Here's what happened, according to the SEC (you can read a PDF of the complaint here - it's 22 pages and most of it is very readable). John Paulson (no relation to Henry Paulson, the former Treasury secretary) runs a hedge fund, and it wanted to go short certain kinds of RMBS.* Paulson asked Goldman Sachs to set this up for him. Goldman Sachs did so by creating a synthetic CDO.** Goldman hired a company called ACA to select the "reference" RMBS to go into the CDO. However, what really happened was that John Paulson, whose economic interest was for the RMBS to fail, selected a list of crappy RMBS and gave it to ACA. ACA added some and subtracted some, but Paulson continued to play a major role throughout the selection process. ACA thought that Paulson was taking a long position in the CDO and that his incentive was for the reference RMBS to perform well, and Goldman knew that ACA thought this. Goldman did nothing to disabuse ACA of this notion.

Goldman took the resulting synthetic CDO (called ABACUS 2007-AC1) and marketed its securities to IKB (a German bank) and other customers (not to AIG). In the disclosure materials, Goldman did not reveal that the RMBS that went into the CDO was actually hand-picked by Paulson, who took the short side of the synthetic CDO.

The SEC alleges that Goldman's failure to disclose Paulson's role in the formulation of the CDO amounts to fraud under Rule 10b-5 and other securities laws and regulations. The crux of the case is that Paulson, whose incentive was to pack the CDO with bad assets, played a major role in selecting its assets. Goldman knew about it and even facilitated it, but then when it marketed the resulting securities it did not disclose this fact. It does claim to have disclosed all of the other salient details about the underlying RMBS, and the SEC has not alleged otherwise.

Here's the SEC's position in a nutshell. Imagine this conversation:

GS: Why don't you buy some synthetic CDO securities, they are awesome.
IKB: Okay. Who picked the RMBS that went into the CDO?
GS: ACA and John Paulson.
IKB: Who is John Paulson?
GS: He's the guy who is taking the short position on the CDO.
IKB: So the crappier the RMBS in the CDO, the richer he gets?
GS: Yes.
IKB: And he picked the RMBS that went into the CDO?
GS: Him and ACA.
IKB: What do you think I am, crazy? You're asking me to invest in assets that were hand-picked by the guy who wants them to fail.

But this conversation never happened, because Goldman never told IKB that Paulson had any role in picking the RMBS. It turns out that the RMBS sucked, Paulson made about $1 billion, and the investors lost about $1 billion. Goldman says it lost about $90 million because it held on to an investment in the CDO. It made $15 million in fees.

So this is a very narrow complaint. It's about the disclosures that Goldman didn't make to particular investors about a particular CDO that it created in 2007. A lot of people are treating this as a sweeping allegation about Goldman's various conflicts of interest, but it's really a question of what investors deserved to know about securities that they bought.

*RMBS refers to residential mortgage-backed securities. For our purposes, picture a company whose only purpose is to hold mortgage loans and use the resulting cash stream to pay its investors. This company issues debt securities in "tranches," each of which gets paid off in order. So the top tranche gets paid first, then the next tranche, and so on. The top tranches are relatively safe, the bottom tranches are relatively risky. The CDO in this case was made from low tranches.

**A CDO is a collateralized debt obligation. Again, for our purposes, picture a company whose only purpose is to own RMBS and to pay the resulting cash stream to its investors. Like MBS, its securities are divided into tranches. The CDO in this case was synthetic, meaning that it didn't actually own RMBS. Rather, it owned credit default swaps on particular RMBS. For our purposes, imagine investors making side-bets on the performance of RMBS. One investor bets that they will perform well (taking the "long" position) and another bets that they will do poorly (taking the "short" position). The long position is economically identical to owning the RMBS itself, and the short position is its mirror image. Paulson took the short position in the CDS that went into the CDO in this case, and IKB and other Goldman customers took the long position. The nature of a synthetic CDO is that there will be someone in the short position, just as you need someone to take the other side of any bet you care to make. However, it would be highly unusual for the investor with the short position to get to choose which RMBS are being bet on. The fact that Paulson did so in this case, and that Goldman didn't inform its customers of this fact, is fraud, according to the SEC.

Oh, and incidentally, this is a civil case. No one is going to jail even if the SEC wins.

@3: Goldman says that it earned a $15 million fee on the transaction and lost $90 million when the synthetic CDO's assets turned out to be crap (Goldman held on to some of the equity in the CDO). So this was not profitable for Goldman regardless of what the SEC does.

@2 that's the biggest failure of all. They need to individually criminalize this behavior, and explicitly craft the law so that if they cannot identify the guilty parties it carries up the line:

Accountant to director to COO to CEO to chairman to shareholder with the most stock by percentage, and if that shareholder is a holding firm or another company, you repeat the chain of responsibility there at the executive level, until you finally hit someone. Any crime by a 'corporation' must have someone ultimately legally responsible.

