The Wall Street Journalstory today about the next Libor domino to fall - RBS, which will be coughing up $700mm or so to regulators in the next few weeks - is full of quietly hilarious lines, perhaps none more so than the Journal's deadpan clarification that "the Justice Department has the power to file criminal charges without the bank’s blessing." For sheer backwardness, though, I think it's hard to top this:
As part of UBS's settlement last month, the Swiss bank's Japanese unit pleaded guilty to wire fraud, a felony. Justice Department officials were heartened by the lack of a negative reaction in the markets and among regulators around the world to UBS's guilty plea. Before the settlement deal, some officials had worried it could destabilize the bank. That has emboldened officials to pursue similar actions against banks like RBS, according to a person familiar with the matter.
The way a lot of people - sometimes even people at the Justice Department! - think about criminal law goes something like this:
- If you do something naughty, we will charge you with a crime.
- If you are convicted of that crime, bad things will happen to you.
- You don't want bad things to happen to you.
- So you won't do anything naughty.
This is called "deterrence." All of the parts of it are important: if you are convicted of a crime and bad things don't happen to you, then the whole system is mostly pointless. When the Justice Department is "heartened by the lack of a negative reaction" to a criminal conviction - when they're like "yay, no deterrent effect!" - then ... then ... gaaah.
Felix Salmon explains the equilibrium here: "The criminal prosecution, in these cases, seems to be little more than a CYA move on the part of the administration, which can now have a slightly straighter face when saying that it’s being tough on the banks." The process is (1) people scream "why are there no criminal prosecutions of banksters etc.?," (2) DOJ offers up some pretty fake criminal prosecutions - "we couldn't jail Jamie Dimon, but we brought you a guilty plea from a Japanese shell subsidiary," and (3) the problem of the awkward press conference is put off for another day.1
That's sort of a strange way to use the tool of criminal prosecutions. Does that suggest that better tools might be available?
Charlie Gasparino has been on fire this week and his column in the Post today on SEC nominee Mary Jo White strikes me as right on2: concerns that White might be too soft on banks because she represented them in private practice are silly, her experience as a prosecutor probably means that she will be an aggressive prosecutor of banks, and - prosecuting banks isn't really what's needed. What's needed is clear, sensible regulation that will achieve good outcomes for markets. A prosecution-first mentality leads to ... I dunno, that thing with UBS.3
Take Libor. If the Libor scandal was about a few bad apples knowingly committing the crime of Libor manipulation, the right response would be something like (1) prosecute and imprison those criminals and (2) maybe take back illicit profits from the banks but don't go too hard on them: they were victims too, in a sense. Or if it was about a few rogue banks that were basically criminal enterprises, with a knowingly illegal culture of Libor manipulation, then I guess you'd (1) jail a lot of people at those banks and (2) criminally prosecute the banks themselves - in a real, punitive way, not a "lack of a negative reaction" way. These two interpretations seem to be false; in any case no one's acting like they believe either one.
A third interpretation would start from the fact that as far as anyone can tell every bank manipulated Libor. BURN THEM ALL DOWN, maybe, but you might consider whether the universality of the manipulation means that the rules against it weren't all that great. For instance: there were none! Libor was just a poll conducted by a private organization; it wasn't clear to a lot of regulators that they even had jurisdiction over it. And the poll was pretty vague, asking banks to submit more of an opinion than a fact; it's not hard to understand why some banks might shade their opinion in a way that might help them. You'd notice also that everyone knew about this - the regulators, the central banks, everyone - but didn't do anything about it when it happened. Just prosecute it later!
Of course on that interpretation a bank-destroying prosecution would seem a bit harsh. So the preferred approach seems to be to make the banks cough up (a small portion of) their illicit profits, and then tack on a fake criminal prosecution just to check that box.
Jesse Litvak might be a similar story. Do his IMs to counterparties lying about his purchase price look bad, deceptive, fraudy? Oh, sure, no doubt. Are they very, very much on a continuum with absolutely standard practices in the OTC market? Definitely. Can they be prevented by after-the-fact criminal prosecution? Umm ... maybe? But where do you draw the line between Litvak's imaginary negotiations with counterparties, and the imaginary negotiations that bond salespeople conduct every day?
Could Litvak's behavior have been prevented by simple up-front regulatory solutions? Yes definitely without any doubt whatsoever. The corporate bond market is not 100% transparent, but trade reporting on TRACE means that what Litvak did to his RMBS customers could never happen in the US corporate bond space. Litvak tells Buyer "Seller is offering me at 80.25, I'll sell to you at 80.5," Seller is actually offering at 78, Buyer says okay, and there are two prints to TRACE at 78 and 80.5. Buyer sees the prints on Bloomberg, Buyer calls Litvak's boss and threatens to stop trading with Jefferies, Litvak is gone by lunchtime.4
TRACE. It just works. It's better than jail!
And real-time trade reporting is slowly coming to other sectors - some OTC derivatives, for instance - over a lot of objection. That'd be a good place for regulators to focus their time and political capital. Or, y'know, put Jesse Litvak in jail for squeezing $2mm from some hedge funds, whatever, either way.
Libor, similarly, is open to a range of solutions, though to be fair they're more complex than TRACE, which is why various blue-ribbon panels are working on them. (My solution, "just make fraud okay," seems not to have found much favor.) The gist of the solutions - have fewer Libors, try to tie them more closely to actual trades, specify what data banks should use in submitting Libors, and clearly say it's illegal to manipulate Libor - seem like a step in the right direction.
The alternative is to have occasional long-after-the-fact prosecutions of things that look like misbehavior in hindsight, as a way to deter future misbehavior. If you think that Libor manipulation and Litvak's scalping were clearly evil at the time, and easily distinguishable from customary and permitted practices, then: I guess that's a fine solution. But you gotta mean it. If the go-to tool of financial regulation is the threat of criminal prosecution, and if you dissipate that threat by preferring criminal prosecutions with no bad consequences, then what are you left with?
U.S. Wants Criminal Charges for RBS [WSJ]
When banks face criminal charges [Reuters / Felix Salmon]
The perils of Mary Jo: The SEC needs more than a prosecutor [NYP / Charlie Gasparino]
RBS in for another round of bonus awkwardness, with added Libor angle [FTAV]
1.I listened to yesterday's Litvak press conference and the prosecutors were asked "why are there no charges against Jefferies?" Answer: "investigation ongoing, etc. etc." The important thing is to cover, very loosely, all the bases. "Why are no banks prosecuted?" "Well, UBS['s Japanese subsidiary] was." "Why are no individuals prosecuted?" "Well, Jesse Litvak was." Etc. Shift the debate.
2.With the caveat that I don't endorse his conclusion that White is "totally unqualified to run the Securities and Exchange Commission." He's right that the hunt for the toughest available prosecutor is the wrong way to staff the SEC, but just because she was hired by a bad process doesn't mean she'll be bad. I'm of the best-athlete school of thought on this one; why judge her solely by her resume? Even former federal prosecutors can turn their lives around and do some good in the world.
3.Here is a caveat to the tune of: I know that the SEC doesn't bring criminal charges (though it helps develop them), and that the CFTC is mostly in charge of Libor. The comment is on a regulatory-prosecutorial mindset, not particular agency jurisdiction.
4.Presumably there are modest ways around this transparency - tell us some in the comments! Certainly it doesn't prevent big markups on bonds in seasoned inventory (but why would that be a goal?). But crossing from one customer to another in a near-instantaneous riskless principal trade for a negotiated markup - the place where Litvak was apparently cheating - that's going to be clearly reflected on TRACE.