Congress Is Ready to Shake the Rating Agencies' World
Congress and others have exposed quite a bit about just how the issuing banks and the rating agencies allegedly worked symbiotically to give gold-standard ratings to toxic investments, resulting in big profits for both sets of institutions. Connecticut Attorney General Richard Blumenthal's lawsuit against the agencies notes that the problem wasn't with all types of ratings. Rather, it was the extremely lucrative "structured-finance" products -- those mortgage-backed securities, collateralized debt obligations (CDOs) and many other flavors of bundled assets -- that only a handful of big players issued. Structured finance changed the ratings power dynamics, enabling issuers to "ratings shop" -- send their business to the agency that provided the highest ratings. And structured instruments allegedly let issuers capture the ratings agencies the way industries try to capture governmental regulatory agencies.
One might expect that the financial meltdown would provide the necessary market discipline to change these dynamics, resulting in newly accurate ratings of structured-finance products without any government regulation. But one would be wrong: As Bloomberg reports, S&P gave AAA ratings to mortgage bonds issued in 2008-2009, two years after the meltdown started -- and has now downgraded them to junk.
Two Key Changes to Watch
By now it seems just a matter of time before some form of financial regulatory overhaul becomes law, and when it does, the ratings world will fundamentally change. Because the House and Senate versions are different, it's not yet clear what that new world will look like. But two key changes that I'm hoping survive, one from the House and one from the Senate, have the potential to make the ratings the meaningful grades of creditworthiness that investors and the public have assumed them to be.
A third change, to eliminate references in laws to "Nationally Recognized Statistical Ratings Organizations" (NRSROs), is in both bills and should be in the final legislation. This change would force federal agencies to develop more diverse and accurate methods of assessing creditworthiness, and remove the government's implicit approval of the rating agencies.
The House provision I'm rooting for would eliminate the rating agencies' special litigation-proof status, at least for lawsuits brought under federal securities law. Currently, the rating agencies are the only category of "expert" involved in issuing securities that is immune from being sued. Other investment experts are liable for the accuracy of their statements about securities. Because of that exemption, plaintiffs suing under federal securities laws have tried hard to characterize rating agencies as playing some other role in which they can be held liable, such as an underwriter or seller of securities. But plaintiffs have consistently failed in those efforts, at least to date.
Adopting the House's provision would ensure that rating agencies had to stand behind their opinions -- which is what ratings are -- just like all the other experts have to stand behind their opinions. Plus, it would cure a related problem: the rating agencies' lack of due diligence into the underlying assets that get bundled into structured-finance securities. If they can be sued for their ratings, I guarantee they'll do the fact-checking needed to be sure the issuers are telling them the truth -- the whole truth and nothing but the truth -- when preparing ratings.
Recently, New York Attorney General Andrew Cuomo opened an investigation to see if the banks lied to the agencies when seeking ratings. If Cuomo discovers that they did lie, that's one more problem that would be minimized if the rating agencies were made liable for their ratings -- a banker's lie would quickly be found out if the agencies did their own due diligence.
Eliminating Conflicts of Interest
The Senate provision I'm hoping to see in the final bill eliminates the conflict of interest regarding structured finance securities that exists with the "issuer-pays" model used by the big three agencies. (The Securities and Exchange Commission has a current list of 10 NRSROs, not all of which use an issuer-pays model.)
The provision, known as the Franken amendment (at S3652-54), would require an issuer that wants its security rated to ask a specially created board for one. That board would then assign a rating agency to do that initial rating. Issuers would be allowed to seek ratings from other agencies, and agencies would be allowed to give independent, unsolicited ratings as well, but the issuer would have no control over the initial rating. The latter point, that issuers can get additional ratings, deflates S&P's criticism that the amendment would reduce competition (this same article quotes a Morningstar analyst suggesting the amendment would increase competition.)
