Interactions between competition and consumer policy

Competition Policy International

Volumn 4, Issue 1, 2008, Pages 97-147

  Armstrong, Mark ^a  

^a UNIVERSITY COLLEGE LONDON   (United Kingdom)

Author keywords

[No Author keywords available]

Indexed keywords

EID: 70349469049     PISSN: 15540189     EISSN: 15546853     Source Type: Journal    
DOI: None     Document Type: Conference Paper

References (138)

1. Timothy Muris, The Interface of Competition and Consumer Protection, Paper presented at Fordham Corporate Law Institute's 29th Annual Conference on International Antitrust Law and Policy, New York (Oct. 31, 2002).
2. MASSIMO MOTTA, COMPETITION POLICY: THEORY & PRACTICE 30 (2004).
3. Muris, supra note 1.
4. J. Vickers, Economics for Consumer Policy, 125 PROC. BRIT. ACAD. 287-310, 289 (2004).
5. C. Camerer, S. Issacharoff, G. Loewenstein, T. O'Donoghue & M. Rabin, Regulation for Conservatives: Behavioral Economics and the Case for "Asymmetric Paternalism", 151(3) U. PENN. L. REV. 1211-54 (2003).
6. For a discussion of recent convergence in the area of consumer fraud, see Muris (2002), supra note 1.
7. The current European Union Unfair Commercial Practices Directive is an important step towards harmonizing consumer policy within Europe. See Directive 2005/29/EC of the European Parliament and of the Council of 11 May 2005 concerning unfair business-to-consumer commercial practices in the internal market, 2005 O.J. (L 149) 22 [hereinafter UCPD].
8. It is conventional to divide products into three classes, depending on the extent of the information problem. "Search goods" are products whose attributes are fully observable at the time of purchase; "experience goods" have attributes which are only revealed after purchase; and "credence goods" have attributes which are not fully revealed even after purchase.
9. See Vickers (2004), supra note 4, at 297.
10. On this last point, see J. Chevalier & D. Mayzlin, The Effect of Word of Mouth on Sales: Online Book Reviews, 43(3) J. MARKETING RES. 345-54 (2006).
11. U. Dulleck & R. Kerschbamer, On Doctors, Mechanics, and Computer Specialists: The Economics of Credence Goods, 44(2) J. ECON. LIT. 5-42 (2006). The result referred to is Lemma 7 in that survey. In addition, in situations where consumers commit to get treatment when they get diagnosis from an expert, and where experts post prices for treatment, competition can, under stringent conditions (such as all consumers having the same probability of needing a serious repair), deliver the ideal outcome (see Proposition 1).
12. Illustrative anecdotes are provided in W. Emons, Credence Goods and Fraudulent Experts, 28(1) RAND J. ECON. 107-19 (1997). For instance, in a region in Switzerland the general population had significantly more medical operations than medical doctors and their families, consistent with a degree of overtreatment among the uninformed. Likewise, see H. Schneider, Agency Problems and Reputation in Expert Services: Evidence from Auto Repair (2007) (mimeo, Cornell University) reports results from a field experiment where the same faulty car was taken to 40 garages for diagnosis and quotes for repair. He found that 27 percent of garages suggested unnecessary repairs, while real faults were missed in 77 percent of cases.
13. See Dulleck & Kerschbamer (2006), supra note 11, at Proposition 4.
14. See the later discussion of the "Diamond paradox" in Section III.C of this paper.
15. P. Bajari & A. Hortacsu, Economic Insights from Internet Auctions, 42(2) J. ECON. LIT. 457-86, §5 (2004), surveys the empirical work on the effectiveness of the reputation mechanism in online auctions, which is mixed in its conclusions.
16. G. Jin & P. Leslie, Reputational Incentives for Restaurant Hygiene, AM. ECON. J.: MICROECONOMICS (forthcoming 2009) provide evidence showing that restaurant hygiene tends to be higher in local markets which have a greater proportion of repeat buyers. Schneider (2007), supra note 12, tests for the impact of reputation by informing some (randomly chosen) garages that he has just moved to the area (thus indicating scope for repeat business) and informing others that he is about to leave the area. He finds that while the potential for repeat business appears to lead to a lower charge for diagnosis, it appears to have no significant affect on the quality of the diagnosis.
