Angrist on Levitt, Instrumental Variables, Identification, and Defending Cuteness

Cite this Article
Joshua D. Wright, Angrist on Levitt, Instrumental Variables, Identification, and Defending Cuteness, Truth on the Market (April 25, 2007),

It is apparently vogue to ask whether Steve Levitt is ruining economics? The serious question behind all of this is whether today’s economists have become too enamored with cute and clever questions rather than the day’s “big” questions.  I’ve already gone on the record on this one with a comment to a post over at Co-Op by Frank Pasquale:

Many people conducting “cute” research are quite aware of the details of legal and other institutions and in fact often rely on these details to construct instruments for their econometric analysis.  I am sympathetic to the view that the economics profession as a whole might be better off with greater attention paid to larger problems. But for what it is worth, do not agree with the claim that Freakonomics is in the least bit responsible for this trend, and disagree strongly with the proposition that Levitt or Freakonomics has made economics worse off.

Greg Mankiw offers a bit of a luke-warm answer to the question, noting that it is “disconcerting, to a degree” that “young economists today are doing Levitt-style economics and fewer are studying the classic questions of economic policy,” but concludes that he is not too worried about it because economics will correct itself because of diminishing returns to call research programs and strong incentives to address the “big ones.”

I prefer Joshua Angrist’s less apologetic response to Scheiber’s attack defending instrumental variables, natural experiments, and even cuteness, in the TNR (HT: Marginal Revolution):

I’ll plead guilty, however, to being especially pleased when students manage to come up with clean identification–that is, they have a convincing strategy for uncovering causal effects.  Clean identification is not a fetish; without it, little of value is learned.  On this score, our students typically do better than the empiricists of H.G. Lewis’s generation … .
Second, in his rush to tar some up-and-comers with the “cute-o-nomics” brush, Scheiber misses a central feature of the clean-identification research agenda, best explained by example.  One of the enduring scientific and policy questions in Labor economics is the sensitivity of hours worked to changes in pay (this matters for tax policy, for example).  The best evidence labor economists have on the relation between wages and hours worked comes from a small experiment (by Ernst Fehr and Lorenz Goette) involving the wages of bicycle messengers in Switzerland.  The second best comes from a study of stadium vendors by Gerald Oettinger.  Who cares about the riders of Veloblitz or snack sellers at Camden Yards?  We care because economics is predicated on the notion that a few simple principles explain behavior in many settings.  These studies produce results that are convincing and may well be general, though, as always in science, it will take replication to know for sure.

I like that answer alot.  While I’ve written here before about the potential harms of fetishizing formal mathematical theory rather, especially with respect to law and economics, I see that as a separate problem altogether.  Economists need not apologize for uncovering clean causal relationships  in less complex settings.   There is no shame in adding to our theoretical and empirical understanding of small subjects with the hopes of generalizing to larger phenomena.

For that matter, though I’m a little late to the party, academic economists dont need to be defensive about playing parlor games from time to time either (If the bar for academic work is immediate policy relevance, how much law review scholarship do you think would qualify?).