Introduction to Industrial Organization PDF

INTRODUCTION TO INDUSTRIAL ORGANIZATION SECOND EDITION  INTRODUCTION TO INDUSTRIAL ORGANIZATION SECOND EDITION luis M. B. CABRAL THE MIT PRESS (AMBRIDGE, MASSACHUSETTS LONDON, ENGLAND © 2017Massachusetts Institute of Technology All rights reserved. No pa...

INTRODUCTION TO INDUSTRIAL ORGANIZATION SECOND EDITION  INTRODUCTION TO INDUSTRIAL ORGANIZATION SECOND EDITION luis M. B. CABRAL THE MIT PRESS (AMBRIDGE, MASSACHUSETTS LONDON, ENGLAND © 2017Massachusetts Institute of Technology All rights reserved. No part of this book may be reproduced in any form by any electronic or mechanical means (including photocopying, recording, or information storage and retrieval) without permission in writing from the publisher. This book was set inMelior by diacriTech, Chennai. Printed and bound in the United States of America. Library of Congress Cataloging-in-Publication Data Names: Cabral, LuisM. B., author. Title: Introduction to industrial organization I LuisM.B. Cabral. Description: Second edition. I Cambridge, MA :MIT Press, I Includes bibliographical references and index. Identifiers: LCCN 2016045858 I ISBN 9780262035941 (hardcover : alk. paper) Subjects: LCSH: Industrial organization (Economic theory) Classification: LCC HD2326.C33 2017 I DDC 338.6-dc21 LC record available at https://lccn.loc.gov/2016045858 10 9 8 7 6 5 4 3 CON TE N TS PREFACE ix ACKNOWLEDGMENTS X C H APTER 1 WHAT ISINDUSTRIALORGANIZATION? 1 1. 1 An example 2 1. 2 Central questions 3 1. 3 Coming next... 9 PART 0 N E MICROECONOMICS fOUNDATIONS 13 C H APTER 2 CONSUMERS 15 2.1 Consumer preferences and demand 15 2.2 Demand elasticity 19 2.3 Estimating the demand curve 26 2.4 Are consumers really rational? 29 C H APTER 3 FIRMS 35 3.1 The firm's production, cost, and supply functions 36 3.2 Pricing 43 3.3 Do firms maximize profits? 52 3.4 What determines a firm's boundaries? 54 3.5 Why are firms different? 56 C H APTER 4 COMPETITIONEQUILIBRIUM , , AND EFFICIENCY 67 4.1 Perfect competition 67 4.2 Competitive selection 74 4.3 Monopolistic competition 79 4.4 Efficiency 81 C H APTER 5 MARKETFAILURE AND PUBLIC POLICY 93 5.1 Externalities and market failure 93 5.2 Imperfect information 99 5.3 Monopoly 102 5.4 Regulation 107 5.5 Competition policy and antitrust 109 5.6 Firm regulation 110 vi 1 CONTE NTS CHAPTER 6 PRICE DISCRIMINATION 121 6.1 Selection by indicators 125 6.2 Self selection 130 6.3 Non-linear pricing 135 6.4 Auctions and negotiations 140 6.5 Is price discrimination legal? Should it be? 144 PART T WO OLIGOPOLY 157 CHAPTER 7 GAMES AND STRATEGIES 159 7.1 Nash equilibrium 161 7.2 Sequential games 167 7.3 Repeated games 172 7.4 Information 174 CHAPTER 8 OLIGOPOLY 185 8.1 The Bertrand model 186 8.2 The Cournot model 194 8.3 Bertrand vs. Cournot 200 8.4 The models at work: comparative statics 202 CHAPTER 9 COLLUSION AND PRICE WARS 217 9.1 Stability of collusive agreements 218 9.2 Price wars 222 9.3 Factors that facilitate collusion 227 9.4 Empirical analysis of cartels and collusion 232 9.5 Public policy 234 PART T HREE ENTRY AND MARKET STRUCTURE 249 CHAPTER10 MARKET STRUCTURE 251 10.1 Entry costs and market structure 253 10.2 Endogenous vs. exogenous entry costs 260 10.3 Intensity of competition, market structure, and market power 264 10.4 Entry and welfare 269 10.5 Entry regulation 273 CHAPTER11 HORIZONTAL MERGERS 281 11.1 Economic effects of horizontal mergers 282 11.2 Horizontal merger dynamics 288 11.3 Horizontal merger policy 292  CONTENTS 1 vii CHAPTER12 MARKET FORECLOSURE 303 12.1 Entry deterrence 304 12.2 Exclusive contracts , bundling, and foreclosure 312 12.3 Predatory pricing 317 12.4 Public policy towards foreclosure 320 PART FO UR NON-PRICE STRATEGIES 331 CHAPTER13 VERTICAL RELATIONS 333 13.1 Vertical integration 334 13.2 Vertical restraints 340 13.3 Public policy 342 CHAPTER14 PRODUCT DIFFERENTIATION 349 14.1 Demand for differentiated products 350 14.2 Competition with differentiated products 354 14.3 Advertising and branding 362 14.4 Consumer behavior and firm strategy 368 14.5 Public policy 371 CHAPTER15 INNOVATION 377 15.1 Market structure and innovation incentives 381 15.2 Diffusion of knowledge and innovations 383 15.3 Innovation strategy 385 15.4 Public policy 391 CHAPTER1 6 NETWORKS 399 16.1 Chicken and egg 401 16.2 Innovation adoption with network effects 405 16.3 Firm strategy 414 16.4 Public policy 419 INDEX 427 PREFACE Sixteen years have passed since the first edition of Introduction to Industrial Organi­ zation (IIO). Meanwhile, a lot has happened in the world and in the field of industrial organization (IO). A second edition of IIO was thus long overdue. There are many things new in the new edition. It has been my experience that IIO can be used as a text both in undergraduate economics and business courses; and in various non-economics masters programs. With that in mind, Part I now includes a more comprehensive overview of the basic microeconomics required for the under­ standing of IO : consumer utility and demand, firm costs and pricing (including price discrimination) , competitive markets, market failure. IO has witnessed a tremendous boost in terms of empirical analysis; however, most textbooks provide little guidance in this area. The present edition provides an introduction to statistical methods in three important areas : demand identification (Section 2.3); analysis of cartels and collusion (Section 9.4); and modeling product differentiation (Section 14.1). In response to many requests , I added selected advanced materials. First, sev­ eral chapters now include advanced sections (in smaller type) that provide an analyti­ cal treatment of ideas previously presented verbally. End-of-chapter exercises help the reader push the boundary, (a) formalizing ideas that are only briefly touched upon in the text; (b) generalizing results that are presented in simple form; or (c) applying concep­ a. In particular, I should tual results to the empirical analysis of particular industries (I included a new section emphasize that the list of at the end of each chapter: "applied exercises"). references in the book does not reflect a balanced Finally, much of what is new in this edition corresponds to new and updated survey of the 10 literature. examples, both in the main text and as separate "boxes." To stress this point, I have Despite all of these innovations and additions , IIO 2nd Edition retains the basic chosen not to make direct reference to sources in the spirit of IIO 1st Edition: a book that is issue driven rather than methodology driven. text, leaving bibliographic Although I make extensive use of models (because I think they are very useful ways of references to notes at the understanding reality) , I only introduce a new model when this is justified in terms of end of the text. {These are the notes marked with the marginal benefit in understanding some issue." The focus on issues is also reflected numbers, as opposed to on the focus on policy implications. While this was already the case in IIO 1st Edition, footnotes as this one, which IIO 2nd Edition includes a policy section at the end of most chapters. are marked with letters. ) X I PREFACE ACKNOWLEDGMENTS IIO 1st Edition included a long lists of acknowledgements. Since 2000, the number of teachers, students and other readers who provided suggestions, corrections - or simply encouragement - has enlarged my debt roll considerably. I apologize for possible omissions in what is now a long list: b Dan Ackerberg, Mark Armstrong, Helmut Bester, Bruno Cassiman, Allan Collard-Wexler, Pascal Courty, Greg Crawford, Leemore Dafny, Jan De Loecker, Kenneth Elzinga, Joe Farrell, Alfonso Gambardella, Joshua Gans , David Genesove, Ben Hermalin, Mitsuru Igami, Jos Jansen, Przemyslaw Jeziorski, Chris Knittel, Tobias Kretschmer, Francine LaFontaine, Ramiro Tovar Landa, Robin Lee, Frank Mathewson, David Mills , GianCarlo Moschini, Petra Moser, Hiroshi Ohashi, Ariel Pakes, Michael Peitz, Rob Porter, Michael Riordan, Daniil Shebyakin, John Small, Adriaan Soetevent, Scott Stern, John Sutton, Frank Verboven, Reinhilde Veugelers , Len Waverman, Matthew Weinberg, Ali Yurukoglu, Peter Zemsky, Christine Zulehner, and various anonymous reviewers who took their time to read previous b. Since much of what was in the first edition is also drafts. Claire Finnigan, Mike Cheely and John Kim provided excellent research assis­ present in the second tance. Finally, I thank various classes at London Business School, London School of edition, the following lists Economics , Berkeley, Yale, and New York University, whom I taught based on previous includes acknowledgements included in the first edition versions of this book and from whom I received useful feedback. Unfortunately, I alone as well. remain responsible for remaining errors and omissions.  