So there is still a lot to discuss about this Goldman Sachs situation. I wrote my initial summary and reaction after reading the SEC's complaint. I found the SEC's logic to be pretty reasonable. Then, I read Goldman's response, which I also initially found to be pretty sensible. Now, after more thought, I have found where I think GS erred in its response. I encourage anyone wishing to have an intelligent discussion about this topic to first read the SEC's roughly 20 page complaint - it's in plain English, not confusing legalese, and will help with your understanding of the issues.
Without doing a line-by-line on Goldman's response, let's take a look at a few key issues:
1) Goldman claims that they lost money on the transaction. This may be true, but isn't relevant to the SEC's complaint. The main complaint is that GS failed to disclose material information to the buyers of the synthetic CDO - not that GS took the other side of the trade and profited. What is odd is why GS was on the wrong side of the trade - that still doesn't make a whole lot of sense to me, considering it seems clear that even if the buyers didn't understand the risks in the underlying reference assets, GS did understand the risks.
2) GS wrote: "ACA, the Largest Investor, Selected The Portfolio." This may be the truth, but it's not the whole truth. We've all seen courtroom scenes on TV where a witness swears to tell "The truth, the whole truth, and nothing but the truth." At the very least, if Goldman is telling the truth here (which I think, by the letter of the law, they probably are - after all, ACA did select The Portfolio) they aren't telling the WHOLE truth. ACA did select The Portfolio, but it was from a list that Paulson presented. I'm not even sure this should matter, since ACA had final say in the portfolio selection. As commenter UncleFester wrote on Henry Blodget's piece: "Furthermore, there is nothing to suggest ACA did not have the ultimate authority on the composition of the portfolio. If ACA didn't like the portfolio, it was free to pull out of the deal prior to its closing."
3) The always brilliant Steve Randy Waldman wrote a whole post today about GS's claim that "These investors also understood that a synthetic CDO transaction necessarily included both a long and short side." Waldman's point is that the investors did NOT understand this - but Waldman doesn't make excuses for the investors, only tries to point out that GS's claim that the investors knew this was false. In my opinion they certainly SHOULD have known this, but they were living in a different bizarro fantasy ignorant world view of the mortgage markets. I think Waldman errs a little bit in that it doesn't matter than investors didn't know - they SHOULD have known. Synthetic CDOs are indeed a type of product where the profits one party earns come directly from the counterparty on the other side of the trade. Nevertheless, perhaps Waldman is correct that investors didn't effectively understand they were making a prop bet verse another party, and thus GS's claim may with respect to this may be false.
4)The most important red herring is GS's claim: "As normal business practice, market makers do not disclose the identities of a buyer to a seller and vice versa." Now, that's completely true, but that is NOT what the complaint here is about. The complaint is that Goldman didn't disclose that the seller was instrumental in creating the portfolio itself - not that GS didn't disclose who the seller was.
Interestingly, the pitchbook for the transaction includes many disclaimers, including one that
"Goldman Sachs may, by virtue of its status as an underwriter, advisor or otherwise, possess or have access to non-publicly available information relating to the Reference Obligations, the Reference Entities and/or other obligations of the Reference Entities and has not undertaken, and does not intend, to disclose, such status or non-public information in connection with the Transaction. Accordingly, this presentation may not contain all information that would be material to the evaluation of the merits and risks of purchasing the Notes."
Now, I am not a lawyer - I don't know if this is the kind of thing you can "disclaim" away (I'd guess not), although it certainly seems like Goldman tried to.
So, now that we've talked about what's potentially wrong with GS's defense, I want to focus on my important point: ACA is not a victim here. Even if GS was guilty of non-disclosure, I find ACA's claim that they would have behaved differently to be a lame excuse. As I wrote in my initial response to the suit: "ACA was providing insurance on the portfolio. It shouldn't matter who they were providing insurance TO, it should matter what they were providing insurance ON."
I think BusinessInsider's Henry Blodget also described the situation very well when he wrote:
"Presumably, like all investors who have made mistakes, ACA would prefer to believe that it was misled than to accept that its analysts blew it. ACA, therefore, has a motive to blame Tourre for misleading it.
In reality, however, to make this case, ACA is going to have to make the embarrassing admission that knowing what Paulson & Co was going to do affected its judgment with respect to the transaction. This information should NOT have affected ACA's security selection process. It should also not have affected ACA's decision to go forward with the deal. ACA is an independent firm staffed with experienced professionals paid millions of dollars to evaluate securities by themselves. What Paulson was or wasn't planning to do, therefore, should have been irrelevant."
This is an essential point. Note that no one is arguing the merits of GS's disclosure (or lack thereof) here, but I am absolutely arguing that the disclosure shouldn't have mattered IF ACA HAD DONE THEIR JOB. The underlying securities in the synthetic CDO are what they are, regardless of who put them on the list, or who takes the other side of the trade. They need to be evaluated based on risk metrics, cash flows, etc. The real issue is that ACA didn't do this work to the level that they needed.
GS may be guilty of insufficient disclosure - let's just pretend they are. My point is that even given this failure to disclose, the buyers of the securities in question were grossly negligent in failing to properly assess the values and prospects of the synthetic CDOs, and they are trying to remedy their bad trade by diverting blame. So hold the Goldman's of the world accountable, but don't let the ACA's of the world off the hook.
I hope to write another post soon about the value of counterparty information in trading markets, how my group used to use it in our trading strategies, and why it fails as a metric in this case. Counterparties definitely matter to traders who might be taking positions on a short term basis, but they should not matter significantly for someone looking for longer term exposure to an asset class which needs to be evaluated based on quantitative risk metrics, like the RMBS synthetic CDOs in question in this case.