A lot of legal issues look like substantive things but are actually things about what institutions can and want to do. Obviously more people want to think about questions like "should the U.S. have universal health insurance?" than about questions like "does the Anti-Injunction Act bar lower federal courts from reviewing the individual mandate until taxes are collected in 2014?," but judges tend to get into the latter question. That's why they're judges. That difference can make judicial decisions sort of hard to interpret.
Today everyone's favorite federal judge, Jed Rakoff, surprised few but pleased many by beating the ever-loving crap out of the SEC's settlement with Citigroup, in which Citi had agreed to pay the SEC $285 million in exchange for the SEC not asking too many questions about its synthetic CDO deals that were maybe not so hot. Here's the gist of it:
Applying these standards to the case in hand, the Court concludes, regretfully, that the proposed Consent Judgment is neither fair, nor reasonable, nor adequate, nor in the public interest. Most fundamentally, this is because it does not provide the Court with a sufficient evidentiary basis to know whether the requested relief is justified under any of these standards. Purely private parties can settle a case without ever agreeing on the facts, for all that is required is that a plaintiff dismiss his complaint. But when a public agency asks a court to become its partner in enforcement by imposing wide-ranging injunctive remedies on a defendant, enforced by the formidable judicial power of contempt, the court, and the public, need some knowledge of what the underlying facts are: for otherwise, the court becomes a mere handmaiden to a settlement privately negotiated on the basis of unknown facts, while the public is deprived of ever knowing the truth in a matter of obvious public importance.
Here, the S.E.C.'s long-standing policy - hallowed by history, but not by reason - of allowing defendants to enter into Consent Judgments without admitting or denying the underlying allegations, deprives the Court of even the most minimal assurance that the substantial injunctive relieve it is being asked to impose has any basis in fact.
Right on! But also maybe just a little disingenuous. Judge Rakoff was not being asked for "substantial injunctive relief," not really. It looks like that on the surface, in the sense that (1) the SEC and Citi worked out a deal where Citi gives the SEC money, promises not to violate the securities laws again, and agrees to do some remedial stuff like telling its salespeople to stop peddling synthetic CDOs structured by the protection buyer without telling anyone because somehow that is still a problem; and in the sense that (2) the SEC was asking Judge Rakoff to enshrine that agreement in an injunction. And then, if Citi didn't keep its agreement - by not doing the remedial things, say, or by violating the securities laws again - the SEC could go back to court and say "hey, Citi violated the injunction" and Judge Rakoff could hold Citi in contempt and fuck. it. up.
And that is the theory, and it is a lovely theory, but as has been rather hilariously documented, companies enter these agreements all the time, and then keep violating the securities laws, and never ever ever ever get held in contempt. So the odds were zero that Judge Rakoff would ever be asked to get "formidable" on Citi's ass just because it kept on violating the securities laws.
The flip side is that, as Judge Rakoff hints, the SEC could get everything it wants without ever going to court. Money, for instance. Judge Rakoff's opinion is about the power of injunctions, because he's been asked for an injunction, but if you hadn't been paying close attention you might think that the SEC/Citi settlement was not about enjoining Citi not to keep breaking the law but was instead about getting Citi to pay the SEC $285 million. You might think that because, among other reasons, that's how the SEC headlined its press release about the settlement.
Now, if (1) Citi wants to give the SEC $285 million, and (2) the SEC wants to take it, and (3) they don't want to have a trial, they can just do that. They don't need a court to sign off on it. Citi writes a check, the SEC drops its lawsuit, the end.*
The injunctions are harder - if Citi and the SEC agree outside of court that Citi will have to retrain its sales force, and Citi doesn't do that, then the SEC can't hold Citi in contempt. But being held in contempt of court is not the only bad thing that can happen to a bank. Citi can agree to do whatever the SEC wants it to do, after negotiation, and the SEC has quite serious mechanisms to enforce that agreement. So if Citi and the SEC signed a settlement saying "Citi will put in big red letters on the front cover 'these mortgages were chosen for you by people who hate you and want you to fail at life,' and it agrees that failing to do that constitutes securities fraud," then if Citi doesn't obey the settlement it's admittedly committed securities fraud and its consequences - fines, criminal liability, what have you - would be at least as bad as the consequences of violating an injunction. They're all hypothetical consequences - the SEC is no more likely to, say, bar Citi from the securities industry than a court would be to imprison Vikram Pandit for contempt - but they're real enough to make Citi try hard to do the things it agreed to do.
