Private Information and Price Regulation in the US Credit Card Market

The 2009 CARD Act limited credit card lenders’ ability to raise borrowers’ interest rates on the basis of information learned during lending relationships. This paper estimates the efficiency and distributional effects of these restrictions using account-level data from a near-universe of US credit cards. I show that two forces drove these restrictions’ effects. First, I show that the Act constrained lenders from adjusting interest rates after learning new information about default risk, which I find exacerbated adverse selection among existing borrowers and caused (partial) market unraveling on new accounts. Second, the Act constrained lenders from adjusting rates in response to new information about demand, which reduced lender rents from inelastic borrowers. To study the net effect of these two forces, I estimate a model that features time-varying consumer risk, flexible correlation between risk and demand characteristics, and differentiated lenders who acquire private information about borrowers over time. When I impose the CARD Act’s pricing restrictions in the model, I find that equilibrium market unraveling is most severe for subprime consumers, but lower lender rents are important throughout the market, so that on net, the Act’s restrictions allow consumers of all credit scores to capture higher surplus on average. Total surplus inclusive of firm profits rises among prime consumers, whereas gains in subprime consumer surplus are greatest among borrowers who were recently prime. ∗Nelson: Department of Economics, MIT ([email protected]), and Office of Research, Consumer Financial Protection Bureau ([email protected]). The views expressed herein are those of the author and do not necessarily reflect those of the Consumer Financial Protection Bureau or the United States. For their generous help at all stages of this project, I am deeply indebted to my dissertation committee: Antoinette Schoar, Jonathan Parker, and especially my committee chair Jim Poterba. For thoughtful discussions and suggestions, I thank Nikhil Agarwal, Alex Bartik, Vivek Bhattacharya, Ron Borzekowsi, Ken Brevoort, Tom Conkling, Amy Finkelstein, Daniel Green, Daniel Grodzicki, Luu Nguyen, Paul Rothstein, Richard Schmalensee, David Silberman, Daniel Waldinger, and Mike Whinston. I am also grateful to Shaista Ahmed, Michelle Kambara, Joe Remy, and Stefano Sciolli for their support and guidance at the CFPB.