Russ Roberts on Scientific Modesty

Tyler Cowen recently interviewed economist Russ Roberts. My favorite parts were Roberts' expressions of intellectual modesty. The first came after Cowen asked him what changes he would make if he were czar of American science:

ROBERTS: Well, I’ll cheat for a minute and say what I’ll change is for people to be more willing to accept the possibility that they’re wrong, which is not a policy lever that we actually have control of. I’m a big fan of Richard Feynman’s quote: “The first thing is not to fool yourself, and you are the easiest person to fool.”

We romanticize science a great deal, and I think we romanticize the objectivity of academics and scientific researchers. They have their own pet theories, their own fads, their own reputations. It’s very hard to do science. It’s very hard to do good science.

And in response to a question about how his views have changed over the past twenty years:

I’ve become disenchanted with economics in general in the following sense: We have this idea, which is a very strange idea when you go deeply into it, which we teach our students, something called utility theory.

We developed utility theory in economics to explain why people buy what they buy. We’re trying to generate demand curves and explain why, when a price of some good goes up, you buy more of a different type of good. We call that a substitute. There might be you buy less of it. We call that a complement.

We were very focused on what I could call commercial behavior — what people do with their money. Then somehow, we made a bizarre leap from that narrow focus to arguing that we have something to say about people’s well-being. You think about how strange that is.

Now, if I said to you, “Does what people buy contribute to their well-being?” Of course it does. We want to buy things that add more to our well-being than things that add less. That’s reasonable. Most people would agree with that. Would you then jump to the conclusion that what people consume determines how happy they are? Now, that would be ludicrous. Adam Smith understood in 1759 that that isn’t the case.

If you asked economists, “Is that true?” “Well, of course not. No, when I mean utility, I mean everything. I mean the nonmonetary aspects of a job, for example, and the nonmonetary aspects of the steak you cook at night for the romantic dinner. It’s the romance that’s more important than the consumption of the steak, of course. We all know that.”

Yet, somehow, we’ve become the arbiters of how policy translates into well-being. I find that really deeply disturbing.