Disenfranchise the ratings agencies.

• Author: Smick, David
• Position: FROM THE EDITORS - Editorial

The House Oversight Committee's hearing on the role of credit ratings agencies lived up to its billing. Great examples of fraud by well-aware participants were brought into evidence by both accusatory witnesses and self-confessing emails. The ratings agencies had sold out to their investment bank clients, giving AAA ratings on securities they did not want to understand for streams of fee-based business. People who trusted the ratings agencies when buying securities were let down. Expressing outrage about the perfidy of intentional over-ratings, however, really misses the point for regulatory reform going forward.

The real issue with ratings agencies is that, even when they are uncorrupted, reliance on them is harmful to the stability of our financial system. Without regulatory encouragement of that reliance, they have no real market for their wares. Even with that regulatory enfranchisement in our markets, they serve no useful purpose except inherently to mislead investors. And that's with the best of intentions. When subjected to the slightest temptation, their incentives lead the ratings agencies to exploit both those rated by them and those who buy rated securities.

Ratings agencies claim that what they assess is default risk, pure and simple, on a comparable and comprehensible scale. Yet, just about every single part of that description is incorrect. Default risk is never pure and simple except for the most vanilla of repeatedly reproduced securities--in which case, economic fundamentals are all it takes to assess the risk, with the ratings adding no value (as a host of econometric studies have demonstrated). If the securities to be rated are more complicated, less frequently replicated, and therefore without clearly comparable securities on which to form a track record, the judgment behind the rating should be worth more. Instead, the ratings are worth less because they fail to capture the variation of the risk over time. They also fail to be comprehensible, because what is AAA for a municipal bond is different from what that means for an emerging market subordinated debt contract which is different in turn from a CDO.

Usually, when this is pointed out, the agencies will insist that is because the purchaser (or observer) is asking about more than default risk, which is unfair to the ratings. The agencies, however, are left with two unappealing alternatives if that is the case. Either default risk is actually unimportant, since ratings...