All The Devils Are Here: Unmasking the Men Who Bankrupted the World (64 page)

At that November 29 meeting, the one in which the top brass for FP and AIG finally met to hash out the situation, Cassano told the others that FP was already in the process of “going to ground” to create a new model that would allow it to value the super-seniors as quickly as possible. Yet at the same time, he once again downplayed the importance of the collateral calls. “Collateral calls are part of the business,” he shrugged, adding that he “does not see this as a material issue with GS or any of the other counterparties,” according to the notes of the meeting.

Then he was asked how the dispute might affect AIG’s profits for the upcoming quarter. “JC noted if we agreed to GS values could be an impact of $5bn for the quarter,” the notes read. “MS”—Martin Sullivan—“noted this would eliminate the quarter’s profits…. JC noted that this was not what he was proposing but illustrative of a worse [sic] case scenario.” Forster would later tell the Financial Crisis Inquiry Commission that, upon hearing the $5 billion figure, Sullivan said the number would give him a heart attack. (Sullivan later testified that he didn’t remember saying that.) And with that, the meeting ended.

Or so Cassano thought. In fact, after the FP executives got off the phone, the accountants stayed in the room with Sullivan and Bensinger to discuss what they had just heard. If the first part of the meeting had been troubling for Sullivan, this latter part was even worse.

One gets the sense, reading the notes of the meeting, that the Cassano conversation was the last straw for the accountants. PWC lead Tim Ryan was not nearly as calm about the collateral calls as Cassano had been; on the contrary, he was quite agitated. He could see, in a way the AIG executives themselves could not, how their poor risk management practices were creating problems. He listed some of the things that bothered him: The fact that FP had posted $2 billion in collateral without bothering to inform headquarters. The way FP was “managing” the valuation process of the super-seniors. The growing exposure at the securities lending program. And the fact that the risk managers had inexplicably allowed the securities lending program to increase its exposure to subprime securities at the same time that FP was reducing its exposure.

“While no conclusions have been reached,” Ryan told Sullivan and Bensinger, according to the notes, “we believe that these items together raise control concerns around risk management that could be a material weakness.”
For Sullivan, there were no two scarier words than “material weakness.” If that wound up being the accountants’ conclusion, it would have to be disclosed to investors—and that would be devastating. He promised to do whatever he had to do to avoid such a declaration. And on that sobering note, the meeting finally ended.

AIG had a long-scheduled investors’ meeting set for Wednesday, December 5, 2007. The planned topic was the company’s life insurance and retirement services businesses. But as the rumors continued to swirl about AIG’s subprime exposure, Sullivan decided to change the focus. The company would talk instead about its credit default swap business, along with the rest of its exposure to the mortgage market.

It was a very long meeting. Sullivan began by noting AIG’s profitability over the past few years, its strong capital position and cash flow ($30 billion in the first nine months of 2007), and its lack of debt. “We have the ability to hold devalued investments to recovery,” he told investors. “That’s very important…. AIG-FP has very large notional amounts of exposure related to its super-senior credit derivative portfolio. But because this business is carefully underwritten and structured … we believe the probability that it will sustain an economic loss is close to zero.”

Over the course of the day (with a break for lunch), fourteen AIG executives made presentations—including Cassano, Forster, model expert Gary Gorton, and Bob Lewis. Every one of them said essentially the same thing: there was little or no chance that the tranches AIG had either insured (in the case of FP) or bought (in the case of other AIG divisions) could ever lose money. Of the fourteen, nobody said this more fervently, or more often, than Cassano. “[W]e have an extremely low loss rate in these portfolios and the underlying reference obligations have a relatively low downgrade migration from the rating agencies,” he said in a typical remark. “It is very difficult to see how there can be any losses in these portfolios.” (Four months earlier, during an earnings call, Cassano had made a similar remark: “It is hard for us, without being flippant, to even see a scenario … that would see us losing one dollar in any of these transactions.” That line would come back to haunt Cassano, as it was quoted ad nauseam in the aftermath of the crisis.) At least half a dozen times he rolled out all the explanations he had been using to push back against Goldman: The due diligence that had gone into assembling the subprime tranches AIG insured. The fact that it had little or no exposure to 2006 and 2007 vintages. The amount of subordination in the CDOs AIG insured, meaning that hell would have to come close to freezing over before any of AIG’s super-seniors defaulted. He acknowledged that FP was in disputes with counterparties over marks but described those disagreements as “parlor games.”

“There is a major disconnect in the market,” he claimed, “between what the market is doing versus the economic realities of our portfolio.” In other words, in Cassano’s view, the market was simply wrong. And since the market didn’t understand the strength of AIG’s underlying collateral, he was damned if he was going to begin marking it down in any meaningful way. (He did tell the gathering that AIG was writing down another $500 million in November, but that was a pittance in the grand scheme of things.) “If you ask me how I manage the business,” he said, “it’s the fundamental underwriting that is the first line of defense, the first line of protection, the first thing that gets you comfortable in this business.” Even now, months after the collateral calls began, Cassano still seemed unable to comprehend that the issue he was facing had nothing to do with the “fundamental underwriting” of the CDOs AIG insured. The issue was that the collateral triggers were putting the entire corporation at risk. AIG may have had plenty of capital, as Sullivan had suggested, but because it was an insurance company, that capital was strictly regulated and very little of it could be used to shore up AIG-FP as it faced the growing onslaught of collateral calls. The notion that FP was invulnerable because of its parent’s financial strength—a notion the market had accepted for years—was suddenly exposed as a giant illusion. It was just the opposite: FP’s sudden vulnerability to liquidity risk was endangering the larger company. That’s what Cassano didn’t understand.

