The real villains of the mortgage crisis.

By this point, everyone knows the rough outlines of the subprime mortgage mess. For much of the past year, we have been told of homebuyers pushed out of houses whose risky mortgages they never should have qualified for in the first place. During that time, bad loans have forced more than 140 mortgage companies out of business. The fallout has spurred shakiness in global markets that has yet to abate.In assigning blame for this fiasco, most people have focused on the mortgage lenders who took advantage of homeowners and the investors who enabled these lenders by buying up bits and pieces of all the mortgages. But one major source of responsibility for the subprime disaster has, until recently, been largely overlooked: the ratings agencies. These government-chartered corporations-- officially known as Nationally Recognized Statistical Rating Organizations (nrsros)--are supposed to provide the public with timely and consistent insights into the credit risks and financial condition of investments and financial institutions. They failed to do so. The reason appears to be a massive conflict of interest between the supposedly objective major ratings agencies--Moody's, S&P, Fitch--and the companies they were assessing. To understand how this happened, you first have to understand the crucial role that ratings agencies play in the loan industry. Basically, lenders make the loans, then sell them off in blocks to investment banks, which slice them up in various ways, recombine them, and resell them to investors. Over the last few years, new corporations were created specifically to buy up and sell interests in those mortgage-backed securities and collateralized debt obligations--that is, the sliced, diced, and recombined mortgages bought and sold on Wall Street. Insurance companies, banks, pension funds, and hedge funds would then invest in the securities of these companies. The reason they were so eager to invest was that these securities (known as "structured" securities) were rated "investment grade"--meaning that they were viewed as carrying very low risk--while providing far higher returns than similarly rated corporate, municipal, or treasury bonds. It was the ratings agencies that had bestowed these labels on the securities. And the labels mattered: Various global financial regulators, treating the ratings agencies as if they were objective onlookers, mandated that banks, insurance companies, and pension managers could only buy securities that were highly rated by nrsros. In effect, regulators were entrusting much of their own oversight responsibility to the ratings agencies. The problem was that the ratings agencies faced a huge conflict of interest. Not only were they vouching for the securities' credit soundness, they were being paid large fees by the issuers of the securities to do so. They were also deeply involved in setting up the structures these companies had to conform to in the first place. In fact, almost half of nrsro revenues come from rating companies that functionally do not exist until the ratings agencies bless them. This wasn't always the way the industry worked. Between 1909, when John Moody began rating the creditworthiness of railroads, and 1970, when the Penn Central Transportation Company defaulted on a commercial paper obligation, the revenues of ratings agencies came almost exclusively from subscription fees of investors who received the firms' research. After the Penn meltdown, however, debt issuers began to pay ratings agencies to evaluate them as a way to stimulate investor confidence in the quality of their credit. Thus began the conflict of interest. These conflicts worsened dramatically during the real estate boom from 2002 to 2007 and the accompanying growth of structured securities. The ratings agencies' profit margins on these deals were enormous. Suddenly, the more deals they could justify rating highly, the better their earnings--and the less incentive they had to rate conservatively. Consider, for example, that the ratings agencies didn't even bother to require issuers to disclose critical pieces of information about the underlying mortgages of the securities they were rating. Between 2002 and 2007--the entire subprime boom and beginning of the bust--Moody's didn't use, as a primary driver of its analysis, one of the three most generally accepted predictors of default: information about the ratio of borrowers' debt to their income. Moody's also didn't use key information about when an adjustable rate mortgage would adjust to a higher rate-- a common cause of defaults. Nor did it require information about the highest level an interest rate could adjust to. There's no way a ratings agency could have made sound judgments about these securities without this information. Yet the agencies continued to determine that the majority of these securities deserved investment-grade ratings. Perhaps of greater importance, the ratings agencies did a poor job of keeping their ratings up to date. The agencies regularly receive hundreds of thousands of dollars for their initial rating of a security, but then charge the issuer only a few thousand dollars a year to monitor those ratings. As a result, the ratings agencies have little incentive to keep their ratings current after the securities are sold. Even where the ratings agencies recognize failures in their models that cause them to change their rating processes, they rarely re-rate existing securities until after the impairments are already obvious to investors. In a nutshell, this fueled the subprime boom and bust. Because lenders knew there would always be investors eager to buy up their mortgages, they never had to worry about holding onto them themselves. When you know you're going to be passing off a loan to someone else, and you know they're going to accept it if it is highly rated, you relax your standards, avert your eyes, get the borrower to sign on the dotted line, and ship the mortgage off to Wall Street, where the ratings agencies will slap a veneer of respectability onto it. If all this sounds familiar, it should. The role of the ratings agencies in the world of structured finance is analogous to the role played by many Wall Street research analysts during the dot-com boom--when they simultaneously vouched for and profited from various companies. But, while the situations are similar, this one is probably worse. In the stock world, SEC filing and disclosure requirements allow investors to do their own analysis. By contrast, there is little public information available and few disclosure requirements on the structured securities the ratings agencies evaluate. What's more, while no one was required to listen to Wall Street analysts during the dot-com boom, regulators require many investors to use the agencies' ratings as the basis for their investments in structured securities. Historically, there hasn't been much recourse against the ratings agencies. The SEC has generally exempted them from liabilities unless they could be proven to have been downright reckless. And the prevailing legal precedent has always treated them as journalists. That's right--journalists. This, of course, is ridiculous. Journalists are expected to test the accuracy of information from all sources in pursuit of the truth. Yet, despite the fact that international securities regulators require the ratings agencies to use all available information that may be relevant to their analysis, the agencies rely almost exclusively on information provided by their customers. Indeed, the ratings agencies themselves generally claim they're not in the business of independent verification. For instance, one agency recently insisted it had "no obligation to verify or audit any information provided to it from any source" or to "obtain any information that the issuer has not otherwise provided." This may be starting to change. In 2002, Fitch and Moody's were issued subpoenas to hand over e-mails and other documents regarding their relationship with a defendant investment bank. Fitch refused, claiming protection under New York's press shield law. But a U.S. appellate court found that Fitch had played a fairly active role in commenting on proposed transactions, even offering suggestions about how to model the transactions to reach the desired ratings. Therefore, it ruled that "Fitch was not entitled to assert the journalist's privilege." This development--combined with a 1995 U.S. appellate court ruling which held that any party that takes steps "necessary" to the sale or distribution of securities can be deemed an underwriter-- suggests that lawsuits growing out of the current crisis might force the ratings agencies to start living up to the public responsibilities that have enriched them. And that would be long overdue.

By Joshua Rosner; Joshua Rosner is managing director of Graham Fisher, an independent research consultancy.