Bezzlin'

John Quiggin has a post about the bezzle, JK Galbraith's[1] idea of the amount of undetected corporate at any given point. 

As a boom continues, and everyone does well, people realise they can siphon off money and use it to make even more money. If they are threatened with detection, the original amount stolen can be returned to the till, and thye are still ahead. But, in a crisis, this can’t be done and, in any case, outside accountants are all over the books. So, embezzlers are caught and the bezzle shrinks. It stays small in the early stages of recovery when most decisions are being made by the cautious types who survived the crisis. But as the boom continues, hungrier and less-risk averse types come to the fore and the bezzle begins to grow again.

The key point about the bezzle is that on the way down it is very much like a bank run. As I see it, the key thing about fraud is what I'll call the oversight timeline. If you are managing someone else's money, at some point they are going to come for it, and knowing how and when they do this is crucial to successful thieving. For instance, a particularly ideal situation for thieving is where you are given a a sum of money to invest, and where the principal (you are the agent) will only ever draw a portion that is lower than the rate of return. As long as they are never planning on drawing more, or, god forbid, moving it to a different money manager, then thieving is pretty easy. You just skim the difference. 


The problem is that these people often have children, and these children have an expectation that they will be able to get the money, and--greedy little buggers--they probably have an expectation  of how much there should be, given a knowledge of how much was given to you to invest and what a reasonable rate of return should have been over this period. This is where the oversight timeline comes into play. If you know that the investor is going to die in 10 years, then you just need to make sure that when 10 years has come and gone you have returned enough of the money to go unnoticed. That is, you need to either get higher returns using the money you siphoned off than the account you were managing, so that you can return a portion while still having made a profit, or you need to have found another principal from where you can transfer funds. The latter starts to look a lot like a Ponzi scheme, juggling your theft between multiple victims.  There is another option, of course. As long as your timeline for success is shorter than theirs, you can just take the money and run. If your "success" involves you getting mixed up in a particular social milieu, having a status in a particular circle, becoming a pillar of the community, then your timeline is almost certainly going to be longer than the oversight timeline. If you want to be an Austro-Hungarian opera house, taking the money and running is not a realistic option.


But key to not getting caught, in all cases, is a knowledge of the oversight timeline: you need to have a good idea when someone is going to look at your books so that you can decide whether you juggle (possible as long as only one set of books is examined at any given time), gamble (which has the added risk that even if you are on average above the amount needed, knowing where you will be when the time has run out is much more difficult to predict), or run. We can rank these option by risk.

1. Juggling
2. Gambling
3. Running

Running is relatively risk free, at least from the perspective of getting out of the country and to a place from where you can't be extradited. You only have one timeline to manage, and as long as your own is shorter than that you are fine. Gambling is risky, because you can't be sure you'll win. But at least you still are only managing one timeline. Juggling requires several oversight timelines to manage. The risk of one coming in early is multiplied. This, of course, is one of the more irritating things about death. We can see it coming in some cases, but in other situations it can really muck up the game plan.


But the uncertainty of the timeline is not the only problem. For the non-running option, you need to be able to secure a higher rate of return through private investments, which usually requires a boom and is tough (but not impossible) in a crisis, or you need to be able to access cash relatively quickly, either by taking from other clients or borrowing against your own assets. That is, you need a boom and/or liquidity. The problem with financial crises is that everything seizes up. Not only is the oversight timeline often dramatically and suddenly shortened (in a crisis, all of a sudden everyone wants detailed financial statements), but your ability to generate sufficient private returns to cover your theft is significantly reduced as correlations approach 1 and liquidity is greatly reduced as banks are less likely to lend you money and as your other clients start being more over-bearing in their oversight, etc.


Which is why financial crises are usually associated with a significant number of prosecutions: fewer people can keep the game running given the conjunction of shortened oversight timelines, decreased returns, and less liquidity. So, why have there been so few this time? Quiggin offers up a few possibilities: 

1. There wasn’t any crime going on in the banking sector this time around....2. Regulators and prosecutors are complicit in the game and are letting financial criminals go free....3. The scale of legal, privately profitable but socially destructive, financial activity is so massive that, even post-meltdown, it’s impossible to detect embezzlement against the general background of dubious financial manipulation.4. The bailout was so comprehensive that the bezzle just kept on growing.

The first I think is nonsense, not because I have any particular animosity but because there is always crime going on in every economic sector. But it does raise the question of the appropriate baseline. On the one hand, we would like to see a perfect correlation between prosecutions and actual fraud. But this isn't going to happen, and while it should inform policy debates about enforcement, it shouldn't necessarily inform a debate about why there are fewer prosecutions this time around. So we use the "last time around" baseline. Why are we seeing fewer prosecutions relative to other financial crises? I assume that any distribution of propensity for fraud among individuals hasn't changed. Again, not because of animosity so much as my priors all point me to thinking that (1) these things are remarkably stable over time, and (2) if anything the financial sector has been recruiting more people willing to take these kinds of risk. So let's assume it's the same. In that case, the incentives or the enforcement must have changed.


Quiggin's 2 and 3 look at enforcement, either complicity or capacity. I also wonder if there was a change in incentives. Not that there was no crime in the banking sector, but relative to other booms there was less. Why? Possibly because the ability to generate a sufficient rate of return to avoid the juggling or the running options (the most risky and the most costly) was too low, given a decade of weak growth. In that case, there may have been less fraud this time around (and accordingly fewer prosecutions) because there was less opportunity to delude yourself into thinking you could get away with it. The likely gains from gambling were smaller, the risks of juggling remained high, leaving running as the best option. But the costs of running were always the highest, and it's not clear that these would have gone down in the last decade.


I have no idea whether any of this is true. Certainly I'm not sure that it's the case that the ability to generate higher than needed returns (gambling) was reduced over the last decade. And my priors and prejudices all point me toward thinking that Quiggin's 2 and 4 are especially important, and that 3 is at least somewhat likely. But I wanted to suggest that there might also have been an ironic reduction in actual fraud based on the lackluster decade that preceded.


[1] Galbraith has only become more enjoyable as we have entered into the age of text and instance messages. I retrospectively see him as cracking even more jokes than he did when  alive.