You don’t have to agree with everything the SEC does to respect the way they do it: passive-aggressively. Felix Salmon this morning discussed “the problematic JOBS Act, where the SEC has done a good job of stalling on various silly yet Congressionally-mandated reforms,” and the SEC’s similar strategy of “being sensible and dragging its feet” on Congressional dedecimalization proposals. If you’re a regulatory agency tasked by Congress with implementing a new law, and you don’t feel like it, the best approach is always to commission a study.

Is that what’s going on with ratings agency reform proposals?

The 2010 Dodd-Frank financial-overhaul law requires the SEC to create a board that would assign a rating firm to evaluate structured-finance deals or come up with another option to eliminate the conflicts that could arise when debt issuers pay rating firms to rate their bonds. … Sen. Al Franken (D., Minn.) proposed the legislation, which is known as the Franken Amendment.

… Mr. Franken defended his proposal Tuesday to attendees at the round table, including SEC commissioners and SEC Chairman Mary Jo White. He also called on the agency to make changes to the credit-rating industry.

“My plea today is that you take action,” Mr. Franken said. “Millions of Americans lost their jobs because the credit rating agencies didn’t do their jobs,” he said.1

The agency published a report in December – six months late – that was widely expected to announce regulatory changes. Instead, the report proposed more discussion and the convening of a round table.

So hahahaha SEC you suck but the problem seems genuinely hard doesn’t it? Here’s the report and it raises a bunch of problems with the Franken Amendment idea of a government system matching issuers to raters, from the operational (“One commenter observes that there would be thousands of structured finance products that would need to be rated at any given time,2 and that the CRA Board would need to have a detailed understanding of each NRSRO’s qualities in order to assign an NRSRO to rate each of those products”) through the constitutional (“government intervention into the selection of who can or cannot render an opinion about the creditworthiness of assets could attract litigation on First Amendment grounds”3) to the symbolic:4

Commenters generally are concerned that the Section 15E(w) System is contrary to government efforts to reduce investor reliance on credit ratings and that investors would rely too heavily on ratings if they were deemed vetted by the CRA Board. One commenter argues that the system would further entrench certain NRSROs “by creating a government-sanctioned special category of NRSROs that are the only NRSROs that are permitted to issue initial credit ratings to issuers.” Another commenter argues that the system is inconsistent with government efforts to reduce reliance on ratings because investors might view the assigned NRSRO producing a rating as a government appointed and regulated entity. Similarly, other commenters argue that the ratings produced by the system could be viewed as having a government seal of approval, running counter to the goal of reducing reliance on ratings generally. Other commenters believe that investors would be less likely to perform appropriate analysis and due diligence if the ratings were seen as indirectly sanctioned by the government through the CRA Board or because the rating produced by the assigned NRSRO is the “right” or “government-sanctioned” rating.

The report discusses other models (issuer-pay, investor-pay, etc.) and mostly hems and haws a lot. Round-table it up! The round table was today and from what I can tell seems to have been inconclusive. (Do round tables ever end with unanimity?) My favorite type of credit rating – the unsolicited passive-aggressive “You don’t pay us to rate your deal? Okay fine we rate it anyway. We rate it F!” – which of course raises its own conflicts, seems to be getting a lot of attention. Issuers continue to shop around for the best ratings.

Obviously issuer-pays has its problems, but I tend to subscribe to the model of credit ratings that goes like:

  • nobody, like, “believes” ratings,5 but
  • lots of big pools of money use ratings to comply with some sort of requirement (imposed by bank capital or liquidity regulation, prime money market fund regulation, investment charter, whatever) that they only invest in AAA or investment grade or rated or whatever securities.

Thus both sides of pretty much every trade – issuers and investors – will want the highest possible ratings on the riskiest (and thus returniest) things: if you don’t like the risk of the AAA tranche, you can always not buy it, but you want the option of buying it and giving it zero risk weight or whatever.6 This explains such basic anomalies as (1) why AAA CDO tranches had higher yields than AA corporate bonds and (2) why CPDOs existed.

It explains another important anomaly too, which is: if investors are all agitated about how bad issuer-paid ratings are, and if they’re on very clear notice that the ratings agencies are conflicted and not to be trusted, then why don’t they just pay for more accurate ratings and ignore the issuer-paid ones? The SEC round table may be concerned with a real problem, but it sort of can’t be a problem of ratings consumers.

This model doesn’t leave much time for worries like “the ratings produced by the system could be viewed as having a government seal of approval,” because the whole point of the ratings produced by the system is to obtain a government seal of approval. On the other hand if, like many people, you dislike the system where a government agency decides how risky various asset classes are and how much capital banks need to have to support them, then you might also dislike the idea of delivering credit ratings into the hands of another government body.

I dunno. If both sides of a trade are winners then an interesting thing to do is to figure out who the losers are.7 If issuers and investors both benefit from inflated ratings, then … well then one answer is “shut up, Al Franken, everything is fine.” (Is that the SEC’s answer? Like: probably, right?) Another is that the investors’ investors are the losers: the shareholders and creditors and taxpayer-bailer-outers of the regulated entities that buy structured credit securities with inflated ratings are the ones hurt by those inflated ratings. Maybe the SEC should find a way to let them select ratings agencies.

S&P Warns Credit-Rating Proposal Could Create ‘New Conflicts’ [WSJ]
Ratings Shopping Revived in Asset-Backed Rebound: Credit Markets [Bloomberg]

1. True? Also, this is good:

“Imagine the pharmaceutical industry having six FDAs, all competing to approve drugs,” said Rob Dobilas, who founded Realpoint LLC, the credit-rating company bought by Morningstar Inc. in 2010, referring to the U.S. Food and Drug Administration. “Everyone would be dead.”

Everyone!

2. Thousands? Seems optimistic no?

3. Also:

One commenter argues that, in addition to being unprecedented in forcing “one private party to deal with another private party of the government’s choosing in a private business transaction,” the Section 15E(w) System would raise Fifth Amendment issues. The commenter opines that the system could violate the Fifth Amendment “to the extent that interference with issuers’ or NRSROs’ contract rights rises to the level of a taking without just compensation or to the extent that this provision is so arbitrary and irrational as to violate the Fifth Amendment’s Due Process Clause.”

That seems … really, really wrong?

4. Also “One commenter observes that political pressure could influence the CRA Board’s initial determination of which NRSROs are qualified to issue ratings and the frequency of assignments to NRSROs in the future,” about which: remember how the SEC seems to really have it in for Egan-Jones for some reason?

5. I mean, you might, whatever, but the marginal user.

6. Separate from the fact that, once a bond has already been placed, obviously no one wants it to be downgraded. Except I guess prospective secondary market buyers. They should pay for ratings. “Anyone who doesn’t buy this CDO has to pay for it to be rated.” That’s a great model.

7. Is structured finance zero sum? Discuss.

2 comments (hidden to protect delicate sensibilities)
Show all comments ↓

Comments (2)

  1. Posted by Guest | May 15, 2013 at 3:21 AM

    But like Investor Pays and Free-rider / Xerox problems.

  2. Posted by QRFAR | May 16, 2013 at 6:58 PM

    The uncaring paying the unknowing for the useless… Sounds right up the governments alley.