Economists have offered all sorts of competing theories about exactly what went wrong to cause the financial crisis. Those theories place the blame on lots of different parties. But the one thing that most everyone agrees on is that the credit rating agencies should shoulder some of the blame. Oddly, however, the rating agencies seem to have been of little concern in the Obama administration's regulatory suggestions released earlier this week. I have three suggestions for reforms Washington might want to consider.
Before getting into those suggestions, what exactly did Obama's financial regulatory proposals recommend for the rating agencies? According to a Reuters article:
The plan urges Moody's Corp's Moody's Investors Service, McGraw-Hill Cos Inc' Standard & Poor's and Fimalac SA's Fitch Ratings and others to bolster the integrity of their ratings, especially in structured finance.
It also calls for reduced conflicts of interest and for regulators worldwide to tighten oversight.
But the blueprint does nothing to address what critics call the industry's key shortcoming: That the biggest agencies are paid by issuers whose securities they rate, creating an incentive to win more business by assigning high ratings.
Those seems like pretty broad strokes to me. Essentially, they have to explain their assumptions better. This doesn't seem to scream: "Problem Solved!" In that same Reuters article, Jonathan Macey, deputy dean of Yale Law School sums up the implications of the new regulations well:
The overall impact of existing and proposed regulatory changes on rating agencies is extraordinarily easy to summarize: They reward abject failure.
I agree. But I'm not sure I buy into the solution that many stand by, described at the end of my first block quote above. It's easy to complain that issuers shouldn't be the ones to pay for the ratings, but now else do the rating agencies get paid? Do you rely on investors? The problem here is that investors can't decide which rating agencies should rate a deal before they buy the securities; they buy those securities at least partially based on what those ratings are. It's a nice little catch-22. One way or another, the issuer needs to decide which rating agencies to use, which can always result in shopping around for the best ratings.
I don't pretend to have a definitive solution to the rating agency problem, but I have three suggestions that might be interesting to consider.
Maybe Rating Agencies Need Some Skin In The Game
One of Obama's new financial regulations would require that mortgage issuers have some skin in the game and keep a portion of the mortgages they originate. I explained that this solution seems nice in theory, but in practice these originators had plenty of skin in the game already. It didn't help. But if you are a believer in this logic, then maybe rating agencies should also have some skin in the game.
How would that work? What if they were paid for their ratings in the securities they rate? Instead of collecting a $10,000 cash fee, maybe a portion of that fee, or even the entire fee, should be paid with bonds from the security they're rating. I can't think of a better way to tie their compensation to their performance. Of course, they would also have to hold those bonds to maturity.
Lift Regulatory Barriers
That's right: maybe less regulation would help. In the case of the rating agency market, there are very high barriers to entry. For a rating agency to be taken seriously, it must be a "Nationally Recognized Statistical Rating Organization." It's not easy to qualify.
The Rating Agency Reform Act of 2006 (which was obviously very effective), requires that the rating agency has been in business for three years before applying. But in order to apply you need to have 20 large issuer subscribers. How do you get those subscribers without first being a NRSRO? You probably don't. That's why the oligopoly of rating agencies in the U.S. is pretty safe.
I'd suggest loosening those requirements. Some might still argue that there is an institutional bias towards using the established rating agencies, so the new guys will be ignored anyway. In 2006, I would have agreed with that. In 2009, I'd say that investors might be very open to some new rating agencies, given the oligopoly's track record over the past few years.
Get Rid Of Them Altogether
When it comes to bonds, everyone relies on the rating agency ratings. But you might have noticed that there's another enormous public security market completely devoid of their ratings: the stock market. How do stocks get by without ratings? Well they don't, at least not entirely. There are equity research firms and groups at banks that examine stocks and publish their analyses. That gives investors some guidance about whether or not to purchase a stock.
Why can't we have the same thing for the bond market? You can say that bonds are more complicated, but I'd argue that bonds are actually much simpler. Bonds are just cash flow payments based on some assumptions. There are at least as many variables that can affect stock prices as bond prices. The data required for legitimate analysis might have made this impossible 15 or more years ago, but where technology and computing are today, I don't believe the data should be a significant barrier.
Let the banks structure the bonds however they feel their assumptions would lead investors to evaluate them. Banks can even offer their view of each tranches' quality. Then, new debt research groups, either privately run or in other banks, can evaluate the bonds for themselves. Investors can then trust their favorite research department, or just do the analysis themselves.