17. In 1984, Kellogg launched an advertising campaign focusing on the health benefits of one of its cereals, All-Bran. This was in direct violation of the U.S. Food and Drug Administration's then-policy which essentially banned health claims on food products. A subsequent relaxation of this ban acted to facilitate information flows to consumers, which led many consumers to change their consumption behavior. Moreover, it seems that government and general information sources before the ban was lifted had little impact on consumer behavior. See P. Ippolito & A. Mathios, Information, Advertising and Health Choices: A Study of the Cereal Market, 21(3) RAND J. ECON. 459-80 (1990).
18. See Kyle Bagwell, The Economic Analysis of Advertising, in 3 THE HANDBOOK OF INDUSTRIAL ORGANIZATION 1701-1844 (M. Armstrong & R. Porter eds., 2007). Likewise, if users of a search engine tend to click on suggested links in the order in which they appear on the page, then websites which are most likely to fit the user's need will often pay the most to be listed first. (And in this case, consumers indeed should click on the links in the suggested order.)
19. See S. Athey & G. Ellison, Position Auctions with Consumer Search (2007) (mimeo, MIT and Harvard University).
20. See, e.g., Money Saving Expert: Consumer Revenge (United Kingdom), at http://www.moneysaving expert.com (last visited Jan. 17, 2008).
21. P. Milgrom & J. Roberts, Relying on the Information of Interested Parties, 17(1) RAND J. ECON. 18-32 (1986). In fact, this "unraveling" argument works even with a monopoly seller, provided that consumers think strategically and have accurate information about the market environment (see Section 2 in that paper). When there is competition, consumers do not need to reason in such a sophisticated way for this result to hold (see Sections 3 and 4 in that paper).
22. Y. Chen, Paying Customers to Switch, 6(4) J. ECON. MGMT. STRATEGY 877-97 (1997).
23. For evidence that consumers have demand for commitment devices, see N. Ashraf, D. Karlan & W. Yin, Tying Odysseus to the Mast: Evidence from a Commitment Savings Product in the Philippines, 121(2) Q.J. ECON. 635-72 (2006).
24. D. Laibson, Golden Eggs and Hyperbolic Discounting, 112(2) Q.J. ECON. 443-77 (1997).
25. S. DellaVigna & U. Malmendier, Contract Design and Self-Control: Theory and Evidence, 119(2) Q.J. ECON. 353-402 (2004)
26. and K. Eliaz & R. Spiegler, Contracting with Diversely Naive Agents, 73(3) REV. ECON. STUD. 689-714 (2006). An alternative and less benign reason why gyms might set low per-visit charges, with respect to exploiting over-optimism on the part of consumers, is discussed in Section III.E of this paper.
27. For discussion of this point, see H. Beales, R. Craswell & S. Salop, The Efficient Regulation of Consumer Information, 24(3) J.L. & ECON. 491-539, §1B (1981). The UCPD, supra note 7, at para. 18, takes the view that a commercial practice is unfair if the average consumer in that market is misled (among other hurdles which need to be passed). If a product is marketed at a particular sub-group of consumers (e.g., the elderly), then the average consumer should be taken with reference to that subgroup. Specifically, the UCPD states that "[t]he average consumer test is not a statistical test. National courts and authorities will have to exercise their own faculty of judgment [ . . . ] to determine the typical reaction of the average consumer in a given case."
28. For a long list of the scams being used in Australia at the time of writing (2008), see ScamNet - Complete Listing of Scams, at http://www.docep.wa. gov.au/ConsumerProtection/ScamNet/content/ pages/full-list.html (last visited Jan. 25, 2008).
29. quot;Claiming that products are able to facilitate winning in games of chance" is one of the 31 practices which are in all circumstances considered unfair in the UCPD, supra note 7.
30. For details of recent intervention against a racing tipster, see Press Release, U.K. Office of Fair Trading, OFT obtains injunction against bogus racing tipster (Jan. 22, 2008), available at http://www.oft.gov.uk/news/press/ 2008/12-08. In return for a membership fee of GBP 590, members received tips on likely winners in horse races. Some 3,000 consumers signed up for the service. The marketing materials for the service were judged to be misleading, and falsely claimed for instance that the tipster owned a team of race horses and had ridden regularly for some of the most successful horse trainers. In addition, members were told they would make a "minimum of £47,000 in 30 days."