1 What is industrial organization? It might help to start by clarifying the meaning of "industrial. " According to Webster's New World Dictionary, "industry" refers to "man­ ufacturing productive enterprises collectively, especially as distinguished from agricul­ ture" (definition 5 a). "Industry" also means "any large-scale business activity, " as in the tourism industry, for example (definition 4 b). This double meaning is a frequent source of confusion regarding the object of industrial organization. For our purpose, "industrial" should be interpreted in the sense of Webster's definition 4 b. That is, industrial organization applies equally well to the steel industry and to the tourism industry; as far as industrial organization is concerned, there is nothing special about manufacturing. Industrial organization is concerned with the workings of markets and industries, in particular the way firms compete with each other. The study of how markets operate, however, is the object of microeconomics ; as a Nobel prize winner put it, "there is no such subject as industrial organization," meaning that industrial organization is nothing but a chapter of microeconomics. 1 The main reason for considering industrial organi­ zation as a separate subject is its emphasis on the study of a firm's strategies that are characteristic of market interaction: price competition, product positioning, advertising, research and development, and so forth. Moreover, whereas microeconomics typically focuses on the extreme cases of monopoly and perfect competition, industrial organiza­ tion is primarily concerned with the intermediate case of oligopoly, that is, competition between a few firms (more than one, as in monopoly, but not as many as in competitive markets). For the above reasons, a more appropriate definition of the field would be some­ thing like "economics of imperfect competition. " But the term "industrial organization" was adopted, and I am not the one to change it. 2 CHAPTER 1 I Industrial organization is concerned with the workings of markets and industries, in particular the way firms compete with each other. 1.1 AN EXAM P L E Examples are often better than definitions. In this section, I look into the market for anti-ulcer, anti-heartburn drugs, a case that touches on a number of issues of interest to industrial organization. The case thus provides a useful introduction to the next sec­ tion, where I look more systematically at the central questions addressed by industrial organization. Until the late 1 9 70s, there was no effective drug to treat ulcers ; severe cases required surgery. Then a research project at Smith, Kline & French (which is now part of GlaxoSmithKline) culminated with the discovery of cimetidine, a wonder drug sold under the brand name Tagamet. The production cost of Tagamet was rather small compared to the price for which it was sold. In cases like this, the huge gap between price and cost is bound to raise a variety of issues. For example, it may seem unfair that many suffering patients be deprived of a cheap-to-produce drug simply because price is so high. Then again, without revenues from drugs such as Tagamet it seems impossible for firms like Smith, Kline & French to continue churning out blockbuster drugs. The fact is that the market for anti-ulcer drugs was enormous. Although Smith, Kline & French had a patent for cimetidine (valid until 1997), this did not stop other pharmaceutical giants from coming up with alternative products : Zantac , introduced by Glaxo in 1983 ; Pepcid, introduced by Merck in 1986; and Axid, introduced by Eli Lilly in 1988. For many industry experts, Zantac , Pepcid, and Axid were little more than a copy­ cats (also known as "me too " drugs). In fact, reviews of clinical trials indicated that there was little difference in the success rates of the various drugs; in other words, one drug could easily substitute for any of the other ones. Why wasn't then price competition more intense? The answer is: advertising. BELLYACHE BATTLES. We knew that the battle for your bellyaches would be big, but we had no idea it would be so bloody. Hundreds of millions of dollars are being p oured into advertising designed to establish brand loyalty for either Tagamet HB or Pepcid AC. Zantac 75 will join the fray shortly. These drugs were blockbusters as prescription ulcer treatments ; now that they are available over the counter for heartburn their manufacturers have really taken off the gloves.2 Nor is this specific to anti-ulcer and heartburn drugs. Overall, the advertising budgets of large pharmaceutical companies are of the same order of magnitude as their research budgets. What matters is not just the product's worth, but also what consumers - and  WHAT IS I N DUSTRIAL ORGANIZATION? I 3 doctors , who frequently act as agents for the final consumer - think the product is worth, Glaxo emerged as the winner of this advertising battle, It "set out to make heart­ burn into an acute and chronic ' disorder' that came with serious consequences if not treated twice daily with the company's two-dollar pill." 3 By 1988, Zantac had overtaken Tagamet as the world's best-selling drug. Zantac's patent expired in the late 1990s, paving the way for competition by gener­ ics. A generic is a chemically equivalent drug that is mainly sold under the chemical name (Ranitidine, in the case of Zantac) rather than under the brand name. Notwith­ standing numerous claims that generic Zantac has the same effect as branded Zantac, the latter still manages to command a large market share while selling at a much higher price. In July 1999, shortly after patent expiry, discount drug seller RxUSA was quoting a 3 0-tablet box of 300 mg Zantac at $85.95. For a little more than that, $95, one could buy a 2 5 0-tablet box of 300 mg generic Zantac Ranitidine - that is, for 7. 5 times less per tablet. More than a decade after patent expiry, price differences remain significant: in January 2014, 1 5 0 mg Zantac cost almost 40 cents per tablet, whereas the corresponding generic sold for about 8 cents. Another testament to the enduring value of a strong brand is that, in 2006, Boehringer Ingelheim paid more than $500 million dollars for the US rights to Zantac. At the time of Zantac's launch, Glaxo was an independent company. Since then, it first merged with Wellcome to form GlaxoWellcome, then with SmithKline (which in turn resulted from a then recent merger) to form GlaxoSmithKline." Frequently, these mergers are heralded as the sources of important synergies. For example, when GlaxoWellcome was formed, the merging parties argued that the combination of Well­ come's AZT and Glaxo's 3TC worked better against AIDS than either drug alone. 4 Crit­ ics, however, see it primarily as a source of greater market power: if you cannot beat the competitor, then buy it. 1.2 C E N T RA L Q U ESTI O N S a. It's a good thing the The example in the previous section illustrates several issues that industrial organiza­ merged firms did not keep tion is concerned with (see below, in italics): for decades, GlaxoWellcome was a firm all of their names, else we that commanded a significant degree of m arket power in the anti-ulcer and heartburn would have to spell GlaxoWellcomeSmith­ therapeutical segment (the relevant market definition). GlaxoWellcome, which resulted KlineFrench. An even more from the merger of Glaxo and Wellcome, established its position by means of a clever impressive example is given R&D strategy that allowed it to enter an industry already dominated by SmithKline; by Sanofi: if every one of its predecessor firms kept its and by means of an aggressive marketing strategy that increased its market share. For name, the company would a period of time, Zantac 's position was protected by patent rights. This is no longer be called - take a deep the case, meaning that differentiating the product with respect to the incoming rivals breath - Sanofi Synthelabo (generics producers) is now a priority. Hoechst Marion Roussel Rhone Poulenc Rohrer In this section, I attempt to formulate the object of industrial organization in a Marion Merril Dow! more systematic way. One can say that the goal of industrial organization is to address (exclamation mark added). 