PDF《Learning in the Credit Card Market》

2 For example, see Zimmerman (1982), Argote, Beckman and Epple (1990), Gruber (1992), Bahk and Gort (1993), Marimon and Sunder (1994), McAfee and McMillan (1996) , Nye (1996), Sargent (1999), Benkard (2000), Thompson (2001), Thornton and Thompson (2001), Evans, Honkaphoja and Marimon (2001), Evans and Honkaphoja (2001), and Barrios and Strobl (2004).

3 For example, Miravete (2003) shows that consumers switch telephone calling plans to minimize monthly bill payments even for very small di?erences in cost. Agarwal et al. (2005) report that

40 percent of borrowers choose suboptimal interest rate contracts, but most eventually switch to cost-minimizing contracts.

4 Lemieux and MacLeod (2000) study the e?ect of an increase in unemployment bene?ts in Canada. They ?nd that the propensity to collect unemployment bene?ts increased with a ?rst-time exposure to this new system via an unemployment spell. Barber, Odean and Strahlevitz (2004) ?nd evidence that individual investors tend to repurchase stocks that they previously sold for a gain.

5 There is a large literature on the magnitude of interest payments and fees in the credit card market: Ausubel (1991), Calem and Mester (1995), Massoud, Saunders and Scholnick (2006), Ausubel (1999), Kerr and Dunn (2002), Shui and Ausubel (2004), DellaVigna and Malmendier (2004), Kerr (2004), Calem, Gordy and Mester (2005).

2 advance fees C since some observers argue that new customers do not optimally minimize such fees.6 We want to know whether credit card holders improve with experience, learning to avoid triggering fees. We ?nd that fee payments are very large immediately after the opening of an account. We ?nd that new accounts generate direct monthly fee payments that average $16 per month.7 However, these payments fall by

75 percent during the ?rst four years of account life. To formally study these dynamics, we estimate a learning model with the Method of Simulated Moments. The data reveals that learning is driven by feedback. Making a late payment ― and consequently paying a fee ― reduces the probability of another late payment in the subsequent month by

44 percent. These learning e?ects may be driven by many di?erent channels. Consumers learn about fees when they are forced to pay them. Alternatively, consumers may pay more attention to their credit card account when they have recently paid fees. Whatever the mechanism, card holders learn to sharply cut their fee payments over time. We ?nd that the learning dynamics are not monotonic. Card holders act as if their knowledge depreciates ― their learning patterns exhibit a recency e?ect.8 A late payment charge from last month is more in?uential than an identical charge that was paid a year ago. The monthly hazard rate of a fee payment increases as previous fee payments recede further into the past (holding all else equal). We estimate an e?ective knowledge depreciation rate of

6 For example, Frontline reports that The new billions in revenue re?ect an age-old habit of human behavior: Most people never anticipate they will pay late, so they do not shop around for better late fees. (http://www.pbs.org/wgbh/pages/frontline/shows/credit/more/rise.html) There is also a nascent academic literature that studies how perfectly rational ?rms interact in equilibrium with imperfectly rational con- sumers. See Shui and Ausubel (2004), DellaVigna and Malmendier (2004, 2006), Miao (2005), Mullainathan and Shleifer (2005), Oster and Morton (2005), Gabaix and Laibson (2006), Heidhues and Koszegi (2006), Jin and Leslie (2006), Koszegi and Rabin (2006), Spiegler (2006), and Ellison (forthcoming) for an overview.