Predatory pricing in antitrust law and economics a historical perspective

§1 Three basic dichotomies 1 §2 The trickiest antitrust problem 3 §3 Further reasons to love predation 5 §4 Lessons in persuasion 8 §5 Treasures in the attic 9 §6 Plan of the book 11 §1

Predatory Pricing in Antitrust Law

and Economics

Can a price ever be too low? Can competition ever be ruinous? Questions like these have always accompanied American antitrust law They testify to the diffi-culty of antitrust enforcement, of protecting competition without protecting competitors

As the business practice that most directly raises these kinds of questions, predatory pricing is at the core of antitrust debates The history of its law and economics offers a privileged standpoint for assessing the broader development

of antitrust, its past, present and future In contrast to existing literature, this book adopts the perspective of the history of economic thought to tell this history, covering a period from the late 1880s to present times

The image of a big firm, such as Rockefeller’s Standard Oil or Duke’s ican Tobacco, crushing its small rivals by underselling them is iconic in Amer-ican antitrust culture It is no surprise that the most brilliant legal and economic minds of the last 130 years have been engaged in solving the predatory pricing puzzle The book shows economic theories that build rigorous stories explaining when predatory pricing may be rational, what welfare harm it may cause and how the law may fight it Among these narratives, a special place belongs to the Chicago story, according to which predatory pricing is never profitable and every low price is always a good price

Amer-Nicola Giocoli is an Associate Professor of Economics at the University of Pisa,

Italy

Sponsored by Michigan State University College of Law

Series Editors:

Nicholas Mercuro

Michigan State University College of Law

Michael D Kaplowitz

Michigan State University

1 Compensation for Regulatory Takings

Thomas J Miceli and Kathleen Segerson

2 Dispute Resolution

Bridging the settlement gap

Edited by David A Anderson

3 The Law and Economics of Development

Edited by Edgardo Buscaglia, William Ratliff and Robert Cooter

4 Fundamental Interrelationships Between Government and Property

Edited by Nicholas Mercuro and Warren J Samuels

5 Property Rights, Economics, and the Environment

Edited by Michael Kaplowitz

6 Law and Economics in Civil Law Countries

Edited by Thierry Kirat and Bruno Deffains

7 The End of Natural Monopoly

Deregulation and competition in the electric power industry

Edited by Peter Z Grossman and Daniel H Cole

8 Just Exchange

A theory of contract

F H Buckley

9 Network Access, Regulation and Antitrust

Edited by Diana L Moss

10 Property Rights Dynamics

A law and economics perspective

Edited by Donatella Porrini and Giovanni Ramello

Implications for economics, accounting and the law

Edited by Yuri Biondi, Arnaldo Canziani and Thierry Kirat

12 The Legal- Economic Nexus

Warren J Samuels

13 Economics, Law and Individual Rights

Edited by Hugo M Mialon and Paul H Rubin

14 Alternative Institutional Structures

Evolution and impact

Edited by Sandra S Batie and Nicholas Mercuro

15 Patent Policy

Effects in a national and international framework

Pia Weiss

16 The Applied Law and Economics of Public Procurement

Edited by Gustavo Piga and Steen Treumer

17 Economics and Regulation in China

Edited by Michael Faure and Guangdong Xu

18 Law, Bubbles and Financial Regulation

Erik F Gerding

19 Empirical Legal Analysis

Assessing the performance of legal institutions

Edited by Yun- chien Chang

20 Predatory Pricing in Antitrust Law and Economics

A historical perspective

Nicola Giocoli

* The first three volumes listed above are published by and available from Elsevier

Predatory Pricing in Antitrust Law and Economics

2 Park Square, Milton Park, Abingdon, Oxon OX14 4RN

and by Routledge

711 Third Avenue, New York, NY 10017

Routledge is an imprint of the Taylor & Francis Group, an informa business

© 2014 Nicola Giocoli

The right of Nicola Giocoli to be identified as author of this work has been asserted by him in accordance with sections 77 and 78 of the Copyright, Designs and Patents Act 1988.

All rights reserved No part of this book may be reprinted or reproduced or utilized in any form or by any electronic, mechanical, or other means, now known or hereafter invented, including photocopying and recording, or in any information storage or retrieval system, without permission in writing from the publishers.

Trademark notice: Product or corporate names may be trademarks or

registered trademarks, and are used only for identification and explanation without intent to infringe.

British Library Cataloguing in Publication Data

A catalogue record for this book is available from the British Library

Library of Congress Cataloging in Publication Data

A catalogue record has been requested for this book.

ISBN: 978-0-415-82252-7 (hbk)

ISBN: 978-1-315-83269-2 (ebk)

Typeset in Times New Roman

by Wearset Ltd, Boldon, Tyne and Wear

Vincenzo

times in danger of forgetting that the conditions of the struggle fix the kind of fitness that shall come out of it; that survival in the prize ring means fitness for pugilism; not for bricklaying nor philanthropy; that survival in predatory com- petition is likely to mean something else than fitness for good and efficient pro- duction; and that only from a strife with the right kind of rules can the right kind

of fitness emerge Competition and its purpose are not individual but social It is

a game played under rules fixed by the state to the end that, so far as possible, the prize of victory shall be earned, not by trickery or mere self- seeking adroitness, but by value rendered It is not the mere play of unrestrained self- interest; it is a method of harnessing the wild beast of self- interest to serve the common good –

a thing of ideals and not of sordidness It is not a natural state, but like any other form of liberty, it is a social achievement, and eternal vigilance is the price of it.

(Clark and Clark [1912] 1914, 200–1) Highly speculative belief about behavior or its consequences does not satisfy [the legal] standard, even when endorsed by expert economic witnesses.

(Demsetz 1992, 209–10)

It would be indeed an extraordinary thing to strike at competition in the name of competition.

(Macrosty 1907, 345)

