There's No Such Thing as an Objective Credit Rating

Includes: MCO, SPGI
by: Felix Salmon

Vickie Tillman of S&P has an interesting letter in the WSJ today, defending the ratings agency against the kind of charges made by Jesse Eisinger in last month's Portfolio. S&P is not conflicted, she says, by the fact that its income comes from the very entities that it is purportedly policing:

Issuers structure deals, and we rate these deals based on our criteria -- criteria that are publicly available, non-negotiable and consistently applied...
Our credit ratings provide objective, impartial opinions on the credit quality of bonds.

The word that jumps out at me here is "objective". And that word, in this context, means something only insofar as it is being contrasted with its opposite, "subjective". A subjective rating, on this view, would be a bad thing, while an objective rating is a good thing.

Now what would the difference between a subjective credit rating and an objective credit rating? A hint is given by the first sentence that I quoted. A subjective credit rating would be one individual's, or one company's, point of view. It would be informed by that individual's, or that company's, expertise, but ultimately de gustibus non est disputandum, and all that.

On objective credit rating, on the other hand, would be something much more rule-based: it would have as its foundation "criteria that are publicly available, non-negotiable and consistently applied". You take the financials of any given issuer, you drop those financials into a black box, and out the other end pops an entirely predictable and certain credit rating; the person doing the dropping has nothing to do with it, and might as well be a trained monkey for all the difference they make.

Now here's the thing: if credit ratings really were objective, then there wouldn't be any need for ratings agencies in the first place, and you wouldn't have different ratings agencies competing against each other, and you certainly wouldn't have ratings agencies making enormous sums of money. After all, anybody could simply use the ratings agencies' publicly-available, non-negotiable and consistently-applied criteria to generate the exact same ratings that S&P, Moody's, and Fitch charge lots of money to provide.

Now in the real world, people who work at credit rating agencies tend to be well paid and highly-qualified individuals. Quite often, they are poached by investment banks, either for their knowledge of how the sausage is made or else just because they're very bright individuals with a lot of insight into the credit markets. I'm sure that Vickie Tillman would agree that she and her colleagues are able and valuable professionals, and that S&P is one of those companies where the real value walks out the door every evening.

But she can't have it both ways. Either S&P hires smart people to make subjective decisions, or else it basically hires anybody who can follow simple rules which spit out a rating according to predetermined criteria. (Although even then, it would need at least a few people to subjectively be in charge of changing and updating those criteria over time.)

In reality, of course, ratings are subjective, not objective. But it's interesting that Tillman seems so keen that we think otherwise.