Almost no one disputes that the rating agencies played a major role in bringing on the financial crisis. To further investigate the mistakes they made, the Senate Committee on Homeland Security and Governmental Affairs held a hearing Friday where many rating agency representatives testified. There was also an enormous exhibits document (giant .pdf) referred to with facts and figures demonstrating the agencies' epic fail, along with scores of e-mails showing their misdeeds. The lessons learned were mostly those we already knew, however: the agencies suffered from deep conflicts of interest and not enough competition.
To see just how poorly a job the agencies did in rating mortgage-related securities, you need look no further than this chart, from the exhibits provided:
You should read this chart as their percentage of failure. In the 2006 and 2007 vintages, that means their ratings ended up being generally incorrect at least 90% of the time. How could they be so wrong? Because they didn't take into account that this could be a housing bubble (click on it for bigger image):
Amusingly, this exhibit appears to come from a Paulson & Co. presentation, the firm famous for betting against the housing market in 2007, and at the center of the Goldman-SEC case. What made the agencies ignore the historical trend line above and claim that the incredibly steep rise in prices was a new normal?
The investment banks and loan issuers were the clients of the rating agencies. If an agency decided to rate deals more conservatively, then they feared their business would flee elsewhere to find one of the other agencies that were happy to be more aggressive to get another paycheck. Numerous intra-agency e-mails in the exhibits demonstrated this point. There was incredible pressure to maintain lax standards to keep the dollars flowing.
This, however, could be easily remedied. One solution would be to revamp the system so that the agencies are no longer paid by banks and issuers. Instead, restructure things so that all agencies can rate deals and still get paid for their work. This could be done by building a fee into deals where investors pay a small fee for rated transactions. Then, divide up that fee between the agencies who choose to rate the transaction.
Not Enough Competition
The other problem stems from the fact that there's an oligopoly of ratings firms -- only three. With that little competition, it's hard to get much diversity of opinion. If one agency decides to relax standards, the others feel pressured to follow.
The clearest solution to this would be to reduce the barriers of entry into the ratings market. Right now regulations make it very difficult to become certified as a rating agency. As long as that's the case, the big three will continue to rule the market.
Instead the current framework should be completely dismantled. Financial firms and third-party research houses should get in the business of evaluating bonds, just like equity analysts do for stocks. In that case, there is often a diversity of views on whether to buy, hold or sell various companies' equity. In much the same way, research arms could evaluate asset-backed securities instead of the market relying on the three agencies. This would also solve the pay problem, since these analysts would derive their income through their research services -- paid by investors.
Unfortunately, the current financial reform effort in the Senate does not include any such solutions to the rating agency problem. It utterly fails to address the incentives issue and the lack of competition. Page 8 of the bill's summary (.pdf) shows how little the legislation would do to fix these problems that clearly contributed to the financial crisis.