§1 Three basic dichotomies 1

§2 The trickiest antitrust problem 3

§3 Further reasons to love predation 5

§4 Lessons in persuasion 8

§5 Treasures in the attic 9

§6 Plan of the book 11

§1  Classic and modern definitions of predatory pricing  13

§2 The basic story 19

§3 The Chicago critique of the basic story 22

§4 It’s a brand new game: predation as strategic

paradox 25

§5 The Stanford connection 28

§6 Madamina, il catalogo è questo 35

§7 Assessing the Bayesian approach to predation 42

§1 The two freedoms 49

§2 The monopoly problem in British common law 50

§3 The classical view of competition 54

§4 Competition in the late nineteenth- century British

economy 58

§5 The dawn of predatory pricing: the Mogul case 60

§6 The Mogul decisions: is predation “a matter contrary to

law”? 62

§7 The new reasonableness test: the Nordenfelt case 66

§8 The legacy of Mogul and Nordenfelt 69

§9 Restraints of trade in American common law 73

3 American economists and destructive competition 81

§1 Monopoly as the inevitable outcome of competition 81

§2 “Let us have peace”: the combination way- out 83

§3 From destructive competition to predatory pricing 86

§4 Economic power and the curse of bigness 88

§5 Playing the trump card: potential competition 90

§6 From Clark to Clarks 94

4 Predatory pricing in the formative era of antitrust law 103

§1 Constitutionalizing freedom of contract 103

§2 The two views in action: the Sherman Act’s Congressional

debate 105

§3 Transcending common law: monopolizing and third- party

actionability 109

§4 The economists’ reaction to the Sherman Act 110

§5 Common law, literalism and reasonableness 112

§6 The predatory side of the 1911 cases 120

§7 The economists’ reasonable dissent 125

§8 The Clayton and FTC Acts 127

§9 Predatory pricing in the formative era: an assessment 130

§1 The decades of neglect (1918–35) 135

§2 Competition strikes back: the end of associationalism 138

§3 The structuralist paradigm 140

§4 Mason’s SCP manifesto 142

§5 Extreme structuralism versus workable competition 145

§6 A “new” Sherman Act? 149

§7 The return of Old Sherman 156

§8 The horror list 160

§9 Intent to exclude intent 163

§10 The worst antitrust decision ever? 166

§11 Conclusion: the divorce between antitrust and

microeconomics 169

6 The Chicago School and the irrelevance of predation 176

§1 The dissolution proposals 176

§2 Chicago to the rescue 179

§3 The two Chicagos 181

§4 Listen to McGee: predation doesn’t exist! 187

§5 Chicago’s peculiar methodology 191

§6 Three Chicago boys 193

§7 Conclusion: a new Chicago story 202

§1 Two reactions to McGee 210

§2 Strategic predation without game theory 211

§3 From the “wilds of economic theory” 215

§4 to a “meaningful and workable” rule 217

§5 A new legal standard 220

§6 The courts’ reaction to the ATR 223

§7 The economists’ reaction to the ATR 226

§8 The post- ATR debate in courts 235

§9 Conclusion: lessons from the ATR saga 237

8 The demise of predatory pricing as an antitrust violation 245

§1 Mr Justice goes to Chicago 245

§2 Predatory pricing case law meets Chicago antitrust 248

§3 Predatory pricing’s last cigarette 253

§4 The Brooke test – Chicago creed or apostasy? 255

§1 Star Wars without Darth Vader 278

§2 It’s the ideology, stupid! 280

§3 Games judges don’t play 284

§4 Chicago rule(s) 285

This volume is part of a broader research project on the history of American antitrust law and economics Along the years I have benefited from the com-ments and help of several colleagues Without involving them in any responsib-ility for remaining mistakes, I wish to thank Robert Albon, David Andrews, Jeff Biddle, Ivars Brivers, Chris Colvin, Carlo Cristiano, Marco Dardi, Luca Fiorito, Salar Ghahramani, Francesco Guala, Dan Hammond, Herrade Igersheim, Bruna Ingrao, Andrew Jewett, Albert Jolink, Robin Paul Malloy, Alain Marciano, Steve Medema, Maurizio Mistri, Ivan Moscati, Lorenzo Pace, Sylvie Rivot, Rodolfo Signorino, Rob Van Horn, Joshua Wright, Alberto Zanni and the editors and anonymous referees of the journals where parts of the book have been published before I am especially grateful to Tony Freyer (our Harvard meeting shows it’s

a small world – really!), Robert Lande, Andrea Maneschi, Henry Manne, Robert

T Masson and Jim Rhodes, who gave me additional suggestions and valuable historical insight

Portions of the book have appeared elsewhere Chapter 2 follows in part Giocoli (2013a); sections of Chapters 6 and 7 are based on Giocoli (2011); Chapter 8 and Conclusion draw on Giocoli (2013b) I am grateful to the editors

and publishers of, respectively, the Research in the History of Economic Thought and Methodology (Emerald Group Publishing), the European Journal of the History of Economic Thought (Routledge, Taylor & Francis Group) and the Supreme Court Economic Review (University of Chicago Press) for granting

permission to reproduce parts of these papers

Librarians at the Historical and Special Collections of Harvard Law School Library were extremely helpful in assisting me while doing archival research on

Areeda, Phillip E., 1930–1995 Papers, 1927–1995 I thank them all – in

par-ticular Leslie Schoenfeld, whose kindness and smile did a lot to warm the sphere in the HSC room (it does need warming, trust me) I am also grateful to

atmo-Ed Moloy for granting permission to quote excerpts from Areeda’s collection The Routledge editorial team, past and present, gave me assistance and encouragement I thank Rob Langham, Emily Kindleysides, Simon Holt and Andy Humphrys

Last, but not least, my collaborators Domenico Fanelli, Tiziana Foresti, and, especially, Francesco Cattabrini offered precious research and teaching

assistantship My deepest gratitude to all of them Simon Cook did an incredible job in polishing my English – and beyond I have no words to express my admi-ration for someone who combines editorial skills, historical knowledge, good

humor and enduring patience at such a high level He trespassed the duties which belong to an editorial assistant (sorry, Simon, I couldn’t resist).

My research benefited from generous financial support by the INET (Institute for New Economic Thinking) INET grant #5190, awarded to the project “Free from what? Evolving notions of ‘market freedom’ in the history and con-temporary practice of US antitrust law and economics,” is gratefully acknowledged

The book is dedicated to the loving memory of my parents, Benedetta and Vincenzo, my aunts, Maria and Rita, and my uncle Enzo, for what they taught and gave me

1 Three basic dichotomies

Antitrust law is about competition It aims at guaranteeing economic agents’ freedom to compete as the best way to promote maximum social welfare Simple

as they may seem, these statements are far from undisputed or self- explanatory Controversy about the meaning and goal of antitrust law is as old as the law itself – and even older, as it dates back to mid- nineteenth-century debates over the British common law restraints of trade

One may legitimately ask what “competition,” “freedom to compete” and

“social welfare” exactly mean Recognizing them as technical terms does not help, because two technical and only partially overlapping jargons apply in the field of antitrust: the language of law and the language of economics The eco-nomic point of view has been on the rise during the last three decades of antitrust law, but still the subject belongs to the legal realm Antitrust enforcement is part

of the legal (and administrative) system; acquainted as they may be with nomics, judges and agencies adjudicate cases following legal rules

Even within the boundaries of economics, unsettled issues exist For example, economists have different ideas of what “competition” actually means What is

competition? Two main characterizations prevail.1 In the first, competition is

viewed as a process, the product of the actions and reactions of sellers and

buyers bargaining in the marketplace It is a force operating in the market that does not coincide with any given market structure Alternatively, competition

may be characterized as a state; that is, as a specific market structure, endowed

with certain desirable properties relating to equilibrium output and price cally speaking, the former view was typical of nineteenth- century classical econ-omists, the latter of twentieth- century neoclassical ones However, in the world

Histori-of antitrust both views have always co- existed

The same can be said of the notion of “freedom to compete.” When is petition free? Once again, two interpretations exist.2 According to the first, com-

com-petition is free when market participants may exercise the utmost freedom of contract – that is to say, when they have unlimited access to every possible

exchange opportunity and the full management of their property rights In the

second characterization, free competition means freedom from market power, or

freedom to trade Loosely speaking, market power may be defined as a market

participant’s capability of increasing her own surplus by constraining or cing other agents’ exchange opportunities or trade In a freely competitive market this power is kept at an insignificant level by the force of competition itself; all agents are therefore on equal footing with respect to the possibility of exploiting trade opportunities (though some agents may still have more oppor-tunities than others)

The two interpretations are obviously related, but do not coincide The key difference resides in the dichotomy between market and non- market coercion The government and the law are the main non- market sources of constraints on freedom of contract, in that they set boundaries to what an economic agent may

do with her own property rights For example, the law may prohibit a merger – that is, the free exchange of property – between two businesses Hence, freedom

of contract may also be characterized as freedom from government or legal cion Together with the right to enjoy the fruits of one’s own work, this was the basic kind of economic freedom for classical liberals at the time the first antitrust statutes were enacted Freedom to trade is on the contrary constrained first and foremost by other agents’ market power This may take the form of, say, supra- competitive prices, denial of access to essential inputs or territorial limitations Apart from the idealized situation of perfect competition, where no such power exists by definition, what causes problems is not market power per se (as it may

coer-be the legitimate fruit of superior talent and ability), but rather the way it is employed to constrain or coerce other agents’ trade

The ideal of freedom from market power is intimately related to neoclassical economics and its idea of competition as a state Market power itself is a techni-cal notion that economists measure in terms of price/cost margins and market shares However, it also one that easily lends itself to a non- analytical extension

In the history of antitrust law, freedom from market power has often been preted as freedom from economic power By the latter term is meant a kind of

inter-power that trespasses upon the market and spreads its negative influence over the whole society A powerful business in this sense is one capable of affecting not only a country’s economy, but also its politics and social life; a threat to demo-cracy in its most basic nature of a system based on equal rights and duties Anti-trust law has a long tradition of looking suspiciously at large concentrations of economic power, usually summarized in terms of sheer business size Simplistic

as it may be, the idea that big is bad has been a driving force for much of the subject’s history

Economic power, as distinguished from market power, is also the reason why even the notion of “social welfare” is problematic Analytically speaking, the notion may be interpreted in purely economic terms, such as allocative efficiency

or total surplus Even broadening the analysis to encompass a dynamic setting so

as to accommodate the long- run efficiencies generated by, say, product tion leaves the basic theoretical framework unchanged Setting social welfare as the goal of antitrust law thus makes antitrust itself a branch of economic policy that must be governed by economic analysis All other concerns, such as

innova-fairness, the plight of small businesses or, crucially, the socio- political sequences of unbalanced economic power, become irrelevant This is how modern antitrust usually proceeds However, when and if economic power is viewed as the main foe, as it has often been throughout history, then the goal of antitrust changes It becomes, broadly speaking, the pursuit of marketplace egal-itarianism, of a Jeffersonian ideal of an economy of “small dealers and worthy men,”3 none of whom are capable of coercing anyone else and, therefore, of neg-atively affecting a country’s socio- political life Social welfare then takes a broader, less rigorous meaning that transcends economics and even the law, in that neither discipline may properly account for loose notions such as market-place fairness or the protection of small businesses.