4 CHAPTER 1 I the following four questions: (a) Do firms have market power? (b) How do firms acquire and maintain market power? (c) What are the implications of market power? (d) What role is there for public policy with regard to market power? Since all of these questions revolve around the notion of market power, it may be useful to make this notion more precise. Market power may be defined as the ability to set prices above cost, specifically above incremental or marginal cost, that is, the cost of producing one extra unit. b So, for example, if GlaxoWellcome spends $10 to produce a box of Zantac and sells it for $50, then we say that it commands a substantial degree of market power. Now for the questions. I s TH E R E MARKET POW ER? Understandably, this is an important question, in fact, a cru­ cial one. If there is no market power, then there is little point in the study of industrial organization. Over the years, many empirical studies have attempted to measure the extent of market power. Assuming that costs are proportional to output, a good approximation to the extent of market power can be obtained from data on prices , output and profit rates.c b. A rigorous definition of One study along these lines found that the extent of market power in the American marginal cost and other cost economy is very low, a conclusion that follows from observing relatively low profit concepts is given in Section rates. 5 This finding is consistent with one of the central tenets of the so-called Chicago 3.1. If costs are proportional to output, then marginal school of economics: as long as there is free entry into each industry, the extent of market cost is equal to unit cost. power is never significant. If a firm were to persistently set prices above cost, a new firm c. The profit rate, r, is would find it profitable to enter the market and undercut the incumbent. Therefore, given by revenues minus market power cannot persist, the argument goes. d cost divided by costs: Not every economist agrees with this view, either at a theoretical or at an empirical r = (R- C) / C, where R is revenues and C is costs. If level. From an empirical point of view, an alternative approximation to the value of costs are proportional to marginal cost is obtained by dividing the increase in cost from year t to year t + 1 by the output, then costs are given increase in output in the same period. Based on this approach, a study estimates that by unit cost times output: UC X Q, where UC is unit prices may be as much as three times larger than marginal case cost and Q is output. Evidence from particular industries also suggests that the extent of market power Revenues, in turn, are given may be significant. Take, for example, the US airline industry. A 1996 US government by price times quantity: report analyzed average fares in 43 large airports. In 10 of these airports, one or a few R = P X Q, where P is price. It follows that airlines held a tight control over takeoff and landing slots. The report found that, on r = (P - UC) / UC, so r is a average, fliers were paying 31% more at these airports than at the remaining 33 airports.8 good measure of the gap In other words , the report suggests that airlines which manage to control the critical between price and unit cost (which in this case is also asset of airport access hold a significant degree of market power. equal to marginal cost). Further examples could be supplied. These would not necessarily be representa­ tive of what takes place in every market. To be sure, there are a large number of indus­ d. The theory of contestable markets tries where firms hold little or no market power (cf Section 4.1). The point is that there formalizes this argument. 6 are some industries where market power exists to a significant extent.  WHAT IS I N DUSTRIAL ORGANIZATION? I s HOW DO FI RMS ACQU I R E A N D MAI NTAI N MARKET POW E R? Market power translates into higher profits. Creating and maintaining market power is therefore an important part of a firm's value-maximization strategy. How do firms acquire market power? One possibility is to be legally protected from competition, so that high prices can be set without new competitors entering the market. For example, in the 1 940s and 1950s Xerox developed the technology of plain-paper photocopying, and patented it. Given the legal protection provided by Xerox's patents , it could raise prices to a significant level without attracting competition (cf Box 1 5. 1). Firm strategy may also play an important role in establishing market power. Take, for example, the case of the British Sky Broadcasting Group (BSkyB) , a leading firm in the British digital TV market (formed in 1990 by the merger of Sky Television and British Satellite Broadcasting). Attempting to preempt the competition, in 1999 B SkyB introduced an aggressive package that included a free set-top decoder box, free Internet access, and a 40% discount on telephone charges. 9 The plan, which was largely suc­ cessful, was to create an early lead in installed base of subscribers , an early lead that would hopefully become permanent. A more recent example is given by Samsung Elec­ tronics. Attempting to break into the lucrative smartphone market, Samsung sold their iPhone-like handsets at significantly lower prices than Apple's; and took a bet by being among the first to use Google's new Android operating system. By 2 0 1 2 , Samsung had already captured a third of the smartphone market. 1 0 Creating market power is only one part of the story. A successful firm must also be able to maintain market power. Patents expire. Imitation takes place. Protected indus­ tries are deregulated. What can incumbents do in order to maintain their position? The airline industry provides an example. American Airlines managed to drive out vari­ ous competitors who attempted to enter into its Dallas/Forth Worth hub : Vanguard, Sun Jet, Western Pacific. Fares on the route between Dallas and Kansas City, for exam­ ple, fell from $ 1 08 to $80 when Vanguard entered the market. After Vanguard exited, American gradually raised fares up to $147. Joel Klein, then head of the Antitrust Divi­ sion at the Justice Department, said that American's strategy achieved more than just driving current rivals out of the market - it also sent a clear signal to potential future entrants: "A sophisticated economist compared it to choosing between two fields with 'no trespassing' signs. One has two dead bodies in it, the other has no dead bodies in it. Which field would you feel ready to trespass?" 11 Reputation for toughness is a reliable means of maintaining a position of market power. A more recent example from the high-tech world is given by Apple Computer. By a combination of constant innovation and clever marketing, the Cupertino giant has managed to maintain a strong market position for a long time. Particularly important is Apple's "ecosystem" of devices and software: "its phones, tablets , computers, and the mobile and desktop operating systems that run them are blending into a single, inseparable whole. " 1 2 6 CHAPTER 1 I In different chapters of this text, especially in Chapters 11-16, I will examine a large set of strategies that firms may deploy in order to create and maintain their market power. WHAT ARE TH E I M P LICATIONS OF MARKET POWER? From the firm's point of view, market power implies greater profits and greater firm value. From a social welfare point of view - or from a policymaker point of view, if we believe policymakers pursue the collective good - the implications are more complicated. The first-order effect of a high price is a transfer from consumers to firms :8 for each extra dollar in price, each buyer is transferring one extra dollar to the seller. If regulators put a greater weight on consumer welfare than on profits, then this transfer should be seen as a negative outcome. In fact, antitrust and competition policies are to a great extent motivated by the goal of protecting consumers from these transfers (see the next question). r But, in addition to a transfer effect, a high price also implies an inefficient allo­ cation of resources. High airfares , for example, mean that there are potential fliers who refrain from buying tickets even though the cost of carrying them as passengers would be very low. From a social point of view, it would be efficient to fly many of these poten­ tial travelers : although the value they derive from flying is lower than the price (hence they don't fly) , that value is greater than the cost of flying (which is much lower than price). The loss that results from the absence of these sales is the allocative inefficiency implied by market power. g "The best of all monopoly profits is the quiet life : " 1 3 A monopolist does not need e. By "first-order" I mean the effect that is to be bothered with competition. More generally, firms with greater market power have quantitatively most less incentives to be cost efficient, one may argue. For example, for many years European significant. airlines were known to be less efficient than North-American airlines. To a great extent, f. An alternative this efficiency gap resulted from the more intense competition in the North-American perspective on antitrust and market. h In other words , market power implies a second type of inefficiency: productive competition policy is that it inefficiency, which we define as the increase in cost that results from market power.i serves to protect the interests of firms. See When market power is artificially maintained by government intervention, a third Section 5. 