ ary losses are compensated, deterrence is reasonably attained, even if a small fraction of the losses—​those that are nonpecuniary—​are not. Thus, in our example, the marginal utility of money argument seems to overcome the countervailing deterrence argument, and the parties would prefer, ex ante, not to have liability for nonpecuniary losses (cf. Rea 1982). If instead, most of the losses expected to result from the breach are nonpecuniary, the conclusion will be different. Take, for example, a contract made between a travel agency and a traveller for providing a package tour by the former to the latter. Here, the efficient rule, which most contractual parties would prefer, is liability for nonpecuniary losses. Although the traveller would never insure against such losses with an insurer, he would insure against them with the travel agency because such “insurance” would provide the agency with efficient incentives to perform the contract. In this example, as opposed to the previous one, most of the expected losses are nonpecuniary, and no liability for those losses would result in severe underdeterrence.13 13 

For the argument that victims might be willing to insure against nonpecuniary losses, see Croley and Hanson (1995).

 

322   ARIEL PORAT

13.4.4. Caps on Consequential Damages The foreseeability requirement, which has been discussed in Section 13.4.1, caps damages by allowing recovery for foreseeable losses only. Contractual parties, however, often cap damages even further, sometimes precluding liability for any consequential losses altogether (Farnsworth 2004, 799). One reason for contractual parties to cap consequential losses relates to the fact that consequential losses are often unique to the promisee, as when the breach causes the promisee a loss of profits coming from third parties. In such cases the promisee can often take steps to mitigate his consequential losses in nonverifiable ways, and making the promisee bear those losses would incentivize him to do so. At the same time, because of their uniqueness, those losses are often not accurately anticipated by the promisor (even if they are foreseeable enough to pass the foreseeability threshold), and, therefore, capping them promotes certainty. Indeed, without liability the promisor’s incentives to perform efficiently decreases because he externalizes some of the costs of the breach to the promisee, but given the advantages described previously, this flaw might be a price worth paying.14 A second reason, which applies mostly to consumer contracts, is that capping damages is an effective tool for avoiding adverse selection. Take an example used in the literature (Epstein 1989). A shipper ships packages for owners and sometimes the packages are damaged. Some packages are worth more; others worth less. The shipper charges a uniform price (discriminating in prices is either too costly or illegal), and when a package is damaged, he is liable for the full amount of the harm done. Under this liability scheme, owners whose packages are worth less than average subsidize owners whose packages are worth more than average because, by assumption, they all pay a uniform premium through the price. This cross-​subsidization is unjust, especially if owners of high-​ value packages are typically wealthier compared with owners of low-​value packages. Moreover, cross-​subsidization results in inefficient consumption and adverse selection. Specifically, because owners with low-​value packages pay more than what they should have paid given their low expected harm, their consumption level is too low; in contrast, because owners with high-​value packages pay less than what they should have paid given their high expected harm, their consumption level is too high. In addition, in the long run, because fewer owners of low-​value packages will consume the shipping services, prices will go up (as the average expected harm will go up). As a result, even fewer owners of low-​value packages will consume the shipping services; prices will go up again; and so on and so forth. At the end, only consumers with very high-​value packages will consume the shipping services, and, in the extreme case, the shipping ser­ vices will shut down. One way to solve the problem is to cap damages. Thus, the shipper could offer contracts to consumers under which damages are limited to the value of low-​ value packages. This would mitigate the inefficiencies, including the adverse selection 14 

Leaving some losses uncompensated also mitigates the promisee’s moral hazard problem: see Shavell (1979).

 

ECONOMICS OF REMEDIES    323 problem, and would be more just. Consumers with high-​value packages would be able to secure full compensation through first-​party insurance, if they choose to do so.

13.5.  The Measure of Recovery Generally, in both torts and contracts, damages are compensatory. But it may sometimes be unclear what compensatory damages are. First, Section 13.5.1 deals with this question in a very specific tort context: damages for bodily injury, and then, Section 13.5.2 deals with one contractual exception to the compensatory damages principle, which is liquidated damages.

13.5.1. Bodily Injury and Lost Income A major component in any award of damages for bodily injury is lost income. The result is that a high-​income (“rich”) victim receives compensation that is higher—​sometimes much higher—​than what a low-​income (“poor”) victim receives, even if both victims suffer from the same bodily impairment because of the wrongdoing. By contrast, when courts set the standard of care, they do not distinguish between rich and poor victims: the injurer is required to take the same level of care towards the victim regardless of his income, even if his type as rich or poor can be anticipated by the injurer.15 This leads to an inconsistency in tort law: there is a misalignment between the standard of care and damages (Porat 2011). The following example illustrates this point. Example 3. Poor and rich neighborhoods. John drives his car at a speed of 30 mph in a rich neighborhood. Unfortunately, he hits a pedestrian as she is crossing the street. Had John driven a bit more slowly, he would have succeeded in stopping his car in time and preventing the accident. A day later, John drives his car again at the same speed, but this time in a poor neighborhood. Once again, he hits a pedestrian. All driving conditions are the same as they were in the rich neighborhood the day before; therefore, in this case, as well, the accident would have been avoided had John driven his car a bit more slowly. Is it possible that, under a rule of negligence, the same court would find John liable for the first accident but not for the second?