These considerations should suffice to reveal how difficult – and fascinating –

a topic the law and economics of antitrust may actually be Even the simplest of statements, such as “antitrust law is about competition” or “antitrust promotes free competition,” conceal a universe of interpretive problems The present book

is an effort to analyze the above- mentioned dichotomic views of “competition,”

“freedom to compete” and “social welfare” from the vantage point of the history

of the economic analysis of antitrust Three general results will emerge: first, that these dichotomies have characterized the entire history of antitrust law; second, that the courts’ evolving attitude towards them has largely determined the way antitrust law has been enforced over the years; third, that the influence of theor-etical economics upon this attitude has been anything but steady, as antitrust courts have oscillated between a total neglect and a partial or full endorsement

of basic economic principles

My analysis will focus on a single antitrust issue and a single country I will deal with the history of the law and economics of predatory pricing in American antitrust The geographical focus hardly requires explanation In every respect, the United States is the cradle of antitrust, the country where the discipline first became a serious matter and where the relationship between the legal and the economic sides of competition has been most intensively studied But why pred-atory pricing? Why should the history of the law and economics of this specific violation of antitrust statutes merit a book of its own? Is there anything special about predatory behavior? And, in any case, why adopt a history of economic thought viewpoint? Are there any lessons the historical method may teach current antitrust enforcers? The remainder of the Introduction tries to answer these legitimate questions

2 The trickiest antitrust problem

Predatory pricing (PP hereafter) is an unlawful business behavior within the broader category of unlawful exclusionary practices A practice is exclusionary when it is

reasonably capable of creating, enlarging, or prolonging monopoly power by limiting the opportunities of rivals [and] either does not benefit consumers at

all, or is unnecessary for the particular consumer benefits produced, or duces harms seriously disproportionate to the resulting benefits.

pro-(Hovenkamp 2005, 152)

A predatory price is then “a price that is profit- maximizing only because of its

exclusionary or other anti- competitive effects” (Bolton et al 2000, 2242–3) In

other words, PP takes place when a firm sets such a low price that its only ale is to damage competitors, current or potential The predator’s eventual goal

ration-is to increase its market power and charge a higher price in the future, after petition has been either disciplined or destroyed

Historically speaking, PP surely ranks high in the catalogue of antitrust tions – at least as high as cartelization And as with cartels, predatory behavior has always symbolized what antitrust law is supposed to fight against The iconic picture of a big business preying upon its smaller rivals by setting the price so low that none of them could survive is second to none in the history of antitrust law, the only possible exception being the image of smoke- filled backrooms in which businessmen secretly agree to fix prices Opposition to predatory behavior was the primary motivating force of the American public opinion’s hostility towards trusts and, therefore, one of the key impulses that led Congress to pass the 1890 Sherman Act

There is more to PP than mere history, though The antitrust problem with cartels is easy to apprehend Cartels are clearly anti- competitive – indeed, their very goal is to avoid competition While even joining a cartel may be seen as an expression of a member’s own freedom of contract as much as the negation of nonmembers’ freedom to trade, the fact remains that the harmful welfare effects

of cartelization are well established and largely undisputed Nobody doubts that fighting cartels means fostering competition This is not so in the case of PP Predatory behavior is the paradigmatic example of using competition to destroy competition – what several American economists at the turn of the twentieth

century called destructive competition We know that defending the utmost

freedom to compete is supposed to be the core of antitrust Mind- boggling tions thus arise Whose freedom to compete deserves antitrust protection – the

ques-predator’s or the prey’s? Can a firm ever be condemned for competing too much? Can a price ever be too low? In short, can competition really be destruc-

tive? Questions like these make anti- PP enforcement one of the trickiest, if not

the trickiest part of antitrust law.

Enforcing law against PP means prohibiting competition to foster tion The danger is clear The law risks discouraging actual competition as freedom of contract for the sake of protecting an abstract ideal of competition as freedom from market power The age- old proscription of predatory behavior indeed reflects a specific policy choice, namely, the idea that using the law to curb a firm’s freedom to set its own price (i.e., to constrain that firm’s contrac-tual freedom) may nonetheless foster freedom to trade, i.e., avoid undue concen-trations of market or economic power This policy choice is, however, far from undisputed

The trade- off between the two freedoms, and between the conflicting views

of what free competition actually means, is therefore intrinsic to the law and nomics of PP – and the chief reason why the subject deserves special scrutiny When still a Judge for the First Circuit, Justice Stephen Breyer once described the trade- off in the following terms:

eco-A price cut that ends up with a price exceeding total cost – in all likelihood a cut made by a firm with market power – is almost certainly moving price in the “right” direction (towards the level that would be set in a competitive marketplace) The antitrust laws very rarely reject such beneficial “birds in hand” for the sake of more speculative (future low price) “birds in the bush.”4

The metaphor captures the power of PP to lay bare the most controversial tures and deepest contradictions of antitrust law “Using competition to destroy competition” and “prohibiting competition to foster competition” are challeng-ing statements that push to the extreme the antitrust project’s goal and method The law and economics of PP is the only framework where these statements really make sense – and, therefore, also the best diagnostic tool for inquiry into the very heart of antitrust

fea-3 Further reasons to love predation

That a low price may not necessarily be a good price is the chief reason for taking PP law and economics as the most exciting vessel for sailing the turbulent waters of antitrust law It is not the only one, though A few more are listed in this section and the next

3.1 The short- run/long- run tradeoff

Breyer’s “birds” highlight another crucial tradeoff Every business practice has short- and long- run effects on social welfare.5 In most cases, economic theory shows that these effects point in opposite directions PP is exemplary in this respect: consumers gain during predation, when prices are low, but suffer after predation is over, when (and if ) the successful predator raises its price to a level higher than before Other instances of the tradeoff are patents, efficiency- driven mergers and essential facilities

When an antitrust case involves a business practice characterized by the short- run/long- run tradeoff, the court may reach a decision only by attributing weights to the practice’s immediate and future welfare effects The choice of weights may seem theoretically driven, but it is not Economics often fails to provide objective grounds for the choice – by, say, proving that short- run effects always (or at least in this specific case) outweigh long- run ones, or vice versa

Assigning weights is frequently a normative operation even for the least

ideologically oriented court It is a peculiar, one- dimensional kind of ness, as it depends only on the court’s preference for short- over long- run

normative-consequences Still a normative judgment it is, one that analytical arguments cannot fully validate.

Not only is PP exemplary of the short- run/long- run tradeoff The history of its law and economics also highlights the inevitably normative nature of the courts’ choices in the face of that tradeoff For a good part of the twentieth century American courts emphasized predation’s negative long- run effects over its pos-itive short- run ones The consequence was a strict enforcement of the per se pro-hibition of predatory behavior More recent PP case law has reversed this attitude, with courts giving decisive weight to the short- run gains guaranteed by any price cut, regardless of its possible strategic motivation and negative long- run effects The reversal mirrors the greater confidence of most contemporary courts in the spontaneous ability of free markets to deliver their efficient out-comes in the long run, without any specific intervention by the law or the state Hence, it is believed that even successful predators will not enjoy their victory long because new competition will surely, and quickly, erode any market power they may have conquered

This belief is clearly a normative judgment that finds no analytical tion in modern economics, exception being made for the most idealized version

justifica-of perfect competition It is a judgment that is typical justifica-of many chapters justifica-of current antitrust case law besides PP – but it is also one the true nature of which

only PP reveals with both clarity and immediacy.

3.2 The predatory paradigm for exclusionary acts

The previous definition of exclusionary acts is not the only one Narrower tions exist that characterize more specifically under what conditions a business practice may be deemed exclusionary Their goal is operational, as they tend to

defini-be formulated in terms of a test that may help courts to identify violations of the anti- monopolization provision of antitrust law – that is, section 2 (§2) of the Sherman Act

According to Herbert Hovenkamp (2008, 114), the recent anti- monopolization literature “has been preoccupied to the point of obsession with the formulation

of a single test for exclusionary conduct.” In the so- called sacrifice test, a firm’s conduct violates §2 when it entails giving up some immediate profits as part of a strategy whose profitability strictly depends on the exclusion of rivals For example, setting a low price today means sacrificing short- run revenues in view

of the benefits that will accrue tomorrow thanks to the high price that follows the creation of a monopoly or the strengthening of a dominant position PP thus fully partakes of the rationale for the sacrifice test, which aims at sanctioning genu-inely competitive conduct, i.e., conduct that is profitable without regard to the creation or preservation of monopoly The sacrifice test is implicit in the way contemporary antitrust courts handle PP cases

A slightly different alternative is the “no economic sense” test, which demns conduct that would be irrational (i.e., would make no economic sense) except when used as a mechanism for excluding rivals and earning monopoly

con-profits The test – which is also frequently used by modern courts – entails that

no single- firm practice should be condemned per se, but only if its sole rational explanation is that the firm used it to eliminate or lessen competition Note that, regardless of their weaknesses,6 both tests are highly operational No calculation

of the given conduct’s net welfare effects is required because the mere existence

of a defendant’s gain that does not come from injuring competitors would suffice

to sanction that conduct

Georgetown economist Steven Salop has classified popular tests like the

sac-rifice or the “no economic sense” under the banner of the predatory paradigm

for exclusionary acts.7 The paradigm is built on the idea that predatory behavior epitomizes §2 violations All kinds of exclusionary practices should be reduced within the boundaries of PP, as they all partake of its key principle of suffering losses now to earn more tomorrow Given the current very lenient enforcement

of anti- PP law, the paradigm’s implication is obvious If all exclusionary tices are like PP, then it is very likely that, exactly like PP, they should cause little, if any, antitrust concern