4. type of inefficiency may result: rent seeking. By rent seeking I mean unproductive g. A rigorous definition of resources spent by firms in an attempt to influence policymakers. Consider, for example, this concept is given in the following news article regarding AT&T's effort to maintain its position in the cable Section 4.4. television market: h. Since European airline This summer, AT&T Corp. faced the specter of cities around the country requiring it to open deregulation in the 1 990s its cable television lines to rival Internet companies... The threat never really materialized. the situation has changed Why not? It depends on whom you ask. considerably. AT&T attributes its success to its ability to explain the issues to local officials... i. Again, we defer the more [Others have a different opinion:] "It comes down to bribery or threats , " says Greg Simon, precise definition to a later co-director of Opennet Coalition, a group of companies that has launched its own lobbying chapter. The discussion of effort to promote open access. 1 4 the above hypothesis (market power leads to productive inefficiency) can Another example of large amounts of resources spent in an attempt to influence deci­ be found in Chapter 5. sionmakers is the 1998 US v. Microsoft case. Netscape, Sun Microsystems, and Microsoft  WHAT IS I N DUSTRIAL ORGANIZATION? I 7 itself would not have spent the vast amounts they did spend if the operating system industry were not as profitable as it is; thus the idea that rent seeking is a consequence of market power. Along similar lines, a more recent example is provided by Amazon's effort to maintain its dominance in the e-book retail market. When its business strategy was chal­ lenged by Apple's "agency model," Amazon approached the US Department of Justice (DOJ) with extensive legal arguments that the DOJ later used to sue Apple for anticom­ petitive practices. 1 5 ·i To summarize, the above paragraphs support the view that market power, good as it might be for firms, is bad for society. First, it makes the owners of firms richer at the expense of consumers. Second, it decreases economic efficiency (allocative and productive efficiency). Third, it induces firms to waste resources in order to achieve and maintain market power. However, from a dynamic point of view, an argument can be made in favor of market power: As soon as we go into the details and inquire into the individual items in which progress was most conspicuous , the trail leads not to the doors of those firms that work under conditions of comparatively free competition but precisely to the doors of the large concerns.16 This argument is one of the central points of the Austrian school, led by its greatest exponent, J. Schumpeter, author of the above quotation. I will examine it in greater detail in Section 15.1. Like the Chicago school, the Austrian school has a very clear position when it comes to market power. However, whereas a Chicago economist would argue that market power does not exist, a Schumpeterian would rather say that market power exists - and it's a good thing that it does, for market power is a precondition for innovation and progress. IS TH E R E A ROLE FO R PU BLIC POLICY AS REGARDS MARKET POW E R? In the context of indus­ trial organization, the primary role of public policy is to avoid the negative conse­ quences of market power. Public policy in this area can be broadly divided into two categories : regulation and antitrust (or competition policy). k Regulation (economic reg­ ulation) refers to the case when a firm retains monopoly or near-monopoly power and its actions (e.g. , the price it sets) are directly under a regulator's oversight. For exam­ ple, ConEdison, a US electricity and gas supplier, needs regulatory approval in order to change its rates. Antitrust policy (or competition policy) is a much broader field. The idea is to prevent firms from taking actions that increase market power in a detrimental way. A j. J will return to this case couple of examples may help. in Section 9.5. k. The terminology For the past two decades , Mars and Unilever have engaged in a series of legal "antitrust" is more common cases in European courts. The issue is the legality of Unilever's exclusivity in the US, whereas "competition policy" is the policies regarding retail ice cream sales. In many European countries , Unilever corresponding European imposes fridge exclusivity: if a store accepts a fridge paid by Unilever, then term; see Section 5.5. 8 CHAPTER 1 I the store can only use the fridge to stock Unilever products. Mars claims that exclusivity effectively makes it impossible for Mars to sell Snickers ice cream and related products as most stores have no space for more than one fridge. Unilever responds that it's their fridges and that they require return on an expensive investment. Different but similar cases in various European countries have come to different conclusions , sometimes favorable to Unilever, sometimes favorable to Mars (see also Section 1 2. 1). In March 2011, AT&T announced that it planned to purchase T-Mobile USA, a smaller wireless operator. Five months later, the US Department of Justice (DOJ) formally announced that it would seek to block the takeover, arguing that it would increase market power substantially. At first AT&T gave signs that it would defy DOJ's decision, but eventually the bid was abandoned. Although the merger might have brought some benefits to consumers , com­ petition between the would-be merging parties has also been a positive force. For example, in 2013 T-Mobile USA announced that it would pay contract can­ cellation fees of AT&T subscribers who wanted to switch to T-Mobile. The above two examples provide an idea of the variety of situations that may fall under the scope of public policy. The overall rationale is to prevent and remedy situations where market power may reach unreasonable levels , to the detriment of society - con­ sumers in particular. Over the course of the next chapters , we will examine several other areas for policy intervention motivated by the goal of curbing market power. As was stated earlier, the Chicago school takes a very different approach. The claim is that, in a world of free competition, market power is never very significant. In fact, the few situations where market power does exist result precisely from govern­ ment intervention. In other words , the Chicago school reverses the order of causation: it's not that market power prompts government intervention but the exact opposite - government intervention creates market power, protecting the interests of firms and not those of consumers. As Milton Friedman, a leader of the Chicago school, put it in the late 1990s: Because we all believed i n competition 5 0 years ago , we were generally i n favor of antitrust. We 've gradually come to the conclusion that, on the whole, it does more harm than good. [Antitrust laws] tend to become prey to the special interests. Right now, who is promoting the Microsoft case? It is their competitors , Sun Microsystems and NetscapeY To summarize, The central questions addressed by industrial organization are: (i) Is there market power? (ii) How do firms acquire and maintain market power? (iii) What are the implications of market power? (iv) Is there a role for public policy with regard to market power?  WHAT IS I N DUSTRIAL ORGANIZATION? I 9 I N D USTRIAL POLICY. In addition to regulation and antitrust (or competition policy) , some countries have followed policies that are intended to promote particular firms or groups of firms. Of particular importance is ind ustrial policy. The goal of industrial policy is very different from regulation and antitrust. Whereas the latter attempt to promote competition, the former is geared towards strengthening the market position of a firm or industry, namely with respect to foreign firms. For example, much of the success of Airbus Industrie, a consortium backed by four European countries, is the result of the support it has received from the respective governments over the past three or four decades. Starting from a market share of less than 1 0 % in the 1970s, Airbus is now fighting head to head with Boeing, the industry's main competitor. Industrial policy is generally not favored by economists. In practice, it amounts to governments picking winners among a number of potential firms and industries. But why should governments know better than the market who the promising firms and industries are? A frequent argument in support of industrial policy is the exam­ ple of MITI, the Japanese Ministry of Industry and Foreign Trade. True, the prowess of the Japanese export sector is a success story and owes a great deal to the role played by MITI. For example, MITI's support was an important factor in the emergence of Japan as a leader in semiconductors. But together with the success stories there are also a fair number of flops: for example, the 1980s project to develop a "fifth gener­ ation computer, " which would leapfrog the American counterparts, lead to very poor results. 