Andrew Sorkin - Too Big to Fail - The Inside Story of How Wall Street and Washington Fought to Save the FinancialSystem--and Themselves

With BISTROs, a bank took a basket of hundreds of corporate loans on its books, calculated the risk of the loans defaulting, and then tried to minimize its exposure by creating a special-purpose vehicle and selling slices of it to investors. It was a seamless, if ominous, strategy. These bond-like investments were called insurance: JP Morgan was protected from the risk of the loans going bad, and investors were paid premiums for taking on the risk.

Ultimately, Cassano passed on buying BISTROs from JP Morgan, but he was intrigued enough that he ordered his own quants to dissect it. Building computer models based on years of historical data on corporate bonds, they concluded that this new device—a credit default swap—seemed foolproof. The odds of a wave of defaults occurring simultaneously were remote, short of another Great Depression. So, absent a catastrophe of that magnitude, the holders of the swap could expect to receive millions of dollars in premiums a year. It was like free money.

Cassano, who became head of the unit in 2001, pushed AIG into the business of writing credit default swaps. By early 2005, it was such a big player in the area that even Cassano had begun to wonder how it had happened so quickly. “How could we possibly be doing so many deals?” he asked his top marketing executive, Alan Frost, during a conference call with the unit’s office in Wilton, Connecticut.

“Dealers know we can close and close quickly,” Frost answered. “That’s why we’re the go-to.”

Even as the bubble was inflating, Cassano and others at AIG expressed little concern. When in August of 2007 credit markets began seizing up, Cassano was telling investors, “It is hard for us, without being flippant, to even see a scenario within any kind of realm of reason that would see us losing $1 in any of those transactions.” His boss, Martin Sullivan, concurred. “That’s why I am sleeping a little bit easier at night.”

The pyramidlike structure of a collateralized debt obligation is a beautiful thing—if you are fascinated by the intricacies of financial engineering. A banker creates a CDO by assembling pieces of debt according to their credit ratings and their yields. The mistake made by AIG and others who were lured by them was believing that the ones with the higher credit ratings were such a sure bet that the companies did not bother to set aside much capital against them in the unlikely event that the CDO would generate losses.

Buoyed by their earnings, AIG executives stubbornly clung to the belief that their firm was invulnerable. They thought they’d dodged a bullet when, toward the end of 2005, they stopped underwriting insurance on CDOs that had pieces tied to subprime mortgage-backed securities. That decision enabled them to avoid the most toxic CDOs, issued over the following two years. The biggest reason, though, for the confidence within the firm was the unusual nature of AIG itself. It was not an investment bank at the mercy of the short-term financing market. It had very little debt and some $40 billion in cash on hand. With a balance sheet of more than $1 trillion, it was simply too big to fail.

Speaking to investors at the Metropolitan Club in Manhattan in December 2007, Sullivan boasted that AIG was one of the five largest businesses in the world. His company, he stressed, “does not rely on asset-backed commercial paper or the securitization markets responding, and importantly, we have the ability to hold devalued investments to recovery. That’s very important.”

He did acknowledge that AIG had a large exposure to underwriting a certain financial product whose future even then seemed dubious: tranches of credit derivatives known as super-seniors. “But because this business is carefully underwritten and structured with very high attachment points to the multiples of expected losses, we believe the probability that it will sustain an economic loss is close to zero.”

By that point in time, however, how AIG saw itself and how everyone had come to view it were rapidly diverging. The clients who bought super-seniors insured by AIG might still be making their payments, but on paper they saw their values falling. Market confidence in CDOs had collapsed; the credit-rating agencies were lowering their rankings on tens of billions of dollars’ worth of CDOs, even those that had triple-A ratings.

In 2007 one of its biggest clients, Goldman Sachs, demanded that AIG put up billions of dollars more in collateral as required under its swaps contracts. AIG disclosed the existence of the collateral dispute in November. At the December conference, Charles Gates, a longtime insurance analyst for Credit Suisse, asked pointedly what it meant that “your assessment of certain super-senior credit default swaps and the related collateral … differs significantly from your counterparties.”

“It means the market’s a little screwed up,” Cassano said, playing on his Brooklyn roots. “How are you, Charlie? Seriously, that is what it means. The market is—and I don’t mean to make light of this—actually just so everybody is aware—the section that Charlie was reading from was a section that dealt with collateral call disputes that we have had with other counterparts in this transaction. It goes to some of the things that James [Bridgwater, who did the modeling at AIG Financial Partners] and I talked about, about the opacity in this market and the inability to see what valuations are.”

The dispute with Goldman had become an irritant to Cassano. Another counterparty, Merrill Lynch, had also been seeking more collateral but wasn’t being as aggressive about it as Goldman. Cassano seemed almost proud of his ability to get these firms to back off. “We have, from time to time, gotten collateral calls from people,” he said on December 5, 2007. “Then we say to them: ‘Well, we don’t agree with your numbers.’ And they go, ‘Oh.’ And they go away.”

At a board meeting that fall, Cassano bristled when questioned about the Goldman collateral issue. “Everyone thinks Goldman is so fucking smart,” he railed. “Just because Goldman says this is the right valuation, you shouldn’t assume it’s correct just because Goldman said it. My brother works at Goldman, and he’s an idiot!”

Even before Willumstad had been given the position of CEO, he had been consumed by FP. Problems at the unit had been simmering at AIG since Greenberg had been forced to resign in 2005 as the result of another major accounting scandal. New York attorney general Eliot Spitzer had even threatened to bring criminal charges against him after launching an investigation into a transaction between AIG and a subsidiary of General Re, an insurer owned by Warren Buffett, that inflated AIG’s cash reserves by $500 million.

In late January of 2008, Willumstad had been sitting in his corner office at Brysam Global Partners when he noticed something startling in a monthly report issued to AIG board members: The FP group had insured some $500 billion in assets, including more than $61 billion in subprime mortgages—most of that for European banks. That piece of business was actually a very clever bit of financial engineering on FP’s part. To meet regulatory requirements, banks could not exceed a certain level of debt, relative to their capital. The beauty of AIG’s insurance—for a short time, at least—was that it enabled banks to step up their leverage without raising new money because they had insurance.

Willumstad did the math and was appalled: With mortgage defaults rapidly mounting, AIG could soon be forced to pay out astronomical sums of money.