The New Republic has the article Remember This Moment When the Next Financial Crisis Strikes: The SEC could have fixed our broken rating agencies. It whiffed.
Too much of the financial industry relies on duping investors about the quality of investments. The SEC is supposed to be the first line of defense against that, but has failed in that mission repeatedly in recent years. That feeds skepticism about its seriousness in combating fraud at the rating agencies, especially since it refused to alter the inherently flawed compensation model. As long as the rating agencies get paid by issuers, they’ll have incentives to please them with high ratings.
“There’s a fundamental business incentive for ratings inflation, and there’s got to be something on the other side,” said Marcus Stanley of Americans for Financial Reform. “This rule could do that, but it’s a very tough challenge.”
One commenter on the article had a really brilliant idea.
A small suggestion that would not be susceptible to any regulatory capture: set a formula for fines to be paid by the raters a[t] the payment rate of the securities they rate at any level falls below a certain level. Some explanation of payment rate, 100% would be a bond issue making all interest payments and returning the principal at the maturity date while 0% would be going bankrupt and an intermediate value would be a default and payment of part of the interest and/or principal. The aggregate payment ratio for a rater would be weighted by the value of outstanding securities under its rating.
Higher ratings would have higher requirements for payment ratios. If a rater’s payment ratio for a particular rating fell below the requirement, the fine would be proportional to the total value of securities rated and the size of the shortfall and possibly higher for higher ratings. It should be set just high enough to dwarf any goodwill that the rater could hope to gain with securities issuers through issuing higher than deserved ratings.
We know that the raters have an incentive to err, to the least, on the side of giving higher ratings. Instead of relying on inspectors, why not lay out exactly what the consequences are for erring high in their ratings, with no chance to argue for a reduced penalty, and that the consequences of such errors are greater than the benefits of such errors?
This is such a good set of controls that I think it would put an end to all ratings agencies. I would change it so that this rule only applies to ratings agencies where they are paid by the people whose financial instruments they are rating. The only way to get out from under these rules would be to have the ratings agency agree to do no business with any entity that it rates.
If a ratings agency has a legitimate need to do business with an entity that it might have to rate, it would just be forced to abstain from rating that entity and publicly say that it abstained. You wouldn’t be able to find ratings for all entities in one place anymore. You would have to go to a different agency for a rating of an entity that your favorite source decided to do business with.