The CDOs That Destroyed AIG: The Big Short Doesn't Quite Reveal What They Knew and When They Knew It

It's been eighteen months since AIG collapsed, and Congress has yet to seriously focus on the most important questions: What did they know and when did they know it?

"What" refers to the fatal flaws in the collateralized debt obligations, or CDOs, that AIG insured.

"They" are the bankers that structured and sold the CDOs, plus the AIG executives who took on the credit risk, plus the rating agencies that handed out AAA ratings.

"When" harkens back to 2005 and 2006, when those toxic CDOs were first issued.

Just before the backdoor bailout of AIG's banks actually closed, Michael Lewis addressed what they knew and when they knew it in Portfolio. He explained why the CDO market was ground zero for Wall Street malfeasance that led to the meltdown:

The funny thing, looking back on it, is how long it took for even someone who predicted the disaster to grasp its root causes... [Fund manager Steve] Eisman knew subprime lenders could be scumbags. What he underestimated was the total unabashed complicity of the upper class of American capitalism. For instance, he knew that the big Wall Street investment banks took huge piles of loans that in and of themselves might be rated BBB, threw them into a trust [i.e. a CDO], carved the trust into tranches, and wound up with 60 percent of the new total being rated AAA.

But he couldn't figure out exactly how the rating agencies justified turning BBB loans into AAA-rated bonds. "I didn't understand how they were turning all this garbage into gold," he says. He brought some of the bond people from Goldman Sachs, Lehman Brothers, and UBS over for a visit. "We always asked the same question," says Eisman. "Where are the rating agencies in all of this? And I'd always get the same reaction. It was a smirk."

The Disaster Created Under Hank Paulson's Watch

That collective smirk reflected more than the bankers' contempt for the rating agencies' analyses. It was a way of maintaining deniability about CDO investments that were obviously designed to become insolvent at the time they were created. In The Big Short, Lewis expands on this point:

[T]here were large sums of money to be made, if you could somehow get [triple-B mortgage bonds] re-rated as triple-A, thereby lowering their perceived risk, however dishonestly and artificially. This is what Goldman Sachs had cleverly done.

This all started at the end of 2004, when:

Goldman was in the position of selling bonds to its customers created by its own traders, so they might bet against them...

According to a former Goldman derivatives trader, Goldman would buy the triple-A tranche of some CDO, pair it off with the credit default swaps AIG sold Goldman that insured the tranche (at a cost well below the yield of the tranche), declare the entire package risk-free, and hold it off its balance sheet. Of course, the whole thing wasn't risk free: If AIG went bust, the insurance was worthless, and Goldman would lose everything. Today, Goldman Sachs is, to put it mildly, unhelpful when asked to explain exactly what it did, and this lack of transparency extends to its own shareholders. "If a team of forensic accountants went over Goldman's books, they'd be shocked at just how good Goldman is at hiding things," says one former AIG FP employee, who helped to unravel the mess, and who was intimate with his Goldman counterparts.

The guy in charge of all this, Goldman's CEO, was Hank Paulson. When he moved over to Treasury, Paulson acknowledged that the subprime bubble, which he helped foment, was central to destabilizing the markets. As he wrote exactly two years ago:

The turmoil in financial markets clearly was triggered by a dramatic weakening of underwriting standards for US subprime mortgages, beginning in late 2004 and extending into early 2007.

[Those are Paulson's italics, not mine.]

Goldman now says that it didn't manipulate anything; it simply responded to market demand. Or as Lloyd Blankfein testified, "what we did in that business was underwrite to [] the most sophisticated investors who sought that exposure." Of course, a lot of so-called sophisticated investors were played for suckers. Just ask Bernie Madoff's clientele.

According to Lewis, the guys at AIG who bought Goldman's deals had no idea that they were so heavily exposed to subprime residential mortgages. I still find this part of Lewis's story too weird to be believable. Most of the deals disclosed investment schedules, like Appendix B for Adirondack 2, which were pretty easy to eyeball.