1 8 Even in the US - the most pro-market western economy - we find examples of failed industrial policy. Recently, the state of Rhode Island approved $75 million in loans for a fledgling video game company led by a former Major League Baseball star pitcher. The ill-fated company went bankrupt two years later and managed to rack up $ 1 5 0 million in debt in the process. 1 9 In sum, although there are success stories (Airbus?) , the overall record of govern­ ments meddling in business strategy is at best poor. For these reasons , and as a matter of consistency, when talking about public policy I will restrict attention to regulation and antitrust (or competition policy). 1.3 CO M I N G N EXT... There are 1 5 chapters to come, divided into four different parts. Part I is introductory in nature. It provides basic tools required for the study of 10 (consumer behavior in Chapter 2, firm behavior in Chapter 3 ) ; and covers the extreme situations when mar­ kets are efficient (Chapter 4) and when they are not (Chapter 5). I conclude Part I by discussing advanced pricing strategies, though still in a context where strategic interac­ tion is absent (Chapter 6). For readers with a background in the field of microeconomics, some of the material treated in Chapters 2-5 may be familiar and can be skipped without much loss. 10 CHAPTER 1 I Insofar as industrial organization is the study of imperfect competition, Parts II through IV make up the core of the text. Within these, Part II plays a central role, as it introduces the basic theory of oligopoly competition. I begin with an introduction to game theory (Chapter 7}, an essential tool for studying strategic behavior; and then cover static models (Chapter 8) and dynamic models (Chapter 9) of oligopoly interaction. Throughout most of the text, I assume a given industry structure. Part III takes one step back and examines the endogenous determinants of industry structure. I begin by looking at how technology and demand conditions influence market structure (Chapter 10}, and then move on to examine the role played by mergers and acquisitions (Chapter 1 1 ) and firm strategy (Chapter 1 2 ). Part I V extends the analysis b y considering firm strategies beyond the simple pricing and output decisions examined in Parts II and III. These include vertical rela­ tions (Chapter 1 3 ) ; product differentiation and advertising (Chapter 14); and innovation (Chapter 15). I conclude with a chapter on networks , a phenomenon of increasing impor­ tance in the "new economy" (Chapter 16). A NOTE O N M ETHODOLOGY. Many economists analyze industries with reference to a framework known as the structure-cond uct-performance (SCP) paradigm. 2 ° First, one looks at the aspects that characterize market structure : the number of buyers and sellers, the degree of product differentiation, and so forth. Second, one pays attention to the typical conduct of firms in the industry: pricing, product positioning and advertising, and so forth. Finally, one attempts to estimate how competitive and efficient the industry is. Underlying this system is the belief that there is a causal chain linking the above components: market structure determines firm conduct, which in turn determines industry and firm performance. For example, in an industry with very few competitors, each firm is more likely to increase prices or collude with its rivals. And higher prices l. It should be clear that the have the performance implications we saw in the previous section. SCP paradigm is not a Causality also works in the reverse direction. For example, a firm that does not model that directly perform well exits the market, so performance influences market structure. Likewise, a provides answers to the questions listed in the firm may price very low in order to drive a rival out of the market, an instance where previous section. It is best conduct influences structure. Finally, government intervention and basic demand and thought of as a guide that supply conditions also influence the different components of the SCP paradigm. allows one to analyze and understand the workings of In Chapter 1 0 I examine the relation between the different components in the different industries. structure-conduct-performance paradigm. However, most of the text centers on the anal­ Alternative frameworks ysis of firm conduct and how it influences firm and industry performance as well as have been proposed for the market structure. 1 same or similar purposes. Examples include Michael Porter's five forces framework for the analysis of industry competition. The SUMMARY five forces are : suppliers, Industrial organization is concerned with the workings of markets and industries , in buyers, substitute products, potential entrants, and competition between particular the way firms compete with each other. The central questions addressed incumbent firms. 21 by industrial organization are: (i) Is there market power? (ii) How do firms acquire and  WHAT IS I N DUSTRIAL ORGANIZATION? 1 11 maintain market power? (iii) What are the implications of market power? (iv) Is there a role for public policy with regard to market power? KEY CONCEPTS market power contestable markets allocative inefficiency productive inefficiency rent seeking industrial policy structure-conduct-performance (SCP) paradigm five-forces framework REVIEW AND PRACTICE EXERCISES 1. 1. COM PETITI O N AN D P E R FORMANCE. Empirical evidence from a sample of more than 600 UK firms indicates that, controlling for the quantity of inputs (that is, taking into account the quantity of inputs) , firm output is increasing in the number of competi­ tors and decreasing in market share and industry concentration. 22 How do these results relate to the ideas presented in this chapter? NOTES 1. Stigler, George J. (1969), The Organization of Industry, Homewood, Illinois: R D Irwin, p. 1. 2. The People's Pharmacy (http :/ /homearts.com/depts/health/kfpeop 1 8.htm). 3. Petersen, Melody (2008), Our Daily Meds: How the Pharmaceutical Companies Transformed Themselves into Slick Marketing Machines and Hooked the Nation on Prescription Drugs, New York: Sarah Crichton B ooks. 4. The Scientist, Vol. 9, No. 1 4 , p. 3 , July 1 0 , 1 9 9 5. 5. Harberger, Arnold C. ( 1 9 5 4 ) , "Monopoly and Resource Allocation , " American Economic Review 44, 7 7-8 7. 6. Baumol, William, John Panzar, and Robert Willig ( 1 9 8 2 ) , Contestable Markets and the Theory of Industry Structure, New York: Harcourt Brace Jovanovich. 7. Hall, Robert E. (1 9 8 8 ) , "The Relationship Between Price and Marginal Cost in US Industry, " Journal of Political Economy 96, 92 1-94 7. 8. The Wall Street Journal Europe, November 1 4 , 1 996. 9. The Wall Street Journal Europe, May 6 , 1999. 1 0. The Economist, September 17, 2014. 1 1. Financial Times, 24 May 1 9 9 9. 1 2. "Apple Strengthens the Pull oflts Orbit With Each Device , " The New York Times, October 2 3 , 2014. 1 3. Hicks , John ( 1 9 3 5 ) , "Annual Survey of Economic Theory: The Theory of Monopoly, " Econo­ metrica 3 , 1-2 0. 1 4. The Wall Street Journal, November 24, 1 9 9 9. 1 5. The Wall Street Journal, September 1 1 , 2 0 1 4. 12 CHAPTER 1 I 1 6. Schumpeter, Joseph ( 1 9 5 0 ) , Capitalism, Socialism, and Democracy, 2nd Ed. (New York) , pp. 82 and 1 0 6. 1 7. The Wall Street Journal Europe, June 1 0 , 1 9 9 8. 1 8. The Economist, August 3 1 st, 1 9 9 6. 1 9. The New York Times, April 2 0 , 2 0 1 3. 2 0. This framework i s based o n the seminal work b y Mason and Bain. S e e Mason, Edward S. ( 1 9 3 9 ) , "Price and Production Policies of Large-Scale Enterprise , " American Economic Review 29, 6 1-74. Mason, Edward S. ( 1 949), "The Current State of the Monopoly Problem in the United State s , " Harvard Law Review 62, 1 2 65-1 2 8 5. Bain, J o e S. ( 1 9 5 6 ) , Barriers to New Competition, Cambridge , Mass. : Harvard University Press. Bain, Joe S. ( 1 9 5 9 ) , Industrial Organization, New York: John Wiley & Sons. 2 1. Porter, Michael E. ( 1 9 8 0 ) , Competitive Strategy, New York, NY: The Free Press. 22. Nickell, Stephen J. (1996), " Competition and Corporate Performance," Journal of Political Econ­ omy 104, 724-746.  