31. See supra note 26.
32. See the discussion of "quacks" in Section III.C.
33. R. Posner, The Federal Trade Commission, 37(1) U. CHI. L. REV. 47-89, 62 (1969).
34. Posner (1969), Id. at 68.
35. J. Brown & A. Goolsbee, Does the Internet Make Markets more Competitive? Evidence from the Life Insurance Industry, 110(3) J. POL. ECON. 481-507 (2002) provide evidence that the growth of pricecomparison websites for life insurance drove prices down for this product. An interesting study in the pre-Internet era had related findings. In 1974, two districts in Canada were chosen for a market transparency experiment in supermarket pricing. In one district, prices were collected but not publicized; in the other district, average price indices for individual supermarkets in that area were publicized in local newspapers. The result was that price dispersion and price levels fell in the second district relative to the first.
36. See G. Devine & B. Marion, The Influence of Consumer Price Information on Retail Pricing and Consumer Behavior, 61(2) AM. J. AGRIC. ECON. 228-37 (1979).
37. OFFICE OF GAS & ELECTRICITY MARKETS (OFGEM), DOMESTIC COMPETITIVE MARKET REVIEW 2004: A REVIEW DOCUMENT Tables 2.14 & 2.16 (Apr. 2004);
38. and U.K. OFFICE OF FAIR TRADING (OFT), INTERNET SHOPPING: AN OFT MARKET STUDY ¶ 9.18 (Jun. 2007), reports that only 47 percent of Internet shoppers had used a price comparison website.
39. M. Baye, J. Morgan & P. Scholten, Price Dispersion in the Small and the Large: Evidence from an Internet Price Comparison Site, 52(4) J. INDUS. ECON. 463-96 (2004).
40. The recent literature on two-sided markets is relevant here, one example of which is M. Armstrong, Competition in Two-Sided Markets, 37(3) RAND J. ECON. 668-91, §5 (2006). Similar issues arise with the (typically non-price) market information found in Yellow Pages directories, where most consumers consult just one directory.
41. See M. Rysman, Competition Between Networks: A Study of the Market for Yellow Pages, 71(2) REV. ECON. STUD. 483-512 (2004).
42. M. Baye & J. Morgan, Information Gatekeepers on the Internet and the Competitiveness of Homogeneous Product Markets, 91(3) AM. ECON. REV. 454-74 (2001) present a model with a monopoly intermediary, and show that the intermediary restricts competition among sellers in order to increase the listing fee it can charge them.
43. It is worth noting that consumer-testing organizations often do not permit a good product review to be used in a product's advertising. This is in part because the consumer organization wishes to maintain revenue from consumers buying its magazine or subscribing to its website, but it does act to impede information flows about product quality to the wider consumer population.
44. Recent papers present evidence that obfuscation strategies, such as presenting a low headline price together with high small-print charges for postage, can confuse consumers and reduce firm-level demand elasticities. See G. Ellison & S. Fisher Ellison, Search, Obfuscation and Price Elasticities on the Internet (2004) (mimeo, MIT) and T. Hossain & J. Morgan, Plus Shipping and Handling: Revenue (Non)equivalence in Field Experiments on eBay, 6(2) ADVANCES IN ECON. ANALYSIS & POL'Y 1429 (2006). OFT (Jun. 2007), supra note 34, at 125-27, documents a survey which revealed that the majority of listed prices do not make clear whether delivery is included, and in a sample of airline ticket bookings the final price was above the original listed price in 47 percent of cases (furthermore, for these cases, the median price increase was 19 percent).
45. Related issues arise with attempts to measure the life-time cost of a product. Such is the case with printers, which may have different per-page costs, or cars, which may have different running costs, where the lifetime cost depends on the usage made of the durable product. Matters are even worse with multi-product markets such as telecommunications and supermarkets, where consumers must key in their estimated demands for many products to get a valid comparison of suppliers.
46. In the United Kingdom, several regulators have accredited price-comparison websites listed on their own websites, and one regulator, the Financial Services Authority, has its own comparison website. See FSA: Money made clear, at http://www.moneymadeclear.fsa.gov.uk (last visited Jan. 28, 2008). In addition, the introduction of an approved price-comparison website was one of the remedies to market failure in U.K. COMPETITION COMMISSION, HOME CREDIT MARKET INVESTIGATION 10 (2006).
47. H. Varian, A Model of Sales, 70(4) AM. ECON. REV. 651-59 (1980).
48. A closely related model is S. Salop & J. Stiglitz, Bargains and Ripoffs: A Model of Monopolistically Competitive Price Dispersion, 44(3) REV. ECON. STUD. 493-510 (1977).
