The Financial Crisis Inquiry Commission grilled Goldman chief operating officer Jonathan Cohn and CFO David Viniar this week, with today’s session focusing on AIG, and in particular, .
Readers know that I have perilous little sympathy for Goldman. However, it is important that investigations focus on matters likely to hit pay dirt. And despite the sabre rattling at the New York Times and various websites on the matter of Goldman’s collateral marks, we think the ire is misguided, and this is one of the few cases where Goldman’s defense is sound. By contrast, the Commission missed other smoking guns.
To recap: Goldman, like other major dealers, had bought credit default swaps from AIG to hedge against some CDO exposures. The case against Goldman, in simple terms, is:
1. Goldman was overly aggressive in marking down the CDOs it had insured with AIG. Remember, the bigger the losses reported on the CDOs, the more cash AIG would have to pony up to Goldman
2. Goldman’s actions contributed to AIG’s demise.
Tom Adams, a former monoline executive, and I have performed considerable, in-depth examination of the AIG CDS on CDOs that were bought out by the Fed at par (the CDOs wound up in a vehicle called Maiden Lane III). We debunked this thesis, presented in a New York Times article, in February:
There is a wee problem with this account. Goldman’s marks were proven correct. With the benefit of hindsight, most players, particularly AIG, were in denial….
In late 2007 and early 2008, the monolines were facing similar issues to AIG….The rating agencies not long afterwards started downgrading AAA asset backed securities CDOs, verifying the “aggressive” position [monoline short Bill] Ackman and Goldman were taking.
The story also repeats the AIG/Fed flattering claim that these CDOs have “rebounded.” We’ve discussed long form in other posts that given the continued, serious deterioration in the underlying mortgages, this notion is simply not credible. The decay in credit quality across the portfolio is severe, and there has been no “rebound” in prices of severely distressed CDOs….
The jig was up for AIG by January of 2008 and the debate was only one of timing, not of what the actual outcome would be. Coincidentally, Ambac, FGIC and XLCA were downgraded in January 2008 directly as a result of high expected losses in their CDO portfolios. Any case against Goldman for aggressive marks against AIG by the SEC or other parties would have take the market environment into consideration. Across the board, CDOs were causing losses and downgrades for the people who insured them. It therefore makes plenty of sense that Goldman would be requesting more collateral for their exposure with AIG.
Yves here. So why were the other AIG counterparties more generous in their marks than Goldman? They held considerable CDO inventory. If they were the packager and had marked down their AIG positions, they’d have to provide similar prices to any customer who had bought a long position in the same CDO from them. And more important, they might be required by auditors or regulators to reduce the prices of similar CDOs, which would result in losses.
While this line of inquiry looks illadvised, others have been neglected. Why has no one questioned any of the banks of the absurdity of relying on guarantees from the monolines and AIG? Insurance on subprime was rife with what traders call “wrong way risk”: if you needed to collect on your insurance policy, the very events that would lead you to put in a claim would be likely to damage the guarantor. (Goldman would assert it did recognize the risk and had bought CDS on AIG, but that is also flawed: as we saw, an AIG default was a probable systemic event. Those contracts suffered from wrong way risk too). Put more bluntly, the idea that you could hedge subprime risk, particularly on the scale Goldman could likely have inferred was taking place, was almost certain to result in non-performance on the insurance. Did this occur to Goldman’s vaunted risk management operation? It might be revealing to follow that thread.
As well as diffusing the spat with the FCIC, Mr Cohn provided new numbers that he said proved the bank did not “bet against its clients” in the market for mortgage derivatives as the credit crisis unfolded, as has been alleged…..
He said Goldman had reviewed all the mortgage securities and derivatives it had created since December 2006, following fraud charges levelled by US regulators earlier this year. It underwrote $47bn (£31bn) of residential mortgage-backed securities and $14.5bn of collateralised debt obligations, and took short positions on the products – which would rise in value if the products fell – of less than 1 per cent of their value.
“During the two years of the financial crisis. Goldman Sachs lost $1.2bn in its residential mortgage-related business,” Mr Cohn told the panel. “We did not ‘bet against our clients’, and the numbers underscore this fact.”
Yves here. This smacks of being the sort of artwork that is technically accurate in its detail, but misleading in the picture it presents.
The role of a financial firm is to facilitate commerce, by helping companies raise money, by allowing investors and savers to deploy funds. But they have lost sight of their role, and many of their activites at best have no social utility and at worst are extractive and destructive. For instance, short sellers have a useful role to correct the pricing of instruments that were created for legitimate uses. But no one until recently would have considered creating positions anew to serve the interests of short sellers was a good idea. It is pouring talent and capital into purely speculative activities. Bear Stearns, far from a vestal virgin, refused to work with subprime short John Paulson to create synthetic CDOs that would enable him to bet against subprime bonds cheaply.
Now let’s look at Cohn’s remarks. They aren’t just a little misleading, they are a lot misleading.
1. Cohn isolates Goldman’s shorting in 2007 and 2008. But Goldman’s Abacus program, which was designed for the firm to establish short positions, started in 2004. Goldman had insured 5 2004 and 2005 Abacus trades with AIG, along with 22 2004 and 2005 CDOs structured by Merrill, 9 2004 and 2005 CDOs structured by other banks, and 2 of its own 2005 cash CDOs. So the roughly $15 billion that Goldman made from AIG is expressly excluded from Cohn’s presentation.
