What does behavioral economics mean for competition policy?

Competition Policy International

Volumn 6, Issue 1, 2010, Pages 111-137

  Bennett, Matthew ^a   Fingleton, John ^a   Fletcher, Amelia ^a   Hurley, Liz ^a   Ruck, David ^a  

^a Fair Trading ^*  (United States)

Author keywords

[No Author keywords available]

Indexed keywords

EID: 77951558451     PISSN: 15540189     EISSN: 15546853     Source Type: Journal    
DOI: None     Document Type: Article

References (77)

1. For a general review of this literature, see S. DellaVigna, Psychology and Economics: Evidence from the Field, 47 J. ECON. LITERATURE 315-372 (June 2009).
2. Note that behavioral economics does not describe a chaotic world in which consumers make random decisions. In general, the behavioral biases exhibited by consumers are systematic and are often boundedly rational.
3. The processing power biases include: choice overload (consumers make choices on sets of information); representational biases (consumers use visible value as a good indicator of hidden value); and rules of thumbs (consumers imitate what other consumers do rather than make their own decisions).
4. The framing biases include: relative utility (a consumer's utility is affected by reference points such as past actions); default biases (consumers adopt the default option); and placement biases (consumers' choices depend on where goods are placed on a list-e.g. a tendency to choose the first).
5. The time inconsistency biases include: projection bias (consumers expect that they will feel the same tomorrow as they do today); over optimism (consumers over estimate how much they will use a good, or underestimate how much it will cost them); and hyperbolic discount biases (consumers value today disproportionately greater than tomorrow).
6. The loss aversion biases include endowment biases (consumers value something more once they have owned it more than before they own it).
7. If there are search costs, Diamond found that consumers may not search the market but simply choose a firm randomly. The best response of firms is then to charge a monopoly price to these consumers. P. Diamond, A Model of Price Adjustment, 3(2) J. ECON. THEORY 156-158 (1971).
8. P.D. Klemperer, Markets with Consumer Switching Costs, 102(2) Q. J. ECON. 375-394 (1987) showed that, in the context of a single period model, switching costs could be thought of as a form of artificial product differentiation, reducing the intensity of competition between competitors.
9. In the two period model in J. Farrell & C. Shapiro, Dynamic Competition with Switching Costs, 19(1) RAND 123-137 (1988) the result is more complicated as firms may compete more intensely in order to exploit their established base in the second period, thus creating a "bargain and rip-off" cycle.
10. See also Office of Fair Trading, Switching Costs, Economic Discussion Paper 5 (April 2003), for a review of the literature.
11. For example, S. Agarwal, J. C. Driscoll, X. Gabaix, & D. Laibson, Learning in the Credit Card Market, Working paper series (2008) investigate learning in the credit card market. They find that although consumers learn (through negative feedback), this hard-earned knowledge does not fully persist (i.e. knowledge depreciates).
12. For example, S. DellaVigna & U. Malmendier, Paying not to go to the gym, 96(3) AMER. ECON. REV, 694-719 (2006), using data from three U.S. health clubs find that consumers frequently choose contracts that appear sub-optimal given their attendance frequency. Members who choose a contract with a flat monthly fee pay a price per expected visit of more than $17, even though they could pay $10 per visit using a 10 visit pass. They suggest this could be driven by consumer overconfidence about gym attendance.
13. For example, V.G. Morwitz, E.A. Greenleaf, & E.J. Johnson, Divide and prosper: Consumers' reactions to partitioned prices, 35 J. MARKETING RES. 453-463 (1998) find that, when prices are presented in parts, consumers' ability to recall the entire price for the good is diminished and demand is increased. This might suggest that consumers may be anchoring to the first piece of information seen (generally the base price) and then attributing less importance to later pieces of information (i.e. surcharges or add-ons).
14. Similarly, T. Hossain & J. Morgan, Plus Shipping and Handling: Revenue (Non) Equivalence in Field Experiment on eBay, ADVANCES IN ECON. ANALYSIS & POL'Y (2005) suggest that consumers treat the base price separately from the handling fee in a natural field experiment they conducted using eBay auctions. The authors found that charging a low reserve price compared to the retail price of the good and high shipping and handling costs resulted in a higher total sales price than the reverse situation (low shipping and handling but high reserve price). This result may also be driven by consumers ignoring or missing the additional costs, although there are alternative behavioral explanations such as endowment bias.