49. For instance, as emphasized by Brown & Goolsbee (2002), supra note 33, the well-informed consumers could be those who use price-comparison websites on the Internet, while the remainder are those who must painstakingly search supplier by supplier. The discussion in the text assumes that suppliers cannot set different prices for online and offline consumers, as was the case in Brown & Goolsbee's data. If suppliers could set different prices to informed and uninformed consumers (e.g., by making the price depend on whether the sale was online or offline), the model would have a very different prediction: informed consumers would pay competitive prices and uninformed consumers would pay monopoly prices, and the prices in each case would not depend on the number of suppliers or the fraction of informed consumers. Thus, informed consumers would not protect the uninformed at all. Here, a policy which boosts the fraction of informed consumers is good for aggregate consumer welfare, but it has no impact on those consumers who remain uninformed.
50. Brown & Goolsbee (2002), supra note 33, find support for this effect in the early part of their data where the fraction of consumers who used price-comparison websites was small.
51. Markets differ greatly in their consumer search costs. For instance, petrol has a rather low search cost since many consumers are already in their car when buying the product. As a result, prices are reasonably competitive and exhibit little dispersion within a local area. Funerals and condoms, by contrast, are often purchased by consumers with little appetite for market research. Car tires are frequently purchased by consumers who are not very mobile. In one study, 71 percent of consumers wanting to buy tires contact no other outlet apart from the one at which they buy. For details and further discussion, see M.Waterson, The Role of Consumers in Competition and Competition Policy, 21(2) INT'L J. INDUS. ORG. 129-50 (2003). However, the case of petrol presents an interesting danger: given that the market is understood by consumers to be reasonably competitive, consumers may not even bother checking the prices when they fill up, and this leaves a profit opportunity for unscrupulous (but legal) suppliers. For instance, in 1996, a Mr. Mole in Lancashire (United Kingdom) operated a number of petrol stations where petrol was nearly twice as expensive as his competitors.
52. See Garage fuels price protest, LANCASHIRE TELEGRAPH, Jan. 19, 1996.
53. See Salop & Stiglitz (1977), supra note 41 and K. Burdett & K. Judd, Equilibrium Price Dispersion, 51(4) ECONOMETRICA 955-69 (1983).
54. Related insights are obtained in the model with sequential search in S. Anderson & R. Renault, Pricing, Product Diversity, and Search Costs: a Bertrand-Chamberlin-Diamond Model, 30(4) RAND J. ECON. 719-35 (1999). In that model, there is no price dispersion and the market price is an increasing function of the search cost.
55. P. Diamond, A Model of Price Adjustment, 3(2) J. ECON. THEORY 156-58 (1971).
56. G. Butters, Equilibrium Distributions of Sales and Advertising Prices, 44(3) REV. ECON. STUD. 465-91 (1977).
57. For instance, empirical studies of opticians and legal services in the United States have investigated how relaxing a state ban on price advertising typically led to lower prices. See L. Benham, The Effect of Advertising on the Price of Eyeglasses, 15(2) J.L. & ECON. 337-52 (1972);
58. L. Benham & A. Benham, Regulating Through the Professions: A Perspective on Information Control, 18(2) J.L. & ECON. 421-47 (1975);
59. and D. Haas-Wilson, The Effect of Commercial Practice Restrictions: The Case of Optometry, 29(1) J.L. & ECON. 165-86 (1986). See also the review of related studies on the impact of advertising on prices in Bagwell (2007), supra note 18, at 1745-46.
60. More generally, while new technologies often facilitate information flows to consumers, they can sometimes also help consumers avoid adverts. For instance, personal video recorders in broadcasting and "pop-up blockers" on the Internet allow consumer to avoid adverts. Since many consumers dislike viewing adverts, but there is a competitive benefit to consumers as a whole when more consumers receive adverts, it is possible that the increased use of ad-avoidance devices may harm aggregate consumer welfare. A simple model of broadcasting in which the ability to opt out of advertising harms viewers can be found in M. Armstrong, Public Service Broadcasting, 26(3) FISCAL STUD. 281-99, 295 (2005).
61. R. Spiegler, The Market for Quacks, 73(4) REV. ECON. STUD. 1113-31 (2006a).
62. R. Spiegler, Competition Over Agents with Boundedly Rational Expectations, 1(2) THEORETICAL ECON. 207-31 (2006b). Spiegler interprets his model in terms of a product with many pricing attributes, where a consumer's inability to process many pieces of information implies that she just looks at a small number of price attributes for the product, while the expected price eventually paid depends on all pricing attributes.
