Credit Rating Agencies and the “Worldwide Credit Crisis”: The Limits of Reputation, the Insufficiency of Reform, and a Proposal for Improvement

Thursday, January 1st, 2009 at 12:00 am by John Hunt
John Hunt, Credit Rating Agencies and the “Worldwide Credit Crisis”: The Limits of Reputation, the Insufficiency of Reform, and a Proposal for Improvement, 2009 Colum. Bus. L. Rev. 109

The “worldwide credit crisis” has thrust credit rating agencies into the spotlight, with attention focused on their ratings of novel structured finance products. Policymakers have undertaken a number of initiatives intended to address perceived problems with such ratings–enhancing competition, promoting transparency, reducing conflicts of interest, and reducing ratings-dependent regulation. These approaches are all broadly consistent with the dominant academic theory of rating agencies, the “reputational-capital” model, which is taken to imply that under the right circumstances a well-functioning reputation mechanism will deter low-quality ratings. The policy initiatives currently under consideration can be seen as efforts to fix discrete problems with the rating market so that the reputation mechanism can work properly. This Article argues that these efforts are fundamentally incomplete, because even a well-functioning reputation mechanism does not generate optimum rating quality on new products: When a new product is introduced, agencies do not have a reputation for high quality in that product so they have nothing to lose from issuing the rating. Even if low quality for a specific product type harms the agency’s reputation for rating other product types, the agency still will be induced to issue low-quality ratings if the new product type in question is large enough in volume. And as long as rating quality across new product types is high enough on average, it is rational for investors to rely on new-product ratings even if they know that some are of low quality.

The incentive problem can be corrected by requiring an agency to disgorge profits on ratings that are revealed to be of low quality by the performance of the product type over time, unless the agency discloses that the ratings are of low quality. Such a system would be superior both to the current regime, which relies on market forces backed by antifraud rules, and to other alternatives, such as the recent proposal to forbid new-product ratings absent prior SEC approval of the products.

Author Information

Fellow, Berkeley Center for Law, Business, and the Economy