The SEC Ranking Ratings Agencies: A Bad, Bad Idea

The U.S. Securities and Exchange Commission seal hangs on the facade of its building in Washington, DC.
Getty Images
The U.S. Securities and Exchange Commission seal hangs on the facade of its building in Washington, DC.

Here’s my nomination for the worst recommendation by the Senate panel on the financial crisis: the proposal for the Securities and Exchange Commission to rate credit ratings agencies based on their “accuracy.”

Everyone agrees that inaccurate ratings of mortgage-related financial products—from straight mortgage-backed securities to complex derivatives—contributed to the financial crisis. And nearly everyone would like credit ratings agencies to perform better in the future.

But charging the SEC with the task of evaluating the “accuracy” of ratings is a recipe for disaster.

In the first place, the SEC has little experience or expertise in evaluating credit risk. It is staffed by lawyers and bureaucrats who are mostly charged with monitoring compliance with established procedures and punishing individuals and companies who commit illegal activities such as insider trading.

Faced with the task of monitoring the accuracy of ratings, the SEC will likely primarily focus on procedural compliance. Was the rating unduly influenced by the financial company purchasing the rating on its product? Did the rating agency keep full and complete records?

This will likely have very little effect on the accuracy of ratings.

The reaction from the ratings agencies will be to tighten procedural regularity to comply with this kind of supervision. But the focus won’t be on more accurate ratings—it will be on a cleaner process for producing ratings.

To the extent that the SEC attempts to develop an expertise in credit risk, it will have to rely on hiring people who have worked for banks and ratings agencies. This is simply a recipe for the further integration of the mindset of bankers and regulators. If part of the problem with ratings was the undue influence of investment bankers over the process, we’re not going to see much better ratings by importing former and future bankers into the offices of the SEC.

By the way, why is the SEC being put in charge of this process at all? The FDIC and the Federal Reserve would seem to be far more natural homes for the evaluation of ratings accuracy. They at least have some experience looking at credit risk.

The deeper problem here is that there’s simply no way for regulators to evaluate the accuracy of ratings. Ratings are an estimate of what is likely to happen in the future. If the forecasted results don’t happen, that doesn’t necessarily mean that the forecast was inaccurate at the time it was made. It might just mean that events were unpredictable or circumstances unknowable. Any view of the "accuracy" of a rating requires the exercise of discretionary judgment rather than the application of an objective standard.

What is likely to happen is that the SEC would develop a model for what counts as an accurate rating. And the ratings agencies will react by rating toward that model rather than their views of the risks of financial products. To the extent that the SEC model doesn’t fully reflect reality, we’ll just end up with another set of inaccurate ratings.

To put it differently, increasing the influence of regulators over ratings will just mean that the ratings will do a better job of reflecting the regulators’ view of risk. But since the regulatory view of risk is no more likely to be accurate than the view of, say, investment bankers or employees of ratings agencies, we won’t end up with ratings that work any better.


Questions? Comments? Email us

Follow John on Twitter @

Follow NetNet on Twitter @

Facebook us @