63. Suppose the unit production cost is c. If all firms offer transparent prices, there is Bertrand competition and price is equal to c. Suppose instead that one firm sets a noisy price, where its true price is 2c and its (unbiased) signal is either zero or 4c with equal probability. This seller will have the lowest signal in the market half the time, and in these cases it will make the sale and make a positive profit while doing so.
64. A. Mathios, The Impact of Mandatory Disclosure Laws on Product Choices: An Analysis of the Salad Dressing Market, 43(2) J.L. & ECON. 651-77 (2000) discusses how a change to a mandatory policy for food labeling affected disclosure of fat content in salad dressing.With a voluntary labeling regime there was substantial disclosure by low-fat suppliers, but the dressings with no disclosure had a wide range of fat content. Mandatory labeling thus improved the information available to consumers.
65. Likewise, G. Jin, Competition and Disclosure Incentives: An Empirical Study of HMOs, 36(1) RAND J. ECON. 93-112 (2005) shows that only some health suppliers chose to make use of a (costly) accreditation service.
66. R. Pitofsky, Beyond Nader: Consumer Protection and the Regulation of Advertising, 90(4) HARV. L. REV. 661-701, 664 (1977). In fact, Posner (1969), supra note 31, at 69, claims that the FTC acted in the early 1960s to prevent the disclosure of tar and nicotine content in cigarette advertising.
67. G. Jin & P. Leslie, The Effect of Information on Product Quality: Evidence From Restaurant Hygiene Grade Cards, 118(2) Q.J. ECON. 409-51 (2003).
68. G. Jin & P. Leslie (2009), supra note 16, at 1. The authors conclude that "even when there is merit to the argument that reputational incentives operate as a market-based mechanism for mitigating information problems, they may be a poor substitute for full information."
69. R. Lal & C. Matutes, Retail Pricing and Advertising Strategies, 67(3) J. BUS. 345-70 (1994).
70. For a survey, see Joseph Farrell & Paul Klemperer, Coordination and Lock-In: Competition with Switching Costs and Network Effects, in 3 THE HANDBOOK OF INDUSTRIAL ORGANIZATION 1967-2072 (M. Armstrong & R. Porter eds., 2007).
71. See M. Eisenberg, The Limits of Cognition and the Limits of Contract, 47(2) STANFORD L. REV. 211-59 (1995)
72. as well as the discussion of that paper and related literature in B. Hermalin, A. Katz & R. Craswell, Contract Law, in 1 THE HANDBOOK OF LAW & ECONOMICS §2.3.4 (M. Polinsky & S. Shavell eds., 2007).
73. This effect is akin to the so-called "waterbed effect" seen in markets such as mobile telephony. There, if one charge is reduced by policy, it does not necessarily have an impact on equilibrium industry profits, but it will cause another price to rise correspondingly.
74. G. Ellison, A Model of Add-on Pricing, 120(2) Q.J. ECON. 585-637 (2005).
75. See Ellison (2005), Id. at Proposition 4 and C. Shapiro, Aftermarkets and Consumer Welfare: Making Sense of Kodak, 63(2) ANTITRUST L.J. 496 (1995).
76. X. Gabaix & D. Laibson, Shrouded Attributes, Consumer Myopia, and Information Suppression in Competitive Markets, 121(2) Q.J. ECON. 505-40 (2006). The numerical example which follows is essentially taken from section 1 of their paper. In a very different model Spiegler (2006a), supra note 50, also shows that firms do not wish to "de-bias" consumers. These recent results go against the claim in Beales et al. (1981), supra note 25, at 502, and many other authors, that "if information dissemination were costless to sellers, theory suggests that disclosure would be complete."
77. In a competitive market, the excess profits made on the add-on will be passed back to consumers in the form of inducements to buy the main item, and industry profits will not be excessive overall. In this case, the consumer harm consists just of the deadweight loss "triangles" rather than the monopoly profit "rectangles" which result from excessive pricing of add-ons. As such, the consumer welfare loss may be relatively small. See Shapiro (1995), supra note 62, at 497-98.
78. As an example of the latter policy, Gabaix & Laibson (2006), supra note 63, report that policy in Singapore required that hotels price their international phone calls at marginal cost plus a maximum of 30 Singaporean cents.
