Home > finance, modeling > Crime and punishment

Crime and punishment

July 22, 2014

When I was prepping for my Slate Money podcast last week I read this column by Matt Levine at Bloomberg on the Citigroup settlement. In it he raises the important question of how the fine amount of $7 billion was determined. Here’s the key part:

 Citi’s and the Justice Department’s approaches both leave something to be desired. Citi’s approach seems to be premised on the idea that the misconduct was securitizing mortgages: The more mortgages you did, the more you gotta pay, regardless of how they performed. The DOJ’s approach, on the other hand, seems to be premised on the idea that the misconduct was sending bad e-mails about mortgages: The more “culpable” you look, the more it should cost you, regardless of how much damage you did.

would have thought that the misconduct was knowingly securitizing bad mortgages, and that the penalties ought to scale with the aggregate badness of Citi’s mortgages. So, for instance, you’d want to measure how often Citi’s mortgages didn’t match up to its stated quality-control standards, and then compare the actual financial performance of the loans that didn’t meet the standards to the performance of the loans that did. Then you could say, well, if Citi had lived up to its promises, investors would have lost $X billion less than they actually did. And then you could fine Citi that amount, or some percentage of that amount. And you could do a similar exercise for the other big banks — JPMorgan, say, which already settled, or Bank of America, which is negotiating its settlement — and get comparable amounts that appropriately balance market share (how many bad mortgages did you sell?) and culpability (how bad were they?).

I think he nailed something here, which has eluded me in the past, namely the concept of what comprises evidence of wrongdoing and how that translates into punishment. It’s similar to what I talked about in this recent post, where I questioned what it means to provide evidence of something, especially when the data you are looking for to gather evidence has been deliberately suppressed by either the people committing wrongdoing or by other people who are somehow gaining from that wrongdoing but are not directly involved.

Basically the way I see Levine’s argument is that the Department of Justice used a lawyerly definition of evidence of wrongdoing – namely, through the existence of emails saying things like “it’s time to pray.” After determining that they were in fact culpable, they basically did some straight-up negotiation to determine the fee. That negotiation was either purely political or was based on information that has been suppressed, because as far as anyone knows the number was kind of arbitrary.

Levine was suggesting a more quantitative definition for evidence of wrongdoing, which involves estimating both “how much you know” and “how much damage you actually did” to determine the damage, and then some fixed transformation of that damage becomes the final fee. I will ignore Citi’s lawyers’ approach since their definition was entirely self-serving.

Here’s the thing, there are problems with both approaches. For example, with the lawyerly approach, you are basically just sending the message that you should never ever write some things on email, and most or at least many people know that by now. In other words, you are training people to game the system, and if they game it well enough, they won’t get in trouble. Of course, given that this was yet another fine and nobody went to jail, you could make the argument – and I did on the podcast – that nobody got in trouble anyway.

The problem with the quantitative approach, is that first of all you still need to estimate “how much you knew” which again often goes back to emails, although in this case could be estimated by how often the stated standards were breached, and second of all, when taken as a model, can be embedded into the overall trading model of securities.

In other words, if I’m a quant at a nasty place that wants to trade in toxic securities, and I know that there’s a chance I’d be caught but I know the formula for how much I’d have to pay if I got caught, then I could include this cost, in addition to an estimate of the likelihood for getting caught, in an optimization engine to determine exactly how many toxic securities I should sell.

To avoid this scenario, it makes sense to have an element of randomness in the punishments for getting caught. Every now and then the punishment should be much larger than the quantitative model might suggest, so that there is less of a chance that people can incorporate the whole shebang into their optimization procedure. So maybe what I’m saying is that arriving at a random number, like the DOJ did, is probably better even though it is less satisfying.

Another possibility to actually deter crimes would be to arbitrarily increasing the likelihood of catching people up to no good, but that has been bounded from above by the way the SEC and the DOJ actually work.

Categories: finance, modeling
  1. Guest2
    July 22, 2014 at 9:43 am

    Probably also should be based on previous history, in other words, what has been meted out in similar cases. This would form the data set (quantification of evidence, individual involvement, period of involvement, layers of corporate culpability, etc.). Multi-dimensional scatter plot, that sort of thing.