The predatory paradigm thus explains why current §2 enforcement is based upon highly permissive tests that would show violation in only a very few cases

It is not just that the sacrifice or “no economic sense” tests are ill- devised The real issue is that the trickiest of all antitrust violations, price predation, has been taken as the foundation for all kinds of exclusionary behavior Studying the law and economics of PP may allow a better understanding of the paradigm’s limits and help in the devising of more effective anti- monopolization tests

3.3 Big is bad

That big business is bad business is an idea that predates antitrust law and has accompanied it throughout its existence To quote from a famous case, the fear that “the vast accumulation of wealth in the hands of corporations and individu-als [ .] and their power had been and would be exerted to oppress individuals and injure the public generally”8 has been a major driver of antitrust legislation and enforcement

It is not by chance that the previous passage came from Standard Oil That

decision by the 1911 Supreme Court established the key precedent for more than seven decades of anti- PP case law Rockefeller’s trust symbolized the “big is bad” mantra in turn- of-the- century American culture, where the popular press depicted Standard Oil as a giant octopus crushing smaller rivals to death The privileged way a trust could “oppress individuals and injure the public generally” was, of course, by pricing at a predatory level, or by under-taking similar exclusionary acts Hence, PP became the leading example of everything that might be wrong with business size The 1911 decision simply acknowledged that

The point is that American antitrust law did not condemn size per se – and would never do so even in later decades So- called “no fault” structural remedies, which called for the dissolution of giant corporations and trusts on account of

their mere size, never reached beyond the status of legislative proposals or demic debate Moreover, despite decades of intense controversy, economic ana-lysis has never offered a clear rationale for condemning size as such Still, fear of the negative consequences that “the vast accumulation of wealth” might cause the economy and, therefore, also American society never disappeared from the anti-trust landscape How to reconcile the dread of economic power with the absence

aca-of any legal or analytical justification for condemning business size per se? Once again, PP came to the rescue If predatory behavior symbolized the evil

of giant business, then what courts had to do was “just” search for evidence

about that very behavior The idea was simply that if a business is big, it must

have preyed, or still be preying, upon its rivals PP thus became a proxy – an excuse, if you wish – to condemn size.9 This does not mean that Standard Oil or other large businesses that, over the decades, have been found guilty of preda-tory behavior were actually innocent Almost surely they were not It is just to

recognize the role that PP has played in supporting the antitrust fight against nomic, as distinct from merely market, power As we said above, this was an

eco-eminently socio- political fight, aimed at preserving no less than American cracy, not just marketplace efficiency The patterns of anti- PP enforcement that courts have applied over the years thus mirror the evolution of socio- political views about economic power at least as much as they track the progress of legal and economic thinking about exclusionary behavior

demo-4 Lessons in persuasion

Reconstructing the history of PP law and economics may have another, broader justification Antitrust law in general, and anti- PP law in particular, are ideal fields to investigate the crucial theme of how economists persuade – that is, of how, when and why theoretical economics may become a tool for concrete policy- and law- making Under what conditions might economists be listened to

by policy- and law- makers? Or, as I like to say, what kind of economic ments have the highest chances of successfully migrating from classroom to policy room or courtroom?

Focusing just on the legal realm, the answer hinges upon the different sional practices of economists and judges Starting (at least) from World War II (WWII), the former have striven for rigor and generality, accumulating new disci-plinary knowledge in terms of mathematical models, that is, by way of idealized representations of reality that by necessity abstract from seemingly irrelevant details and specific circumstances By contrast, judges’ interest and practice reside

profes-in fprofes-indprofes-ing the most effective way to admprofes-inister law profes-in any given case, by takprofes-ing into account every single fact and detail of the trial record and in pursuit of what-ever goal the law may have been assigned It follows that, regardless of their intrinsic theoretical or empirical validity, economists’ models have succeeded in migrating from classrooms to courtrooms only when they enjoyed two properties: that of being general enough not to depend upon idiosyncratic or heroic hypo-theses, and that of being easily translatable into workable rules or principles

The history of antitrust law offers several examples of that migratory pattern, and even more so of cases where the migration failed – that is, where econo-mists’ arguments were either totally neglected by judges and legislators or endorsed on a purely ad hoc, inconsistent basis So, for instance, it is well known that American economists opposed in vain the enactment of the first antitrust statute in 1890, severely criticized the Supreme Court’s rule of reason formu-

lated in Standard Oil, and had a major direct influence in the passing of the 1914

Clayton Act The ebbs and flows of economists’ efforts to persuade antitrust judges and legislators continued over the following decades.10 Antitrust is there-fore an invaluable source of teachings for grasping the power of economics to affect reality

The law and economics of PP is especially rich in this respect As the ing chapters show, the evolution of theoretical reflections about predatory behav-ior offers a whole array of different styles of economic reasoning: from basic price- theoretic models to semi- automatic rules, from vague “economic- flavored” arguments to rigorous game- theoretic analysis Understanding how, when and why these different approaches managed to persuade courts – or failed to – may thus provide a useful case study In particular, the erratic fortune of PP models in courtrooms seems to validate the previous claim that economists’ practices can

follow-be compatible with those of judges only in those special cases where the product

of the former’s theorizing is at the same time general enough and sufficiently operational

5 Treasures in the attic

Common law systems naturally lend themselves to an appreciation of the ical evolution of legal doctrines and enforcement practices Due to the rule of precedent, entire areas of the law may be presided over by age- old principles A few of these principles may sometimes have been inspired, either directly or indirectly, by economic reasoning As a consequence, traces of old economic ideas may still survive in common law This is clearly a boon for historians of economic thought The law is one of those rare fields where historians have an edge with respect to theoretical or applied economists

Antitrust law is the best example Its statutes have gone almost unchanged for

a century or more Key case law doctrines have provided the ruling precedents for many decades Old ideas, sometimes dating back to the late nineteenth and early twentieth century, still hold sway in various chapters of antitrust law Save for a few later amendments introduced by Congress, changes in enforcement have always been due to changing interpretations of the same norms Hence, one may understand current enforcement patterns by tracing back what particular interpretation of the law underlay a given doctrine at the time it was formulated Saying that history is essential for understanding antitrust law is hardly a novelty Since 1890 antitrust scholars, judges and practitioners have continually turned to history as a fundamental source of guidance, insight and justification Historical research has helped illuminate both the original legislative intent and

the meaning of earlier case law Courts at all levels have always based their trust opinions on historical evidence and interpretation – and they still do.11

anti-However, the term “history” in antitrust has always meant legal history, as

written by lawyers themselves for their own direct use or, more recently, by fessionally trained legal historians.12 While the latter have questioned the nạve view that “the legal past speaks authoritatively to the legal present” (Ernst 1990, 879), thereby paving the way to a more mature understanding of the inevitable historical contingency of the legal system, it remains true that the history of anti-trust has so far been entirely reconstructed using the language and methods of legal history.13

This is somehow unfortunate because antitrust law has always been related

to, when not directly affected by, the economic analysis of firms, markets and competition But economic theory has never been a datum Just think of the evolving meaning, since 1890, of key concepts such as market, technology, welfare, efficiency and, of course, competition Economists over the years have shaped, dissected and rebuilt each of these concepts Through a multiplicity of avenues the fruit of these economists’ efforts has eventually reached antitrust enforcers, or legislators, influencing their own work “Because the [Sherman Act’s] vital terms directly implicated economic concepts, their interpretation inevitably would invite contributions from economists,” wrote Bill Kovacic and Carl Shapiro “As economic learning changed, the contours of antitrust doctrine and enforcement eventually would shift, as well” (Kovacic and Shapiro 2000, 43) If we accept that changing economic ideas have played an important role in shaping antitrust statutes and case law, then it is only natural to turn to the dis-cipline the professional goal of which is precisely to investigate the historical evolution of those very ideas, namely, the history of economic thought (HET) Think of a basic question one may legitimately raise in the face of every anti-trust decision: what kind of economic “theory” (in the broadest sense of the word) did the court have in mind when making that decision? I claim this is a question that only HET may properly answer; and, equally importantly, one that

contemporary economists cannot answer Acknowledging what today’s antitrust

economics says about a certain business practice that the given decision demned falls short of explaining why that practice was prohibited in the past, why that prohibition held for a long time, and why it still does Only historians are equipped to asses what the “contextual” (read: theoretical, political, social, legal, etc.) conditions and goals were that led an old court to first prohibit that practice, and that justified maintaining the prohibition every time the same prac-tice was challenged in later cases And only historians of economic thought may fully appreciate a circumstance that contemporary economists often forget, namely, that economic theories are never developed in a vacuum, that we may never take the universal value of economic principles for granted, that some economic arguments are compatible only with some contextual conditions and goals while others are not What economic ideas were compatible with, say, the

con-socio- politico-legal context of 1911 America, when Standard Oil was decided

and PP prohibited – a prohibition that lasted for the next seven decades? What

about those economic ideas that were not compatible, and thus were dismissed

by antitrust courts regardless of their theoretical “correctness” in terms of today’s economics?