But even smart people can be fooled by CDO terminology, which is Orwellian by design. Consider the super-senior tranches of high grade multi-sector CDOs that AIG insured via credit default swaps. Where else in the English-speaking world does "multi-sector" translate into "singularly invested in risky real estate mortgages"? Where else are "super-senior" tranches exclusively invested in deeply subordinated claims? And how is it that "high grade" CDOs are differentiated their mezzanine counterparts by a 2% sliver of capitalization, a virtual rounding error?

Lewis also writes that these CDO deals were never seriously questioned by AIG's then-CEO, Martin Sullivan. In June 2008, Sullivan was fired and replaced by Bob Willumstad, an outsider who had first joined AIG's board in April 2006, after the AIG had decided to stop insuring subprime CDOs.

In September 2008, the one thing that AIG had going for it was a CEO who had no reason to defend the toxic CDO deals that closed in 2005 and 2006. Willumstad could look regulators and investors in the eye and agree with Lewis's assessment:

Goldman created a bunch of multi-billion dollar deals that transferred to AIG the responsibility for all future losses from $20 billion in triple-B-rated subprime mortgage bonds. It was incredible: In exchange for a few million bucks a year, this insurance company was taking the very real risk that the $20 billion would simply go poof.

So Paulson unilaterally replaced Willumstad, and installed a crony, Goldman director Ed Liddy, who would never challenge the dodgy CDOs. In On the Brink, Paulson's lobotomized financial history, he goes after Willumstad with a passive aggressive smear. He recounts a comment from a former Goldman partner, billionaire investor Chris Flowers, at a meeting to discuss financing options for Lehman, on Saturday, September 13, 2008:

As everyone got up to leave, Chris Flowers motioned me aside and said, "Hank, can I tell you what a mess it is over at AIG?" He produced a piece of paper that he said showed AIG's day-to-day liquidity...Flowers told me that according to AIG's own projections the company would run out of cash in ten days.

"Is there a deal to be done?" I asked.

"They are totally incompetent," Flowers said. "I would only put money in if management was replaced."

I knew AIG was having problems--its shares had been pummeled all week--but I didn't expect this. In addition to its vast insurance operations, the company had written credit default swaps to insure obligations backed by mortgages. The housing market crash hurt AIG badly, and it had posted losses for the past three quarters.

With his "I-didn't-expect-this" story, Paulson expects us to believe that he was surprised to learn the exact problems that were laid out by AIG to the Fed, which kept Treasury fully apprised at all times, 48 hours earlier. On September 11, 2008, AIG approached the New York Fed, which simultaneously informed Treasury, to inform all concerned that AIG was running out of cash because it was facing a ratings downgrade that was caused by credit default swaps on subprime mortgages. On that very day in that very building, New York Fed employees were trying to determine if AIG's bankruptcy would have presented an unacceptable systemic risk.

"I have been blessed with a good memory, so I almost never needed to take notes," writes Paulson, who also testified, "I did not know -- I had no knowledge of the size of the [CDO] claim of any bank." That must mean that the topic never came up during the 24 different phone calls he had with Lloyd Blankfein during the week that AIG was bailed out. Apparently, the information was never conveyed by his personal proxy, Dan Jester, who "was calling many of the shots at the insurer between mid-September, when the New York Fed decided to go ahead with the bailout, and the end of October 2008, when Jester was replaced at A.I.G. by another Treasury official because, according to The New York Times, of Jester's 'stockholdings in Goldman Sachs.'"

Paulson's little hit-and-run smear against Willumstad was intended to distract us from the source of the mess and to conflate blame on to those tasked with the cleanup. On the Brink never recounts Paulson's personal role in role in destroying AIG, his decision to replace Willumstad with Liddy, or his own analysis dated March 13, 2008. In typical "Who me?" fashion, Paulson decries the problems with opaque CDOs, but never mentions Goldman's pivotal role in creating the disaster. Lewis writes:

Goldman Sachs had created a security so opaque and complex that it would remain forever misunderstood by investors and rating agencies: the synthetic subprime mortgage bond-backed CDO, or collateralized debt obligation...[I]t didn't require any sort of genius to see the fortune to be had from the laundering of triple-B-rated bonds into triple-A-rated bonds. What demanded genius was finding $20 billion in triple-B-rated bonds to launder...To create a billion-dollar CDO composed solely of triple-B-rated subprime mortgage bonds, you needed to lend $50 billion in cash to actual human beings. That took time and effort. A credit default swap took neither.

