The U.S. Congress held a hearing yesterday on the role of ratings agencies in the financial crisis. The ratings agencies are supposed to provide independent estimates of the risk of various financial products. It turns out that they have not been so independent or informed. The ratings agencies failed substantively, in their use of complex risk models, but also procedurally, in their conflicts of interest and failure to critically assess investment products.

The following exchange, summarized by Rachel Pulfer in the Canadian Business Online, is really amazing:

“Let me read this email to you. It’s a series of instant messages between Standard & Poor’s officials that were sent on April 7 2007.”

The speaker is John Yarmuth, a Democratic Congressman from Kansas. It’s Oct. 22, 2008, and Yarmuth, a member of the House Committee on Oversight and Government Reform in Washington, D.C., is in the process of grilling two former executives at credit rating agencies, and one current one. (Ratings agencies represent just one of many players being assigned blame in the ongoing financial crisis. For the full list of official blame-recipients being fingered by Congress, check this out.)

Yarmuth’s question is directed at Mr. Frank Raiter, the former head of the division in charge of rating mortgage backed securities at Standard & Poors.

“These employees are from the structured finance division,” Yarmuth goes on. “One writes: ‘By the way, that deal is ridiculous.’”

“The other writes back: “The model does not reflect even half of the risk.’”

“The first employee writes back: “Yeah, but we rate every deal. It could be structured by cows, and we would be rating it.’

“What exactly does this mean, Mr. Raiter, that the model ‘does not even capture half the risk?’”

Raiter takes a breath. “Collateralized Debt Obligations were driven by the diversity index,” he explains. “They are supposed to tell you if bonds are highly correlated or not. But I can remember days when I was requested to put ratings onto transactions I had not even seen.”

Raiter’s co-panelists are Jerome Fons, formerly of Moody’s, and Sean Egan, the managing director of Egan-Jones Ratings, a small independent ratings agency. Unlike Raiter’s former employer, Standard & Poors, or Moody’s, Egan-Jones Ratings receives no compensation from the issuers of securities for its ratings. The bulk of its revenue is from investors.

Yarmuth goes back to the transcript; this time, he directs his comments to Mr. Egan. “The transcript says the models ‘could be structured by cows, and we would still rate them,’” he quotes. “What does that mean to you?”

Egan takes the microphone. “Um, perhaps that cow is particularly talented?”

The committee laughs, quietly.

Seriously, Egan goes on, what it means is that the rating is ridiculous. If the investment is unsound, you don’t rate it. “It’s as simple as that. If you are a meat inspector,” Egan continues, “you don’t rate tainted meat — you have an obligation to your consumers to get rid of the meat.” But in this case, he finishes, agencies went ahead and rated the “meat” — a.k.a. the securities — because they were being paid to do so.

The problem, Egan explains, is that back in the 1970s, ratings agencies stopped aligning their interests with investors. Instead, they started accepting fees from issuers.