Thursday, October 13, 2011

Testing Theory

Matt Yglesias:

Last week I was outside my office and I saw a $5 bill on the ground. Famously, economists say you never see a $5 bill on the ground because someone would pick it up. But instead of picking it up, I stood around watching to see if anyone else would. A bunch of people walked by not noticing it. Then one guy saw it, saw me, and asked if it was mine. I said no it wasn’t, I was just curious what would happen. He laughed and made a joke about economists. Then a second guy came by, picked it up, and said I’d dropped five dollars. I said no, actually it was there before me. He looked around, noticed a homeless guy across the street, said “I think he needs it more than me,” walked over and gave it to him.

While a single test is not proof of anything (expect the strongest of the possible theories), I have certainly had people tell me that I have dropped money. That seems to go against the assumption that rational people will always act to increase their wealth (after all, they could just pick up the bill).

I think that it is worth keeping in mind that the assumptions of economic models are just that. Models can be useful but are rarely ever completely correct.

And Matt's stuff in the same post on the Night Watchman state is simply not to be missed.


  1. This may be a fine example of the Hawthorne effect. Did the fact that he was being watched influence the altruistic impulse? Hard to tell from this experiment.

  2. I completely agree that the experiment is not proof of anything. But the idea that the events reported do not seem grossly unrealistic should encourage us to consider more complex models of human behavior. Even if donating the $5 was a status play, it makes economic rationality a more complex construct.