Those synthetic CDOs, including the notorious Abacus CDOs, were not sold by AIG to the New York Fed, which only financed securities holding "real" assets. They remain on AIG's balance sheet, shrouded in secrecy.

The CDO Market Remains A Bunch of Black Boxes

The bankers and hedge fund managers who made billions selling these toxic CDOs are still smirking. They made billions by shorting those subprime bonds and CDOs, but almost all of their handiwork remains hidden, concealed from public view. The Big Short, The Greatest Trade Ever and The Quants never give us specifics. The authors never identify the particular CDOs that Greg Lippman, John Paulson or Alec Litowitz bet against. Without the actual details on the trades, we must rely on the hearsay narratives of three journalists; we cannot examine the hard evidence to trace through to what they knew and when they knew it.

CDOs are not like regular mortgage bonds, which may be scrutinized via their initial prospectuses registered with the S.E.C. Bonds such as GSAMP Trust 2005-HE4 are structured so that the mortgage pool is essentially fixed at closing. What you see is what you get. Actual bond performance is available, for a price, from ABSNet.

CDOs are different. Everything is concealed. Aside from a relative handful of cases, the public has no access to the initial prospectuses. Even if a CDO prospectus were retrievable through the Irish Stock Exchange, that CDO's investment portfolio is still likely to be kept secret. Unlike subprime mortgage bonds, these CDOs had no legitimate business purpose. They neither financed the mortgages, which had been financed through the bonds, nor did they add to liquidity in the marketplace, since the CDOs were non-tradable black box investments.

You'll never figure out a CDO by reading a rating agency analysis, which offers a few cryptic comments of substance buried amid the boilerplate.

In addition, the CDOs are set up so that the asset manager can do all sorts of bait-and-switch maneuvers, within broad credit-rating based parameters, after the deal closes. CDO performance cannot be tracked, because the performance data is only accessible to CDO investors.

Hundreds of billions of fatally flawed subprime CDOs were created, but, with a handful of exceptions, we still do not know who bought what under what circumstances.

That's why the investigation into AIG's CDO exposure is such an important opportunity. For the first time in February, we had a schedule where we could match up the CDOs with the relevant exposure amounts with the insured counterparties. It sure looks like Societe Generale "bought" CDOs for the sole purpose of acting as a front for Goldman, which created most of the CDOs that AIG insured on SG's behalf. When The New York Times pointed out the suspicious circumstances of SG's CDO positions, Goldman spokesman Lucas van Praag responded with a non sequitur denial:

NYT assertion: "In addition, according to two people with knowledge of the positions a portion of the $11 billion in taxpayer money that went to Societe Generale, a French bank that traded with A.I.G, was subsequently transferred to Goldman under a deal the two banks had struck."

The facts: The assertion is false and misleading. Goldman Sachs provided financing to many counterparties, but in that role we would not have known whether a counterparty had obtained credit default protection, let alone from whom or in what amount.

Neither van Praag, nor the Goldman lawyers who reviewed his statement, are confused. They simply want to confuse us. The Times didn't allege that Goldman provided financing to SG; it alleged the opposite--that SG provided financing to Goldman. By acting as the middleman in two back-to-back transactions, SG bought the credit risk from Goldman and simultaneously sold the same risk, in the form of a credit default swap, to AIG. In other words, SG acted like the character in the Edward Jones commercial who, after submitting the highest bid at an art auction, says, "I want to go ahead and sell it now."

The best way to start to get to the bottom of all this is to pass a law that requires all performance reports of all private label mortgage securitizations, past and present, be made public. Second, as I wrote previously, there should be a national registry for every ownership claim, including every derivative claim, on a mortgage securitization. We won't get anywhere until these transactions are fully exposed to sunlight.

So far, Neil Barofsky, the TARP Inspector General appointed by Paulson, has shown no curiosity in finding out what they knew and when they knew it. His report on the backdoor bailout of the CDO banks ignores the subject completely. The selective pursuit of evidence is a big topic for another piece.