You might think this passage would be from Econ Journal Watch‘s “Inspecting the Mechanism” column, but you’d be wrong:
A recent experience says it all. I was asked to discuss three papers presented at a session of the American Economic Association meetings. Each of the papers was presented by a very prominent economist and each was regressing the interest rate on the deficit using different adjustments to the deficit and different time periods. Two of the economists were trying to show that deficits raised interest rates and one was trying to show the opposite.
In my comments, I pointed out that the exercise was nonsensical—that one could produce any desired effect of “the” deficit on interest rates by simply labeling receipts and payments to generate a deficit time series with the “right” (desired) statistical relationship. After the session concluded, each of the presenters came up to me to tell me they agreed entirely with me but that they were running their regressions in order to counteract regressions showing the opposite effect. In effect they were saying, “I’m doing this because he’s doing this.”
This isn’t economics. This is sophistry and a sorry commentary on our profession. Moreover, it is incredibly dangerous at a time when the true deficit, the growth in the fiscal gap, is monumental. If established academic economists won’t focus on fundamentals, who will? My hope lies with new economic Ph.D.s. They are beholden to no one, not even to their past selves, and are free to join the child—the hero of The Emperor’s New Clothes—in shouting “He’s naked!”
The actual source: Laurence Kotlikoff in the latest issue of the Economist’s Voice.
Dan Klein may think he’s just a voice crying in the wilderness, but I’m starting to think that his candor is making a difference.
READER COMMENTS
Michael F. Martin
Jun 10 2008 at 12:50pm
The problem is that economists gave up trying to model the dynamics of markets in the 1930s and 1940s. They switched to modeling markets as wholly decoupled, with exogenous shocks driving market dynamics.
Any realistic treatment of markets should take account of the coupling between different supply and demand cycles. In the defense of economists in the 1930s and 1940s, physicists hadn’t even developed the mathematical techniques to build such models until the 1950s. (This is what Feynman, Schwinger, and Tomonaga won a Nobel prize for in the 1950s.)
But we have it now! Let’s go back to doing some Fourier analysis of markets, and dig into the mechanisms whereby supply and demand couple.
There are so many useful insights that regulators could get from even simple models of dynamics.
Scott Wood
Jun 10 2008 at 12:52pm
Surely the best way (maybe the only truly intellectually honest way) to approach this sort of thing is to make your regression and then do whatever you can to make it come out the other way. The paper can then report what you had to do to pull that off.
It’s been ages since I’ve read any real economics (and even then it wasn’t applied), so I don’t know the extent to which anyone does that.
TGGP
Jun 10 2008 at 8:09pm
When is Dan Klein going to get a blog of his own? I’d actually be more interested in seeing Jeff Friedman get a blog, but beggars can’t be choosers.
Comments are closed.