2 Markets are made of buyers and sellers. In many cases. buyers are consumers and sellers are firms. In this and in the next chapter I lay the behavioral foundations of consumers and firms, respectively. 2.1 CO N S U M E R P R E F E R E N C ES AN D D E MA N D Creating Demand: Move the Masses to Buy Your Product, Service, or Idea. You will find this and many other such titles at your local bookstore (if it still exists , which is an interesting demand-related question). If there is no demand there is no business. Therefore, before deciding what to do in business it's important to have some knowledge of what the demand for your product is. The economist's answer to this is to think of consumers as having preferences; and from this to derive their demand function: how much they want to buy of a certain product as a function of a variety of factors , including price. CO N S U M E R TASTES. The traditional theoretical setup for thinking about demand starts with the tastes or preferences of an individual consumer. To see how this works , suppose there are two products, A and B. We might express different combinations of purchases in a diagram with quantities of product A on the vertical axis and quantities of product B on the horizontal axis. The top left panel in Figure 2.1, for example , shows four possible combinations , from £ 1 to E 4 A consumer's preferences can be described by a ranking over all possible combinations of A and B. For example, a consumer might like the bundle £ 2 best among the four possibilities; be indifferent between £ 1 and E 4 ; and like E 3 the least. 16 PART 1 CHAPTER 2 I I qA qA creasing utility /; I n c rease in Y I n itial budget constraint Yj pA I n c rease i n p, 2 E3 q, q, 2 Yj p, q, q, q, q, FI GU R E 2.1 From consumer preferences to consumer demand We can represent these preference orderings (that is, tastes) by drawing the con­ sumer's indifference curves , as shown on the top left panel in Figure 2.1. For example, we might think of two units of product A and one of product B (point E 1 ) as equally desirable as the reverse (one unit of product A and two units of product B, point £ 4). A line connecting all such points about which we are indifferent (i.e., like equally well) is called an indifference curve. Generally, indifference curves are downward-sloping, since we need more of a product to compensate us for less of the other. (We call this the "no satiation" principle, which really means that "more is better. ") We also assume they are curved away from the origin, since as we consider combinations with more and more of one product, we need increasing amounts of it to keep us equally satisfied. If we put lots of indifference curves on our graph, we can get a complete descrip­ tion of our tastes. Since "more is better, " indifference curves that are up and to the right (farther from the origin) represent higher levels of satisfaction. The other ingredient in our analysis is what the consumer CO N S U M E R B U DG ETS. can afford: the budget set. With two products, the budget might be expressed by the inequality: PAqA + Ps qs :S Y where p; is the price of product i, q; the quantity purchased of product i (i = A, B) , and Y the available income. If we solve for qA we see the budget set corresponds to the area below the following downward-sloping line: Y Ps _ qA- --- qs PA PA  17 CONSUM ERS I which is line b on the top right panel in Figure 2.1. This is a straight line whose position and slope depend on income and prices. If you increase income, for example, the budget line shifts out. This is illustrated by line b' in the top right panel of Figure 2.1. If you increase p8 , the line rotates clockwise (around the vertical intercept, Y /pA). This is illustrated by line b" in the top right panel of Figure 2.1. And if you increase pA, the line rotates counterclockwise (around the fixed horizontal intercept, Y jp8). D E MAND. Putting together tastes (represented by indifference curves) and possibilities (represented by the budget line) we can find out what our hypothetical consumer should do. Suppose that the consumer's goal is to maximize utility. Then the consumer's best choice corresponds to the point where the highest indifference curve has a common point with the budget line. This is illustrated in the bottom left panel in Figure 2.1, where E is the best combination of A and B that the consumer can afford with a budget represented by the budget line b (that is, the area under b). The optimal point, E , corresponds to quantities q and q;. These are the quantities demanded by the consumer. Implicitly, these demands depend on tastes (which are built into the indifference curves); they also depend on income (since a change in income shifts the budget line and therefore leads to a change in demand) and prices (for the same reason). We can summarize and abstract from the underlying indifference curves and bud­ get sets by writing down a demand function, qi (pi , z ) , denoting the quantity demanded, qi , for a given price of the good, P i · and for given values of other variables, z, which might include income, the prices of all other goods, and any other relevant factors that affect the demand of good i. For example, if i is "gasoline consumption" then z might include variables such as consumer income and the price of cars. (Can you think of other ones?) The demand curve gives the quantity demanded of a given good as a function of its price and of other factors ; it can be derived from consumer preferences. Instead of qi(P i · z ) , we can equivalently write an i nverse demand function, P i ( q i , z ) , which denotes what the price must be if the quantity demanded is to be equal to qi. To summarize, the quantity of a product demanded by an individual consumer depends on: The individual's tastes, expressed by indifference curves. The price of the product. Generally, the lower the price the higher the quantity demanded. Depending on the curvature of the indifference curves , a change in price might have a small or aj large impact on the quantity demanded. The price of other products. Decisions are not made in isolation: if we spend less on one product, that necessarily leaves more to spend on others. Income. At higher levels of income, we can buy more of everything (and gener­ ally do). 18 PART 1 CHAPTER 2 I I Frequently, we graph q; as a function of p; only." We refer to the curve that shows q; as a function of p; as the demand curve. Basically, it corresponds to the demand function under the assumption that all variables other than p; are constant (that is, z is constant). The bottom right panel of Figure 2. 1 illustrates the process of deriving the demand curve. Keeping Y and PA fixed, we change the value of pa. For example, we increase the value of pa from p to p. This implies a clockwise rotation of the budget line around the qA axis intercept. The consumer's optimal point is now given by E'. In other words , by increasing the value of p8 from p to p , the quantity of q8 demanded decreases from q to q. Repeating this exercise for all possible values of p8 , we get the demand curve for product B. It is important to distinguish between movements along the demand curve and shifts in the demand curve itself. If p; changes , then the quantity demanded of product i changes as we move along the demand curve. If however some other variable such as Pi (the price of another good) changes , than q; changes because of a shift of the demand curve. A change in price leads to a movement along the demand curve; a change in other factors leads to a shift in the demand curve itself. Suppose that before going to watch a movie you stop at a pizzeria Co N S U M E R S U R P LUS. and place your order. Pizza comes at $1 a slice. Imagine what would be the maximum price you would pay for one pizza slice. Perhaps $ 3 , especially if you are very hungry and there is no alternative restaurant in the neighborhood. Consumers don't usually think about this; all they need to know is that they are willing to pay at least $1 for that pizza slice. But, for the sake of argument, let us suppose that the maximum you would be willing to pay is $3. How about a second slice of pizza? While one slice is the minimum necessary to survive through a movie, a second slice is an option. It makes sense to assume you would be willing to pay less for a second slice than for the first slice; say, $ 1. 5 0. How about a third slice? For most consumers, a third slice would be superfluous. If you are going to watch a movie, you might not have the time to eat it, anyway. If you were to buy a third slice, you would probably only eat the toppings and little else. You wouldn't be willing to pay more than, say, 20 cents. Putting all of this information together, we have your demand curve for pizza. The left panel in Figure 2. 