The SEC's Lawsuit Against Goldman - Revised

As I mentioned on Friday, my initial impression was that the 13% haircut Goldman's stock took in reaction to the SEC's lawsuit was an excellent opportunity to buy heading into their earnings report on Tuesday.  After having some more time to reflect and read up further, my conviction has been strengthened.

Forget about reading confusing, conflicting news reports about the SEC's allegations against Goldman.  The SEC has released a surprisingly readable (not full of legal jargon) report detailing their case.  If you don't want to read the 22 pages, I'll give you a summary.

Goldman was approached by John Paulson, who was looking to short the housing market.  Goldman figured that the best way to do this was to create a synthetic CDO which replicates the return on a reference portfolio of residential mortgage-backed securities (RMBSs).  To quote the SEC's report, the difference between a CDO and a synthetic CDO is that

"CDOs are debt securities collateralized by debt obligations including RMBS. These securities are packaged and generally held by a special purpose vehicle (“SPV”) that issues notes entitling their holders to payments derived from the underlying assets. In a synthetic CDO, the SPV does not actually own a portfolio of fixed income assets, but rather enters into CDSs that reference the performance of a portfolio" (point 13 of SEC report)

The implication of this difference is that in a synthetic CDO, there must, by definition, be a long and a short party.  Remember this, it is an important point.  Back to the story.  Goldman got ACA Management LLC, a firm quite experienced with this type of transaction (having performed upwards of 25 similar transactions prior to the one under discussion) to sign on as the manager of the proposed synthetic CDO (from this point forward referred to as Abacus).  Intending to short Abacus, Paulson had an incentive to have what he considered poor-quality RMBSs to be included in the reference portfolio.  ACA and Paulson met on a number of occasions and after some back and forth, a portfolio was finally agreed upon which included a large number of Paulson's suggestions.  Another very important point is that ACA had the final say of which mortgages went into the portfolio.  At one point ACA wanted more information on what exactly Paulson's role in the transaction was.  This is where things get particularly contentious.  Point 47 of the SEC report details the charge that Goldman misled ACA on this issue.  Since it is so important, I will include the entire point.

"On January 10, 2007, Tourre emailed ACA a “Transaction Summary” that included a description of Paulson as the “Transaction Sponsor” and referenced a “Contemplated Capital Structure” with a “[0]% - [9]%: pre-committed first loss” as part of the Paulson deal structure. The description of this [0]% - [9]% tranche at the bottom of the capital structure was consistent with the description of an equity tranche and ACA reasonably believed it to be a reference to the equity tranche. In fact, GS&Co never intended to market to anyone a “[0]% - [9]%” first loss equity tranche in this transaction." (point 47 of the SEC report)

What exactly does the SEC mean by "referenc[ing] a 'Contemplated Capital Structre' with a '[0]% - [9]%: pre-committed first loss' as part of the Paulson deal structure"?  While I am not privy to the jargon of the CDO industry, this looks more like obfuscation than misrepresentation by Goldman.  It is clear when reading the report in its entirety that the SEC had excellent access to Goldman's internal communications, as well as their communications with parties in the deal.  I feel that if Goldman had unequivocally indicated that Paulson was going long Abacus, the SEC would have been able to present a quote more condemning than this.  To me, point 47, the crux of the case, says effectively nothing.  Especially the final sentence.  May I have the rest of the quote to substantiate the SEC's claim that Goldman never intended to sell an equity tranche?  The next few points of the SEC's report detail evidence of ACA being of the impression that Paulson was long the equity tranche and Goldman failing to correct this misconception.

Operating under the incorrect assumption that Paulson was an equity investor, ACA went ahead with the deal.  It was subsequently marketed to other parties with ACA identified as the portfolio selection agent.  Ultimately the majority of the credit exposure was sold to a few European banks and they ended up taking a combined $1B in write downs, with the proceeds collected by Paulson's fund.

The SEC takes issue with Goldman's conduct on two fronts:

1. Goldman misled investors by failing to mention Paulson's role in the selection of securities in Abacus' reference portfolio
2. Goldman misled ACA over Paulson's interest Abacus

I should make it clear here that I am no expert in securities law, but it appears to me that since ACA had the final say on what was included in the reference portfolio, Goldman is in all likelihood following the letter of the law in making the now contentious statement that ACA selected the portfolio.  Whether Goldman followed the spirit of the law will not be the debate in the courthouse.  It should also be noted that when this deal was closed, people who were bearish on the US housing market were a small minority, and Paulson was a little-known M&A arbitrage hedge fund manager, rather than the prophet he is regarded as today.  I do not consider it unrealistic that had Goldman disclosed the fact that Paulson was the short and had been involved in the portfolio selection process, it would have actually been a selling point for Abacus.  What does a M&A arb guy know about the complex MBS-backed synthetic CDO market?  You want to talk about style drift.  But I digress.  The fact that Goldman lost money on the deal also undermines the SEC's charges.  This implies that Goldman themselves did not consider Paulson's role in the portfolio selection process a threat, and in fact put so little stock in his convictions that they actually bet against him.  This final point will go a long way in court (if the case ever gets there) as it indicates that Goldman did not intentionally mislead investors.

As for the charge that Goldman intentionally misled ACA re: Paulson's interests in Abacus, does anyone at the SEC really believe the evidence in support of this charge (as laid out in point 47) is going to stand up to Goldman's legal defense team?  I think that more substantial evidence of Goldman intentionally misleading ACA will prove necessary to return a guilty verdict on this charge.  However, I do think that the case as laid out certainly casts Goldman as grossly negligent in the very least.  Proving beyond a reasonable doubt that Goldman behaved in a fraudulent manner may prove quite difficult, but the damage to their reputation will be done regardless of the verdict.

I think that the timing of this announcement is another interesting consideration.  Over the last few weeks, Goldman has engaged in an aggressive PR campaign to exonerate themselves of responsibility for the crisis.  Highlights of this campaign include a cover story in Business Week complete with interviews with upper management as well as a letter to shareholders defending their conduct in the crisis which appeared in their 2009 annual report.  Perhaps the SEC was looking to undermine this PR blitz?  Maybe, but I think it a more likely explanation is linked to the push for financial reform. Some politicians are becoming complacent and reluctant to back proposed reform to reign in the big banks.  Now the supporters can point to the SEC lawsuit and say "See, these guys are defrauding investors, we need to come down hard on them to teach them a lesson", thereby acquiring support for reform.  By the time this case is settled in any fashion (court decision, settlement or dropped) it will have served its purpose in this regard.  Bruce Krasting at Seeking Alpha does a better job fleshing out this argument than I do.

It is also quite possible that I am dead wrong in my analysis.  Let us examine the consequences in this situation.  Goldman gets taken to task by the SEC and ends up paying a huge fine.  How big could it be?  Dick Bove, an influential analyst who covers the financial sector, estimates as much as $2 billion.  Other industry analysts offer considerably lower estimates.  Compare this to the drop in Goldman's market cap Friday ($12 billion).  That implies there needs to be six cases of this magnitude filed and won against Goldman to justify the drop. There is an argument that Goldman will lose a lot of business on the back of this announcement.  Maybe, but who is going to stop looking to Goldman when they need to buy or sell fixed income on the basis of this story?  That's where Goldman is earning all their money these days anyways.  At the end of the day, Goldman losing a some business isn't going to change their earnings figure (the medium term impact is more uncertain) and on this basis I stand by Friday's call to buy.