If, like me and David Kotz, you get some sad pleasure from getting annoyed at SEC incompetence, then you might enjoy the filings that the SEC and Citi made today in a federal court defending their settlement over charges that Citi did some bad stuff with CDOs.

The quick background: Citi decided to make a big prop bet against some mortgages, so it structured a synthetic CDO with the exposures it wanted to short and sold it to some dopes, keeping virtually all of the short side of the trade on its books. This was a good idea and Citi made $160mm, but it worked out less well for the dopes. The SEC sued Citi for not telling the dopes certain things, like that it had picked the mortgages involved because of their exceptional badness, and they signed up a $285 million settlement. Unfortunately for them, the federal judge hearing the case is Jed Rakoff, who is as high on the enemies list of the SEC’s employees as it is possible to be without standing between them and their tranny porn. Judge Rakoff had some questions about the settlement. Questions like “Why, for example, is the penalty in this case less than one-fifth of the $535 million penalty assessed in SEC v. Goldman Sachs & Co.,” or “How can a securities fraud of this nature and magnitude be the result simply of negligence?” Today Citi and the SEC filed answers those questions. They’re a fun read.

For Citi, the incentives here are pretty obvious. Given the hostility of the judge’s questions, you need to mount a full-on defense of your conduct: basically you want it to sound like Citi was totally innocent, and fully disclosed all its conflicts of interest to informed and sophisticated buyers, but was force to settle because the SEC’s overreach has put it at risk of a disastrous verdict. So that’s what they did, and as a sympathetic observer I thought they did a nice, maybe somewhat passive-aggressive job of it:

For instance, the Class V offering documents provide specific disclosures concerning [Citi's] roles in the transaction, although the SEC alleges that those disclosures were incomplete. Among other things, the offering materials informed investors that, in its role as the initial short counterparty, CGMI “may be expected to have interests that are adverse to the interests of the Noteholders.” The disclosures further stated that [Citi] “may provide CDS Assets as an intermediary with matching off-setting positions requested by the Manager or may provide CDS Assets alone without any off-setting positions.” …

This Court has cataloged the many risks faced by a public company that chooses to engage in protracted litigation with its regulators, including private litigation risk, reputational harm, and the risk of collateral regulatory consequences. See [a conference speech by Judge Rakoff].

And, of course, if you want to dispel any concerns that Citi was an evil genius twirling its mustaches while laughing at its sucker clients whose loss was its gain, there’s always:

Citigroup did not predict or profit from the subprime crisis, the collapse of housing prices, or the collapse of the CDO market. Precisely to the contrary: over a period of 18 months beginning in late 2007, Citigroup’s CDO-related losses totaled more than $30 billion—more than any other financial institution in the world. Notwithstanding the $160 million in profits CGMI allegedly earned in connection with the Class V transaction, Citigroup lost tens of billions of dollars in its CDO-related investments during this period because it retained significant long positions in the CDOs it structured.

The SEC, on the other hand, has to walk a finer line. On the one hand, it wants the judge to agree that what Citi did was a Bad Thing and needs to be punished severely, to the tune of $285mm (or $95mm, whatever*). On the other hand, it can’t make things sound so bad that the judge says “screw this settlement, we’re going to trial and taking them for everything they’re worth” – because the SEC doesn’t seem all that confident that it would win a trial. As a further complication, the SEC needs to look like it has some idea of what it’s doing, to preserve some credibility for the next case.

The result is occasionally a mess, as when answering the “why is this settlement different from a Goldman Sachs settlement?” question:

The proposed settlement with Citigroup – like the settlements with Goldman Sachs and J.P. Morgan – resulted from an extensive, industry-wide investigation into certain abuses that contributed to the recent financial crisis. Given these substantial investigative efforts, the SEC is well-positioned to make comparative judgments regarding the relative culpability of the entities and individuals involved. With regard to the penalty imposed in SEC v. Goldman Sachs & Co., Goldman Sachs was charged with scienter-based violations of the securities laws. As a general rule, scienter-based violations are worthy of a more significant sanction. In addition, other factors, including whether a defendant clearly has articulated a willingness to settle prior to the filing of a complaint, is a factor that may be considered.

Ooh words. But these are very silly words. “Scienter-based violations of the securities laws” just means that GS, unlike Citi, was charged with intentionally stuffing dopes with bad CDOs. True! The SEC charged GS with doing that intentionally, and it only charged Citi with doing the same thing “negligently,” i.e., something a little bit north of “accidentally.” But, like, the SEC just decided to charge it that way – and they don’t explain why. It is sort of unimaginable that anyone could accidentally create a mortgage-backed security filled with loans you know are going to fail so that you can sell it to a client who isn’t aware that you sabotaged it by intentionally picking the misleadingly rated loans most likely to be defaulted upon. Like, you have to pay attention in order to do that. You have to pick the loans, and write stuff down, and tell people about the thing, and convince them to buy the thing. The SEC gamely tries to explain how this could in fact all be due to negligence, but it doesn’t really matter – the point is that GS and Citi did pretty much the same thing, so if it’s negligence for Citi it’s negligence for GS. The only explanation of why Citi gets off easier than GS is “because we chose to let Citi get off easier than GS.”

Same – even more ludicrously – with taking into account “whether a defendant clearly has articulated a willingness to settle prior to the filing of a complaint.” Citi articulated that willingness because the SEC went to them and negotiated a settlement before filing a complaint. The SEC apparently did not tell GS before filing a complaint, so they never had a chance to settle.

The real explanation is probably much closer to the one Citi’s lawyers hit upon: that Citi is the all-time league table leader in losing tons of money on CDOs. To a lot of people, Goldman really does look like the evil genius who went heavily net short the housing market and made a ton of money. Citi are just some goofballs who lost a ton of money on a ton of CDOs, and half-accidentally made some money on one CDO. That pattern does sort of make Goldman look like they had a diabolical plan to screw everyone, while Citi’s screwing a few people looks, well, negligent.

It’s just that this is a very bad legal theory. Shorting the housing market to your customers when you have no inside information about particular mortgages, but just did better analysis than them of the macro data, is … it’s kind of unpleasant behavior, maybe, but it’s probably not fraud. Being generally long mortgages but short one particular trade because you’ve secretly cherry-picked it to be the single worst CDO you can conceive of with your somewhat limited imagination, that’s – I mean, that’s stupid, but it’s also a lot closer to actual fraud.

SEC Defends Citi Settlement [WSJ]

Citi Memorandum of Law

SEC Memorandum of Law

* Because Citi made $160mm on the trade, the SEC calculates the $285mm settlement as $160mm of profits, $30mm of interest, and only a $95mm “penalty.” Compare to Goldman’s Abacus trade, where GS was not on the short side and made only relatively modest fee income on the trade, so that $535mm of the $550mm settlement was “penalty.”