Speaking of ripoffs...

http://finance.yahoo.com/q/it?s=AAPL

5-Apr-10 OPPENHEIMER PETER
Officer 850 Direct Disposition (Non Open Market) at $0 per share. N/A

1-Apr-10 SERLET BERTRAND
Officer 10,000 Direct Automatic Sale at $237.67 per share. $2,376,700

25-Mar-10 COOK TIMOTHY D
Officer 159,960 Direct Automatic Sale at $226.87 - $230.69 per share. $36,596,0002

25-Mar-10 JOHNSON RONALD B
Officer 107,075 Direct Automatic Sale at $230.09 per share. $24,636,886

25-Mar-10 OPPENHEIMER PETER
Officer 107,075 Direct Automatic Sale at $230.32 per share. $24,661,514

25-Mar-10 SCHILLER PHILIP W
Officer 80,415 Direct Automatic Sale at $230.15 per share. $18,507,512

24-Mar-10 FORSTALL SCOTT J
Officer 75,000 Direct Option Exercise N/A

24-Mar-10 FORSTALL SCOTT J
Officer 34,575 Direct Disposition (Non Open Market) at $229.37 per share. $7,930,467

24-Mar-10 MANSFIELD ROBERT J
Officer 75,000 Direct Option Exercise N/A

24-Mar-10 MANSFIELD ROBERT J
Officer 34,575 Direct Disposition (Non Open Market) at $229.37 per share. $7,930,467

24-Mar-10 COOK TIMOTHY D
Officer 300,000 Direct Option Exercise N/A

24-Mar-10 COOK TIMOTHY D
Officer 140,040 Direct Disposition (Non Open Market) at $229.37 per share. $32,120,974

24-Mar-10 JOHNSON RONALD B
Officer 200,000 Direct Option Exercise N/A

24-Mar-10 JOHNSON RONALD B
Officer 92,925 Direct Disposition (Non Open Market) at $229.37 per share. $21,314,207

You present zero evidence that some Obama-driven transformation at the SEC is behind this suit. The available facts would seem to argue against such a narrative. (1) Low and mid-level career staff are the same. (2) High-level staff hasn't changed much, and the replacements are just as close to Goldman and Wall St. as their predecessors. (3) The national mood is a huge confounding factor in this experiment; the SEC would be feeling massively increased pressure to produce high-visibility enforecemnt actions whether under an R or D administration.

@7: Oh, I don't know. There is this. And, well, this. And then there's also this. And then there's also this.

The new SEC chief - who has been significantly more aggressive in pursuing enforcement than the old one - was appointed almost immediately by Barack Obama after he took office.

In other words, (1) might be true, (3) might even be true, but (2) is at least partially false. In addition, when the Administration changes, the boss changes no matter what. And if the new boss starts doing things in a dramatically different way, it's probably not totally unreasonable to assume they had something to do with it. Also, the timing alongside financial reform couldn't be more obvious, unless we're supposed to believe that's a coincidence.

The Republicans practiced enforcement too, to be sure, but Democrats fundamentally believe in regulation of the markets more than Republicans do. This isn't to say market regulation is a "good thing" (most Republicans believe its not) but it's sort of absurd to argue Democrats aren't doing it. One could argue they're not doing it enough, and some do; you don't typically vote Republican to see more enforcement of financial regulation, so...

Unless its broad and criminal, this is just the cost of doing business for Goldman. Thinking this is some vindication of justice is just naive.

Goldman has claimed it is innocent by the facts and by the law.

The only way I can interpret this: We didn't do what you said we did and even if we did do what you said we did, it isn't illegal.

Paulson comes out smelling like roses, even though he created the CDO specifically so he could take money from the suckers.

Paulson: Have you noticed no one seems to question the yields on mortgage-backed securities?

Goldman: Why would they? We only pay credit-rating agencies who promise to rate them good-as-gold.

Paulson: Let's figure out how to take advantage of them.

Goldman: Heck, we're already defrauding them with the credit ratings. Why don't we just forget to tell them the CDO was designed to fail.

Doesurmindglow @ 8: You present a good rejoinder to (2). I'll certainly grant you that Schapiro and Khuzami don't have Wall St. backgrounds. Of course, the real question on (2) is whether on average across all appointments Obama's appointees are more less connected to Wall St. that Bush's.

But the general sentiment that "R's are for business, D's are against it" I won't grant as evidence. In fact, that narrative is exactly what we're supposed to be testing here. The existence of that narrative shouldn't be submitted as evidence to support its own truth.

Institutionalized rape is not funny under any circumstance, Jonathan (re: your clip).

You're better than that.

@12

Will somebody please think of the children! Either you have thin skin and no sense of humor, you just like to see yourself type, or c) all of the above.

No, I don't have a sense of humor about rape, #13.

Damn

@11: I guess this depends on what question you are actually asking. One is a question of policy, ie. whether it leads to a better outcome to regulate large banking operations or to deregulate them. Another is a question of efficacy, ie. if the Democrats could be doing a more effective job of regulating large banking operations, if we choose that as the proper course.

I would agree that a McCain Administration would be likely, as a result of overwhelming public pressure, to take a liberal policy stance - in this case, the position in support of comprehensive government regulation of large banking operations. This is because, even though our country does not self-identify as "liberal," they do tend to self-identify as supportive of modern liberal policies (funding public infrastructure, reducing the debt, healthcare as a right, regulation of the market to level the playing field, etc.).

To this extent, we could expect Republicans to, in regard to this issue, respond in a similar manner as Democrats. But my argument was that, if that were the case, it would only be because they would be pressured to support a Democratic approach to this problem.

On top of both of those things, we have a serious problem here: we're trying to prove or disprove a total hypothetical. McCain didn't win the election, and we can't run an alternate reality where he did. What we do know - which was the basis of the argument - was that a candidate whose liberal ideology emphatically supports regulation defeated a candidate whose conservative ideology does not. Subsequently, we've seen policy actions taken toward regulation. Thus, we can say it's likely - though not certain - that these policy actions extend from the platform of the person who is enacting them.

@11: Oh shit, sorry David, I forgot to mention essentially that I've been using "post hoc, ergo propter hoc" not because it actually makes deductive sense, but because we can't really draw any deductively valid conclusions from the information we have (that is, the lack of credible information about what would have happened under a McCain Administration were one to exist). I'm suggesting that we can, however, draw inductive conclusions that are "likely" to be true.

Goldman may have sacrificed a few million on Abacus to provide plausible deniability, but they made billions of dollars busting-out AIG through short-sales and swaps.