Assuming that in the rich neighborhood, most people have a higher income than the residents of the poor neighborhood do, one could argue that different standards of care should be applied in the two neighborhoods. It is quite possible, even reasonable, then, that the same court would find that: (a) John failed to take due care in the rich

15 

When a victim’s type as rich or poor cannot be anticipated in advance by the injurer, there can be no question that the standard of care should be set according to the average victim.

 

324   ARIEL PORAT neighborhood and, therefore, should be held liable towards his victim; and (b) John took due care in the poor neighborhood and, therefore, should be exempt from all liability. Courts, however, do not set different standards of care for driving in rich and poor neighborhoods. Similarly, they also do not set different standards of care for doctors treating rich and poor patients. If a court were required to explain the application of the same standard of care for the rich and the poor, it would reason that the lives and limbs of the rich and poor have identical social value and, therefore, are deserving of the same level of legal protection. But such reasoning, convincing as it may be, is inconsistent with the practice of awarding higher damages to rich victims. This practice suggests that rich people’s lives and limbs are more highly valued by the law relative to those of poor victims. To be consistent with this practice, so it seems, injurers should take greater care towards the rich than towards the poor, just as they should be more careful in their interactions with high-​value property. Therefore, to restore consistency to the law, courts should have chosen one of two routes: to either apply different standards of care to rich and poor victims (contrary to what they actually do), coupled with different levels of compensation (as they actually do), or, alternatively, to apply the same standard of care to rich and poor victims (as they actually do), coupled with the same level of compensation (contrary to what they actually do). From a social perspective, there is no compelling reason why lost income should be the main criterion for valuing people’s lives and limbs,16 but this question is beyond the scope of this chapter. Therefore, let us assume first that lost income is the right criterion, and consider the efficiency of the law under this assumption. If lost income is the right criterion, and given that the standard of care is set uniformly according to average income but damages are awarded for the victim’s lost income, injurers will comply with the standard of care when they expect a rich victim but undercomply when they expect a poor victim. To see why, assume that the expected harm of rich victims is 15, the expected harm of poor victims is 5, and average expected harm is 10. Further, assume that the standard of care is set at 10, namely, injurers are required to take precautions up to 10 to reduce the risk to either rich or poor victims. With these figures, injurers will take precautions up to 10 towards rich victims (and pay no damages if harm occurs) and up to 5 towards poor victims (and pay damages if harm occurs). As the expected harm for rich victims is 15 and for poor victims is 5, injurers will be underdeterred towards rich victims and optimally deterred towards poor victims. Assume now that lost income is not the right criterion for awarding damages for bodily injury, and that rich and poor people’s lives and limbs have the same value. Further, assume that people’s lives and limbs are determined by average income.17 Now, taking precautions by injurers up to 10 towards rich victims is efficient because their 16  Lost income could be correlated with people’s productivity. It is questionable, however, whether this correlation is strong, and whether productivity is the main value of people’s lives and health. 17  This is a simplifying assumption: a plausible argument is that average income should also not play a major role in valuing victims’ lives and limbs (Friedman 1982; Porat and Tabbach 2011).

 

ECONOMICS OF REMEDIES    325 expected harm is assumed to be 10. Conversely, taking precautions up to 5 towards poor victims is inefficient because 10, rather than 5, is assumed to be their expected harm. The analysis so far has implicitly assumed no courts’ errors in setting the standard of care and awarding damages and no injurers’ errors in anticipating courts’ decisions. With the risk of errors, the analysis becomes more complex (Porat 2011), but it does not change the basic conclusion: efficiency wise,18 the value of people’s lives and limbs should be reflected in both the standard of care and damages in a consistent manner, otherwise inefficiencies will result.