15. For example, M. Baye, J. Morgan, & P. Scholten, Price Dispersion in the Small and in the Large: Evidence from an Internet Price Comparison Site, 52(4) J. INDUS. ECON. 463-496 (2004), using a UK data set of consumer click throughs from Kelkoo.co.uk for 2003 to 2004, found that even though Kelkoo does not order results by price by default, a firm listed first on a search results page still benefited from 17.5 percent higher demand on average than when it was listed second. This is despite the ease with which the consumer can usually reorder the results by lowest price.
16. More generally, A. Tversky & D. Kahneman, The Framing of Decisions and the Psychology of Choice, 211 (44810) SCI. 453-458 (1981) show how psychological principles govern the perception of decision problems and the evaluation of options.
17. For example, as well as finding consumers are overconfident about gym use (as suggested above) DellaVigna & Malmendier (2006), supra note 10, suggest that consumers might overestimate their propensity to cancel automatically renewed contracts.
18. For a discussion of these and other consumer issues which can arise, see M. Armstrong, Interactions between competition and consumer policy: A report prepared for the OFT, OFT991 (April 2008).
19. The research uses a controlled economic experiment to test five pricing frames, whereby the true price is provided in a complex way. The pricing frames investigated are drip pricing, "sales," complex pricing, bait pricing, and time limited offers. Drip pricing is where the consumers see only part of the full price up front and price increments are dripped through the buying process. "Sales" occur where a sale price is given and a pre-sale price is also given as a reference to the consumer, for example "was £2 is now £1." Complex pricing is where the unit price may be difficult to determine, for example "3 for the price of 2." Bait pricing is where sellers may promote a special price but there are only a limited number of goods actually available at that price. Time limited offers are where a price is advertised as only being available for a pre-defined short period of time. The report found that all of these pricing practices have some adverse effect on consumer choice and that most of them do significantly impact on consumer welfare. It suggests that the root of the errors can be found in the existence of the behavioral biases, largely the endowment effect and cognitive errors. Office of Fair Trading, The Impact of Price Frames on Consumer Decision Making, Economic Discussion paper, (Forthcoming, April 2010).
20. For example, P. Klemperer, Competition when Consumers have Switching Costs: An Overview with Applications to Industrial Organization, Macroeconomics, and International Trade, REV ECON. STUDIES 515-539 (1995) finds that in a simple multi-period model with two firms both firms are able to maintain higher prices and earn higher profits in the presence of switching costs than without switching costs.
21. See also A. Banerjee & L. Summers, On frequent flyer programs and other loyalty-inducing arrangements, H.I.E.R. DP no.1337 (1987).
22. For example DellaVigna & Malmendier (2006), supra note 10, argue that a reason why gyms favor the use of term contracts with upfront payments may be to lock in overconfident consumers.
23. See also M. Grubb, Selling to Overconfident Consumers, 99(5) AMER. ECON. REV. 1770-1807 (2009) who analyzed U.S. mobile phone data to investigate whether the three part tariffs seen within the U.S. mobile phone industry were developed as a means of capturing consumers' overconfidence. He found this was the most plausible of different explanations for the tariff structure. Grubb argues that the model can be reinterpreted more widely to explain the use of flat rates and late fees in rental markets, and teaser rates on loans.
24. See, for example, M. Eisenberg, The Limits of Cognition and the Limits of Contract, 47(2) STAN. L. REV. 211-259 (1995).
25. Several commentators have argued that the low-cost airlines are particularly effective in using drip pricing to exploit the fact that consumers are more likely to buy the product after they have invested time in it, see, for example, D. Milmo, Ryanair Scraps Airport Check-in Desks, GUARDIAN (Sept. 30, 2009).
26. See also G. Ellison & S.F. Ellison, Search, Obfuscation, and Price Elasticities on the Internet, 77(2) ECONOMETRICA 427-452, 03 (2009) who argue that economists should think about firms' active incentives to obfuscate as well as consumers' incentives to search.
27. Indeed one strategy consultancy advertises courses on how to minimize banking competition by increasing the difficulties for consumers to compare across products, stating that: "The likelihood that banks continually try to undersell one another is greater if their price structures make it easy for customers to compare offers. In order to prevent easy comparisons, a bank should create price structures that are clearly distinguishable from those of its rivals. Price systems with several price components are especially effective." G. Wuebker & J. Baumgarten, Strategies against Price Wars in the Financial Service Industry, Simon-Kucher & Partners.
28. See, for example, C. Shapiro, Aftermarkets and Consumer Welfare: Making Sense of Kodak, 63(2) ANTITRUST L. J. 496 (1995).
29. See also H. Beales, R. Craswell & S. Salop, The Efficient Regulation of Consumer Information, 24(3) J. L. & ECON. 491-539, §1B (1981).
30. See X. Gabaix & D. Laibson, Shrouded Attributes, Consumer Myopia, and Information Suppression in Competitive Markets, 121(2) Q.J. ECON. 505-540 (2006).