Applying the HET viewpoint allows the answering of questions like these, and even more As the following chapters show, the history of the law and eco-

nomics of PP also helps us to understand that economic ideas may and, on sion, do resurrect, and thus that economics does not proceed linearly, as Whig

occa-historians would claim.14 Progress in economics, especially in applied fields like antitrust, may come from unexpected quarters, including some seemingly dead- and-buried ones For their part, contemporary economists should therefore reject theoretical foreclosures and remain open to the possibility of surprising returns Indeed, it would be only rational for them to do so Records show that the true test of economic ideas in antitrust law has never been their conformity to the most up- to-date analytical frontier, but rather their concrete applicability to real world cases Big money usually awaits those economists who, working as expert counsel in antitrust litigations, are able to select the right theory for the right case HET is there to remind them never to forget the old treasure trove of ideas their forefathers left in the attic

6 Plan of the book

The remainder of this book is organized into eight chapters and a conclusion

In the first chapter I offer a brief survey of the economic models of PP, from the so- called basic story to more sophisticated Bayesian game theory The exposition is deliberately non- analytical, to help readers devoid of advanced mathematical skills (a knowledge of basic microeconomic principles is the only prerequisite)

Chapter 2 deals with the late nineteenth- century British common law of restraints of trade, the area of the law in which PP cases, like the early 1890s

Mogul Steamship, were first litigated British law fostered the utmost freedom of

contract and was the benchmark for American courts when Congress enacted the first antitrust statute

The American economists’ heterogeneous views at the turn of the twentieth century about antitrust in general, and so- called destructive competition (of which PP was the main example) in particular, are the theme of Chapter 3 The fourth chapter covers the formative era of American antitrust, starting with Congressional debates about the Sherman Act (1890) and ending with the approval of the Clayton and FTC Acts (1914) In between came the first two decades of antitrust enforcement, an extraordinary period when landmark cases

were decided, including the 1911 Standard Oil.

The focus of Chapter 5 is on so- called structuralist antitrust, the period of most active enforcement between WWII and the late 1960s These years are characterized by the structure- conduct-performance approach on the theoretical side and the Warren Court’s antitrust doctrines on the jurisprudential side

The period ended with a serious crisis of antitrust, of which the Court’s anti- PP

doctrine (as in the 1967 Utah Pie case) was among the clearest examples.

In Chapter 6 I present the reaction to this crisis by the Chicago School Not surprisingly, the reaction started from criticism of PP case law Launched by John McGee in 1958, the attack culminated with the declaration that PP was de facto impossible as a business strategy and should therefore be ignored as an antitrust violation

Simultaneously, two Harvard law scholars, Phillip Areeda and Donald Turner, proposed in 1975 a different way out from the crisis of anti- PP enforcement in terms of a straightforward price/cost rule that courts could easily apply Chapter

7 deals with the development of that rule, how courts swiftly adopted it, and the big debate it raised in antitrust law and economics

The eighth chapter covers both the triumph of the Chicago interpretation of

PP, symbolized by the Supreme Court’s decisions openly endorsing it, and the failure to date of the so- called Post- Chicago approach to win the courts’ atten-tion despite the recognized superiority of its game- theoretic analysis

A short conclusion offers some final thoughts on what the Post- Chicago fiasco may teach about the possibility of sophisticated economic arguments suc-cessfully migrating from classroom to courtroom

Notes

1 See, for example, McNulty (1968); Blaug (1997); Salvadori and Signorino (forthcoming).

2 See Peritz (1996); Steiner (2007); Page (2008).

3 To borrow Justice Peckham’s formula in United States v Trans- Missouri Freight

Association, 166 U.S 290 (1897), at 323.

4 Barry Wright v ITT Grinnell Corp., 724 F.2d 227 (1st Cir., 1983), at 234.

5 See Devlin and Jacobs (2010).

6 On which see Hovenkamp (2008, 114–16).

7 Salop (2008, 142–4) As an alternative, Salop proposes his own raising rivals’ costs paradigm.

8 Standard Oil Co v United States, 221 U.S 1 (1911), at 50.

9 This especially until 1950, when Congress eventually amended antitrust law to allow

a more effective anti- merger policy Challenging mergers that would result in a large concentration of power then became an alternative method of fighting size than pursu- ing predation.

10 Once again then- Judge Breyer nicely summarized the issue: “While technical nomic discussion helps to inform the antitrust laws, those laws cannot precisely repli- cate the economists’ (sometimes conflicting) views For, unlike economics, law is an

eco-administrative system” (Barry Wright, at 234).

11 See May (1990, 857).

12 For the evolution of the legal history of antitrust, see Ernst (1990); May (1990).

13 No intention to downplay that literature, though Legal history of antitrust includes gems like Hovenkamp (1991), Freyer (1992) and Peritz (1996) that will be our guiding lights in the following chapters.

14 According to the so- called Whig, or incrementalist approach to the history of nomics (Samuelson 1987), present economic theory is the outcome of the accumula- tion of “correct” ideas, while all other, forgotten ideas have been rightly abandoned as

eco-“incorrect.”

1 The economics of predatory

pricing

1  Classic and modern definitions of predatory pricing

The second section of the Sherman Act reads:

Every person who shall monopolize, or attempt to monopolize, or combine

or conspire with any other person or persons, to monopolize any part of the trade or commerce among the several States, or with foreign nations, shall

be deemed guilty of a felony, and, on conviction thereof, shall be punished

Legally speaking, the offense of monopolization – or §2 offense, as it is often

called – requires proof of a dominant firm with substantial market power and at least one qualifying exclusionary practice The latter may be defined as an act that

1 is reasonably capable of creating, enlarging, or prolonging monopoly power by limiting the opportunities of rivals; and 2 either does not benefit consumers at all, or is unnecessary for the particular consumer benefits pro-duced, or produces harms seriously disproportionate to the resulting benefits

(Hovenkamp 2005, 152) Predatory behavior is one such practice It is therefore a §2 offense, explicitly forbidden by antitrust law While the legitimate aim of any competitive act is to increase one’s own profit and while this may often legitimately happen only at the expense of one’s own competitors, what characterizes predation as an unlaw-ful exclusionary practice is the circumstance that the predatory act would not be profitable absent its exclusionary effect This feature is emphasized in the fol-lowing classic definitions:

Predation may be defined, provisionally, as a firm’s deliberate aggression against one or more rivals through the employment of business practices that would not be considered profit maximizing except for the expectation either that (1) rivals will be driven from the market, leaving the predator with a market share sufficient to command monopoly profits, or (2) rivals

will be chastened sufficiently to abandon competitive behavior the predator finds inconvenient or threatening.

(Bork 1978, 144)Predatory behavior is a response to a rival that sacrifices part of the profit that could be earned under competitive circumstances, were the rival to remain viable, in order to induce exit and gain consequent additional mono-poly profit

(Ordover and Willig 1981, 9–10) Several business practices may fall within such broad definitions This book will focus on the most frequent, and famous, of them, predatory pricing (PP

hereafter) The practice is defined in the Oxford English Dictionary as “the

setting of uneconomically low prices in order to damage one’s competitors or put smaller rival companies out of business.” US law gives a similar definition

In Title 15, Ch 1, §13.a of the so- called US Code we read: “It shall be unlawful for any person engaged in commerce, in the course of such commerce [ .] to sell, or contract to sell, goods at unreasonably low prices for the purpose of destroying competition or eliminating a competitor.”1

This prohibition did not feature in the Sherman Act, nor in the original lation of the Clayton Act, to which it has been added only in 1936, as §3 of the Robinson–Patman Act PP may therefore violate two different laws: the Clayton Act, as amended by the Robinson–Patman Act, and, as an instance of exclusion-ary practices, §2 of the Sherman Act The two violations are treated in the same way in modern case law As the US Supreme Court clarified in a landmark decision:

formu-Competitive injury under the Robinson–Patman Act is of the same general character as the injury inflicted by predatory pricing schemes actionable under §2 of the Sherman Act [ .] The essence of the claim under either statute is the same: A business rival has priced its products in an unfair manner with an object to eliminate or retard competition and thereby gain and exercise control over prices in the relevant market

(Brooke Group Ltd v Brown & Williamson Tobacco Corp., 509 U.S 209

[1993], at 221–2)The notion of “unreasonably low price,” mentioned by US law, or that of “unfair pricing,” used by the Supreme Court, are given concrete meaning by comparing price to production costs For example, in the OECD glossary of industrial eco-nomics we read that PP is:

A deliberate strategy, usually by a dominant firm, of driving competitors out

of the market by setting very low prices or selling below the firm’s mental costs of producing the output (often equated for practical purposes with average variable costs) Once the predator has successfully driven out

incre-existing competitors and deterred entry of new firms, it can raise prices and earn higher profits.