13.5.2. Liquidated Damages While the default rule in contract law is that damages are compensatory, the parties are free to opt out of the default rule by incorporating a liquidated damages clause. If, however, damages are set too high, courts are authorized to strike down the liquidated damages clause as being a penalty, and to award compensatory damages instead. When deciding whether to uphold a liquidated damages clause, courts are instructed to consider the anticipated loss at the time of contracting, the actual loss, and the difficulties of proving the actual loss (Restatement 2d Contracts, §356, [1981]). As liquidated damages are closer to both anticipated and actual losses, and as there are more difficulties in proving actual loss, courts are more willing to uphold the liquidated damages clause. The Uniform Commercial Code (“UCC”) adds a fourth consideration to be taken into account, which is “the inconvenience or nonfeasibility of otherwise obtaining an adequate remedy” (UCC, §2-​7 18). It is puzzling why courts are authorized to scrutinize liquidated damages clauses, while they are generally not authorized to do the same thing with other clauses (Posner 1979, 290). Indeed, at least when the parties are rational and well informed, there is no reason to assume that the liquidated damages clause they incorporated into their contract sets damages too high (Schwartz 1990). In the next paragraphs, both the advantages and disadvantages of liquidated damages are discussed, when the question that arises is whether the disadvantages can justify the wide discretion granted to courts to strike down liquidated damages clauses. Incorporating a liquidated damages clause into their contracts has several advantages for the parties. First, it saves litigation costs. Indeed, since the actual loss is a consider­ ation for the court whether to uphold the clause, the parties might litigate as to the exact magnitude of the actual loss. But that would not be necessary in most cases because as long as the gap between the actual loss and liquidated damages is not large, courts will tend to uphold the clause; on many occasions, the defendant will not even try to argue that the gap is large, and it would not be necessary to litigate about actual losses (Goetz and Scott 1977). 18 

Distributive justice considerations could provide at least a partial explanation for the inconsistency in the law (Porat 2011, 105–​107).

 

326   ARIEL PORAT Second, a liquidated damages clause promotes certainty because, with this clause, the parties can accurately anticipate the amount of damages to be paid by the promisor to the promisee in case of a breach. This advantage is prevalent if the liquidated damages clause sets both a floor and a ceiling, as is the case when the parties have not agreed otherwise. Third, a liquidated damages clause protects interests that otherwise are not adequately protected by contract law. Thus, if the parties anticipate that a breach would result in nonpecuniary losses, or losses that are hard to prove (such as reputational losses), they might incorporate into their contract a liquidated damages clause relating to such losses (Goetz and Scott 1977). Consequently, not only will compensation be secured, but the promisor’s incentives will improve as well.19 Fourth, a liquidated damages clause is a solution to the promisee’s overreliance problem: as has been explained, with liquidated damages, the promisee internalizes both the costs and benefits of his reliance and, therefore, relies efficiently.20 Fifth and last, a liquidated damages clause enables the promisor to signal his credibility to the promisee: by undertaking to pay a large enough amount of damages in case of a breach, he is able to signal to the promisee that the probability of a breach is low (Posner 2011, 160). Thus, a landlord might hesitate to lease his property to a tenant he hardly knows, fearing the latter may damage the property or fail to evacuate it on time. Given that the level of enforcement is lower than 100%, and given litigation costs, the landlord’s entitlement to compensatory damages might not be a satisfactory guarantee for the tenant’s performance, and the landlord will not lease the property to him. If, however, an enforceable liquidated damages clause, stipulating damages in an amount that is much higher than expected losses, is incorporated into the contract, the landlord might be convinced to lease his property to the tenant after all. This advantage, however, is generally unattainable because in order to attain it, the parties must stipulate damages in an amount that is much higher than both expected and actual losses; courts would not enforce such a stipulation and would strike it down as a penalty. Liquidating damages clauses come at a cost: sometimes they might result in inefficiencies. First, they might lead to inefficient breach or inefficient performance (and to inefficient investment in precautions); the former will occur if liquidated damages are lower than actual losses; the latter, if they are higher than actual losses.21 If this were the main disadvantage of liquidated damages clauses, courts’ power to strike them down should have been a default rule. After all, it is for the parties to decide whether the liquidated damages clause should be set aside if it largely deviates from actual losses.

19  However, undercompensating some losses could improve the promisee’s incentives. See Shavell (1979). 20  Supra Section 13.3.3. 21  On the other hand, the promisor who is generally reluctant to breach, even efficiently, for moral reasons, might be willing to breach when damages are stipulated in the contract. Cf. Wilkinson-​Ryan (2010). Thus, liquidated damages clauses might sometimes encourage efficient breaches.

