Henderson's First Law of Econometrics
By David Henderson
When I was a colleague of Richard Thaler at the University of Rochester in the late 1970s, I learned from him Thaler’s First Law of Dieting. The law: Don’t eat beyond the point where it hurts. Somehow, for someone who won the Nobel prize for his insights about people’s limitations in acting rationally, that law seems particularly a propos. I think I asked him what his second law is and that he answered that he didn’t have one.
Similarly I have Henderson First Law of Econometrics, but I don’t have a second. It’s this:
When you read an econometric study done after 2005, the probability that the researcher has failed to take account of an objection that a non-economist will think of is close to zero.
In the last few years, I’ve gone to a number of presentations by empirical economists at the Naval Postgraduate School, usually by labor economists. The economists presenting are either my (now former) junior colleagues or visiting economists. One reason the presentations take so long is that the presenters carefully lay out all the ways that they have covered or corrected for various factors that a casual listener or even a fairly careful listener would think of.
As a result, my opinion of the quality of empirical work has increased substantially.
Why did I choose the year 2005? It’s a little arbitrary, but I seem to have noticed an increase in care in the last 10 years or so.
One thing that made me think of this was commenter Justin’s excellent response to commenter Alan Goldhammer on this recent post of mine.