Schneier on Security

Jonadab • June 26, 2012 1:27 AM

I’m not sure that I did know
that the returns were low.

Then you hadn’t really thought it through.

Banks keep enough cash on hand to cover expected withdrawals, with a bit of a margin in case somebody unexpectedly wants to close out a particularly large account. (Note that, unlike in the movies, where a particularly large account could be a hillion quadjillion dollars, in real life that’s no more than the limit of what the FDIC will cover in the event the bank goes under.)

Yes, I know, in bank-robbery movies the criminals always know that Bank X is going to have a quarter of a billion smackers lying around in the vault over the weekend because, umm, it’s being stored there on that date especially for the MacGuffin Event the following Monday. In the real world, however, there’s never any reason for a bank to have that kind of money laying around in cash. I don’t think it has happened once in the entire history of banking.

Although it is well know that banks, at least in the US, are required to retain liquid assets sufficient to cover a percentage (10% last I recall) of deposits, they are NOT required to keep this much on site in cash, and they don’t. They have it available to them, e.g., they have funds on deposit (electronically these days) with a federal reserve bank. That counts. (What doesn’t count is money they’ve lent back out as loans and other investments of that nature. That all has to come out of the other 90%, which is not required to be liquid.)

Mundane bank theft, then, of the kind that involves walking up to the teller and flashing a black squirt gun and demanding a sackful of green paper, is never going to be able to get more than what cash the bank has physically on hand. Indeed, as most banks are located inside municipalities (and thus are relatively close to police coverage), time constraints frequently require the perpetrators to settle for what’s in the teller’s drawer, which is generally a small percentage of the cash the bank has on hand. (In that case, you might actually get more from the cash register of a busy retail business, if you catch it near the end of the cashier’s shift.)

White-collar bank theft (of the kind depicted in the movie Ghost, though of course the movie depiction is not strictly realistic) can potentially net much more, but it also requires a great deal more skill to even attempt, let alone succeed, so it’s always going to be the minority case.

And yeah, all bank robbers get caught (unless they, like, have a heart attack and die first, or something). If they don’t get caught the first time, they blow through the money rather more quickly than they expected and find themselves trying it again, and it doesn’t take very many times to get caught. As I said above, almost all banks are located in towns and cities, so the police are seldom very many minutes away. Larger banks (which are tempting because they’re more likely to have a slightly larger amount of cash on hand) are especially likely to be located near police stations.

Reality is a harsh mistress for bank robbers.

Schneier on Security

A blog covering security and security technology.

« Far-Fetched Scams Separate the Gullible from Everyone Else | Main | Colbert Report on the Orangutan Cyberthreat »
June 22, 2012
Economic Analysis of Bank Robberies

Yes, it’s clever:

The basic problem is the average haul from a bank job: for the three-year period, it was only £20,330.50 (~$31,613). And it gets worse, as the average robbery involved 1.6 thieves. So the authors conclude, "The return on an average bank robbery is, frankly, rubbish. It is not unimaginable wealth. It is a very modest £12,706.60 per person per raid."

"Given that the average UK wage for those in full-time employment is around £26,000, it will give him a modest life-style for no more than 6 months," the authors note. If a robber keeps hitting banks at a rate sufficient to maintain that modest lifestyle, by a year and a half into their career, odds are better than not they'll have been caught. "As a profitable occupation, bank robbery leaves a lot to be desired."

Worse still, the success of a robbery was a bit like winning the lottery, as the standard deviation on the £20,330.50 was £53,510.20. That means some robbers did far better than average, but it also means that fully a third of robberies failed entirely.

(If, at this point, you're thinking that the UK is just a poor location for the bank robbery industry, think again, as the authors use FBI figures to determine that the average heist in the States only nets $4,330.00.)

There are ways to increase your chance of getting a larger haul. "Every extra member of the gang raises the expected value of the robbery proceeds by £9,033.20, on average and other things being equal," the authors note. Brandishing some sort of firearm adds another £10 300.50, "again on average and other things being equal."

We all kind of knew this — that’s why most of us aren’t bank robbers. The interesting question, at least to me, is why anyone is a bank robber. Why do people do things that, by any rational economic analysis, are irrational?

The answer is that people are terrible at figuring this sort of stuff out. They’re terrible at estimating the probability that any of their endeavors will succeed, and they’re terrible at estimating what their reward will be if they do succeed. There is a lot of research supporting this, but the most recent — and entertaining — thing on the topic I’ve seen recently is this TED talk by Daniel Gilbert.

Note bonus discussion terrorism at the very end.

Posted on June 22, 2012 at 7:20 AM • 41 Comments

To receive these entries once a month by e-mail, sign up for the Crypto-Gram Newsletter.
Comments

Some people do these things for the thrill of it, and some times desparate people do desparate things.

Posted by: Dan at June 22, 2012 7:34 AM

I think the distribution is the key.