2 illustrates this. On the horizontal axis, we have the number of a. Although we think of qi pizza slices you buy. On the vertical axis, we measure the willingness to pay, that is, the as a function of Pi we maximum price (in dollars) at which you would still want to buy. normally measure price on the vertical axis. British There are two things we can do with a demand curve. First, knowing what the economist Alfred Marshall price is ($1 per slice), we can predict the number of slices bought. This is the number started doing it this way of slices such that willingness to pay is greater or equal to price. Or, to use the demand back in the nineteenth century and things haven't curve, the quantity demanded is given by the point where the demand curve crosses the changed ever since. line p = 1 : two slices.  19 CONSUM ERS I $ p 1.5 p' price = 1. 0 +-------'- 0.2 2 3 q' FI GU R E 2.2 Consumer surplus: (a) demand for pizza slices and (b) general case. A second important use ofthe demand curve is to measure the consumer's surplus. You would be willing to pay up to $3 for one slice of pizza. That is, had the price been $ 3 , you would have bought one slice of pizza anyway. In fact, you only paid $1 for that first slice. Since the pizza is the same in both cases , you are $ 3 - $ 1 $2 better off than = you would be had you bought the slice under the worst possible circumstances (or not bought it at all). Likewise, you paid 50 cents less for the second slice than the maximum you would have been willing to pay. Your total surplus as a consumer is thus $2+50cents = $2.50. Consumer surplus is the difference between willingness to pay and price. More generally, consumer surplus is given by the area under the demand curve and above the price paid by the consumer. This is illustrated in the right panel in Figure 2. 2. I f price i s p* , then quantity demanded i s q * and consumer surplus i s given b y the shaded area A. 2.2 D E MA N D E LAST I C I TY Having established the foundations of the demand curve, I am now interested in char­ acterizing its shape, that is, understanding how it depends on a variety of factors : how it depends on the price of the product in question, but also on the prices of other goods, as well as on the consumer's income. For this purpose, economists use the concept of demand elasticity. PRICE E LASTICITY OF D E MAND. We presume (based on lots of evidence) that the demand for a product increases if we lower its price. This doesn't have to be the case, but it invariably is. The question is how sensitive demand is to price. If I were to tell you that the world oil demand decreases by 1. 3 million barrels a day when price increases from $50 to $60 per barrel, would you consider the demand for oil very sensitive or not very 20 PART 1 CHAPTER 2 I I sensitive to price? It's hard to tell, unless you have a very good idea of the size of the world oil market. What if I told you that the demand for sugar in Europe decreases by one million tons per day when average retail price increases from 80 to 90 euros cents per kilo: can you compare the demand for sugar in Europe to the worldwide demand for oil? It's even more difficult. The problem is that, by measuring the slope of the demand curve, we are stuck with units : barrels, dollars , kilos, euros , and so on. The solution is to measure things in relative terms, that is, in terms of percent changes. Specifically, we measure the sensi­ tivity of demand to changes in price by the price elasticity of demand: E..:l q d q p_ E = -- dp dp q p where d q represents a differential variation of q and d q / d p stands for the derivative of q with respect to p (sometimes denoted q ' (p)). In words : The price elasticity of demand is the ratio between the percentage change in quantity and the percentage change in price, for a small change in price. A note on terminology: as I will show later, there are many elasticity concepts , including but not limited to the concept of price elasticity of demand. When economists simply refer to " demand elasticity" they really mean "price elasticity of demand. " In the rest of the chapter (and in the rest of the book) I may fall in the same abuse of language. In the above definition, by "small" change in price we mean, strictly speaking, an infinitesimally small change in price, something we represent by d p. In practice, we observe small but not infinitesimal changes in p, which we denote by Ll p. We thus have: where the sign stands for "approximately equal to. " Note that elasticities are generally negative, since quantity demanded declines when price increases. The question is how negative. We say that products for which I E I > 1 have an "elastic" demand, meaning the quantity demanded is very sensitive to price; the higher I E I is, the more sensitive to price. Conversely, if 0 < I E I < 1 (elasticity is "small"), we say that demand is "inelastic ," meaning that quantity demanded is rela­ tively insensitive to price; the closer E is to zero , the less sensitive demand is to price. b. Frequently, if product 1 Later we will see why the value 1 (the threshold between inelastic and elastic demand) e1 = -3 and product 2 is so important when classifying demand elasticity. b has has €z = -2 we say, with some abuse of language, Table 2.1 presents the values of the demand elasticity for selected products. that the demand for product Would you expect the demand for coffee to be inelastic? Would you expect the long­ 1 is more elastic. Strictly run demand elasticity for natural gas to be greater than the short-run elasticity? Would speaking, e1 is smaller than ez , but in absolute value the you expect the demand for foreign luxury cars in the US to be elastic? Why? I will return opposite is true. to some of these questions in Section 2.3.  21 CONSUM ERS I Note that the elasticity is defined at a point: generally speaking, its value varies along a demand curve. The left panel in Figure 2.3 considers the case of a linear demand curve. As we go from the extreme when p is equal to 0 to the extreme when q is equal to zero , the value of E varies from 0 to - oo. (You can check this by looking at the definition of elasticity.) At some intermediate point (the midpoint, if the demand curve is linear) , we have I E I = 1. Although the value of demand elasticity varies from point to point, there is such a thing as a demand curve with constant elasticity, that is, with the same value of demand elasticity at every point. The right panel in Figure 2.3 depicts several examples. There are two extreme cases: a vertical demand curve ( E = O), such that for any price the quan­ tity demanded is always the same; and a horizontal curve (E = - oo } , the extreme case such that even a very small change in price leads to an infinite increase in quantity demanded. These extreme examples are not found in any real-world situation, though some market demands may be close to it. (Can you think of examples?) For the major­ ity of real-world markets, demand elasticity lies somewhere between the two extremes. Table 2.1 provides a few examples. T A B L E 2.1 Price elasticity of demand for selected prod ucts and services. 2 Product a n d m a rket N o rwegian s a l m o n in S p a i n - o.8 N o rwegian s a l m o n i n Italy Coffee i n t h e Netherlands - o.2 N atural gas i n Europe (s h o rt· r u n) - 0.2 N atural gas in Europe (lo n g· r u n) US luxury cars in U S Foreign luxury cars i n U S - 2.8 Basic c a b le T V i n U S Satellite T V i n U S -sA Ocean s h i p p i n g services (wo rldwide) p p E = - oo E = 0 q f i GU R E 2.3 Demand elasticity 22 PART 1 CHAPTER 2 I I At this point, it should be clear that elasticity is not the same thing as slope, although slope is an input to it (recall that the slope is given by dq/ dp). The difference between slope and elasticity is illustrated by the left and right panels in Figure 2. 3. On the left panel, we have a curve with constant slope but varying elasticity; on the right panel, we have a series of curves with constant elasticity but varying slopes. A useful variant of the elasticity formula uses log­ TECH N ICAL PO I N T: LOGAR ITHMS. arithms. You might recall that: dq d ( !og q ) = q where log stands for natural logarithm. Therefore, dq C) _ d log q dp - d log p p If we do our usual approximation, replacing d's with ·s, we have: log q E --- log p The approximation is exact if the elasticity is constant along the demand curve; see Exercise 2. 8. N U M E R ICAL EXAM PLE. To s e e how the concepts of demand elasticity can be used in practice, consider the following example. Suppose you have access to historical sales data from the Yellow Note jazz label. When price was set at $ 1 0 , $1 1 , and $ 1 3 , total units sold (thousands) were 6. 3 1 , 5. 6 3 , and 4. 