When the time came for questions, most analysts seemed to accept Cassano’s version of reality. Several, however, did not. One investor—unidentified in the transcript of the meeting—while acknowledging to Cassano that “you’ve clearly demonstrated no economic loss,” asked what should have been an obvious question: “[W]hat if you did use the ABX index and the counterparties? What would your marks be?”

“It’s nonsensical,” Cassano replied curtly.

“But what would the nonsensical number—?”

“I don’t know,” Cassano cut him off. “It’s nonsensical.”

“Could it be north of $5 billion?” the investor pressed.

“You know I have no— Do you have any idea? I don’t know. Look, we’re in the business of going to the core of the fundamentals. The ABX is just not representative of the pool of business that we have.” And that was that.

A few minutes later, Josh Smith, an analyst at TIAA-CREF, the financial services giant, posed another important question. “I noticed that some of the underlying collateral has been replaced with ’06/’07,” he began. “I think people take a lot of comfort that you stopped writing the ’06/’07. Can you quantify the risk that the underlying collateral from the earlier vintages gets replaced with this ’06/’07 stuff, which isn’t as good?”

Here was something else almost no one had noticed before—either inside or outside of AIG. For all of FP’s pride in having ended its multisector CDO business in 2005, it simply was not true that the referenced securities didn’t include those terrible 2006 and 2007 vintages. A number of the CDOs that AIG insured allowed for the CDO manager to replace older subprime bonds with newer ones—bonds that would invariably generate higher yields precisely because they were riskier. AIG didn’t even have to be informed that the collateral was being swapped out.

Take, for example, the $1.5 billion CDO known as Davis Square III, which Goldman Sachs underwrote in 2004. The CDO manager, Lou Lucido, worked for the Los Angeles investment firm TCW Group. During much of 2006 and 2007, Lucido was busy boosting the yield on Davis Square III by putting in subprime bonds from later vintages and kicking out many of the bonds that had been in the CDO when AIG agreed to insure it. Bloomberg estimated that, by 2008, “Lucido’s team, following criteria set by Goldman Sachs, changed almost one-third of the collateral in Davis Square III.” By May 2008, Davis Square III had been downgraded to junk, costing AIG $616 million in additional collateral calls—which came, of course, from Goldman Sachs.

At the investor meeting, however, none of this was divulged. When asked point-blank what percentage of AIG’s collateral was 2006 and 2007 subprime vintages, Forster—whom Cassano had kicked the question to—said he didn’t know.

Still, in the immediate aftermath of the meeting, the market seemed pleased. AIG’s stock had been around $58 a share in the week preceding the investor meeting; after the meeting, it got a nice little pop, to $61 a share. And the meeting seemed to have energized Cassano as well. Two days after the meeting, on December 7, FP sent Goldman a letter demanding the return of $1.5 billion in collateral. Goldman, of course, refused. Several of the new marks that AIG-FP provided showed FP valuing the securities at par. David Lehman would later tell the Financial Crisis Inquiry Commission that AIG-FP’s valuation was “not credible.” He was right.

Though he didn’t realize it, Cassano’s biggest problem wasn’t Goldman Sachs. It was Tim Ryan at PricewaterhouseCoopers. All through November and December, in meetings with management, with the audit committee, and with the full board of directors, Ryan continued to raise concerns about the way FP was valuing the super-seniors, about the way it was managing the process, and about the inability or unwillingness of AIG management to get involved. While Cassano was focused on fending off more collateral calls—by the end of the year counterparties were demanding $2.7 billion, of which $2.1 billion were demands from Goldman—Ryan was making the case that AIG could not continue to allow Cassano and his FP team to manage the situation themselves.

It wasn’t until the beginning of 2008 that headquarters finally got involved, but by then it was too late. In a mid-January meeting with the audit committee, according to the notes of the meeting, “Mr. Habayeb believes that he is limited in his ability to influence change, and the super-senior valuation process is not going as smoothly as it could.” Ryan responded, essentially, that this was not acceptable.

Meanwhile, Cassano was scrambling to come up with a value for the portfolio in time to report year-end results in early February. It was clear that there were going to have to be more write-downs. Using a theory he called a “negative basis adjustment,” Cassano estimated the write-down would be $1.2 billion. (Essentially, he was claiming that this adjustment reflected the difference between the way the swaps were priced and the way the underlying securities were priced.) Without this adjustment, the write-down would be $5 billion. The board—and the accountants—first learned about Cassano’s theory in a January board meeting. The auditors were not pleased, and they would have the final say. Over the next few weeks, Cassano attempted to convince the auditors that the negative basis adjustment was a legitimate valuation method. But Ryan wasn’t biting. It had no basis in accounting rules, he said.

In late January, Ryan dropped the hammer, declaring that AIG had “a material weakness in its internal control over financial reporting and oversight relating to the fair value of the AIG-FP super-senior credit default swap portfolio.” On February 5, AIG released the news of the material weakness in an SEC filing. The stock sank. Counterparties that had previously sat on the sidelines began demanding collateral. Cassano was furious. The “material
weakness” announcement had “weakened our negotiation position as to collateral calls,” he wrote in an e-mail.

But it was over for Cassano. The board no longer trusted him and insisted that Sullivan fire him, something Sullivan was still reluctant to do, according to a former AIG executive. Cassano made it easy for him.

“Joe, we have these issues,” Sullivan said.

“Should I retire?” Cassano replied.