79. This was a remedy in U.K. COMPETITION COMMISSION, EXTENDED WARRANTIES ON DOMESTIC ELECTRICAL GOODS (2003).
80. OFT (Jun. 2007), supra note 34, at 127.
81. This assumes the printer controls the retail price of its cartridges, and would also require a standardized "page" to facilitate comparisons. The general issue of establishing standardized scoring systems, and how a public body might act as a useful coordination device for this, is discussed in Beales et al. (1981), supra note 25, at 523-27.
82. U.K. FINANCIAL SERVICES AUTHORITY, LEVELS OF FINANCIAL CAPABILITY IN THE UK: RESULTS OF A BASELINE SURVEY (2006) reports that when buying a financial product, about half of survey respondents read the terms and conditions "in detail", while 10 percent did not read them at all.
83. See also the references in R. Korobkin, Bounded Rationality, Standard Form Contracts, and Unconscionability, 70 U. CHI. L. REV. 1203-95, n. 45 (2003) and the references listed in note 38 supra.
84. DellaVigna & Malmendier (2004), supra note 24, and S. DellaVigna & U. Malmendier, Paying Not to Go to the Gym, 96(3) AM. ECON. REV. 694-719 (2006). Their 2006 paper provides evidence showing that many consumers choose a gym membership, even though they end up going so rarely they would be better off with pay-per-visit arrangement.
85. A related feature is observed in a monopoly context in Milgrom & Roberts (1986), supra note 20, at Proposition 3.
86. See, respectively, N.Weinstein, Unrealistic Optimism About Future Life Events, 39(5) J. PERSONALITY & SOC. PSYCHOL. 806-20 (1980);
87. and O. Svenson, Are We All Less Risky and More Skillful than our Fellow Drivers?, 47(2) ACTA PSYCHOLOGICA 143-48 (1981).
88. L. Ausubel, The Failure of Competition in the Credit Card Market, 81(1) AM. ECON. REV. 50-81 (1991).
89. M. Grubb, Selling to Overconfident Consumers (2007) (mimeo, MIT). As well as providing the theoretical analysis outlined here, Grubb presents empirical data in support of the hypothesis that consumers are overconfident in their predictions of future usage. See Eliaz & Spiegler (2006), supra note 24, for closely related theoretical work in the context of time-inconsistent consumers.
90. DellaVigna & Malmendier (2004), supra note 24, at §II.G, discuss possible regulatory intervention in the contracts offered by, say, gyms. They come to the pessimistic conclusion that "it requires extensive information which the government is unlikely to have [ . . . ]. A better policy for the government, in general, is to educate partially naive users and make them aware of their naiveté." No specific guidelines for how to do this are given, however.
91. Here, the form of the warning is likely to be very important. See C. Jolls & C. Sunstein, Debiasing Through Law, 35(1) J. LEGAL STUD. 199-241 (2006).
92. A. Sandroni & F. Squintani, Overconfidence, Insurance, and Paternalism, 97(5) AM. ECON. REV. 1994-2004 (2007).
93. Taken from Camerer et al. (2003), supra note 5, at 1231.
94. G. Loewenstein, T. O'Donoghue & M. Rabin, Projection Bias in Predicting Future Utility, 118(4) Q.J. ECON. 1209-48 (2003).
95. Camerer et al. (2003), supra note 5. Of course, another reason for cooling-off periods is when there is a danger that the contract may have been signed under duress. In this case, the cooling-off period will both protect the consumer from being locked into a contract she did not freely sign, and lessen the incentive to engage in aggressive selling practices in the first place.
96. See DellaVigna & Malmendier (2004), supra note 24, at §IV.
97. When Korobkin (2003), supra note 69, at 1235-36, discusses this point, he does not analyze a monopolist's incentive to supply the low-quality product.
98. M. Armstrong & Y. Chen, Inattentive Consumers and Product Quality (2007) (mimeo, University College London and University of Colorado at Boulder).
99. Y. Chen & M. Riordan, Price-Increasing Competition (2006) (mimeo, University of Colorado at Boulder and Columbia University).
100. This is discussed in Armstrong & Chen (2007), supra note 83, rather than in Varian's article itself.
101. A famous psychology experiment is described in S. Iyengar & M. Lepper, When Choice is Demotivating: Can One Desire Too Much of a Good Thing?, 79(6) J. PERSONALITY & SOC. PSYCHOL. 995-1006 (2000).
