The economics of cheating

This essay is inspired by the chapter in Levitt’s book Freakonomics (see my review) in which he writes about cheating. But I think his book was long on stories and short on analysis. So here is some of my analysis.

The obvious reason why people cheat is that they gain some kind of benefit from it.

You can cheat on the SAT and thereby get into a better college. You can download music files from the internet and get to listen for free instead of having to pay for a CD. You can hold up a 7-11 and get all the money from the cash register. These are all benefits.

So why not cheat all the time? Because with these benefits come costs. Let’s examine what these costs are. The most obvious cost that people think about is punishment for getting caught. So when people consider cheating, what they consider is the perceived chance of getting caught multiplied by the perceived punishment.

I use the word “perceived” here to emphasize that this is probably the area that potential cheaters are the most likely to misjudge. It is a known fact that people are very bad at judging probabilities.

For example, it seems to me that the guy who holds up a 7-11 is making an extremely poor judgment. The severe punishment for armed robbery coupled with the high priority that police place on finding armed robbery suspects makes holding up a 7-11 seem like a very bad deal to me. People who engage in this behavior have to be pretty stupid.

On the other hand, there are probably other cases where people are overestimating the chance of getting caught. It may be that the average taxpayer has such a tiny chance of being audited that he’s foolish for not cheating on his income taxes. And I know that most people would cheat on their taxes if they knew they wouldn’t get caught. Is there really a net benefit to cheating on your taxes? I don’t know, and this is just a great example of how even the economically sophisticated are often ignorant of the true probabilities involved.

Another cost of cheating is opportunity costs. Cheating often takes effort, and such effort could otherwise be used for other pursuits. For example, the primary reason why I don’t download music from the internet is because it’s easier for me to buy the CD at Amazon.com. My opportunity costs are higher than the value of the free music because I’m an adult with a decent source of income. If I were a kid in high school without a job, then the opportunity costs of downloading music would be a lot lower, so illegal downloads of music would be my primary means of obtaining music.

In his book Freakonomics, Steven Levitt wrote about the social stigma costs of cheating. I think this is just a type of harm that comes from getting caught. Some types of cheating, like cutting in a line, involve a high chance of getting caught, but the punishment received is zero from a pecuniary standpoint. However the cheater feels like a real schmuck or schnorrer, and it’s this social stigma, or shame if you want to call it that, which keeps people honest.

Some types of cheating have no social stigma at all: such as driving faster than the posted speed limit on I-395 in Northern Virginia. In fact, not cheating and driving 55 mph will cause the stigma, because every other driver will probably be pissed at you for slowing down traffic. I don’t think that downloading or copying music has much social stigma either.

The final cost of cheating, and the most interesting because it’s the least rational, is the moral pain people feel when doing something wrong. Because of the moral pain involved, people will often avoid cheating even when the benefit of cheating far outweighs the opportunity costs and there is zero chance of getting caught.

When society tells you that something is wrong, you internalize the moral lesson and feel moral pain, or guilt if you prefer to call it that, while doing and after doing a wrongful deed. Some people feel moral pain more strongly than others. If you are too far above the norm with respect to feeling moral pain, people call you a “goody two shoes” and they don’t like you.

One general rule about moral pain is that it’s related to being able to identify a victim. When you steal from a person, the victim is obvious. When you steal from a big corporation, the victims are the corporation’s shareholders, but their identity is so spread out that it feels more like a victimless crime. Then there are true victimless crimes, such as downloading music; this crime is victimless because if you weren’t going to buy the CD anyway, no one is losing any money because of your actions. And even if you would have bought the CD, the victim is still a big corporation. This explains why the music industry is having such a hard time convincing people that downloading music is wrong. According to the way most people view the world, it’s not wrong.

In conclusion, we learned that a person’s net benefit from cheating is equal to the benefit he gets from cheating, minus the opportunity costs, minus the perceived probability of being caught times the punishment, minus the social stigma, minus the moral pain. A person will cheat if the net benefit exceeds zero. But because people have differing abilities to correctly ascertain the probability of being caught and because people have differing sensitivities to moral pain, when faced with the same cheating temptation not all people will behave the same way: some will cheat and some will not.

Comments

[Steve]

"Case in point: workers who have completed a degree earn much more than those who dropped out right before completion. Signalling explains this phenomenon. Human capital theory does not."

Brilliant point, if the statistic is true (and it sounds believable). But I also think that dropping out of college right before getting a degree is also a CLASS issue. Someone from an upper middle class family with upper middle class expectations wouldn't do it.

I'm glad you liked my point! I think class is not talked about enough in this country. Those class expectations reinforce economic status.

[meep]

Well, some people also misperceive the benefit from cheating -- consider the idiots who buy crap papers on the internet, just to fail anyway. When I taught calculus at a college, a fellow calc teacher caught a student cheating on exams... but didn't do anything about it because the kid cheated off the 2nd-worst student in the class, and didn't cheat enough to get the D the 2nd-worst guy got.

Being able to cheat (and lie) effectively is a skill, something many people don't realize until it's too late.

Oh, and moral pain does not require a victim, by the way. I'm Catholic, and I can tell you that there are plenty of "victimless" sins in the Catechism. But even we recognize different levels of culpability.

[Michael H.]

Hi Half Sigma
Interesting post.
I think the reason why some people cheat and other do not has less to do with their ability to correctly determine the probability of getting caught than the "moral pain" that these people have when they cheat others. In other words, some people are just bad and being smart has nothing to do with it.

[Half Sigma]

Both are factors. Moral pain might be a bigger factor. The authors of The Bell Curve showed that being caught committing crime was negatively correlated with intelligence. Less intelligent people, in addition to not being as good with estimating probablities, also are more likely to overweight immediate gratification (the money you get from stealing) and underweight the future consequences.