31. This point is discussed in respect of switching costs within Klemperer (1987), supra note 8. But the point is more general.
32. This result can be derived from either the switching/search literature (supra note 8) or the behavioral literature. With regards to the behavioral literature, Ellison and Ellison (2009), supra note 19, examine price data for internet retailers. They show that some retailers engage in obfuscation in order to frustrate consumer search, thus resulting in much less price sensitivity on other products. At the extreme, R. Spiegler, Competition Over Agents with Boundedly Rational Expectations, 1(2) THEORETICAL ECON. 207-31 (2006) showed that under certain circumstances firms' prices may be entirely independent of competition.
33. In standard competition issues, prices above the competitive level result in underconsumption of the product (the Harberger triangle). In aftermarkets there may be two distortions: the underconsumption of the secondary product, but also overconsumption of the primary product sold below cost. In large markets the resulting allocative loss may be significant. For example, in the UK the Competition Commission estimated that the cross subsidy from the price of insurance on loans to loans resulted in an allocative inefficiency in excess of £200m. See Competition Commission, Market investigation into payment protection insurance, Final Report, ¶10.494 (January 2009).
34. While the relationship between competition and innovation is complicated; in general, competition before the innovation takes place drives faster innovation. For a discussion of the links among competition, productivity, and innovation see Productivity and Competition, an OFT Perspective on Productivity Debate, OFT877 (2007).
35. See, for example, J. Hausman & G. Leonard, The Competitive Effects of a New Product Introduction: A Case Study, 50(3) J. INDUS. ECON. 237-263 (2002).
36. See also A Petrin, Quantifying the Benefits of New Products: The Case of the Minivan, 110 J. POL. ECON. 705 (2002).
37. Sale and rent back is a product which allows consumers who are in difficulties in making mortgage payments to sell their home to a company and then rent it back from them. Sale and rent back: An OFT Market Study, OFT (October, 2008).
38. Personal Current Accounts in the UK: An OFT Market Study, OFT (July 2008).
39. For example, the UK market investigation reference into payment protection insurance ("PPI") is discussed later in this paper. In this case, the OFT could potentially have attempted to bring an Article 102 abuse case against each of the suppliers of credit and PPI. Alternatively, action under consumer legislation could have been used. However, the OFT took the view that a reference to the Competition Commission was most appropriate, since the CC could consider all aspects in considering whether a problem existed and, if so, how best to remedy the problem both from the competition and consumer perspective. See, Payment protection insurance: Report on the market study and proposed decision to make a market investigation reference, OFT869 (October 2006).
40. It is noteworthy that the Vertical Restraints Block Exemption Guidelines explicitly mention that tying may lead to supra-competitive prices, especially "in the case of long-term contracts or in the case of aftermarkets with original equipment with a long replacement time [as] it becomes difficult for customers to calculate the consequence of the tying." Commission notice of 13 October 2000: Guidelines on Vertical Restraints, COM(2000/C 291/01), Official Journal C 291 ¶217 (October, 2000).
41. Case COMP/C-3/37.792 Microsoft.
42. See Gabaix & Laibson (2006) supra note 22.
43. See also S. DellaVigna & U. Malmendier, Contract Design and Self-Control: Theory and Evidence, Q. J. ECON. 119: 353-402 (2004) who describe a model in which consumers make their purchase decisions based on naïvely low estimates for their use of the secondary good. Once consumers are locked into the primary good they find they use the secondary good more often than initially estimated. Firms can exploit this naivety by extracting the consumers' rents upfront, but then extracting an additional rent when consumers are locked in. The result is more than one monopoly rent.
44. There are a few exceptions to this. First, in Spiegler's model firms respond to a greater number of competitors by hiding their prices even more, supra note 24. Second, in other general models involving product differentiation all profits are competed away only when the market is fully covered-that is when industry demand is independent of price. When this unrealistic assumption is relaxed, the amount of profits the firm can retain in exploiting the secondary market is proportional to the degree of primary competition. See, for example, C. Genakos & T. Valletti, Testing the 'Waterbed' Effect in Mobile Telephony, CEIS Working Paper No.110 (January 2008).
45. See Gabaix & Laibson (2006), supra note 22.
46. Federal Trade Commission, Personal Protective Armor Ass'n, 59 Fed.Reg. 19,019 (1994).
47. This section draws from a recent paper by Armstrong & Huck. See M. Armstrong & S. Huck, Behavioral Economics as Applied to Firms: A Primer, 6(1) COMPETITION POL'Y INT'L 3-45 (Spring, 2010)