 

ECONOMICS OF REMEDIES    327 Second, if liquidated damages are set too high, the promisee’s expectation for overcompensation might tempt her to induce inefficiently a breach in nonverifiable ways (Clarkson, Miller, and Muris 1978). This risk, however, is not necessarily a reason for courts to intervene in liquidated damages clauses, as long as the parties are assumed aware of such risk when stipulating damages in their contracts. Third, because of lack of information, irrationality, or bounded rationality, the promi­ sor might not be aware of the harsh consequences of a liquidated damages clause, and might agree to set it too high. For example, he might be overly optimistic about his ability to perform the contract, believing he will almost never be subject to the liquidated damages clause. Liquidated damages set too high might encourage the promisor to overinvest in precautions and perform even when a breach is efficient. If this were the main disadvantage of liquidated damages clauses, courts’ power to strike them down should have been limited to cases where asymmetric information or irrationality is a real concern, such as in some consumer contracts. Forth, sometimes, through a liquidated damages clause, the parties might try to externalize costs to third parties by setting damages much higher than what efficiency requires. Thus, suppose that in our previous example, the tenant is under a substantial risk of bankruptcy and the landlord is aware of this. The parties might reach an agreement, setting damages five times higher than anticipated losses, knowing that if the tenant eventually goes bankrupt, part of the costs of the liquidated damages will be borne by his creditors. In such cases, intervention by courts is essential regardless of the parties’ wishes, even if the parties are well informed and fully rational.22

13.6.  Partial Recoveries In this part, two concepts of partial recoveries are discussed: probabilistic recoveries and offsetting risks. The first concept is well known and has been applied by courts in some specific categories of cases, whereas the second is less known and has not been applied by courts so far.

13.6.1. Probabilistic Recoveries In civil cases, the plaintiff succeeds in trial if he proves his case by a preponderance of the evidence. The question that arises is whether in certain cases tort victims should be entitled to a probabilistic recovery if they have failed to prove causation by a preponderance of the evidence. Typical cases where this question emerges are medical malpractice cases when the doctor’s negligence reduced the patient’s chances of recovery and the 22 

A liquidated damages clause that is above expected harm might have anticompetitive effects because it makes the breach for the promisor very costly (Chung 1992; Spier and Whinston 1995).

 

328   ARIEL PORAT patient eventually did not recover (King 1981; Levmore 1990; Porat and Stein 2001, 122–​ 125). For example, suppose that the patient’s chances of recovery were 30%, but because of negligent misdiagnosis by the doctor, those chances were reduced to zero. The probability that the doctor’s negligence caused the patient’s non-​recovery is 30%. Should the patient recover for 30% of his losses because the doctor deprived him of his chances of recovery?23 In some jurisdictions, the answer is yes, while in others the answer is no (Porat and Stein 2001, 74–​76). Does efficiency require compensation in this case? Not necessarily, if we assume that in the specific category of cases doctors handle not only cases of less than 50% chances of recovery, but also cases of more than 50% chances of recovery, and there is symmetry between the two groups of cases. With such symmetry, a doctor’s liability would be the same under both a probabilistic recovery and a preponderance of the evidence rule. To see why, imagine that a patient has either a 30% chance of recovery or a 70% chance of recovery, with equal probabilities. The doctor is negligent and reduces the patient’s chances of recovery to zero. The patient does not recover and suffers harm of H. Under a probabilistic recovery rule, the negligent doctor’s expected liability is 50% × 30% × H + 50% × 70% × H = 50%H.24 Under a preponderance of the evidence rule, the negligent doctor’s expected liability is the same: 50% × 0 + 50% × H = 50%H.25 Liability of 50%H in this example is also efficient because the expected harm of the doctor’s negligence is 50%H.26 Things are different if there is an asymmetry between the two groups of cases. Consider an extreme example where in a certain hospital department all the patients have a less than 50% chance of recovery. Under a preponderance of the evidence rule, those patients would never be entitled to compensation because, in each and every case when harm occurs, the probability that the negligent doctor caused the harm is less than 50%. Therefore, under the latter rule, doctors will not be deterred. In contrast, under the probabilistic recovery rule, patients who suffer harm will always be entitled to compensation from negligent doctors, and the latter’s expected liability will equal expected harm, as required for efficiency. Note that even when there is symmetry between the two groups of cases, probabilistic recovery is essential, as long as the doctor can identify in advance whether her patient has a less than or more than 50% chance of recovery. 23  Suppose that chances of recovery were reduced by the doctor’s negligence from 70% to 40% and the patient has not recovered; under a probabilistic recovery rule the patient should recover 50%—​not 30%—​ of the ultimate harm because the probability that she suffered that harm from the doctor’s negligence is 50% (Porat and Stein 2001, 124; Restatement 3d Torts: Liability for Physical and Emotional Harm, § 26, cmt. n, 2010). 24  There is a probability of 50% that the patient has a 30% chance of recovery, and then liability would be 30%H, and there is a probability of 50% that the patient has a 70% chance of recovery, and then liability would be 70%H. 25  There is a probability of 50% that the patient has a 30% chance of recovery, and then liability would be zero, and there is a probability of 50% that the patient has a 70% chance of recovery, and then liability would be H. 26  There is a probability of 50% that the patient has a 30% chance of recovery, and then expected harm is 30%H, and there is a probability of 50% that the patient has a 70% chance of recovery, and then expected harm is 70%H. Thus, expected harm is 50% × 30% × H + 50% × 70% × H = 50%H.

