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

By: Peter Schwartz

Peter Schwartz — Fri, 04 May 2007 03:24:19 +0000

Regarding Kate Litvak’s comment, why do we need to assume we can find, or must have, patterns? Why is it so important to generalize, particularly if it may be true that efforts to generalize are not generally possible, perhaps because human nature is not irreducible? Maybe it is in the ungeneralizable particulars that the truth is to be found? Is Levitt’s study of abortion rates and crime rates simply cute, or is it also elegant and clever? Admittedly, this is the comment of someone who finds more truth in stories than in data (even though my day job exposes me to an unending river of data).

By: Kate Litvak

Kate Litvak — Sat, 28 Apr 2007 04:40:43 +0000

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.

And why do we think that the data on bicycle messengers in Switzerland tell us anything useful about lawyers in New York or drill sergeants in Iraq or call center employees in Bangalore? Because of the irreducible human nature? Because we assume so? Granted, if we had lots and lots of papers in different representative settings, with similar results, we may have informally concluded that there is a pattern. But the problem is that these studies are rare (per issue) and not representative. And they cannot be made systematic or representative because the settings are selected for the sole reason of involving a cute instrument (the availability of which may or may not be correlated with the factors that we study). The generalizability of those findings is highly questionable.

By: TRUTH ON THE MARKET » How Steve Levitt Might Save Economics

TRUTH ON THE MARKET » How Steve Levitt Might Save Economics — Thu, 26 Apr 2007 16:32:25 +0000

[...] Geoffrey Manne on Angrist on Levitt, Instrumental Variables, Identification, and Defending Cuteness.M. Hodak on Angrist on Levitt, Instrumental Variables, Identification, and Defending Cuteness.Puzzles and Problems Redux « Organizations and Markets on Angrist on Levitt, Instrumental Variables, Identification, and Defending Cuteness.Michael Guttentag on Congratulations to Kate Litvak!.The Harvard Law School Corporate Governance Blog » An Inconvenient Location for the Annual Shareholders’ Meeting on Halliburton Annual Shareholders’ Meeting Location.M. Hodak on North Dakota Publicly Traded Corporation Act.Jerry Kohl on A New Defense of the Per Se Prohibition Against RPM?.D. Daniel Sokol on A New Defense of the Per Se Prohibition Against RPM?.TRUTH ON THE MARKET » A New Defense of the Per Se Prohibition Against RPM? on A Response to Commissioner Harbour’s “Open Letter” on Leegin.Heidi Feinstein on Rent a CFO?. [...]

By: Geoffrey Manne

Geoffrey Manne — Thu, 26 Apr 2007 15:58:05 +0000

I would echo Marc’s comment. “Cute-o-nomics” is the real and sophisticated (just because the ignorant call it “cute” doesn’t mean it isn’t rigorous and complex) application of a mode of inquiry, of a system of analysis, to real life problems. These are problems the solution to which, one hopes, will illuminate the human condition.Here’s why I think a lot of detractors actually detract: The underlying principle of this work is that “people respond to incentives” and they do so in fundamentally similar ways. It is the application of basic, classical intuitions to the sort of problems many like to point to as transformative exceptions to neoclassical analysis. Frank Pasquale loves to construct unreal hypotheticals to demonstrate a set of facts where traditional economic intuition yields the “wrong” result. But the more these sorts of quirky situations are shown to be “solved” by classical economic intuition creatively applied, the less force there is in the arguments that the entire edifice should be scrapped.Not entirely unrelated is the fact that this mode of analysis is real economics in the purest sense. The folks like Levitt who do this work, who can perceive problems and think through creative solutions to them, are applying real economic intutition. They may use a lot of math, to be sure, but the underlying logic is generally quite simple (not as in “easy” but as in “not-complicated”) and complex mathematics is not necessary to explain or to understand what’s going on. It is economics in the style of Tullock and Alchian and Coase. These are real lessons learned through application of a powerful system of analysis. In contrast, too much of economics today seems to be little more than mathematic gymnastics. It is largely devoid of real fundamental understanding of human behavior and of the analytic power of a few simple rules to explain it. The debates don’t turn on seemingly fundamental questions like, “how well does this explain observed behavior?” Rather the debates are about the elegance of complex models and the proper use of this variable or that equation. Important endeavors, to be sure, but hardly deserving of more accolades than the work of incisive “natural” economists.

By: M. Hodak

M. Hodak — Thu, 26 Apr 2007 14:32:36 +0000

I agree that the critique of clean ID is misplaced. I would go a step further and say it’s way off base. The idea that economics would be better off sticking to “big problems” of inflation, unemployment, poverty, economic growth, regulation, political economy, etc. is actually a reflection of the hubris of academics who really want to be “big-issue” policy-makers.

Here’s a check their egos: Everyone can’t come up with the Theory of Relativity. Someone needs to actually measure the warping of a specific wavelength of light in the gravitational field of a specific star.

IMHO, one of the reasons economics is a “dismal science” is that the big questions have been so controversial, and their big answers have been so unconvincing. They’re unconvincing because they are trying to attack entrenched opinion with messy data and inherently weak controls. How many times have we heard that, “Sure, minimum wages may, in theory, reduce overall employment, but this study in New Jersey showed it didn’t?”

Until we develop a hundred “cute” studies showing that that mainstream economics works in ‘this’ and ‘this’ and ‘this’ and ‘this’ and ‘this’ situation, even if it doesn’t work as well in ‘that’ one, we will be hearing about ‘that’ in New York Times editorials arguing for more enlightened intervention.

I think Levitt and his clones are doing God’s work.

By: Puzzles and Problems Redux « Organizations and Markets

Puzzles and Problems Redux « Organizations and Markets — Thu, 26 Apr 2007 12:50:13 +0000

[...] from Levitt (very unhappy), Joshua Angrist (also unhappy, though not for personal reasons), Josh Wright (mildly peeved), and Greg Mankiw (neutral). Sadly, no one has picked up my (very clever) reference [...]