48. also published as M. Armstrong & S. Huck, Behavioral Economics as Applied to Firms: A report prepared for the OFT, OFT 1213.
49. For a discussion of the literature see Armstrong & Huck, Id.
50. Id.
51. See M. Levenstein & V. Suslow, Cartel bargaining and monitoring: The role of information sharing, THE PROS AND CONS OF INFORMATION SHARING (Mats Bergman ed.), Swedish Competition Authority (2006).
52. This may also address some of the criticism of the "cheap talk" literature. This literature argues that there is no point in exchanging information in order to monitor an agreement if it can't be verified, since no firm is going to cheat on the agreement but still provide the true information that reveals they have cheated. However, the role of private information exchange may not be to monitor the cartel but rather to create and sustain the trust needed to maintain the cartel. See J. Farrell & M. Rabin, Cheap Talk, 10 J. ECON. PERSPECTIVES 103-118 (1996).
53. N. Al-Najjar, S. Baliga, & D. Besanko, Market forces meet behavioural biases: Cost misallocation and irrational pricing, 39 RAND 214-223 (2008) describe recent physiological and experimental literature suggesting that firms may confuse fixed, sunk, and variable costs. They show that even, if pricing strategies that increase profits do better in the market, there may be equilibrium in which there are non-standard pricing practices.
54. See, for example, U. Malmendier & G. A. Tate, Who Makes Acquisitions? CEO Overconfidence and the Market's Reaction, AFA 2004 San Diego Meetings (March 15, 2003).
55. See Gabaix & Laibson (2006), supra note 22.
56. See Judgment: The Office of Fair Trading v Abbey National plc & Others, UK Supreme Court, UKSC 6 (2009).
57. Of course where a mistake leads to irrevocable and significant detriment (for example, buying the wrong pension) learning may not be sufficient to prevent serious harm.
58. See DellaVigna (2009), supra note 1.
59. Self-regulation may also create competition concerns as it may provide opportunities for anticompetitive practices, such as foreclosure or price-fixing. For more details, see Office of Fair Trading, The Economics of Self Regulation in Solving Consumer Quality Issues, Economic Discussion Paper OFT1059 (March 2009).
60. JOHN STUART MILL, ON LIBERTY (1859).
61. For a popular explanation of liberal paternalism see R. THALER & C.R. SUNSTEIN, NUDGE: IMPROVING DECISION ABOUT HEALTH, WEALTH AND HAPPINESS (2008).
62. Recent Article 9 decisions include Rambus, RWE, Eon, Ship Classification, and Microsoft. See, Commission accepts commitments from Rambus lowering memory chip royalty rates, Rambus IP/09/1897 Brussels, (December 2009);
63. Commission opens German gas market to competition by accepting commitments from RWE to divest transmission network, RWE, IP/09/410 Brussels, (March 2009);
64. Commission opens German electricity market to competition, EON, IP/08/1774 Brussels, (November 2008);
65. Commission accepts commitments by GDF Suez to boost competition in French gas market, GDF Suez, MEMO/09/536 Brussels (December 2009);
66. Commission paves way for more competition in ship classification market by making IACS' commitments legally binding, Ship Classification, IP/09/1513 (October 2009);
67. Commission welcomes Microsoft's roll-out of web browser choice, Microsoft, IP/10/216 Brussels, (March 2010).
68. Such an approach has been articulated in the context of enforcing exploitative abuses. See A. Fletcher & A. Jardine, Towards an Appropriate Policy for Excessive Pricing, 12th Annual Competition Law and Policy Workshop (June 2007).
69. Commission welcomes Microsoft's roll-out of web browser choice, supra note 51.
70. For example, see Warning: Too much information can harm, A final report by the Better Regulation Executive and National Consumer council on maximising the positive impact of regulated information for consumers and markets, (November 2007).
71. See Federal Trade Commission Bureau of Economics Staff Report, The Effect of Mortgage Broker Compensation Disclosures on Consumers and Competition: A Controlled Experiment, (February 2004).
72. See Office of Fair Trading, Assessing the effectiveness of potential remedies in consumer markets, Economic Discussion Paper OFT994, ¶¶ 4.39-4.42 (April 2008).
73. The concepts of stated and revealed preference are well understood in competition analysis. For a discussion of the implications of these consequences, see Office of Fair Trading / Competition Commission, Good practice in the design and presentation of consumer survey evidence in merger inquiries, Forthcoming for consultation March 2010.
74. See Competition Commission, Market investigation into payment protection insurance, ¶4.97 (January 2009).
75. Barclays Bank PLC v Competition Commission, 2009, Competition Appeal Tribunal, Case: 1109/6/8/09.
76. See Competition Commission and Office of Fair Trading, Road testing of consumer remedies, Economic Discussion Paper OFT1099 (July 2009).
77. For a recent report that provides a toolkit for introducing choice and competition into public service markets, drawing on the 'access, assess, act' framework set out above, see Office of Fair Trading, Choice and Competition in Public Service Markets, Economic Discussion Paper OFT (March 2010).