(OECD 1993, 67)This definition calls into play a list of elements – a dominant firm, a price below the rival’s costs, two phases, two possible goals (forcing exit or deterring entry) – which feature in all PP definitions given by modern industrial economists A few examples will suffice

One of the most popular Industrial Organization (IO hereafter) handbooks defines PP as:

a strategy in which a dominant firm cuts price below rivals’ average cost, even if this means accepting short- run losses, to drive rivals from the market Once rivals leave the market, the incumbent raises price and col-lects sufficient economic profit to make the present- discounted return from predatory pricing positive A variation is that a dominant firm cuts price as a way of encouraging rivals to compete less vigorously, without necessarily driving them from the market

(Martin 2002, 246)Another recent textbook on competition policy explains that predatory pricingoccurs when a firm sets prices at a level that implies the sacrifice of profits

in the short- run in order to eliminate competition and get higher profits in the long- run This definition [ .] contains the two main elements for the identification of predatory behaviour in practice: first, the existence of a short- term loss; second, the existence of enough market power by the pred-ator so that it can reasonably expect to be able to raise prices so as to increase profits in the long- run once a rival (or more rivals) has been driven out the market

(Motta 2004, 412)The conditions making PP profitable are stated in another popular volume: when incumbents lower their prices,

this is defined to be predatory pricing only when the intent is exclusionary;

that is, profitable only because it will result in a higher future price in this or some other markets by virtue of inducing exit or deterring entry in those other markets

(Viscusi et al 2005, 307–8, original emphasis)

In light of the aim of the present book, two further definitions may help What makes these additional definitions especially attractive is that they present the alternative views of PP – the classic and the game- theoretic one – which, as the next sections show, have animated the debate about the economics of predatory

behavior In his influential 1970 exposition of IO, Mike Scherer gave a lengthy explanation of PP in the following terms:

One of the purported traits of large, financially powerful conglomerate firms

is their greater ability and willingness to engage in sustained price- cutting with the intent of disciplining smaller competitors or even driving them out

of the market [ .] The most extreme form of predatory pricing takes place when a seller holds price below the level of its rivals’ costs (and perhaps also its own) for protracted periods, until the rivals either close down opera-tion altogether or sell out on favorable terms The predator’s motivation is

to secure a monopoly position once rivals have been driven from the arena, enjoying long- run profits higher than they would be if the rivals were per-mitted to survive Even if it must itself sustain losses during the price war, the predator can afford to do so because it can draw upon profits earned selling the same product in other geographic territories, or from selling dif-ferent products In other words, it subsidizes its predatory operations with profits from other markets until the predation creates conditions which will repay the original subsidy

(Scherer 1970, 273)

It is convenient to juxtapose this long passage with the definition provided by Patrick Bolton, Joseph Brodley and Michael Riordan (BBR) in their oft-quoted paper sponsoring the adoption by US courts of the game- theoretic approach:Predatory pricing is defined in economic terms as a price reduction that is profitable only because of the added market power the predator gains from eliminating, disciplining, or otherwise inhibiting the competitive conduct of

a rival or potential rival Stated more precisely, a predatory price is a price that is profit- maximizing only because of its exclusionary or other anticom-petitive effects The anticompetitive effects of predatory pricing are higher prices and reduced output – including reduced innovation – achieved through the exclusion of a rival or potential rival

However, the differences between the two definitions are even more ficant than their commonalities First of all, Scherer’s definition only applies to a

signi-big firm trying to achieve monopoly starting from an already dominant position

Bigness is deemed necessary for sustaining the predatory investment, because size (to be understood as “presence in other markets”) warrants the availability

of the financial resources enabling the predator to survive the losses As trial economist Corwin Edwards reportedly said: “The large company is in a position to hurt without being hurt” (quoted by Scherer 1970, 273) Thus, a pre-

indus-condition of PP is that the predator be endowed with economic – as distinguished from mere market – power While market power is a strictly theoretical notion –

to be measured via analytical tools, such as market shares or the Amoroso-

Lerner index – economic power is a broader, non- measurable notion, which

encompasses all the powers – financial, political, social – bestowed upon a firm

by its sheer size and the assessment of which goes beyond the limits of nomic analysis.2 Second, Scherer identifies PP as selling below cost – by which

eco-he surely means below rivals’ costs, and possibly also below one’s own Third,

his definition explicitly mentions the requirement of intent: the big firm’s price

cut must be deliberately aimed at harming its competitors Finally, Scherer’s

def-inition views PP as a strategy only directed against existing rivals.

The alternative definition by BBR is more general under all these respects

First, their definition only requires that the predator be endowed with some market power, to which the predatory strategy is said to add Nowhere, not even

implicitly, is the predator required to be either a big or a dominant firm, nor to be active in multiple markets, nor to aim at full monopoly Economic power falls outside of their discourse, which remains entirely within the boundaries of eco-nomic analysis Second, there is no necessary relation between price and costs: all that is required is that the price charged by the predator be a seemingly irra-tional one, were it not for its power- augmenting effect Even the notion of the predator’s losses is generalized from actual losses (negative profits) to the broader idea of foregone profits Third, this definition intentionally eschews any reference to the predator’s intent, let alone some specific intent to monopolize

As the next chapters show, the intent requirement has always caused trouble to the enforcement of PP legislation It is not surprising that a modern definition eschews it Finally, BBR explicitly refer to the possibility that PP may be dir-ected at curbing potential competition by deterring entry This is consistent with

the notion of anticompetitive effects (in terms of higher prices and reduced

output) which features as another key novelty of their definition While Scherer’s definition may also encompass a characterization of antitrust law as directed at

protecting a big firm’s competitors, especially small ones, BBR unambiguously depict the law’s goal as the protection of competition, rather than competitors, to

be achieved by preventing the competitive harm caused by PP

Historically speaking, Scherer’s narrower version corresponds to what we

may call the classic definition of PP, while that of BBR is the modern one The

word “classic” here denotes that this has been the definition which, more or less explicitly and more or less exactly, several generations of industrial economists have made their own, commencing in the last two decades of the nineteenth century, proceeding with full force after WWII and ending only with the rise of

the game- theoretic approach to IO in the early 1980s The latter approach would provide a broader characterization of PP, properly represented by BBR’s

“modern” definition

However, even the classic definition of PP never fully overlapped with the narrative that US courts had in mind when asked to apply antitrust law to alleg-

edly predatory episodes In what we may dub the classic legal standard of PP,

the specific theoretical elements of Scherer’s definition – market dominance, monopoly position, below- cost pricing, loss subsidizing, etc – were dissolved into a vague narrative that emphasized its non- analytical components, like the

“unfair use” of pricing policy against smaller rivals or the multi- faceted risks (economic, political, social) implicit in any concentrated market structure In that story, PP just meant that a big firm had unfairly exploited its power to undercut its smaller rivals, with the intention of further increasing that power by causing them to exit the market For more than 60 years, from 1911 to 1975, US courts have been enforcing the anti- PP prohibition guided by a legal standard that hinged on that story, rather than on economic analysis Accordingly, the goals to

be pursued were the protection of small firms from unfair price cuts made by bigger rivals and the preservation of a competitive market structure, understood

as absence of excessive agglomerations of economic power In short, protecting competitors and diffusing economic power were the real drivers of anti- PP enforcement Economic efficiency was never specifically articulated as a distinct goal.3

Following the classic legal standard, courts looked for the two essential ments of its underlying narrative: a considerable economic power and the inten-tion to prey upon smaller rivals, i.e., to unfairly exploit that power Power and intent thus constituted the two necessary ingredients of any accusation of preda-tory behavior Curiously, both the standard and the underlying story were closer under a crucial respect to BBR’s definition of PP than to Scherer’s: US courts could apply the standard without, strictly speaking, any comparison between the predator’s price and its or its rival’s costs Though in practice such a comparison did represent the core issue of most PP cases, it was possible for a big firm to be found guilty of predatory behavior despite pricing above cost What really mat-tered was, first, that it was a big firm, and, second, that the court was convinced that it had unlawfully intended to prey upon its smaller rivals

