A reassessment of market power among credit card banks

Abstract

An improved empirical specification of credit card conduct agrees with several prior studies in rejecting perfectly competitive equilibrium, indicates structural disequilibrium in the industry and is consistent with monopolistic competition. Measures of liquidity management costs, omitted from prior studies, are shown to be important factors in a properly specified model of pricing conduct in credit card lending.

Acknowledgements

The authors are grateful to Mark Taylor and an anonymous referee for helpful comments.

Notes

¹ While many empirical studies have explored the degree of imperfect competition, a few have begun to focus on the specific mechanisms or facilitating practices underlying the observed imperfection; see e.g. Ayogu and Dezhbakhsh (2004).

² Commercial loans made under commitment are another credit product that can share this feature.

³ Evans and Schmalensee (1999) provide a more detailed description of the credit card industry.

⁴ The number of banks issuing credit cards has been put at some 4000 during the 1980s (Ausubel, 1991) and 6000 during the 1990s (GAO, 1994).

⁵ In 1994, the average outstanding balance per credit card account was less than $1400 (Simmons, 1995).

⁶ Aggregate liquidity constraints have been shown to have significant macroeconomic consequences even in industrialized countries, such as prolonging recessions due to endogenous ‘credit crunches’ (Stanton, 1998) or prolonging the recent banking crisis in Japan (Konishi and Yasuda, 2003).

⁷ Reliable data on credit card pricing are notoriously difficult to obtain for many countries, to the point where most cross-country studies do not even attempt to quantify finance charges (see e.g. Humphrey et al ., 1996, pp. 937f.). Unofficial data on various web sites suggests that finance charges remain diverse but mostly high across several countries including the US, UK, Canada, and Japan as of early 2006; see e.g. www.credit-land.com and www.jcbusa.com/help_cFAQ.htm.

⁸ At the midpoint of our sample, data from the Federal Reserve's G.19 Consumer Credit Statistical Release (www.federalreserve.gov) indicates that consumer credit outstanding at finance companies, thrift institutions and nonfinancial business together exceeded total consumer credit outstanding at commercial banks, and several of those nonbank products (such as retail merchant charge accounts) are close substitutes for bank card loans.

⁹ If input prices and total costs decline, competition will likewise compel such a firm to pass along the full decrease by reducing its output prices.

¹⁰ Another commonly used empirical test developed by Bresnahan (1982) would not reveal the true degree of market power in this case, because it relies on the estimated gap between marginal revenue and marginal cost, which is systematically distorted when the firm adjusts its total costs to match total revenue as input prices change.

¹¹ This period straddles the introduction of the Fair Credit and Charge Card Disclosure Act of 1988, intended by Congress to reduce consumer search costs, but Shaffer (1999) has shown that this Act had no measurable effect on the degree of competition in the credit card industry.