Freakonomics: Cheating

This is the second of a series of blog posts reviewing Freakonomics by Steven Levitt and Stephen Dubner. Today we look at Chapter 1, “What Do Schoolteachers and Sumo Wrestlers Have in Common?”

What they have in common is that they both cheat.

There’s a fascinating story about how Steven Levitt figured out that cheating is rampant in Japanese sumo wrestling, but unless you’re Japanese you probably won’t get why this is a big deal.

So lets move on to schoolteachers. It’s especially outrageous when schoolteachers cheat, because their job is to teach their students not to cheat on their tests and school assignments. But cheat they do.

How exactly are teachers cheating? Well we’ve created a system of rewards and punishments for teachers based on how well their students do on standardized tests. So teachers an have incentive to boost their students’ scores on such tests.

Levitt did a study of test answers in the Chicago Public School system, and discovered that 5% of the teachers were cheating in order to give their students a boost. Based on Levitt’s finding, a dozen teachers were fired, and the next year cheating went down by 30%.

So if sumo wrestlers, schoolteachers, and day-care parents all cheat, are we to assume that mankind is innately and morally corrupt?

That’s exactly the message I got out of this chapter, even though Levitt tried to assure us that it’s not so bad. In the future I plan to write about the economics of cheating. Levitt has certainly given me something to think about.

* * *

The last story in this chapter is about a guy who sells bagels in offices using the honor system. He puts out the bagels and a collection box where people are supposed to put in money for what they eat. About 87% of the people pay, but 13% of the bagels disappear.

What’s especially interesting is the relationship between honesty and the size of the office:

The data also show that smaller offices are more honest than big offices. An office with a few dozen employees generally outpays by 3 to 5 percent an office with a few hundred employees.

Levitt suggests that small offices are more honest because there’s greater “shame” in being dishonest at a small office; people there are more likely to know you.

I think Levitt is wrong about that. Small offices are more honest because people at smaller companies tend to be happier employees. I previously blogged about a Harris poll which showed that people working at small companies are happier with their job. I concluded that small companies provide an inherently more rewarding work experience.

The bagel seller’s data is reflecting this phenomenon because mood contributes to honesty. For example, “unseasonably pleasant weather inspires people to pay at a higher rate.” And there were other examples of mood being related to payment rate. Workers at smaller offices are probably in a better mood.

So hidden in this chapter is a very important piece of evidence regarding the superior work experience of small offices and small companies, yet Levitt doesn’t even realize it.

UPDATE

Michael Higgins has some thoughts about the bagel guy story.

Comments

[Michael H.]

Hi Half Sigma
I was intrigued by the bagel guy story. We have a bagel guy who comes to our office twice a week. These are really low-quality day-old bagels so I don't know why people would want them. Anyway, a lot of people are apparently cheating this guy.
I think it is easier for people to free ride if there are more people. The cheater figures that his contribution to the box will not make much difference to the total kitty and no one will know that she cheated him. In an office of 5 people, one cheat is immediately noticed and all five are shamed (they are all one-fifth cheats ex ante). In an office of 100 people, one cheater only makes everyone else a 1 percent cheat.
Cheating is much easier in large anonymous crowds.

[Half Sigma]

The bagel seller story was supposed to tell us that 87% of us are honest. However I didn't get that at all. It only says that 87% are honest when a $1 bagel is at stake. And maybe it isn't so easy to pilfer a bagel without co-workers noticing.

What would happen if someone tried to sell Rolex watches using an honor system like the bagel seller's?

[Michael H.]

Hi Half Sigma
I think that it is terrible that people where I work with their high income (and top secret security clearances) would cheat a guy out of a buck. I do believe that the incentives for cheating are very different in a large office. Click on my name and you can read about it.