August 31, 2010
UC Berkeley study calls for elimination of credit rating-contingent regulation by governments
In order to prevent inflated credit ratings and an economic crisis similar to that in 2008, the government should eliminate rating-contingent regulation, according to Marcus Opp, finance assistant professor at the University of California, Berkeley's Haas School of Business.
"If the government didn't have these rating-contingent regulations, I believe the ratings would automatically become more informative," says Opp, describing the findings in "Rating Agencies in the Face of Regulation: Rating Inflation and Regulatory Arbitrage," co-written by Christian Opp, assistant professor, The Wharton School, University of Pennsylvania, and Milton Harris, professor at the University of Chicago's Booth School of Business.
Many observers blame the 2008 financial crisis on credit rating agencies for inflating ratings and knowingly underestimating the risk of various securities, such as those backed by subprime mortgages. On the heels of the passage of the Dodd-Frank financial regulatory reform bill that makes rating agencies liable for ratings' quality, this research offers data that supports a new approach to regulation.
The ratings issued by credit rating agencies are supposed to measure the likelihood of default for debt products. In reality, ratings may not provide an accurate picture to investors because the agencies profit from handing out high ratings. Investors are willing to pay more for an asset with a favorable rating. The goal of the research was to determine why ratings are still valued by investors even when the ratings do not depict actual risk. The study found the government's reliance on ratings to determine the required capital held by institutional investors, such as banks, plays a key role in exposing the system's faults.
Since the mid-1970's, national governments have increasingly used these same ratings for regulating institutional investors. "If the importance of the regulation is too high, then a rating agency will not provide any real information to investors. It will simply inflate ratings," Opp explains "The solution is actually quite simple. Get rid of rating-contingent regulation. Let rating agencies do their jobs without serving two purposes at the same time."
Opp and his co-authors studied a non-competitive business model to replicate the actual market in which (effectively) only three ratings agencies (Moody's, S&P and Fitch) exist and are virtually non-competitors. The model indicated that rating inflation is more likely in the case of complex securities that are costly to evaluate. Opp says prior to the recent economic crisis, only one percent of bonds in the corporate bond market received a "triple A" rating while in the subprime market, 70% of assets were given this highest rating. "Our theory tells us that business grew in the subprime market because regulation did not distinguish between different asset classes," says Opp.
Based on the paper's conclusions, Opp believes any proposed reforms being debated in Congress will fall short if rating-contingent regulation persists and does not at least take into account the perverse feedback effects of the current regulation in place.