So the SEC can get (1) money and (2) reasonable assurance that Citi will try really hard not to do more terrible things without any court approval. In fact, the SEC did exactly that with Credit Suisse in this case. (CS were the ones going around claiming that they'd picked the underlying mortgages to be the very best mortgages that they could be, which was maybe less accurate than "Citi picked the underlying mortgages that it wanted to have a short position on.") The SEC, on the same day it announced its settlement with Citi, announced its administrative settlement with Credit Suisse, which involved fines, disgorgement, "we promise not to be bad any more" agreements, and suspending one Credit Suisse guy from the securities industry for six months.
So why didn't they do that with Citi? Why didn't the SEC and Citi just agree on what the deal was and announce it in a press release, rather than agree on what the deal was and announce it in a press release and submit it to a judge who just hates, hates, hates the SEC and who they had to know would call them out for being generally all-around shoddy re: this whole Citi thing?
I don't know. Part of it is that the SEC's enforcement mechanisms are complex, and there are some benefits to the SEC in getting a court-approved settlement rather than an SEC-only "cease and desist order" settlement. Perhaps I'm missing something that makes a court order critical - in particular, the SEC-only cases have caps on civil penalties, though they seem pretty malleable (they're per day of violation, which, y'know, find a day that Citi wasn't in violation of a securities law), and though I doubt anyone would be able to object if the SEC and Citi just agreed on a bigger fine than those caps allow.
But I suspect that the main answer is symbolic: there's a sense that you go to court in the Big Cases, and you make a Big Public Deal of them, and - of course - the judge goes along with the SEC's Exhaustively Researched Case. I mean, it worked with Goldman and JPMorgan, and the Citi case is not really less compelling than those cases. It just got a less favorable judge.
Now, of course, the symbolism has broken down, and the SEC is stuck. It probably still could, legally, drop this case and settle everything up privately - Judge Rakoff probably can't literally force the SEC to go to trial, all he can do is refuse to issue the injunction that the settlement calls for - but then the SEC would look even worse than it usually does. Which means it has to either go to trial or negotiate a better deal - maybe a higher fine, certainly more disclosure and admissions - that Rakoff can sign off on. That will be tough, given that Rakoff clearly thinks that Citi did intentional badness - and that admitting intentional badness could lead to hundreds of millions or billions of liability to private purchasers of these CDOs. So Citi can't really settle the way Rakoff wants, the SEC can't really settle the way Citi wants, and we may actually see a trial.
Which might push the SEC to think a little more carefully in the future - not about whether its settlements are in the public interest (whatever Judge Rakoff may think, the SEC clearly believes that getting a $285 million unlitigated settlement out of Citi is a good deal here - and I'm not sure they're wrong), but about whether the legitimacy it can get from having a federal court approve these settlements outweighs the pain of having them reviewed by judges. If the SEC does everything in its own administrative actions, then everyone involved is a repeat player at the SEC or in the securities industry: they understand each other, and they can work out settlements that are good for the industry and seem like rough justice to everyone closely involved.
Judges, on the other hand, tend to be smart, independent, not deeply in the weeds of the securities industry, not so concerned with keeping up relationships between banks and regulators, not as attuned to Citi's capital position - and used to a little bit more public airing of grievances than the SEC might like. Here's how Rakoff concludes his opinion:
In much of the world, propaganda reigns, and truth is confined to secretive, fearful whispers. Even in our nation, apologists for suppressing or obscuring the truth may always be found. But the SEC, of all agencies, has a duty, inherent in its statutory mission, to see that the truth emerges; and if it fails to do so, this court must not, in the name of deference or convenience, grant judicial enforcement to the agency’s contrivances.
Yeah, but. The SEC minute by minute, hour by hour, in all sorts of ways fails to see that truth emerges. Sometimes that gets noticed, resulting in a public scolding from Rakoff, or David Kotz, or Matt Taibbi. Most of the time, though, it doesn't get noticed. Bringing lawsuits in the Southern District of New York is a good way to get the SEC's failings noticed. Bringing SEC administrative actions, announcing big fines with upbeat press releases, and not having them subject to any review, is a good way to sweep them under the rug. The more scrutiny that the SEC gets when it goes to court, the more tempted it will be to handle cases like this on its own.
One thing that might give them pause, though. The SEC has, in the wake of recent expansions of its powers under Dodd-Frank, experimented with moving away from the courts and towards administrative actions when its case against supposed wrongdoers maybe isn't so hot. It's already been slapped down for doing so. By Jed Rakoff.
Rakoff Causes More Heartburn at the SEC [WSJ Law Blog]
* Of course if a cop says to you "well, I was going to arrest you for pepper-spraying your fellow holiday shoppers, but if you give me $285 I'll let it go," that would look, um, bad, but the SEC actually does have the power to collect out-of-court fines. The biggest I've found outside of court was $34 million on top of $34 million disgorgement, which is not nothing. These have a more formal process than just writing a check to Mary Shapiro, but not by much - and a court never gets to touch them.