6 1 , respectively. (As you can see, Yellow Note has not been a great commercial success.) Suppose that these historical data corre­ spond to different points of the demand curve (that is, suppose that the demand curve has remained constant over these periods). This is a big - and possibly unrealistic ­ assumption, an assumption to which I will return later; but it will make our lives eas­ ier for the time being. Given this big if, how do we estimate the value of the demand elasticity at the price of $10? We can approximate the value of the demand elasticity by the "percent change" formula applied to prices $10 and $ 1 1 : E q P._ = p q ( 5. 61 31 -- 61 0. 3 1 ) ( __2Q_ 6.31 ) = - l.08 This i s an approximation, since we're using discrete changes. We could get a better approximation by using a finer price grid. In fact, the numbers in the present example were generated by a demand curve with constant elasticity e = - 1. 2. So, just as the true value, the estimate of - 1.08 is greater than one in absolute value (that is, we correctly estimate that the demand for Yellow Note CDs is elastic) ; however, there is a consider­ able estimation error ( - 1.08 as opposed to - 1. 2).  23 CONSUM ERS I How could I use the elasticity estimate to predict quantity demanded when price is $ 1 3 ? Since E ::::: ;; , we conclude that: fi q fi p - :=::i E - q p Specifically, consider a variation in price from $ 1 1 to $ 1 3. This corresponds to fipjp = 2 / 1 1 = 1 8. 1 8%. We therefore estimate that fiqjq = - 1.08 X 1 8. 1 8 % = - 1 9.64%, which implies a value of q given by 5.63 X ( 1 - 1 9.64%) = 4.53. This overestimates the decrease in q: as we know, the true value is 4. 6 1. LOGAR ITHMS (CO NT). Alternatively, w e can d o the above calculations using loga­ rithms. First, we estimate the demand elasticity as: log 5.63 - log 6. 3 1 -. 1 140 _ E ::::: = = 1.20 log 1 1 - log 10.0953 In fact, given that the demand has constant elasticity, the logarithmic formula yields the exact value of the demand elasticity. Next, we can estimate the new value of q when p = 13 by solving the equation: t: :=::: log 5.63 - log q = -1.20 log 1 1 - log 1 3 The result i s q = 4. 6 1 , the true value o f q. Once again, w e see that i f the demand has constant elasticity, then the logarithms approach delivers more precise results. The above numerical example shows how the value of the demand elasticity can be used to predict the change in quantity demanded following a change in price. Similarly, the value of E can be used to predict the change in revenue (the product price X quantity). It can be shown that, if I E I < 1 (that is, if the demand is inelastic} , then an increase in price leads to an increase in revenue; whereas , if I E I > 1 (that is, if the demand is elastic} , then an decrease in price leads to an increase in revenue. This is a basic point, but one that some have missed: that to increase revenue in markets with elastic demand, you need to lower price, not raise it. E LASTICITY A N D REVE N U E. Formally, the percent change in revenue induced by a (small) change in price is: d (p q) q dp + p d q dp dq dp dq dp dp = = + = + E_ = (1 + E) pq pq p q p dp q p p That is, the percent change in revenue following a price change is (1 + elastic­ ity) X (% change in price). Since E < 0, the direct effect of the price change (the " 1 " term) and the indirect effect through changes in quantity demanded (the " t: " term) have opposite signs. If demand is elastic ( I E I > 1 } , then the demand effect is larger and an increase in price leads to lower revenue. 24 PART 1 CHAPTER 2 I I CROSS-PRICE E LASTICITY. We have seen that demand for a product depends not only on its own price, but also on the prices of other goods. For example, the demand for ski boots depends on the demand for skis: if skis get more expensive, we expect people to buy fewer skis - and fewer boots , too. As an alternative example, we expect the demand for commuter rail tickets to be influenced by the price of gasoline: if gasoline becomes more expensive, people drive their cars less and take the train more often. Notice that, in the first example, the quantity demanded of good 1 (ski boots) decreases when the price of good 2 (skis) increases; whereas, in the second example, the quantity demanded of good 1 (rail tickets) increases when the price of good 2 (gaso­ line) increases. More generally, we summarize the sensitivity of demand to the price of another product with the cross price elasticity: E.!ll , q E1 2 ==: -- P2 That is, the cross price elasticity of product 1 with respect to product 2 is given by the ratio of the percent change in quantity demanded of product 1 to the percent change in the price for product 2. The essential distinction here i s between substitutes and complements. I f the cross-price elasticity is positive, we say that the products are substitutes. Hence Coke and Pepsi are substitutes: if Coke becomes more expensive, we'd expect some people (but not all) to switch to Pepsi, thus increasing the quantity demanded of Pepsi. (I know that you would never do such thing, but some consumers just don't have any princi­ ples.) Conversely, if the cross-price elasticity is negative, we say the products are com ­ plements. For example, if gasoline price increases we expect people to drive less, which in turn decreases the quantity demanded of cars. In this sense, cars and gasoline are complement products. Speaking of cars , Table 2. 2 presents the values of direct and cross-price elasticities for selected car models. For example, the number 0. 2 on the third column, second row means that a 1 % increase in the price of the Accord leads to a 0. 2 X 1% increase in Cavalier sales. Are the different car models complements or substitutes? What is the model closest to the Cavalier, in terms of demand? See Exercise 2. 5 for more details. T A B L E 2.2 Automobile demand elasticities Model Mazda 323 - 6.4 0.6 0.2 0.1 0.0 0.0 Cava lier 0.0 - 6-4 0.2 0.1 0.1 0.0 Accord 0.0 0.1 - 4.8 0.1 0.0 0.0 Taurus 0.0 0.1 0.2 - 4. 2 0.0 0.0 C e n t u ry 0.0 0.1 0.2 0.1 - 6.8 0.0 BMW 735 i 0.0 0.0 0.0 0.0 0.0 - J,S  25 CONSUM ERS I Changes in income also affect demand. Higher income generally I NCOM E E LASTICITY. means greater demand for all products , but some products benefit more than others. We define the income elasticity of a product by: dq q Ey = -- - E.x_ y That is, the income elasticity of demand is given by the percent change in quantity demanded induced by a 1 % change in income (represented by y). Economists have names for products with different income elasticities. I n ferior goods have negative income elasticities. Although inferior goods aren't all that com­ mon, it's fun to think of examples. Spam comes to mind, on the assumption that any­ one with enough money would buy something else. Normal goods have positive income elasticities. Within normal goods , those with elasticities between zero and one are referred to as necessities, and those with elasticities greater than one as luxuries. Can you explain why? APPLICATI O N : D E MA N D FOR GASOLI N E. Let US consider a specific example where the above concepts play an important role. Using historical data on gasoline price and con­ sumption in the US from 2001-2006, the following relation was estimated: log q = - 1.697 - 0.042 log p + 0. 5 3 0 log y where q is gasoline consumption, p is the price of gasoline , and y is income. 2 What is the price elasticity of gasoline demand? From the analysis above, d log q E= = - 0. 042 d log p Is gasoline demand elastic or inelastic? Well, since I E I < 1 , we conclude that it is inelas­ tic. No major surprises here: with the exception of New Yorkers , Americans need to drive to work, and there aren't many reasonable alternatives to get there (at least not in the short run). What is the income elasticity of gasoline demand? The answer is: d log q 11 = = 0.530 d log y Is gasoline a normal good? Yes , since the income elasticity is greater than zero. In fact, since 11 < 1, gasoline is a necessity. Notice that the threshold from necessity to luxury is not purely arbitrary. In fact, if the income elasticity is greater than 1 then an increase in income implies that the fraction of income spent on that good increases. So, on average richer people spend more on gasoline (17 > 0) but spend a lower proportion of their income on gasoline (17 < 1). This example helps explain an apparent "paradox, " which is illustrated in the left panel of Figure 2.4. In the first decade of the millennium, the price of gasoline increased and so did consumption levels: compare, for example, the 2000 and 2010 data points. Doesn't this violate the regularity that demand curves are downward sloping; that is, 26 PART 1 CHAPTER 2 I I p i ndex (1 983 =100) p ($) 350 Dzooo Dzo10 Dt998 000 \ ·.. 300 30 250 2010 200 200 20 150 1 998