Wharton finance prof's role in AIG debacle

Did a Wharton finance professor's mathematical models seduce and betray American International Group into taking on massive financial risk it couldn't handle? Or was AIG's 2008 failure, and the credit-market freeze, federal bailouts and economic slump that followed, all the auditors' fault, for belatedly demanding devastating write-offs that have turned out to be unnecessary?

Joseph Cassano, the high-paid executive who formerly headed that unit and has been blamed for blowing up the economy, told his side of the story yesterday to the federal Financial Crisis Inquiry Commission here with other ex-AIG and Goldman Sachs bosses and academics. Full testimonies here.

How did Cassano's group believe it could safely guarantee the home loan business? Cassano said he relied on "University of Pennsylvania Professor Gary Gorton, who served as a consultant to AIG-FP (and) worked with our experienced analysts to refine the deal structure." Gorton "used a sophsiticated actuarial model to make sure (each) proposed deal was fundamentally sound... This process was designed to minimize risk."

Despite the model, AIG traders realized the home loan business was getting dangerous back in 2005. Cassano's group stopped doing subprime mortgage deals.. The problem then became how to deal with the risk AIG had already taken on. Cassano and his colleagues convinced the company not to write off billions in potential losses; he and his colleagues "remained confident" in the assets, thanks to the analysis they'd conducted with Gorton's help.

Auditors finally overruled Cassano in early 2008, and he agreed to leave the company. But Cassano still says they were wrong. The decision "to unwind the credit-default swap contracts (was) due largely, so far as I can tell, to the proliferation of collateral calls" by nervous swaps investors, he told the commission. "The adjustment had a huge impact of the company," eventually leading to the government bailout and takeover.

Doesn't AIG's failure speak for itself? No, said Cassano: AIG's suspect CDOs ended up taken over by the government and packed into its Maiden Lane III investment portfolio, whose results are public information. Said Cassano: "As I look at the performance of some of these same CDOs in Maiden Lane III, I think there would have been few, if any, realized losses on the CDS contracts had they not been unwound in the bailout."

Ex-AIG chief risk officer Robert E. Lewis was humbler: "We weer wrong about how bad things could get," he testified. Prof. Gorton is now at Yale University, where he didn't immediately respond to my queries about what he would have done differently, if he'd known in the mid-2000s what we know now about subprime home loans and credit-default swaps.