 

ECONOMICS OF REMEDIES    329 It is important to note that probabilistic recovery is more essential if the typical case is less than 50% chance of recovery rather than more than 50% chance of recovery. When chances are less than 50%, with a preponderance of the evidence rule, underdeterrence will result. When chances are more than 50%, with the same rule, overdeterrence might result, but only if we assume courts’ errors in setting the standard of care and awarding damages or injurers’ errors in anticipating courts’ decisions. (Porat 2011, 112–​114). To see why, suppose that in a certain hospital department all the patients have a 70% chance of recovery, and the harm, if it occurs, is H. Bearing these figures in mind, the patient’s expected harm is 70%H, and the doctor should be required—​efficiency wise—​to take precautions up to 70%H. Would the doctor take higher precautions because if harm occurs, he bears a liability of H? The answer is no: the doctor will take precautions up to 70%H, thereby satisfying the standard of care and bearing no liability. However, the doctor might overcomply if there is a risk of courts’ errors in setting the standard of care and awarding damages or injurers’ errors in anticipating courts’ decisions. Since such risk is common, the probabilistic recovery rule might be superior to a preponderance of the evidence rule, not only in less-​than-​50%-​chance cases, but also in more-​than-​50%-​chance cases. Beyond lost chances of recovery cases, the Market Share Liability doctrine (MSL), which is a form of probabilistic recovery rule, has been applied by courts in some jurisdictions. In one case to which the doctrine was applied, numerous manufacturers produced the same generic drug, which later was found to be defective and harmful. Since in most cases plaintiffs could not tell which manufacturer’s drug caused their harm, a preponderance of the evidence rule allowed all manufacturers in most suits to escape liability. Under MSL, however, each manufacturer was required to bear liability according to his market share at the relevant time with respect to the harmful drug. Thus, if one manufacturer’s market share was 10%, he was required to compensate each and every plaintiff who suffered harm of H, in the amount of 10%H (Porat and Stein 2001, 58–​69). Although MSL does not provide fully efficient incentives to injurers, it is more efficient compared with no liability. Inefficiency might still result because the manufacturers’ behaviours are often nonverifiable, and even with MSL, a manufacturer might refrain from taking costly but efficient precautions because most of the benefits of his precautions would be captured by the other manufacturers (Cooter and Porat 2007).

13.6.2. Offsetting Risks Consider the following example (Porat 2011): Example 4. Choosing between two medical treatments. A doctor must decide between Treatment A and Treatment B for his patient.27 Each treatment entails different risks but produces the same utility if the risks do not materialize. This utility is much 27 

Note that one of the treatments could be an omission, such as not operating on the patient or not administering a certain medicine.

 

330   ARIEL PORAT greater than the respective risks of each treatment. The costs of administering the treatments are the same, and the costs of choosing between them are low. Treatment A entails a risk of 500 to the patient’s left arm (there is a probability of .01 that the treatment will produce harm of 50,000), and Treatment B entails a risk of 400 to the patient’s right arm (there is a probability of .01 that the treatment will produce a harm of 40,000). The risks of Treatments A and B are not correlated: the realization of the risk from one treatment has no bearing on the probability of the realization of the risk from the other treatment. The doctor negligently chooses Treatment A, and a harm of 50,000 materializes. Should the doctor be held liable? If so, in what amount?