  2. Josh
    July 22, 2014 at 9:56 am

    I think the quantitative approach makes sense.
    I think the solution to your concern that the expected cost will be factored in is for the fine to be a multiple of the damage AND ALSO for the executives to be held personally responsible.
    They should pay restitution plus a fine. There is lots of precedent for this (triple-damages is comment) and common sense to justify it. In fact, that is the form of Citi’s penalty (though, as you noted on your podcast, it’s odd that the restitution is going to homeowners, not investors, more on that below).
    Of course, if the probability of getting prosecuted is < 1/3, triple is not enough. But it is a good start. The added penalty of real repercussions for executives would be both necessary and just. I wonder what price Jamie Dimon would have the modellers put in for "risk I might go to jail". I bet it would be a high number.
    As far as guilty e-mails, I agree this is problematic. But, if you are issuing hundreds of millions of securities with false information, isn't that a crime? Do you really have to prove you knew they were false. And, statutes such as RICO and SOX should make the culpability move up the executive ladder.
    I realize proving it was not just negligence ups the ante, but there are ways to do that without e-mails. Prosecutors make deals with lower level people to give evidence. E-mails haven't been so helpful, anyway. A good lawyer will introduce reasonable doubt about what "It's time to pray" meant. If that's the best they've got, we are lucky Citi forked over as much as they did.
    Though, I think the cost is more like $4 billion than 7. Maybe less. I think the compensation to homeowners is stuff they've already agreed to do. It was just tacked on to make the settlement look bigger than it is. That's why it is homeowners that are being compensated, not investors. It is because it is really unrelated. Also, I believe some of what Citi is paying is tax-deductible.


  3. someonesomewhere
    July 22, 2014 at 10:04 am

    You could make the fine be >100% of the damage they did?
    That would make it really punitive and makes gaming the system more tough. If they go bankrupt because of the fine, just liquidate the bank?


    • cat
      July 22, 2014 at 10:18 am

      I believe when you make punishments to severe people will double down and end up doing more damage to avoid the penalty because A) The reward has to be that much greater to justify the risk B) if you are going to lose everything there is no point in holding back. If this doesn’t seem like rational behavior to you its because you aren’t the kind of person who would cheat someone out of a dollar let alone a billion dollars.


      • July 23, 2014 at 12:47 pm

        That’s what you “believe”. But it’s been studied. Punishment works best when it’s swift, sure and relatively mild (in terms of incarceration).


        • July 23, 2014 at 7:43 pm

          Not quite correct. Punishment’s effectiveness is regulated by the perceived probability of getting caught and suffering the consequence, whatever it is. Draconian punishments have two problems and one big benefit.

          First, the society in which the infraction occurs makes moral judgments on the validity of the law. To the extent the perceived punishment is considered overly harsh, the society around the perpetrator(s) will fail to help and fail to report. Second, the harsher the punishment, the less the perpetrator has to lose. A probably heavy prison sentence may drive flight, attempts to bribe officials, murder of witnesses, etc.

          The big benefit of a draconian punishment is that it is an excellent negotiating chip for obtaining cooperation. If the probable punishment is 20 years in prison, it is possible to incentivize cooperation by granting leniency. If the probable punishment is a year’s probation, or just a fine, then there isn’t much incentive for the perpetrator(s) to turn in others, etc.


  4. cat
    July 22, 2014 at 10:23 am

    “To avoid this scenario, it makes sense to have an element of randomness in the punishments for getting caught. ”

    I’m 100% behind getting some game theory involved here that way even if the Gov’t published the model for punishment you couldn’t optimize because your competitors could pick a legal strategy which should give a better result.


    • josh
      July 22, 2014 at 10:52 am

      Cat: I agree that there is a doubling down risk (Nick Leeson / Baring’s bankruptcy being the clearest example of this I know of). Which is one reason the accountability should go up the executive ladder.

      As far as introducing randomness: our system of justice is (unfortunately) too capricious already. No need to add to it and some moral issues to doing so. (Of course, those moral concerns apply to the system as it is)..


  5. Min
    July 22, 2014 at 11:01 am

    To me this underscores the importance of jail time for white collar crime.

    Even lesser personal punishment can have an effect. I was struck by the difference in attitude of the CEOs of big banks in the UK and the US in the aftermath of the financial crisis. Testifying before Congress, the US CEOs came off as at least a bit arrogant, while the UK CEOs, testifying before, I guess, some parliamentary committee, appeared to be contrite and penitent. Aside from any cultural differences, the UK CEOs had all lost their jobs.


  6. July 22, 2014 at 4:33 pm

    Why do you assume that the punishment should be a fine? If it’s fraud let’s send em to prison!


  7. July 22, 2014 at 6:56 pm

    The problem is that your approach is not consistent with the law. It is not fraudulent to sell low quality mortgage backed securities if you have adequate disclosure. The fraud occurs when you knowingly misrepresent the relevant facts.


    • July 22, 2014 at 10:06 pm

      Yes, the sentencing ought to in some way consider the intent, not just the outcome, in the same way that there are range of penalties for unlawfully causing someone’s death conditional on the level of intent.


  8. July 22, 2014 at 8:19 pm

    Why does everybody believe that the fine has anything to do with the crime. Maybe it has everything with the amount of free money the banks received through the FED zero percent interest program. In a sense the FED has been paying the banks to pay the fine since 2008. And now it’s just a transfer of money to the U.S. Gov.


