The Goldman vs. SEC Story That No One Has Written ...

When I saw that the SEC had finally decided to go after Goldman Sachs, I immediately rejoiced: Yes. At last. What the hell took you so long?

Then, when I started sifting through the strange case of Abacus 2007-AC1 (fairly trips right off the tongue eh?), I had a "pullback" moment, especially after seeing Goldman's defense (see the bottom of the page).

First, I have no love for Goldman. Far from it. I think it's rather creepy the way they release their ideological spores throughout our political system by practicing "civic responsibility" and occasionally shipping a handful of executives off to the Treasury Department, to keep the U.S. approach to the financial system appropriately capitalist at all times. But this Abacus case -- ah well, it doesn't make sense, unfortunately. I think the SEC will lose unless Goldman wants to pay up to make the bad publicity go away.

Here's why.

Read the SEC complaint. Read Goldman's denial. Observe the Venn diagram point where the two fact sets overlap in a significant way.

From Goldman: ACA had the largest exposure to the transaction, investing $951 million.

From the SEC: On or about May 31, 2007, ACA Capital sold protection or “wrapped” the $909 million super senior tranche of ABACUS 2007-AC1, meaning that it assumed the credit risk associated with that portion of the capital structure via a CDS in exchange for premium payments of approximately 50 basis points per year.

Think about that for a second. Whether it's $909 million, $951 million or $927.33311 million -- ACA, both sides agree, had a huge exposure to this deal. This was a $2 billion synthetic CDO. The German bank IBK, the other banner investor, only had $150 million of exposure (though to riskier tranches, true).

So ponder this a bit: why would ACA, whose duty was "portfolio selection agent," allow itself to be duped into stuffing the CDO sausage with the RMBS equivalent of rat tails and nose parts, if it was so hugely on the hook for the losses? Because consider this (all part of the SEC's own fact set in its complaint):

1. Paulson was a known short on subprime mortgages at this point. In 2007, the synthetic CDO was set up. Here's what happened a year earlier, according to the SEC:

Beginning in 2006, Paulson created two funds, known as the Paulson Credit Opportunity Funds, which took a bearish view on subprime mortgage loans by buying protection through CDS on various debt securities.

2. ACA wasn't some newbie from Canoobie when it came to setting up CDOs. The SEC tells us:

ACA previously had constructed and managed numerous CDOs for a fee. As of December 31, 2006, ACA had closed on 22 CDO transactions with underlying portfolios consisting of $15.7 billion of assets.

And, what's more, ACA knew that Paulson was heavily involved in helping pick the securities for Abacus. Again, the SEC:

On February 5, 2007, an internal ACA email asked, “Attached is the revised portfolio that Paulson would like us to commit to – all names are at the Baa2 level. The final portfolio will have between 80 and these 92 names. Are ‘we’ ok to say yes on this portfolio?”

So get a load of this: You're ACA. You're among the best at structuring CDOs. You should be able to evaluate the mortgage bonds being assembled for the security pretty well. You should know how the risks work. You should also know what's going on in the larger market: who's bullish on these things, who's bearish (Paulson, Paulson, Paulson).

And you swallow risk on about half of a synthetic CDO that you let Paulson fill with, um, crap?

I think there's more to this story than meets the eye. My guess is that ACA is much more guilty (of stupidity or something else) than anyone is suggesting. I think (1) they got caught being idiots, essentially making a longish bet on residential mortgages by insuring the super-senior tranche of the CDO (this is the last one to take losses, when the defaults start to mount) (2) they may have been making a cynical play, on the bottom part of the synthetic CDO, letting Paulson pick some crap, thinking that the super-senior tranche would be amply protected if the housing market deflated a little OR they were simply grossly negligent and unbelievably stupid by not reviewing the bonds that a known subprime-mortgage short was stuffing into a CDO they were insuring almost half of.

So I'm doubtful the SEC will win this one, unless there's something big I'm missing. But I think the SEC's case will be the perfect stalking horse for achieving the financial system reform that we do need pretty badly -- so maybe it's not so bad to put Goldman on the rack for a year or two.