Under prevailing tort law, the doctor in Example 4 would be found liable because he was negligent: he could have reduced the total risk to the patient by 100 (500–​400) at a low cost, but failed to do so. The negligent doctor’s liability under prevailing tort law would amount to the entire harm, which is 50,000, because that is the harm caused by his negligence. Thus, while the net risk created by the doctor’s negligence is 100, his expected liability is five times higher:  .01 × 50,000  =  500. The reason for the misalignment between net risk and expected liability is that tort law ignores that the negligent doctor, by choosing treatment A, not only increased the risk to the patient’s left arm (of 500), but also decreased the risk to the patient’s right arm (of 400). If the doctor bears liability for the increased risk (internalizes the negative externalities), without being credited for the decreased risk (does not internalize the positive externalities), his expected liability will be higher, even much higher, than social costs, which are the net rather than the gross risk. Liability for far more than social costs is likely to result in overdeterrence, which in the medical context often takes the form of defensive medicine. Overdeterrence and defensive medicine will result if there are courts’ errors in setting the standard of care and awarding damages or doctors’ errors in anticipating courts’ decisions (nonverifiability of some harms but not others is a typical problem leading to defensive medicine). As doctors pay more in damages, overdeterrence and defensive medicine become more severe (Porat 2007). To restore the alignment between social costs and expected liability, the doctor’s liability in Example 4 should be 10,000 rather than 50,000. Liability of 10,000 would result in this example in expected liability of 100, which is the social cost of the doctor’s negligence. In more general terms, a doctor’s liability should be H(Ra–​Rb)/​Ra, where H stands for the materialized harm, Ra for the risk that was increased by the doctor’s negligence and materialized into harm, and Rb for the risk that was decreased by the doctor’s negligence. The same formula can be applied to other cases where an injurer increased the risks to person A and decreased the risks to person B (Porat 2007) (although such application is expected to raise objections mainly from noneconomic lawyers). In such cases, as long as a restitution claim against person B is legally or practically impossible, there is room for the argument that, efficiency wise, the injurer’s liability towards person A should be decreased in accordance with the risks reduced to person B. A similar

 

ECONOMICS OF REMEDIES    331 argument applies to cases where person B received a certain benefit (as opposed to a probabilistic benefit in terms of decreasing B’s exposure to risks): as long as the benefit is social rather than private, and some other conditions are met, offsetting the benefit from damages awarded to person A would be justified (again, under the assumption that a restitution claim against person B is legally or practically impossible) (Porat and Posner 2014).

References Abraham, Kenneth S. 2011. “Catastrophic Oil Spills and the Problem of Insurance.” Vanderbilt Law Review 64, pp. 1769–​1791. Avraham, Ronen. 2004. “Modular Liability Rules.” International Review of Law and Economics 24, pp. 269–​297. Ayres, Ian. 1996. “Legal Entitlements as Auctions: Property Rules, Liability Rules, and Beyond.” Yale Law Journal 106, pp. 703–​750. Ayres, Ian and Gertner, Robert. 1989. “Filling Gaps in Incomplete Contracts: An Economic Theory of Default Rules.” Yale Law Journal 99, pp. 87–​130. Baron, Jonathan and Wilkinson-​Ryan, Tess. 2009. “Moral Judgment and Moral Heuristics in Breach of Contract.” Journal of Empirical Legal Studies 6, pp. 407–​423. Bebchuk, Lucian A. 2001. “Property Rights and Liability Rules:  The Ex Ante View of the Cathedral.” Michigan Law Review 100, pp. 601–​639. Bigoni, Maria, Bortolotti, Stefania, Parisi, Francesco, and Porat, Ariel. 2014. Unbundling Efficient Breach. Chicago:  University of Chicago Law School. http://​ssrn.com/​ abstract=2477973 [Accessed 22 August 2016]. Bishop, William. 1982. “Economic Loss in Tort.” Oxford Journal of Legal Studies 2, pp. 1–​29. Bovbjerg, Randall R., Sloan, Frank A., and Blumstein, James F. 1989. “Valuing Life and Limb in Tort: Scheduling ‘Pain and Suffering.’” Northwestern University Law Review 83, pp. 908–​976. Calabresi, Guido and Melamed, Douglas A. 1972. “Property Rules, Liability Rules, and Inalienability: One View of the Cathedral.” Harvard Law Review 85, pp. 1089–​1128. Chang, Yun-​chien. 2014. Optional Law in Property:  Theoretical Critiques. http://​ssrn.com/​ abstract=2351651 [Accessed 22 August 2016]. Chung, Tai-​Yeong. 1992. “On the Social Optimality of Liquidated Damages Clauses:  An Economic Analysis.” Journal of Law, Economics, and Organization 8, pp. 280–​305. Clarkson, Kenneth W., Miller, Roger L., and Muris, Timothy J. 1978. “Liquidated Damages v. Penalties: Sense or Nonsense?” Wisconsin Law Review 1978, pp. 351–​390. Coase, Ronald H. 1960. “The Problem of Social Costs.” Journal of Law and Economics 3, pp. 1–​44. Cohen, David and Knetsch, Jack L. 1992. “Judicial Choice and Disparities between Measures of Economic Values.” Osgoode Hall Law Journal 30, pp. 737–​770. Coleman, Jules L. and Kraus, Jody. 1986. “Rethinking the Theory of Legal Rights.” Yale Law Review 95, pp. 1335–​1371. Cooter, Robert D. 1985. “Unity in Tort, Contract, and Property: The Model of Precaution.” California Law Review 73, pp. 1–​51. Cooter, Robert D. and Porat, Ariel. 2002. “Anti-​insurance.” Journal of Legal Studies 31, pp. 203–​232.

 

332   ARIEL PORAT Cooter, Robert D. and Porat, Ariel. 2004. “Decreasing Liability Contracts.” Journal of Legal Studies 33, pp. 157–​197. Cooter, Robert D. and Porat, Ariel. 2007. “Total Liability for Excessive Harm.” Journal of Legal Studies 36, pp. 63–​80. Craswell, Richard. 1988. “Contract Remedies, Renegotiation, and the Theory of Efficient Breach.” Southern California Law Review 61, pp. 629–​670. Croley, Steven P. and Hanson, Jon D. 1995. “The Nonpecuniary Costs of Accidents: Pain-​and-​ Suffering Damages in Tort Law.” Harvard Law Review 108, pp. 1785–​1917. Danzon, Patricia M. 1984. “Tort Reform and the Role of Government in Private Insurance Markets.” Journal of Legal Studies 13, pp. 517–​549. Dari-​Mattiacci, Giuseppe and Schäfer, Hans-​Bernd. 2007. “The Core of Pure Economic Loss.” International Review of Law and Economics 27, pp. 8–​28. Dillbary, Shahar. 2013. “Tortfest.” University of Chicago Law Review 80, pp. 953–​1005. Dobbs, Dan B. 2000. The Law of Torts. St. Paul, MN: West Group. Epstein, Richard A. 1989. “Beyond Foreseeability:  Consequential Damages in the Law of Contract.” Journal of Legal Studies 18, pp. 105–​138. Farnsworth, Allan E. 2004. Contracts. 4d ed. New York: Aspen Law & Business. Friedman, David D. 1982. “What is ‘Fair Compensation’ for Death Injury?” International Review of Law and Economics 2, pp. 81–​93. Goetz, Charles J. and Scott, Robert E. 1977. “Liquidated Damages, Penalties, and the Just Compensation Principle: Some Notes on an Enforcement Model and a Theory of Efficient Breach.” Columbia Law Review 77, pp. 554–​594. Goetz, Charles J. and Scott, Robert E. 1983. “The Mitigation Principle:  Toward a General Theory of Contractual Obligation.” Virginia Law Review 69, pp. 967–​1024. King, Joseph H. 1981. “Causation, Valuation, and Chance in Personal Injury Torts Involving Preexisting Conditions and Future Consequences.” Yale Law Journal 90, pp. 1353–​1397. Kronman, Anthony T. 1978. “Specific Performance.” University of Chicago Law Review 45, pp. 351–​382. Landes, William M. and Posner, Richard A. 1987. The Economic Structure of Tort Law. Cambridge, MA: Harvard University Press. Levmore, Saul. 1990. “Probabilistic Recoveries, Restitution, and Recurring Wrongs.” Journal of Legal Studies 19, pp. 691–​726. Luppi, Barbara and Parisi, Francesco. 2011. “Toward an Asymmetric Coase Theorem.” European Journal of Law and Economics 31, pp. 111–​122. Markovits, Daniel and Schwartz, Alan. 2011. “The Myth of Efficient Breach: New Defenses of the Expectation Interest.” Virginia Law Review 97, pp. 1939–​2008. Markovits, Daniel and Schwartz, Alan. 2012. “The Expectation Remedy Revisited.” Virginia Law Review 98, pp. 1093–​1107. Porat, Ariel. 2007. “Offsetting Risks.” Michigan Law Review 106, pp. 243–​276. Porat, Ariel. 2009. “Private Production of Public Goods: Liability for Unrequested Benefits.” Michigan Law Review 108, pp. 189–​227. Porat, Ariel. 2011. “Misalignments in Tort Law.” Yale Law Journal 121, pp. 82–​141. Porat, Ariel and Posner, Eric. 2014. “Offsetting Benefits.” Virginia Law Review 100, pp. 1165-​1209. Porat, Ariel and Stein, Alex. 2001. Tort Liability under Uncertainty. Oxford:  Oxford University Press.

 