This approach brought troubles to PP case law Too many decisions by courts guided by the classic legal standard showed a weak economic rationale – if any

at all The situation became explosive with the post- WWII intensification of trust activity, and all the more so in the 1960s, when sheer business size became the declared target of antitrust enforcement Things changed only in the mid- 1970s, when courts replaced the classic legal standard with a new approach, founded upon mainstream price theory and, above all, upon a clear- cut rule, rather than a vague story Yet, by the time the change took place, the economic analysis of PP was on the verge of another quantum leap, away from Scherer’s classic definition and towards BBR’s strategic one

anti-2 The basic story

The distance separating the vague narrative underlying the classic legal standard from the more specific story supporting the economists’ classic definition of PP becomes clearer when the latter is more closely analyzed Notwithstanding its shortcomings, that story provides a good benchmark for comparing the way anti-

PP law has been enforced by US courts for many decades and how it should have been, if only judges and juries had paid more attention to what industrial economists were telling them This section and the next aim at reconstructing the

economists’ classic narrative of PP – what we may call the basic PP story – and

the critiques raised against it from the late 1950s onwards

Any PP story founded upon economic analysis must answer two fundamental questions What market structure is necessary for price predation to be a profitable strategy? What market structure is necessary for price predation to be actually undertaken by a rational firm? The first question aims at determining under what conditions PP is a strategy with a positive net present value (NPV),

i.e., is profitable in an absolute sense The second question, which only

makes sense in case of a positive first answer, looks at whether predation is more

profitable than any other alternative strategy, i.e., it deals with PP’s relative

profitability

The basic PP story is a stylized narrative of the events and conditions

required for an affirmative answer to the first question The story runs as follows:

1 A market leader exists who has market power in two or more markets, separ ated by either geography or product categories In short, the leader is a multi- market firm earning supra- competitive profits in all its markets

2 In one (or more) of these markets the leader has one or more competitors, selling the same product The rivals are active only in that market, i.e., are single- market firms Hence, the leader is not a monopolist in these markets [As a variant of the first two points, the leader may also be a single- market firm that used to be a monopolist in that market, but now faces the entry of a rival and thus has to give up its monopoly.]

3 To further increase its profits, the leader may behave as predator in the market where it faces competition [In the variant, the goal is to preserve the monopoly position.] To do so, it sets a price lower than its own marginal (or average variable) cost This is the predatory phase On the assumption that the rival has the same cost structure (or worse) than the leader’s, both the leader and the rival suffer losses from such a low price

4 The leader may bear those losses for a significant amount of time, due to the supra- competitive profits earned in his other markets – in PP jargon, due to

its deep pocket or long purse This is not so for the single- market rival,

which has no deep pocket and is thus forced to exit the market, or go bust (In the variant, the leader survives the losses thanks to the reserves accumu-lated when it was a monopolist.) Note that it is not required that the leader

raises its price in any of its other markets to survive the predatory phase (the assumption being that in those markets it is already pricing at the profit- maximizing level).

5 As soon as the rival leaves the market, the leader may raise the price to the monopoly level, earning higher profits than before (In the variant, the leader regains its initial monopoly.) These extra profits allow it to more than recover the losses suffered during the predatory phase This is the recoup-ment phase, which ends the strategy

Simplistic as it may be, this basic story has driven the economics of PP for more than half a century, from the late nineteenth century to, at least, 1958, when it was first seriously questioned Despite the subsequent critiques, the story remained the benchmark in IO textbooks well into the 1980s, when it was even-tually supplanted by more sophisticated game- theoretic narratives Its key element, namely, that predation means pricing below cost, has constituted the theoretical pillar for the score of price/cost rules that economists and law schol-ars have developed since 1975 to help courts enforce the anti- PP prohibition.4

Even the ruling doctrine in US case law, formulated by the Supreme Court in a

1993 case, is founded upon the basic PP story – if only to express the Court’s skepticism about its real- world significance

Three remarks on the basic story are in order First, the story highlights the intertemporal dimension of PP The strategy may be profitable only as long as the leader earns extra profits in the recoupment phase that are big enough to more than compensate the losses suffered during the predatory phase Predation thus partakes of the investment logic, i.e., bearing costs “today” in order to earn

more “tomorrow.” PP is an investment in the creation of market power (or, in

the variant, in its preservation) The second remark has to do with the leader’s cost and production structure An implicit assumption is that the leader may satisfy the additional demand which follows the price cut by boosting its supply

at non- increasing average cost This is to exclude the possibility that the leader’s strategy be defeated by its own rising costs depleting even a very deep pocket Finally, from the consumers’ viewpoint, the basic story entails a short- run increase in consumer welfare, lasting as long as the predation phase, followed by

a long- run (possibly permanent) welfare loss when the leader becomes (or, in the variant, returns to being) a monopolist

Returning to the two above- mentioned questions, we may note that the basic story deals only with the first Depending on the length of the two phases, and on the amounts of the gains and losses attached to them, the story identifies the con-ditions warranting the strategy’s positive NPV, i.e., that PP be profitable in an absolute sense However, the story is silent about any alternative strategy that the leader may find more profitable than price predation, i.e., we do not know whether PP is the most profitable strategy

The determinants of PP’s absolute profitability may be easily stated The strategy’s NPV depends, in the basic story, on three elements: the amount of the losses during the predatory phase, the amount of the extra profits during the

recoupment phase, the discount factor The latter element is the easiest to handle

Any investment’s NPV diminishes ceteris paribus with the increase in the

dis-count factor, that is, with the preference for a dollar today over a dollar row It follows that a myopic leader, which discounts future gains a lot, would

tomor-be less likely to undertake a predatory strategy, while a far- sighted leader would

be more willing to implement it

The other two elements are trickier The losses suffered by the leader depend

on two key factors: the size of the price cut and the length of the predatory phase There is a clear tradeoff between the two: the lower the price, the heavier the losses on each unit sold for both the leader and the rival, the shorter the lat-ter’s resistance and thus the shorter the predatory phase itself The leader must therefore set the price to optimally handle the tradeoff between its own losses and the time required to force the rival’s exit The length of the predatory phase also depends on the structure of the rival’s costs, in particular on the relative

amount of its sunk costs, i.e., costs which cannot be recovered if the firm quits

production If much of its costs are sunk, the rival will be harder to “kill” because it will try to resist as much as possible in order to avoid turning its un- retrievable costs into pure losses

Sunk costs also provide a first counter- argument to the basic story At least part of the rival’s sunk costs are usually embodied by some physical assets – say, its plant – which could be purchased by another firm at a low, but positive, price The new firm would be lured into the market by the supra- competitive profits warranted by its structure – say, a leader/follower duopoly – as well as by the possibility of avoiding any entry or other sunk cost, because they have already been paid by the original rival As long as the rival’s assets are “out there” – and

if they are sunk, it is going to be a long time – this entry possibility always exists The entry of a new competitor purchasing the original one’s assets would doom the PP strategy, by recreating competition and indefinitely postponing the recoupment phase.5

The gains to be earned during the recoupment phase depend on the price the leader- turned-monopolist can charge and the duration of its newly acquired monopoly position The existence of entry barriers protecting the leader from new competitors plays a key role here Yet the basic story does not even mention these barriers A second critique of that story is therefore that the absolute profit-ability of PP crucially depends on a neglected feature Ignoring entry barriers is perhaps the worst drawback of the basic story In the limiting case of a perfectly contestable market, where no such barriers exist,6 PP would never be profitable because new rivals would always enter as soon as the leader raises its price fol-lowing the original rival’s exit Only the existence of absolute barriers, protect-ing it from new entry regardless of price, would guarantee the leader the possibility to price at the monopoly level In the intermediate, and most common, case of non- negligible barriers, a new trade- off ensues between the price set by the leader in the recoupment phase and the length of the phase itself: the higher the price, the larger the revenues, but also the easier the entry of new rivals inter-rupting recoupment The same sunk costs that make it harder to “kill” the rival

during the predatory phase may, after the rival’s “death” with no surviving asset, provide a considerable barrier protecting the leader’s recoupment.

3 The Chicago critique of the basic story

Following John McGee’s opening salvo in 1958,7 the Chicago School of trust has raised several objections against the basic story A first collection of critiques aims at demonstrating that PP can never be a profitable strategy and therefore will be never undertaken by a rational firm A second group of objec-tions attack the basic story for neglecting alternative strategies that are always

anti-more profitable than PP Both sets of critiques conclude that rational firms would

never implement a predatory strategy But if this is so, antitrust law should not pursue a non- existent behavior If a low price can never be a predatory price, then any price cut is always a sign of genuine competition – a kind of behavior that antitrust law should encourage, rather than punish In short, according to the Chicago School a proper account of the basic story and its limits shows that anti-trust law should stop worrying about PP

As we know from the previous section, the first group of critiques concern the missing elements in the basic story’s description of the two phases, predation and recoupment For instance, a classic Chicago argument is that the predation phase will never end because the rival will never “die.” Despite undercutting by the leader, the rival will resist indefinitely thanks to the support received by either its customers or the financial market Help from financial markets is justi-fied, according to Chicago scholars, by the extra profits the firm will necessarily earn if it survives predation The existence of these extra profits we may take as certain, first because the market is, by assumption, less than perfectly com-petitive and, second, because only the expectation of earning such profits would justify the firm’s costly efforts to resist predation in the first place Financial markets should always be willing to finance the firm’s resistance, in the reason-able expectation of recovering their credit once the predation phase is over and the price returns to the initial, supra- competitive level This financial help makes

up for the firm’s lack of a deep pocket by exploiting its future earnings.

