Nothing Should Be Random About Rating Agency Assignments

Franken's measure was creative, but the diverse range of qualifications among the approved rating agencies is frightening to contemplate if a random assignment was the norm.
This post was published on the now-closed HuffPost Contributor platform. Contributors control their own work and posted freely to our site. If you need to flag this entry as abusive, send us an email.

Sen. Al Franken's proposal to have rating agencies assigned to securitizations randomly by the SEC was rejected by a House-Senate Conference Committee this week, to my great relief.

According to The New York Times "Deal Book" blog, the measure gained traction because it addressed the problem of issuers selecting which agency will rate their transactions. While the Franken amendment passed in the Senate, it died in conference due to "practicalities."

I'd say that practicalities don't begin to describe how bad the suggestion of random assignments was.

We've worked with the rating agencies on securitizations in the re-emerging mortgage securitization market. I've met with them all and have studied their approaches and requirements for firms such as Allonhill, which they refer to as "Third Party Review" firms, or TPRs. In fact, based on these discussions, we've developed our own securitization guidelines for issuers, explaining in great detail how to meet new, stricter rating agency requirements.

To understand why random assignment of rating agencies to deals is a bad idea, first you need to know about the rating agencies. There are 10 SEC-approved "Nationally Recognized Statistical Ratings Organizations," or rating agencies. Two are Japanese and do not rate MBS in the U.S.; one rates insurance companies; one rates equities; and two are new and operating under an investor subscription format and do not publicly disclose their ratings. There are three that the market knows fairly well, and one is an up-and-comer that has privately rated a lot of transactions in recent months.

The three that are well-known are Moody's, Standard & Poor's and Fitch. Add to that DBRS, a Toronto-based firm that has been prolific in this space, but, like all but Moody's, has yet to publicly rate a private mortgage securitization in the new market.

Careful analysis of these four firms will show you this:

Moody's and S&P both have issued fairly detailed and well thought-out securitization requirements that result in independent data on transactions, produced by a qualified TPR. The two firms have different approaches to certain key areas of the review, but fundamentally, they both have put a lot of work into crafting an approach that digs much deeper into transactions than in the past, ensuring investors that the data produced is factual and independent. Moody's has been on the only publicly-rated private securitization that has come to market since the fall of 2008.

Fitch has produced materials that indicate a strong core understanding of the review process for loans going into a securitization. But Fitch has not yet emerged as an active participant in the RMBS space. When I last spoke to them, they had taken the position -- wisely, I believe -- of waiting for the market to open back up a bit before plunging into it.

DBRS has a set of guidelines for its securitizations process, but these guidelines only touch on the role of the due diligence process in passing; DBRS does not have a set of guidelines specifically to ensure that the TPR firm and review are independent and follow an established approach. That's because DBRS wasn't subject to the New York Attorney General's settlement agreement in 2008, under which the rating agencies which dominated the market --Fitch, Moody's and S&P--had to develop requirements for the work of the TPR. To gain investor acceptance on publically-rated transactions, DBRS must do what the others have done, and develop its own methodology for TPRs that shows that securitizations rated by DBRS consistently offer data that was reviewed independently and in accordance with published criteria that investors can obtain and review for themselves.

If Franken's amendment had been left in Sen. Chris Dodd's financial reform bill, investors would have to rely on data provided by firms with a broad range of competence and readiness -- from the agencies that have embraced independence guidelines and loan criteria, represented by Moody's, and S&P, to the others on the SEC's list who have not yet executed or articulated a solid approach that brings independence and measures against published criteria.

As it is, the measure did not pass. But the conflict of interest issue remains. Sen. Dodd's take: "How is it that you [issuers] get to pick your own rating agency, where you don't do any due diligence and just rely on the data and information given to you by the very company that has hired you to give them a rating? I mean, just saying it alone, it screams out for a resolution."

I disagree with Sen. Dodd. The issuer does choose the rating agency, but at least in the case of S&P, Moody's and Fitch, the rating is not based on data and information provided by the issuer without due diligence. These firms have guidelines in place (which Dodd's bill requires of all agencies), under which loan data goes from the issuer to an independent TPR firm like Allonhill. Allonhill reviews the data independently and then provides it directly to the rating agency, not back to the issuer. The data is only used by the agencies to rate a transaction once it has been verified independently by us.

I think Sen. Franken's measure was creative, but I'm surprised it was taken seriously. The diverse range of qualifications among the approved rating agencies is frightening to contemplate if a random assignment was the norm. The conflict of interest situation still needs to be addressed, but not because the data isn't independently reviewed. At least in theory, the issuer still can choose the rating agency that will give it the best ratings levels. But if you study the criteria for the loan data that feeds into those ratings, you'll see that what issuers really need to be shopping for isn't a rating agency, it's good loans to securitize. The criteria are stringent, and the requirements of the TPR will result in bad ratings for pools with bad loans.

The conflict needs to be resolved, and I hope it will be without causing more harm than good, as Franken's amendment would have done. But one thing is clear: provided issuers of rated transactions use one of the agencies with guidelines and requiring a qualified TPR, investors can know, with certainty, that the data behind those securitizations is independent and that it was scrutinized against a published set of criteria that set a high standard. That gives the investor the base information on which to make a decision, which is the meaning of "transparency." How the transaction is rated, what price it sells for, and whether an investor should or should not buy it isn't up to us at Allonhill; our role is to review and monitor loans. But the new and evolving role of independent TPR firms like ours, and our interaction with Moody's and S&P, represents a tremendously important change for the better in our industry.

Postscript: Franken's proposal isn't entirely dead, by the way. According to "The new measure leaves the door open for the SEC to figure out a better way to match rating agencies with financial firms. But if it can't, the SEC is required to follow the original plan proposed by Sen. Al two years."

Go To Homepage

Popular in the Community