ECONOMICS OF REMEDIES    333 Porat, Ariel and Tabbach, Avraham. 2011. “Willingness to Pay, Death, Wealth, and Damages.” American Law and Economics Review 13, pp. 45–​102. Posner, Richard A. 1979. “Some Uses and Abuses of Economics in Law.” University of Chicago Law Review 46, pp. 281–​306. Posner, Richard A. 2009. “Let Us Never Blame A Contract Breaker.” Michigan Law Review 107, pp. 1349–​1363. Posner, Richard A. 2011. Economic Analysis of Law. 8th ed. New York: Aspen. Rea, Samuel A., Jr. 1982. “Nonpecuniary Loss and Breach of Contract.” Journal of Legal Studies 11, pp. 35–​53. Rizzo, Mario J. 1982a. “A Theory of Economic Loss in the Law of Torts.” Journal of Legal Studies 11, pp. 281–​310. Rizzo, Mario J. 1982b. “The Economic Loss Problem: A Comment on Bishop.” Oxford Journal of Legal Studies 2, pp. 197–​206. Schwartz, Alan. 1979. “The Case for Specific Performance.” Yale Law Journal 89, pp. 271–​306. Schwartz, Alan. 1990. “The Myth that Promisees Prefer Supercompensatory Remedies:  An Analysis of Contracting for Damages Measures.” Yale Law Journal 100, pp. 369–​407. Shavell, Steven. 1979. “On Moral Hazard and Insurance.” Quarterly Journal of Economics 93, pp. 541–​562. Shavell, Steven. 1987. Economic Analysis of Accident Law. Cambridge, MA:  Harvard University Press. Shavell, Steven. 2009. “Why Breach of Contract May Not Be Immoral Given the Incompleteness of Contracts.” Michigan Law Review 107, pp. 1569–​1581. Shiffrin, Seana V. 2009. “Could Breach of Contract be Immoral?” Michigan Law Review 107, pp. 1551–​1568. Spier, Kathryn E. and Whinston, Michael D. 1995. “On the Efficiency of Privately Stipulated Damages for Breach of Contract: Entry Barriers, Reliance and Renegotiation.” Rand Journal of Economics 26, pp. 180–​202. Ulen, Thomas S. 1984. “The Efficiency of Specific Performance: Toward a Unified Theory of Contract Remedies.” Michigan Law Review 83, pp. 341–​403. Wilkinson-​Ryan, Tess. 2010. “Do Liquidated Damages Encourage Breach? A  Psychological Experiment.” Michigan Law Review 108, pp. 633–​671. Zamir, Eyal and Medina, Barak. 2010. Law, Economics, and Morality. Oxford:  Oxford University Press. Zamir, Eyal and Ritov, Ilana. 2010. “Revisiting the Debate over Attorney’s Contingent Fees: A Behavioral Analysis.” Journal of Legal Studies 39, pp. 245–​288. Zipursky, Benjamin. 1998. “Rights, Wrongs, and Recourse in the Law of Torts.” Vanderbilt Law Review 51, pp. 1–​100.

 

 

Pa rt  I I

C OR P OR AT E , C OM M E RC IA L , A N D E N V I RON M E N TA L L AW

 

 

Chapter 14

THE EC ONOM IC NAT U RE OF THE C ORP ORAT I ON Lynn Stout

14.1. Introduction Corporations are among the most powerful and economically important institutions in modern society. This chapter examines the nature and structure of corporate entities (a category that includes not only business corporations but also nonprofits and many limited liability companies [LLCs]). It discusses the basic characteristics of corporate entities; surveys different models or theories of the corporation; and explores the question of corporate purpose. The chapter pays special attention to how corporate law allocates economic and control rights to, and among, both the corporate entity itself and the natural persons involved with corporations, such as shareholders, directors, and executive officers.

14.1.1. The Corporation versus the Firm As a preliminary matter, it is essential to note that while the words corporation and firm are often used interchangeably, they refer to very different things. The careless but unfortunately common habit of treating them as synonyms confuses and misleads (Robe 2011). Firm is an economic concept that refers to the organization of economic activity involving more than one person, outside of formal markets. By contrast, corporation is a legal concept that carries important economic consequences. In particular, a corporation is a specific pattern of formal rights and responsibilities created by law and distributed between and among both human persons, and a legislatively created, artificial legal person (the corporate entity).

 

338   LYNN STOUT It is possible to create a firm that is not a corporation. For example, a law firm or an accounting firm organized as a partnership is still (as the name implies) a firm in the economic sense. Similarly, a sole proprietor who hires employees has created a firm, but has not created a corporation. In theory, it is equally possible to create a corporation that is not a firm. Modern corporate codes typically allow a single human person to create a corporate entity, contribute all of its capital, and then serve as that entity’s sole shareholder, director, officer, and employee. Such a corporate entity is just that—​a corporate entity—​but it is not a firm. The existence of a firm is neither necessary nor sufficient for a corporation to exist. A theory of the firm is not a theory of the corporation.

14.1.2. Corporations and Agency Cost Analysis Because the firm and the corporation are so frequently confused, economic analysis of corporations often focuses, almost exclusively, on the so-​called agency cost problem (see, e.g., Ciepley 2013). However, agency costs are an issue whenever economic activity is organized in firms, not only (or always) in corporations. Agency costs also arise in partnerships and indeed in any project involving more than one person. Because most corporate act