What the stats appear to say is you have an e.g. 1/3 chance of getting away with a large haul. Most people plan a robbery assuming they will succeed, rather than get caught, so they don’t make the full assessment of expected value, only an assessment of expected value assuming everything goes to plan.

Posted by: sam at June 22, 2012 7:38 AM

I think the answer is actually really simple. Unlike the diabolical masterminds you see in the movies, real criminals are (by and large) just plain stupid.

Posted by: Peter at June 22, 2012 7:43 AM

Erm, I’m not sure that I did know that the returns were low. I kind of assumed that the risk of capture was high, and more importantly that it’s morally wrong.

I think you do your fellow humans a disservice by assuming that the low rate of return is a significant disincentive. It’s not at all a disincentive to honest people who have a satisfactory standard of living.

Posted by: Ian Eiloart at June 22, 2012 7:48 AM

“Every extra member of the gang raises the expected value of the robbery proceeds by £9,033.20, on average and other things being equal,” the authors note. Brandishing some sort of firearm adds another £10 300.50, “again on average and other things being equal.”

Ok, so at my next robbery, I’m gonna bring 100 guys with 100 guns.. That should do the trick, no?

Think about it like this, 100 guys would be harder to round-up than four. And, they are mostly ‘extras’ anyways so they get a smaller cut. In the end I make out better.

Posted by: Ninja at June 22, 2012 7:50 AM

These figures are well known in criminology/sociology. Gottfredson and Hirschi did a pretty serious study of violent criminals 20 years ago and found (as I recall) that the primary contributing factor for bank robbery (or any robbery) was lack of impulse control. Offenders didn’t even case the joint in many instances; the robbery was literally a spur-of-the-moment choice driven by immediate need — with no consideration of any longterm consequences either positive or negative.

The same is equally true of assaults and murders; violent attacks are almost exclusively driven by an immediate impulse with no forethought or planning. (This is also why the UK has far more attempted murders than murders relative to the US; the lack of easy access to firearms means that these violent impulses are acted on with knives and bludgeons, which have a much higher survival rate.)

For a full rundown, I recommend “A General Theory of Crime” — it’s dated but still very relevant. http://criminology.wikia.com/wiki/…

Posted by: Jay Garmon at June 22, 2012 7:54 AM

Wherever someone can calculate a risk/ reward ratio, theyll have a crack at it. But statistically they look like crap risk managers.

Posted by: Taipan at June 22, 2012 7:55 AM

Some people are not in a position make a purely economic choice. For example people with a substantial drug debt might be given an ultimatum: Rob a bank, or lose your kneecaps.

Posted by: Kasper at June 22, 2012 8:01 AM

I think you are figuring the comparison wrong. You are figuring a person in stable employment vs bank robbery, when these are the kind of people unlikely to take up bank robbery. Whereas what you want to consider is more like the earning potential of someone in unstable employment, with no doubt a range of other problems. People take up bank robbery because they don’t want or can’t get the stable employment in the first place and see crime as a way of getting the money and lifestyle they want. Just like other people with low and unstable incomes, they are poor at long term considerations, treating all income as a short term windfall to be spent immediately.

Posted by: Paul at June 22, 2012 8:09 AM

those robberies are just grab and run, so what do you expect, the people who can rob a bank of billions and get away with it and do it again next year are called CEO’s and they do it with “derivitives” not guns.

Posted by: dimon at June 22, 2012 8:27 AM

There are a couple of points I’ve seen in related discussions.

Firstly, these data are apparently averages. A highly skilled bank robber who plans carefully to hit a rich target may get much better returns, and also, if his planning is good, have a lower risk of getting caught.

Secondly, the article is not clear on whether the data are about bank losses, or about arrests and convictions. The difference between these is, of course, the losses for which there is no corresponding arrest and conviction — the ones that get away with it. It’s possible that these are the economically successful bank robbers, who may take a much smaller number of much larger hauls.

And thirdly, of course, for those who are cynical about the financial sector, the most economically successful bank robbers of all are right before our eyes. The absolutely best way to rob a bank is to own one.

Posted by: Urban Garlic at June 22, 2012 8:38 AM

I believe that these low numbers are simply the result of various security measures put in place by banks over the decades.

50 years ago, robbing banks was probably a lot more profitable, and that’s why the idea of robbing a banks as a one-off action that makes you rich if you don’t get caught lingers on.

Posted by: brazzy at June 22, 2012 8:55 AM

Identity theft is much more
profitable and less dangerous.

Fraud in general is more profitable and less dangerous than outright theft. It’s possible to make a VERY nice living selling worthless junk (or worthless investments) under false pretenses, and the legalities in many cases are sufficiently blurry that you can avoid going to jail by spending most of the take before anyone figures out how to prosecute you, apologizing and giving back what little is left, then drive a hundred miles and start a new scam the following Tuesday.

If you take the time spent into
consideration… The hourly wage
must be highly superior to the one
of the average employee.

You forgot to count the time spent in prison after getting caught. Throw that in, and you can expect to make less than minimum wage, even if you only count the prison time for forty hours a week (as opposed to 168).