2. The comparison is further misleading by comparing its activity over a period of time (underwriting over a two year period) versus its short position that was presumably measured at a single point in time
3. What does “short positions on the products” mean, exactly? Technically, if you take down the short sided of a synthetic CDO, you have short positions in tranches of subprime bonds and other assets. If you only kept the short side on particular RMBS in that CDO, not all, you are arguably not short the CDO, but short some bonds. Similarly, Goldman may also have used the ABX or the TABX indices to establish short positions, so it could be taking a view against the market without being short the specific transactions it was pedalling.
4. Pray tell, how was this “less than 1%” arrived at? The dollar amount of the short position was CERTAIN to be small because Goldman used credit default swaps. The cost of establishing a short position was only 100-140 basis points until spreads started blowing out at the end of 2006 (and they tightened again in March 2007). The proper comparison would be the notional amount insured versus the cash position.
The FCIC also bears other signs of being badly unprepared, witness this exchange reported in the (hat tip reader Francois T):
The panel created to investigate the roots of the financial crisis escalated the government’s assault on Goldman Sachs on Thursday, criticizing the Wall Street firm for failing to turn over basic documents and accusing it nearly lying under oath.
For a second consecutive day, the bipartisan Financial Crisis Inquiry Commission reiterated its request for additional data from Goldman, namely figures regarding the firm’s derivatives activities. And for a second consecutive day, Goldman’s top executives demurred.
“We generally do not have a derivatives business,” David Viniar, Goldman’s chief financial officer, told the panel Thursday under oath.
Goldman Sachs holds more than $49 trillion in notional derivatives contracts, making it the third-largest derivatives dealer among U.S. banks, according to first quarter figures from national bank regulator the Office of the Comptroller of the Currency. The commission has found that Goldman is a party to more than 1 million different derivatives contracts, Commissioner Brooksley Born disclosed Thursday.
“We don’t separate out derivatives and cash businesses,” Viniar clarified under questioning. The derivatives units are “integrated” into the firm’s cash businesses, making it difficult for the firm to isolate its derivatives data, he said.
In January, the panel asked Goldman chairman and chief executive Lloyd C. Blankfein for a breakdown of the firm’s revenues and profits from its derivatives activities. He said the firm would comply. The commission reiterated that request Wednesday and Thursday.
Viniar said the firm doesn’t “keep” records outlining its revenues from its derivatives dealing.
“I am very skeptical that you can’t measure these revenues and profits,” Born told Viniar. “I urge you to provide us with this information. It’s been about six months we’ve been asking for it… and it makes one wonder also why Goldman has the incentive or impetus not to reveal this information.
“You’re suggesting you don’t give it to your regulators. You don’t put it in your financial reports… so you don’t give it to the market… [or to your counterparties],” Born continued. “And you’re refusing to give it to us. I hope very much that we will see this very shortly.”
Viniar took exception to that last comment.
“Commissioner, again, we’re not refusing anything,” Goldman’s chief financial officer said. “We don’t have a separate derivatives business.”
Viniar then said that Goldman isn’t alone in not breaking out its derivatives-specific revenues and profits.
Born quickly shot back.
“They don’t,” Born, the nation’s former top derivatives regulator, conceded. “But some other firms have provided us with that data when we’ve asked for it, and Goldman Sachs hasn’t.”
Phil Angelides, the panel’s chairman, could barely contain his incredulousness.
“Are you telling me you have no system at your company that tracks revenues or assets of contracts, and liabilities and payments under contracts?” Angelides asked. “You have no management reports, no financial reports that track these contracts?”
“I’ve never seen one,” Viniar responded. Pressed further, Viniar added that the firm doesn’t track these things because it’s “not meaningful.”
Viniar again was asked to provide the data.
Yves here. I have to tell you, this is a ridiculous line of questioning. What the hell is the FCIC trying to get at? There is NO SUCH THING as a “derivatives business”. This in fact illustrates how financial services lobbyists have managed to muddy policy debates, to the advantage of the industry, by lumping a lot of disparate activities under the derivatives banner.
Goldman no doubt has commodities futures businesses, FX and currency swaps, and corporate and asset backed credit default swaps activities. I’m sure it also engages in stock and bond index and futures trading in a number of markets. I’m a big believer in knowing what questions you are trying to answer when drilling into data, and I see no utility in having an aggregate figure across these activities.
And some firms do manage the cash and derivatives businesses of related businesses on an integrated basis. In particular, it appears from the voluminous Goldman documents released by the Senate that Goldman ran its cash and synthetic CDO packaging business from the same business unit. This would not be unusual.
Now the flip side is Goldman clearly does have transaction level information and could no doubt provide analyses to address specific FCIC questions . But it isn’t clear at all what the FCIC wants. This reminds me of the sort of exercise I’d fight tooth and nail as an associate at McKinsey, because it was a complete waste of client time and money, that of simply taking whatever data the client had and cutting it various ways to see if anything emerged. It would provide a lot of charts for a progress review, and if it produced any insight, it was completely random and could have been arrived at much more cheaply with a more deliberate approach.
So as much as Goldman is a deserving target, the FCIC appears to be quite overmastered by them, in part due to insufficient preparation (a function of insufficient budget, staffing, and unrealistic deadlines) and lack of well honed interviewing skills on behalf of its commissioners.