Tuesday, August 26, 2008

Conference Presentations By the Numbers

Once you've gone to a half dozen academic presentations, you begin to notece the same comments appearing over and over. Along those lines, here's a classic piece from George Stigler's in the 1977 Journal of Political Economy — “The Conference Handbook.” Stigler suggested that econ seminar participants could speed things up by making objections to the speaker's presentation using the following numbered list:
  1. Adam Smith said that.
  2. Unfortunately, there is an identification problem which is not dealt with adequately in the paper.
  3. The residuals are clearly non-normal, and the specification of the model is incorrect.
  4. Theorizing is not fruitful at this stage; we need a series of case studies.
  5. Case studies are a clue, but no real progress can be made until a model of the process is constructed.
  6. The second-best consideration would, of course, vitiate the argument.
  7. That is an index number problem (obs., except in Cambridge).
  8. Have you tried two-stage least squares?
  9. The conclusions change if you introduce uncertainty.
  10. You didn’t use probit analysis?
  11. I proved the main results in a paper published years ago.
  12. The analysis is marred by a failure to distinguish transitory and permanent components.
  13. The market cannot, of course, deal satisfactorily with that externality.
  14. But what if transaction costs are not zero?
  15. That follows from the Coase Theorem.
  16. Of course, if you allow for the investment in human capital, the entire picture changes.
  17. Of course, the demand function is quite inelastic.
  18. Of course, the supply function is highly inelastic.
  19. The author uses a sledgehammer to crack a peanut.
  20. What empirical finding would contradict your theory?
  21. The central argument is not only a tautology, it is false.
  22. What happens when you extend the analysis to the later (or earlier) period?
  23. The motivation of the agents in this theory is so narrowly egotistic that it cannot possibly explain the behavior of real people.
  24. The flabby economic actor in this impressionistic model should be replaced by the utility-maximizing individual.
  25. Did you have any trouble in inverting the singular matrix?
  26. It is unfortunate that the wrong choice was made between M1 and M2.
  27. That is alright in theory, but it doesn’t work out in practice (use sparingly).
  28. The speaker apparently believes that there is still one free lunch.
  29. The problem cannot be dealt with by partial equilibrium methods; it requires a general equilibrium formulation.
  30. The paper is rigidly confined by the paradigm of neoclassical economics, so large parts of urgent reality are outside its comprehension.
  31. The conclusion rests on the assumption of fixed tastes, but (of course) tastes have surely changed.
  32. The trouble with the present situation is that the property rights have not been fully assigned.
HT: The Freakonomics blog.

Some of the commenters on the piece added more options. Here are some of the better ones:
  1. How did you handle endogeneity problem?” (Note: this almost always works well at finance conferences, particularly for corporate finance pieces)
  2. Your standard errors are too small because you failed to cluster (or clustered at the improper level).
  3. At Fed banks, certainly one of the items for this list would be, “How is this of any relevance to monetary policy?”
  4. The results are driven by unobserved heterogeneity.
  5. Did you try using a Difference-in-Difference technique? Did you try using a non-parametric estimation?
  6. Experiments conducted by Kahnmen and Tversky have clearly demonstrated that people do NOT choose rationally under those conditions.
  7. Is there a weak instruments problem?
  8. But what if the actors aren’t rational?
  9. Isn’t this just Modigliani-Miller?
  10. How is your model identified?
  11. Have you included fixed effects?
  12. That’s ok in practice, but it won’t work in theory.
  13. This theory is only valid in the static case and won’t work in the dynamic one.
  14. Why do we care about this? or “Why is this question important anyway?"
  15. That’s true, but not very interesting
  16. That’s not a large effect you’ve found, it’s a small effect.
  17. That’s not a small effect you’ve dismissed, it’s a large effect.
  18. Did you try first-differencing?
  19. What happens if you estimate it by GMM?
  20. You just ran a bunch of regressions. What have we learned from your analysis?
  21. Here's a popular one, in the “I did this back in..” vein:“I was troubled that you didn’t cite my work in the field.”
  22. Your empirical results are obviously biased by a troubling sample selection issue.
  23. But what if we view this as a 2-stage game?.
  24. ‘In an efficient market, that type of arbitrage isn’t possible’.
  25. I believe your correlation is a spurious one unless you convince me you checked for co-integration.
If you can think of others to add to the list, feel free to put them in the comments section.