As for the firm’s customers, the Chicago claim is that they should continue purchasing from it at a price higher than that of the leader Rational customers should understand that the short- run gain of buying from the leader at a below- cost price would be more than compensated by the future monopoly price, to be paid as soon as the rival is “dead” and the leader has become a monopolist Pro-vided they do not discount the future too much, customers would prefer to bear today the cost of “keeping competition alive” in view of the welfare gains such competition would grant them tomorrow Assuming its customers’ rationality and far- sightedness, the rival could avoid following the leader in the price cut and charge its usual price, thus defeating predation

Both objections to the basic story are themselves open to criticism Consider the supposed aid of financial markets Asymmetric information and other market imperfections may hinder such help Potential lenders may be unable to foresee

the rival’s future profitability Or they may doubt the rival’s possibility of viving predation and repaying its loans Financial help may therefore be unavail-able Though Chicago economists have countered these objections by requiring that critics specify the alleged imperfections in the capital market,8 it is undeni-able that a (usually) small firm subjected to a predatory attack by a (supposedly) big market leader is not exactly the kind of business real- world lenders would rush to finance.

Even more compelling is the objection against customers’ help The Chicago argument neglects the free riding opportunity every rational customer could exploit, buying at a low price from the leader in the expectation that other cus-tomers will ensure the rival’s survival by purchasing from it at a higher price Preserving competition in the market is a kind of public good which every cus-tomer would be happy to “consume,” but which none is willing to pay for As with every other public good, it is up to the policy- maker to guarantee the “pro-duction” of competition in the marketplace The Chicago counter- argument here

is that a firm under predatory attack may circumvent free riding by offering its customers a more complex contract For example, it could enter into a long- run commitment to always sell the good at a price lower than the monopoly level (i.e., the price a successful predator would charge), provided the customer buys from it in the short run Such a contract should entice rational (and not too myopic) customers to overcome free riding temptations and support the firm’s resistance to predation Leaving aside the complexity and cost of those long- term contracts, the implicit Chicago assumption is that customers entering into them must possess, like lenders in financial markets, a considerable degree of ration-ality and forecasting capability Moreover, they must trust the firm’s ability to survive predation and be still around to honor its long- run commitment Such customers forfeit an immediate bargain (buying at low price from the predator)

in the hope of a future gain, which might never materialize

Finally, Chicago scholars claim that, even accepting that PP may sometimes

be a profitable strategy, it will never be undertaken by a rational firm because at least one other strategy exists that is always more profitable: the takeover strategy, i.e., the direct and immediate purchase of the competitor The predator may offer to acquire its rival’s business at a price between the value of the rival’s future discounted profits, at the minimum, and the value of the leader’s post- takeover future discounted profits, at the maximum The takeover is surely more profitable for the leader than predation; at the same time, a rational rival will always accept the leader’s offer rather than risk predation The rival is being offered for the takeover a sum that is, at the minimum, equal to the value of its future profits.9 Any predatory episode, regardless of its length and eventual outcome, would surely diminish this value, so every firm fearing predation would willingly pocket it for certain, accepting the leader’s proposal As to the leader itself, takeover is always more profitable than predation because, by immediately getting rid of competition, it can charge the monopoly price right away, without having to suffer the losses of a (possibly long) predatory phase The future discounted value of the post- takeover profits is surely larger than that

of the post- predation profits By granting the rival a fraction of these additional monopoly profits, the leader will be certain to pocket the rest That fraction may

go from zero to one The latter figure corresponds to the maximum takeover price, when the leader bestows upon its rival the entire extra profits This is an unlikely outcome The leader will often acquire the rival by offering it a price very close, possibly equal, to the minimum, especially if it may accompany this proposal with a credible predatory threat

Once again, the Chicago argument aims at demonstrating that PP is a non- rational strategy that will never take place in the real world and which, therefore, requires no policing by antitrust law Once again, the argument is open to criti-cism Beyond the usual caveats concerning the alternative strategy’s rationality and information requirements, a key objection is that a so- called merger- to-monopoly (i.e., a takeover by the market leader of its only competitor) is an even more flagrant violation of §2 of the Sherman Act than PP.10 The law prohibiting monopolization would be applied even more forcefully against such a takeover than against PP – if only because, while price predation may always be disguised

as a competitive price cut, it is much harder for defendants to justify a merger- to-monopoly

In conclusion, neither the basic story nor the Chicago critiques are fully vincing, theoretically speaking Both suffer from over- simplifications and an idealized characterization of competition between a market leader and its rivals Key elements of real- world predatory episodes are missing, first among them the necessary strategic character of any such business behavior With the benefit of hindsight, we now know that cutting prices below cost may find its rationale in sending a message to future rivals, rather than in “killing” existing ones – the

con-message being: “this is my market, don’t try to enter it, ’cause I’m ready to

squander bags of money to defend it.” Messages of this kind will represent the core of the next generation of PP stories, explicitly based on game- theoretic reasoning

Still, the controversy between the basic story and its Chicago critics may teach a useful lesson It casts light on the two poles of the century- long debate about PP On the one side, a reasonable argument showing, though often too nạvely, that predatory behavior may well be explained within a price- theoretic framework On the other, a series of critiques that, regardless of their intrinsic robustness, deserve credit for having revealed the worst danger of too restrictive

a policy against price cuts by dominant firms – namely, the danger of chilling

genuinely competitive behavior Cutting prices is the competitive strategy par excellence, the one that every antitrust enforcer should enhance and protect The

Chicago critiques warn us that preventing big firms from cutting their prices means depriving them of their most pro- competitive weapon Indeed, distin-guishing between pro- competitive and anti- competitive price cuts is one of the

most – if not the most – difficult task for antitrust enforcers The clash between

the basic story and Chicago is an acute reminder of this simple fact

4 It’s a brand new game: predation as strategic paradox

The development of strategic models of PP in the early 1980s was a landmark event in the history of modern game theory These were among the earliest

models where the new techniques and solution concepts of Bayesian game theory (BGT) were applied to analyze games without the traditional assumptions

of perfect and complete information.11 Until the late 1960s, these unrealistic assumptions had undermined the applicability of game theory to real worlds problems, including IO ones Explaining the late 1970s/early 1980s game theory boom within neoclassical economics would exceed this book’s limits But none can doubt that the game theorists’ new ability to handle more realistic games with either imperfect or incomplete information played a key role in the devel-opment of modern economics Both the rise of modern information economics (including mechanism design theory) and the establishment of game theory as the core of mainstream microeconomics (a result the first generation of game theorists had missed) descended from this technical improvement.12

Two names stood behind this chain of events, Reinhard Selten and John sanyi, both recipients of the 1994 Nobel Prize in Economics, together with John

Har-Nash Harsanyi taught game theorists how to deal with games of incomplete

information, where players lack some information about the game’s structure

He suggested turning such games into games of imperfect information, defined

as games where players make at least one of their moves without being fully informed about the previous moves made by other players The trick that allows the transformation lies in assuming that a fictitious player, called Nature, ran-domly selects a player’s “type,” i.e., a specification of the player’s relevant char-acteristics As a consequence of Nature’s choice, the other players’ uncertainty and lack of information may then be captured by their beliefs over the prob-ability distribution of that player’s types – that is, by a kind of mathematical object, subjective beliefs, which game theorists may handle with standard Baye-

sian tools Hence the name Bayesian games for this class of models à la sanyi The solution concept for these games is the so- called Bayesian Nash equilibrium, an extension of the standard Nash equilibrium to the agents’ beliefs

Har-over the distribution of types.13

Selten’s role is even more crucial for our story Beyond contributing on the technical side with the development of backward induction and (subgame) perfect equilibrium, the German mathematician was responsible for drawing game theorists’ attention to PP In his classic 1978 paper, “The Chain Store Paradox,” he demonstrated that in a finite game under complete and perfect information a predatory strategy is never part of the game equilibrium, i.e., that

PP is a behavior no strategically rational firm would ever undertake

It is essential to understand what Selten’s result actually meant The 1978 paper was neither about PP nor, more generally, about any other IO problem The story of the chain store having to decide whether to behave aggressively or cooperatively in the face of potential entry was just an “expositional device [that] should not be misunderstood as a model of a real situation” (Selten 1978,