102. M. BERTRAND, D. KARLAN, S. MULLAINATHAN, E. SHAFIR & J. ZINMAN,WHAT'S PSYCHOLOGYWORTH? A FIELD EXPERIMENT IN THE CONSUMER CREDIT MARKET (Nat'l Bureau of Econ. Research,Working Paper No. 11892, 2005).
103. For data indicating that many consumers do not choose the best electricity supplier for their needs even when they switch supplier, see CHRIS WILSON & CATHERINE WADDAMS PRICE, DO CONSUMERS SWITCH TO THE BEST SUPPLIER? (Centre for Competition Policy,Working Paper No. 07-6, 2007).
104. All of this is taken from U.K. NATIONAL AUDIT OFFICE, DIRECTORY ENQUIRIES: FROM 192 TO 118 (Mar. 2005).
105. Waterson (2003), supra note 44, at Table 7, presents a survey in which one-third of consumers thought it would take at least a day to switch electricity supplier, when in fact it would probably take less than an hour.
106. As quoted in RONALD COASE, BRITISH BROADCASTING: A STUDY IN MONOPOLY 186 (1950).
107. See COASE (1950), Id. at 177.
108. See Beales et al. (1981), supra note 25, at 497.
109. Pitofsky (1977), supra note 54, at 674. The report by Professor Posner alluded to was expanded to become Posner (1969), supra note 31. 95
110. See note 17, supra.
111. See SCOTTISH EXECUTIVE, REPORT OF THE RESEARCH WORKING GROUP OF THE LEGAL SERVICES MARKET IN SCOTLAND ch. (2006)
112. See Muris, supra note 1. Muris quotes the FTC as writing in a court statement that "[r]ather than protect[ ing] consumers by exposing funeral directors to meaningful competition, the [law] protects funeral directors from facing any competition from third-party casket sellers."
113. PRODUCTIVITY COMMISSION OF AUSTRALIA, REVIEW OF AUSTRALIA'S CONSUMER POLICY FRAMEWORK 9 (2007).
114. However, information remedies which educate naive consumers can also make sophisticated consumers worse off in the B-type situations described in Section III.D. But there is a difference between correcting an "unfair" cross-subsidy caused by inattentive consumers, and harming sophisticated consumers in order to support inattentive consumers.
115. Also, in democracies majorities may vote to outlaw choices made by minorities. For instance, in 1998 a California referendum voted to outlaw the sale of horsemeat in restaurants. See Alvin Roth, Repugnance as a Constraint on Markets, 21(3) J. ECON. PERSP. 37-58 (2007).
116. J. Klick & G. Mitchell, Government Regulation of Irrationality: Moral and Cognitive Hazards, 90 MINNESOTA L. REV. 1620-63 (2006).
117. Beales et al. (1981), supra note 25, at 513.
118. Likewise, in 2005 the European Commission required airlines to compensate consumers for cancelled flights, regardless of the cause, thus bundling together the flight and travel insurance.
119. H. Leland, Quacks, Lemons and Licensing: A Theory of Minimum Quality Standards, 87(6) J. POL. ECON. 1328-46 (1979).
120. As Korobkin (2003), supra note 69, at 1251-52, puts it: "ex ante mandatory terms are desirable when a simple rule or an only moderately complicated rule can insure that the mandated content will be efficient for a relatively large proportion of contracts."
121. See Armstrong & Chen (2007), supra note 83.
122. Of course, then such a firm would not put the charge in the small print. Alternatively, a credit card company might require its customers to agree to pay a minimum monthly payment (or even the entire balance) by direct debit each month. The supplier may thus hope to attract credit-worthy customers and also to remove the need to run a large "debt collection" office. However, this point conflicts with the discussion in Section III.D of this paper, where I discussed how a credit card company might make more money from its less solvent customers.
123. In 2006, the OFT announced that it regarded penalty charges (for all kinds of default, including spending beyond the agreed credit limit) higher than GBP 12 as being unfair, and that such charges would lead it to intervene. The legal framework for this decision is provided by the 1999 Unfair Terms in Consumer Contracts Regulations. A "core term" (e.g., the headline price, or interest rate in the case of credit cards) in a contract can essentially be set at any level, but other items in a standard contract (to which some consumers will not pay attention when they sign) are subject to a "fairness" constraint. In the case of breach of contract (which is what late payment technically is), the fairness constraint broadly allows the supplier to have its additional administrative costs covered by the penalty charge. The OFT estimated these administrative costs at GBP 12, and gave some broad (but unquantified) indications of how it estimated these costs. It states that "on the basis of our analysis we consider that the threshold is robust and there are unlikely to be grounds to consider any higher threshold for our action over the short or medium term." See U.K. OFFICE OF FAIR TRADING, CALCULATING FAIR DEFAULT CHARGES IN CREDIT CARD CONTRACTS: A STATEMENT OF THE OFT'S POSITION (Apr. 2006).