  9. Savonarola
    July 23, 2014 at 4:05 am

    Sure, “lawyerly” is a bad word and all. But you realize, when people (and even corporations) get sentenced in actual court, it is done with a shockingly quantitative method embodied in the Sentencing Guidelines. Most statutes mandate something like forfeiture of either the harm caused by the crime or the profit gained from the crime (at least with “white collar” crimes). In sentencing memorandums, the court gets down into the nitty gritty of how it calculated the amount and the defendant’s lawyer works to change the formula that will go into the presentencing memo.

    Here’s the problem with the Justice Department: they aren’t nearly lawyerly enough. They are afraid to take these crooks to court because they sometimes lose big trials through sheer incompetence. It’s incredibly difficult to prove anything beyond a reasonable doubt, especially the intent in a corporation’s “head” at the time of some particular malfeasance. But the DOJ don’t exactly go at it with fantastic acuity. They are bluffing with lions, who know full well that the gun is not really loaded.


    • July 23, 2014 at 6:49 am

      Super interesting information, thanks! And I am pretty convinced too.


      • Savonarola
        July 27, 2014 at 6:39 pm

        An interesting place to see this in action is antitrust. It’s a criminal statute, so sometimes there are guilty pleas and even trials, but the government often brings its cases under the civil side of the statute instead to improve its odds. There have recently been a couple of criminal trials of Japanese executives (and their corporations) for some worldwide cartels. If you take a look at news on TFT-LCDs or on some of the automotive parts cases, you can find interesting stuff about how this works on the ground. I remember seeing a recent opinion on post trial motions in one TFT-LCD case that kind of highlights the complexity of getting to “guilty” in a relatively simple case.
        For some examples of major prosecutorial mistakes in a high-profile criminal financial case, check out the ancillary Enron cases – the so-called “barges” case is instructive.


  10. July 23, 2014 at 12:44 pm

    It’s bad enough that the discipline of Econ has been overrun by math majors. That’s how we got to a point where business people do a cost benefit analysis on abrogating contracts. It wasn’t always like that!

    The last thing we need is math majors doing criminal justice. Richard Posner is in retreat and we need to keep it that way.

    The purpose of punishment is three-fold: societal vengeance, deterrence (specific and general), and rehabilitation. That’s how you calculate the proper penalty.

    Kathy, I’m a fan from the podcast! But don’t let shiny objects distract you from the big issue: turning morality into math is how we got to a Wall Street that needs occupying!


    • July 23, 2014 at 11:00 pm

      I realize “math majors” sounds rather personal given the context of the particular blog. It is my term for the model based Econ that is an infectious disease in the body of capitalism, not math per se.


      • July 23, 2014 at 11:17 pm

        What is your term for the infectious disease in the body of communism?


        • July 24, 2014 at 7:14 am

          The conventional wisdom is that communism died from a genetic disorder, not an infection.


        • July 24, 2014 at 7:28 am

          OK, and your term for the infectious disease in the body of socialism?


        • July 24, 2014 at 2:03 pm



  11. July 23, 2014 at 7:43 pm

    In punishing corporations, especially banks, who should go to prison? The CEO or the employee? Nick Leeson or the Barings CEO? Joseph Jett or the Kidder Peabody CEO? I chose those two cases, because in the former, Nick Leeson succeeded in fooling his won people. In the latter, Jett’s bosses were all in on the scam. Was E. Stanley O’Neal complicit in Merrill’s actions or was he really ignorant of derivatives? Dick Fuld was probably highly knowledgeable in derivatives. But do we really know? Can Justice or SEC really assess this? If we don’t know the answers to these questions how do we set penalties? Are the shareholders to blame? Maybe not, but they “own” the risk. Just some thoughts.


    • davidflint
      July 24, 2014 at 2:31 am

      Well, abekohen, those are questions for a jury. Crimes are committed by people – individually or in groups – and those people should be punished. In banking those people will be presumably be traders, loan officers, etc.

      It should be (perhaps it is) a crime to encourage or tolerate fraud or to be reckless in controlling the risks. The guilty people are likely to be senior executives including those who specified or approved the IT systems and mathematical models.


    • Josh
      July 24, 2014 at 7:52 am


      Nick Leeson was an isolated person who got away with what he did for a short period of time.

      The lawbreaking at the banks were department-wide, at least, often global and went on for years. Read the settlements on the banks. So, the CEO and other executives were negligent, at best. That is what RICO was supposed to address.

      I also agree with davidflint that a jury can decide these questions.


      • July 24, 2014 at 8:03 am

        Nick Leeson was not the only lone wolf who did bad things at a bank. There were lots more cases like his. Some were made public. Others weren’t. So if someone’s boss was hoodwinked by such a player, should the boss be punished?

        As for juries deciding such questions, are they really equipped to decide whether the boss was a conspirator or a victim? In most cases, I think not. Yet, somehow we feel good when a jury has made a decision. Why is that?


  1. July 23, 2014 at 11:59 am
  2. July 23, 2014 at 1:09 pm
  3. August 4, 2014 at 12:01 pm
Comments are closed.
%d bloggers like this: