Fabrice Tourre testified in his SEC trial late last week and many perplexing things came out, with the most perplexing being a tie between:
- Fab's claims that the key disputed email, in which he maybe-defrauded maybe-victim ACA, was "not accurate" but not "false," an epistemologico-semiotic dispute that probably sounds better in French, and
- the fact that Fab got a base salary of $738,000 for his 19 months of paid leave, which is just sort of an odd base salary for a Goldman VP.1
Perhaps less perplexing is that Fab's feelings about Abacus seem to have been less about bamboozling one client on behalf of another and more about just printing a trade, whichever direction it went in. John Carney reports that Goldman was taking too long getting ABN Amro to intermediate ACA's guarantee of the super senior tranche of the deal, Paulson was getting antsy, and Fab, ever servicey, was trying to assuage their antsiness by just getting Goldman to do the deal naked:
Tourre suggested to his desk that Goldman just sell protection directly to Paulson without waiting for the other part of the transaction to be put in place. As an alternative, he suggested that Goldman waive its collateral rule and deal directly with ACA. Either move would have required Goldman to take on a lot more risk.
"Would like to take down their super senior risk tomorrow in order to avoid losing the order," Tourre said in an email to the heads of his unit at Goldman. ... The email was put into evidence by Tourre's lawyers. It suggests that Tourre didn't really believe that Abacus was "designed to fail," as he was advocating that Goldman go long.
Ehhh I don't know about that last part, Fab's lawyers; they were still working on finding protection sellers so going naked short $1bn of protection to Paulson for a short period would just have motivated them to get the other side of the deal done. (If they lied to investors when they were just buying protection for Paulson, imagine how much they would lie if they were buying protection for themselves, etc. etc.)
I'm intrigued, though, by Fab's willingness to take uncollateralized credit exposure to ACA. For one thing it allows for some psychologizing of Fab, though I don't know which way that cuts. From the vantage point of 2013, ACA looks to have been a terrible credit risk - more on that in a bit - and so Fab's willingness to take up to $909mm of uncollateralized credit exposure to them seems like the sort of recklessness that would lead someone to also cut corners on disclosure. On the other hand, in 2007, ACA didn't look that terrible, particularly if you didn't think that the mortgage market was about to blow up (and take ACA with it), or if you didn't think that your actual exposure to ACA would ever approach the $909mm notional. So maybe it proves that Fab was, in his heart, building Abacus to succeed, sure.
My view has alwaysbeen that ACA look both villainous and foolish in this deal,2 since any deception that Goldman visited on other investor was intentionally aided by ACA's claims to have picked the portfolio that blew up. But if you actually look at the list of victims the biggest money loser on Abacus is not really ACA, which ended up paying $30 million, or Goldman, which actually lost $80-90mm on a slice of protection it sold to Paulson but never hedged, or IKB, investor who was deceived by the offering memo, bought Abacus notes, and lost "almost all of its $150mm investment."
Rather it was ABN Amro, which intermediated the $909mm super senior tranche guarantee between ACA and Goldman in exchange for a 17bps (~$1.5mm) annual fee. That is, ABN wrote protection on the Abacus super senior to Goldman, and ACA wrote it to ABN. And Goldman wrote it to Paulson. And when the super senior tranche went to zeroish, and when ACA turned out not to have any money, ABN was on the hook. Here:
You don't hear a lot from ABN in these Abacus cases and I wonder how they feel. Bad, I mean, but like: their risk was that both (1) these mortgages would crap out and (2) so would ACA. That is what you'd call a wrong-way risk, of course, and when the mortgages crapped out so did ACA. Because ACA's whole business was writing guarantees on structured credit securities that would eventually crap out. From the SEC's original complaint:
At the end of2007, ACA Capital was experiencing severe financial difficulties. In early 2008, ACA Capital entered into a global settlement agreement with its counterparties to effectively unwind approximately $69 billion worth of CDSs, approximately $26 billion of which were related to 2005-06 vintage subprime RMBS. ACA Capital is currently operating as a run-off financial guaranty insurance company.
So the $909mm policy it wrote on Abacus represents around 1-4%, depending how you count, of the trouble that ACA got in. The rest was mostly on Abacus-like securities, but without the fraud charges. Designed to fail, designed to succeed, designed to whatever: if ACA wrote CDS on it, it failed. ABN's $840mm loss might have something to do with Paulson building a CDO to fail, sure, and something to do with Goldman (and ACA!) falsely telling ABN that ACA had actually built the CDO to succeed - but it had a lot to do with the fact that ACA ended up paying ABN about 3 cents on the dollar on its own obligations.
And those are the guys Fab was willing to trust! I dunno, it's just a neat test of his good faith, as well as a useful reminder that, if you owe the bank a billion dollars, you own the bank. How solicitous do you think Fab was of ACA's best interests? His willingness to take a billion dollars of uncollateralized credit risk to them seems like a good fact. For this case, I mean. On the other hand, the fact that Goldman wouldn't go near that risk is a good fact for them. Not particularly for this case, I mean - they're out of the lawsuit now anyway - but because it would have been a terrible decision. If there was one thing in this transaction that was really, though unwittingly, designed to fail, it wasn't Abacus. It was ACA.
1.And which apparently sounds better in pounds, because for some reason Reuters says "Goldman Sachs Group Inc paid Tourre a base salary of 480,000 pounds ($738,000) after he was put on leave following the filing of the SEC lawsuit in 2010," even though Fab is a Frenchman living in the US.
2.A view that is not changed by this exchange:
On Friday, Pamela Chepiga, Tourre's lawyer, displayed a series of documents including deal term sheets sent to ACA after the fact listing the first loss tranche as "NA."
[ACA deal person Laura] Schwartz had testified she believed "NA" meant "not available," a reference to Paulson having already committed to it. But Tourre had a different definition, "not applicable."
So one of those is right and the other one is wrong! Did she really run around her whole career thinking "NA" meant "not available"? Context clues, c'mon.