When the subprime mortgage bubble burst in August 2007, many immediately blamed credit rating agencies which had affixed reassuring AAA labels on subprime-backed securities.
Standard and Poor's, Moody's Investors Service and Fitch Ratings stood accused of negligently or even deliberately misleading investors; calls rang out for tougher regulation of the firms, which enjoy the government-designated status of Nationally Recognized Statistical Ratings Organizations, with all the competitive advantages and public responsibilities that role implies.
In the months since, however, the performance of the ratings agencies receded as an issue - probably because financial authorities have been busy staving off economic collapse.
Thus Wednesday's promise by Mary L. Schapiro, the new chairman of the Securities and Exchange Commission, to undertake an "intense review" of the $5 billion-per-year securities rating industry was a welcome development. The SEC and the agencies themselves have taken some steps toward reform, but "there is still more work to do," as Ms. Schapiro put it.
In a roundtable session including the firms and regulators, she asked tough questions, such as whether certain securities are simply too complex for the agencies to rate and whether the government might usefully reduce its reliance on agency ratings. Above all, she focused on the agencies' business model, which calls for them to be paid by the same companies whose securities they are evaluating.
This conflict of interest, and the "ratings shopping" by securities issuers that it allegedly promotes, are long-standing issues whose resolution would help market confidence. But even conflict-free agencies might still contribute to the market's failure to anticipate systemic crisis. By their nature, credit bubbles inflate the perceived value of cash flows and collateral.
Credit analysts, assessing companies individually while taking the effervescence of the overall economy as a given, perceive corporate profits and asset prices as more durably robust than they really are. The result is credit grade inflation, which persists late into the boom phase of the business cycle. Of course, during the inevitable bust, the opposite dynamic kicks in. Bottom line: less economic stability.
British financial analyst George Cooper suggests a possible fix: instead of handing out AAAs as they see fit, agencies should have to grade securities on a relative scale - just as some college instructors grade students "on a curve."
Mr. Cooper recommends that agencies group their clients' bonds into 10 equally sized "quality buckets," with only the best 10 percent getting the highest rating and so on down the line. This would reduce the incentive to "sell" favorable ratings, since for every upgrade there would have to be an offsetting downgrade. Average credit ratings would be stable across the business cycle.
As she develops her reform agenda, Ms. Schapiro would do well to put this idea on the discussion list.