Yesterday, the Treasury sent some legislation to Congress in an effort to reform rating agencies. Many believe that the rating agencies played a major part in the financial crisis for giving good ratings to bad financial products. In the past, I've recommended some changes. The Treasury's are virtually useless.

One good way to assess the Treasury's recommendations is to look at each one and determine whether it would have prevented incorrect ratings. Out of the 17 suggestions they list, approximately one might have helped. I'll go through some of the highlights:

Conflicts of Interest

In general, I think conflicts of interest had very little to do with bad ratings. I have never known any rating agency analyst who would have given a deal I worked on more favorable ratings than they believed it deserved under any circumstances. We'd disagree sometimes, but ultimately, what they believed was all that mattered. Here are a few specifics from this aspect of the Treasury's plan:

Bar Firms From Consulting With Any Company That They Also Rate

Okay, but this possibly never resulted in artificially favorable ratings. If anything, by an agency seeing more closely how a company works through consulting, its assumptions would have been better, not worse.


Disclose Fees Paid By An Issuer Along With Each Rating Report.

Really? Because investors don't already know that rating agencies are paid by issuers?


Look-Back Requirement To Address The Conflicts From A "Revolving Door": If a rating agency employee is hired by an issuer and if the employee had worked on ratings for that issuer in the preceding year, the rating agency will be required to conduct a review of ratings for that issuer. . .

The idea here is that rating agency employees who quit to work for an issuer might have rated its deals more favorably. I doubt it. Issuers would want to poach a ratings analyst because s/he is smart, not shady. But even if you think it's possible, I find it unlikely that such an analyst only did so for that one issuer -- so you should probably look at the ratings for all issuers that analyst worked on.


Transparency & Disclosure

Transparency and disclosure are fine. But these measures lack teeth:

Require Disclosure Of Preliminary Ratings To Reduce "Ratings Shopping"

I'm not sure exactly how this would work. But in my experience, deals can be tweaked and made safer after preliminary ratings to achieve better ratings. Issuers can just claim that poor initial ratings don't reflect the final deal. If you're really worried about shopping, why not instead force banks to accept final ratings from any agency they initially solicit?


Require Different Symbols To Be Used To Distinguish The Risks Of Structured Products

Oh please. Anyone buying a AAA structured product who doesn't already know it's different from a AAA unsecured corporate bond shouldn't be buying it in the first place.


Require Qualitative And Quantitative Disclosure Of The Risks Measured In A Rating:

Sure, but prior to the crisis, risk measures would have been incorrect too. They would have suffered from the same bad assumptions. Ratings within ratings seem like nonsense to me. Instead, why not provide the data to everyone so investors can more easily use their own assumptions and conclusions.


Reduce Reliance on Credit Rating Agencies

I find this section particularly laughable. One easy way to do this would be to get rid of the rating agencies altogether and let investors do their own due diligence. But let's look at one of the Treasury's specific suggestions:

SEC Recently Requested Public Comment on Whether to Remove References to Ratings in Money Market Mutual Fund Regulation

So what you're telling me is that you'd rather potentially have Money Markets buy bonds of any rating? Oh, that makes much more sense.


Strongly Support SEC Actions on Credit Rating Agencies

These aren't original ideas by the Treasury -- just hat tips to the SEC. But, lo and behold, a good one:

Enable Additional Ratings On Structured Products: Because structured products are often complex and require detailed information to assess, it can be difficult for a rating agency to provide "unsolicited ratings" - ratings on products it was not paid to rate. These ratings, while in existence previously, were ineffective because investors understood that these unsolicited ratings did not benefit from the same information as the fully contracted ratings. The SEC has proposed a rule that would require issuers to provide the same data they provide to one credit rating agency as the basis of a rating to all other credit rating agencies. This will allow other credit rating agencies to provide additional, independent analysis to the market.

What's this? An idea that actually seems to be heading in the right direction! Unfortunately it doesn't go far enough. Why not provide that information to everyone, instead of just rating agencies? Then investors could do as much due diligence as they want and avoid reliance on ratings entirely.


For further proof that the Treasury's proposals are barely even a slap on the wrist, the Financial Times provides the rating agencies' reactions in an article on the topic. They don't seem particularly worried:

S&P said it was studying the proposal. Moody's said it supported the goals of "increased transparency and enhanced ratings quality''. Fitch said the plans were consistent with its views on transparency.

We want to hear what you think about this article. Submit a letter to the editor or write to letters@theatlantic.com.