124. In addition to this list, one could point out that in the United Kingdom there has been so much recent publicity about credit card penalty charges that the number of consumers who are now unaware of such charges has shrunk substantially, thus mitigating to some extent the need for further intervention in this area.
125. This view is emphasized in Klick & Mitchell (2006), supra note 101. In particular, §1A of that paper surveys the ample, though not unanimous, evidence that subjects in laboratory experiments make better decisions when their stakes are higher.
126. Posner (1969), supra note 31, at 67.
127. However, just because there is moral hazard does not mean insurance should not be offered at all. One might balk at permitting sales to the general public of Japanese pufferfish, which is fatal if prepared even slightly incorrectly. Staying with the food theme, another related issue is the widespread use of "use-by" dates on food. Many consumers never use food beyond its use-by date. Given that the use-by date is chosen so that the foodstuff is almost certain to be edible regardless of local conditions (e.g., how often the consumer's fridge is opened), it is plausible that inefficiency arises from this policy. If use-by dates were less widespread (say, in the days when many consumers purchased meat from a butcher rather than a supermarket), consumers would likely have better skills in detecting whether food is edible (e.g., by smell). This is another instance of how arguably excessive protection leads to consumers possessing too few market skills. Of course, though, one cannot sniff a "ready meal".
128. There is a plausible lifecycle effect here. Younger consumers may not yet have learnt their market skills, whereas some of the elderly may have learnt, but forgotten, their skills. In particular, policy might sometimes exempt the latter group from this caution about over-insulating consumers against market risk. See S. AGARWAL, J. DRISCOLL, X. GABAIX & D. LAIBSON, THE AGE OF REASON: FINANCIAL DECISION-MAKING OVER THE LIFECYCLE (MIT Department of Economics,Working Paper No. 07-11, 2007) for some evidence that consumers reach their decision-making peak for financial products at around 53 years of age.
129. The first model presented in the appendix is similar to (but simpler than) D. Dranove & M. Satterthwaite, Monopolistic Competition when Price and
130. S. Huck, G. Lünser & J.-R. Tyran, Pricing and Trust (2008) (mimeo, University of College London), presents experimental evidence which is consistent with this discussion.
131. This insight is similar to that in the principal-agent literature, where when an agent's overall performance depends on several kinds of effort, excessive focus on one kind of effort may lead to poor overall performance. See B. Holmstrom & P. Milgrom, Multitask Principal-Agent Analyses: Incentive Contracts, Asset Ownership, and Job Design, 7(special issue) J.L. ECON & ORG. 24-52 (1991).
132. D. Dranove, D. Kessler, M. McClellan & M. Satterthwaite, Is More Information Better? The Effects of "Report Cards" on Health Care Providers, 111(3) J. POL. ECON. 555-88 (2003).
133. See ARCHON FUNG, MARY GRAHAM & DAVID WEIL, FULL DISCLOSURE: THE PERILS & PROMISE OF TRANSPARENCY (Cambridge University Press, 2007) for a detailed analysis and evaluation of various transparency policies such as this one.
134. S. Albæk, P. Møllgaard & P. Overgaard, Government-Assisted Oligopoly Coordination? A Concrete Case, 45(4) J. INDUS. ECON. 429-43 (1997).
135. See, e.g., Mark Armstrong, Recent Developments in the Economics of Price Discrimination §3.3, in II ADVANCES IN ECONOMICS & ECONOMETRICS 97-141 (R. Blundell et al. eds., 2006).
136. The philosophy behind such policies is sometimes termed "asymmetric paternalism" by Camerer et al. (2003), supra note 5, or "libertarian paternalism" (or even "anti-anti-paternalism") by R. Thaler & C. Sunstein, Libertarian Paternalism, 93(2) AM. ECON. REV. 175-79 (2003).1
137. See Camerer et al. (2003), supra note 5, at 1226. 122
138. See, e.g., MOTTA (2004), supra note 2, at §1.3.1.