Ensuring that the initial rating comes from a firm not chosen by the issuer solves the agencies' conflict-of-interest problem, even though the agencies can still charge issuers for secondary ratings. With a presumably accurate benchmark rating existing, second or third ratings couldn't be wildly higher simply because the issuer is paying the rater. Big deviations from the initial rating would have to be objectively justifiable, or the rater would lose all credibility with investors, destroying the rating's value to the issuer.
Evaluating Ratings Performance
Severing the financial connection between ratings agencies and issuers is critical, but the Franken amendment goes further in several key ways. First, the board it creates has an investor bias: A majority must be members of the investment community and not represent issuers. If investor interests are central, agencies have no incentive for inflated ratings.
Second, the amendment requires the new board to evaluate the rating agencies' performance at least annually. In evaluating their performance, the board must consider, among other things, the accuracy of each agency's ratings and the effectiveness of each agency's methodology. If an independent board is looking over the raters' shoulders every year, I find it hard to believe that massive, sudden downgrades from AAA to junk (below BBB- at S&P) will ever happen again.
This annual evaluation also will force rating agencies to do due diligence because it's hard to issue accurate ratings if you don't have accurate facts. Similarly, with the board checking the agencies' work, it seems to me that competition isn't eliminated -- it's just redirected into maximizing accuracy, instead of maximizing how favorable a rating can be.
Finally, the Franken amendment narrows the revolving door between industry and regulatory body before it opens, by prohibiting board members or employees from negotiating for industry jobs unless they immediately announce such negotiations and recuse themselves from relevant transactions. It also prohibits the members and employees from receiving loans or gifts from issuers or investors.
Will the Provisions Survive?
The Franken amendment passed with strong bipartisan support (64 - 35 with 11 Republican votes), which should give both chambers incentive to keep it in the final bill. In addition, the House is routinely perceived as more progressive and reformist than the Senate, so it's hard to imagine that chamber watering down the provision. Finally, the amendment just makes a lot of sense, so I can only hope that it carries the day. All that said, nothing's a lock, and the provision is lacking one key ally: Senate Banking Committee Chairman Christopher Dodd (D-Conn.).
However, I'm not as hopeful about the House provision that removes the agencies' liability shield. That's because the Senate knew about it, and decided not to include a similar provision in its bill.
Talking Points Memo reports that the Senate is scheduled to vote on the bill this week. While it may not pass on the initial vote because of various political maneuvers, it's likely to pass ultimately, so the process of reconciling the versions should begin shortly.
The New Rating Agency Landscape
Although the big three ratings agencies are often discussed as if they're they only game in town, seven other agencies already have SEC approval to rate at least certain types of securities. In addition, corporate sleuth Jules B. Kroll announced last year that he plans to start a new agency, possibly by this July.
Kroll brings a strong brand to the market, and given the damage to the big three's brands, a real opportunity seems to exist for the new company. Although their market share hasn't declined yet, it's hard to see the big agencies maintaining their position over the medium term even if the Franken amendment doesn't survive. Considering the damage the big three have already done to their credibility, competitors should make some headway no matter what.
If the Franken amendment does survive, I imagine the major agencies' market share could shrink much more, and far more quickly, for two potential reasons. The first, and smaller one, is that the new initial ratings assignment process would send more business to other firmst. The other, more powerful, reason is the annual appraisals of ratings performance. Unless the big three do noticeably better than the other initial raters at assessing risk, their only claim to disproportionate market share will be established relationships and the force of habit.
If other ratings agencies prove themselves with ratings that perform well, the market-share landscape will likely undergo earthquake-style rearrangement. And if the big three outperform the other raters, they'll maintain most of their market share for the right reasons. Who could complain about that?
Of course, the Franken amendment could end up having little impact on ratings agencies, market share or otherwise, if structured finance products go away, because the amendment applies only to structured finance. But the market for asset-backed securities, whether those assets are mortgages, credit card debt or student loans, isn't likely to vanish completely. Nor are CDOs about to disappear. So hold on tight, Moody's, S&P and Fitch, your world is about to turn upside down.