Rating Agencies' 'Business Model' Didn't Cause the Crisis

 |  Includes: MCO, SPGI
by: Matt Stichnoth

University of Minnesota law professor Claire Hill explains why, the yawping of the Sean Egans and Paul Krugmans of the world notwithstanding, the rating agencies’ “issuer pays” business model isn’t what led the agencies to mis-rate all those subprime CDOs and nearly blow up the financial system:

[T]he view that conflicts of interest caused the too high ratings in some straightforward way—an issuer told a rating agency that if the agency did not give the desired rating, the issuer would go to another agency, and the agency succumbed to the pressure—embeds a strong and rather startling assumption. Consider that rating an instrument much higher than its quality warrants almost certainly can’t work in the moderate term. A rating agency is consulted for its ratings only because those ratings are deemed by investors to be accurate. Issuers are paying for ratings, but if investors think high ratings can be bought, the ratings will be worthless, and rating agencies will lose all their business. Thus, the strong conflict argument turns on enormous agency costs—many people at the agency being willing to go for the quick buck, knowing that in the moderate term the agency would suffer enormously, as would, presumably, their long-term job prospects at the agency (and perhaps elsewhere). The short-term payoffs were high—the agencies were indeed making huge quantities of money.32 But they weren’t making enough money to look as foolish as they do now. . . . [Emph. added]

Plus, issuer-pays seems to be working just fine for other instruments, such as corporates and municipals, that the agencies have been rating for years. . . .

You’ll get nor argument here that, if there were a single villain in the whole housing-bubble-and-crackup, it would be the rating agencies.

But it wasn’t the agencies’ business model that led them to come up with all those crazy AAAs. It was something else, more subtle, that began with a ridiculously misplaced confidence in their own models. . . .