For the Department of Justice's basic theory of the case to make sense you have to believe that the investment banks creating collateralized debt obligations wanted high ratings at any costs-even if the ratings were highly inaccurate.
This hasn't proved and probably cannot be. Wall Street believed the ratings.
The weakest point of the Department of Justice's case against Standard & Poor's is the government's assumption that there was something wrong with the credit rating agency changing its standards to win more business.
The 119-page civil complaint filed Monday night by the Justice Department abounds with evidence, much of it from emails and instant messages between S&P employees, that concern over the company's market share influenced its ratings decisions. For example, one employee directly complains that S&P had lost market share because its standards required 10 percent more collateral than Moody's.
But this only looks bad if you make the additional assumption that the issuers of the credit products S&P was rating did not care about the quality of the ratings. That is, for the Department of Justice's basic theory of the case to make sense you have to believe that the investment banks creating collateralized debt obligations wanted high ratings at any cost-even if the ratings were highly inaccurate.
This idea that issuers recklessly demanded high ratings for even the worst bundles of mortgage-related assets is deeply ingrained in public discussions about the financial crisis. It has become part of the conventional wisdom.
It has not, however, ever been established in any legal arena. Which means that the Justice Department will have to prove that issuers demanded fraudulently high ratings in order to establish the claim that the agencies engaged in fraud by changing ratings in pursuit of market share.
There's not much by way of evidence in the Department of Justice's complaint that issuers demanded fraudulently high ratings. This is surprising since the entire case turns on this question. It may be that the government is so enthralled with the conventional wisdom on this question that it does not believe this can seriously be challenged.
To get an idea of how hard this will be to prove just imagine who could be called upon to provide evidence. Will the issuers themselves admit that they wanted high ratings regardless of the actual quality of the assets? Of course not. This would instantly open them up to lawsuits from the purchasers of those assets. It seems unlikely that the government has evidence of direct collusion between the issuers and S&P. If it had this, it would have made it into the complaint.
Nonetheless, S&P will have to provide an answer this central question: how can the company's evident concern with market share be squared with the idea of providing accurate ratings?
The answer is quite straight forward: S&P could have seen deteriorating market share as a market vote on the accuracy of its ratings.
That is to say, S&P can argue that it wasn't engaged in a race to the bottom with rival firms such as Moody's and Fitch. Instead, it might have been in a race to quality and accuracy. Or, at least, what the customers for credit ratings believed were quality and accuracy. If another ratings agency's standards were winning more business, perhaps that ratings agency was just viewed as having more accurate standards. In that case, it makes sense for S&P to adjust its standards.
There's nothing shocking about this. When almost any major company in America notices it is losing market share, it at least considers adjusting the product. To do this, it will look at what the competitors are doing to win market share. If it decides to make adjustments to its major products, it performs tests on the ideas. Ask customers, take some surveys, focus group the thing. This is what S&P seems to have been doing.
The Justice Department is so blind to this possibility that it includes in its complaint an email from an executive-known in the complaint as "Executive H"- upset that proposed changes to rating criteria would include consideration of "market insight," ratings implications, and interviews with "3 to 5 investors in the product."
"What do you mean by 'market insight' with regard to a proposed a proposed criteria change? What does rating implication have to do with the search for truth?" Executive H complains.
This isn't the voice of someone who believes in market processes. It's more like the voice of a socialist planner who thinks that his "search for truth" gives him unique access. It is the voice of someone who believes he knows better than the market because his mathematical models tell him the market is wrong. (For more on this, read Matt Levine at Dealbreaker's post, where he describes Executive H as a "quant truther.")
The Justice Department's case depends on this kind of contempt for the idea of paying attention to market insight. And this contempt may wind up undermining the entire case against S&P.
Follow me on Twitter @Carney