by Tanta on 9/27/2007 09:48:00 AM

From the International Herald Tribune, we see that Congress is getting up on its hind legs and barking over the perennial issue of who pays for bond ratings:

Democratic and Republican senators said they were particularly concerned with one aspect of the agencies' business models: They get paid by the companies whose bonds they rate. That is like a film production company paying a critic to review a movie, and then using that review in its advertising, said Senator Jim Bunning, Republican of Kentucky.

Right. Because we all know that film critics' salaries are paid by surcharges on movie tickets, not newspaper advertising revenue. Whatever.

I really do want to know why those who believe that rating agency fees should be paid by investors, not issuers, believe that this is going to improve anything. I'm not defending the status quo; I'm legitimately curious. There's no context we can imagine in which, say, large institutional investors who pay the lion's share of those fees might not put pressure on the RAs to keep them from downgrading an outstanding issue, against which the investor doesn't really want to hold more reserves or capital? There are no transparency concerns when only those "accredited investors" who pay subscription fees can see rating benchmarks? (It's working so well for the hedge fund industry . . .) Issuers would somehow become unable to pressure the agencies for the "right" rating if they didn't directly pay those invoices? You know a clear, bright-line distinction between "issuers" and "investors" you could share with me? If Bear Stearns, say, buys tranches for its own trading account, and also issues them off its own shelf, what difference does it make which cost center cuts the checks to Moodys?

Let's contrast this situation with the highly satisfactory smooth working of consumer credit reports. The information in those reports is provided by creditors, not the individuals who are subject to them. The fee for use of that information is paid by subscribers--creditors, mostly--although they turn around and pass that cost onto applicants and borrowers. You pay the cost of lenders getting your credit report from a for-profit company who considers its information on you to be its property, not yours. In an environment of rampant identity theft and privacy violation--an environment caused, basically, by the existence of electronic repositories of sensitive data that consumers don't profit from, although they take the incalculable privacy risk--there can be no "transparency" of FICOs, and you wouldn't want there to be.

However, in an attempt to fight back, a whole industry of manipulating consumer credit ratings has sprung up, driven by both consumers wanting more or cheaper credit and lenders wanting more or more profitable borrowers. In all of this, the mortgage industry in particular is reeling from the results of its overreliance on FICOs, which were designed more for evaluating the risk of unsecured lending than mortgage lending, and that can rather easily be used to find not the most credit-worthy borrowers but the most profitable ones: those who do not default, but who run up balances sufficiently and pay enough periodic late fees to be a credit card issuer's dream come true. As more people are granted more unsecured credit by savy users of FICOs, the FICOs themselves "improve" for a class of borrowers who now want mortgage loans based on those FICOs.

Some of you may not know this, but there used to be a chronic problem with subprime mortgage servicers refusing to report data to the credit repositories. That meant that those borrowers who got a subprime loan and then paid it on time were not benefitting, since their credit record didn't show the on-time payments; when they went to refinance, they got put into another subprime loan. You would also find subprime servicers not reporting current balances, only original ones, making these borrowers look more indebted than they really were. It took Fannie and Freddie coming out with the declaration that they would no longer buy loans from seller/servicers who did not "full file report" to the credit bureaus to put a stop to this practice. Anyone who is convinced that users of ratings always have an interest in the accuracy of ratings should ponder this: there's always money to be made off of information asymmetry. It's fashionable these days to see bond investors as the hapless innocent victims of issuers and their lapdog rating agencies, but I'm old enough to have seen lenders who foot the bill for consumer credit reports going out of their way to make sure those reports didn't tell the whole picture.

I really want to know why we think subscriber-paid fees in the bond rating world is going to result in something we'll be happy with.