Arnold Kling on Model Building and Graduate School

Arnold Kling has two posts highly worth reading.

The first is commenting on a revised book by David Colander on Graduate Study in Economics. (See Colander's book on Amazon.com. I saw it in GMU's bookstore yesterday and will have to pick-up a copy.) Kling states that if he could change graduate education in econ, he would do the following:

1. Require that students take a course for one quarter in "How to teach economics." This would include examples from popular books, classic economic fallacies, and so on.
2. Require that students take at least two semesters in an applied field. This might be health economics, environmental economics, development economics, or the economics of education, for example.
3. In tenure decisions, put less emphasis on published journal articles and more emphasis on books as well as on shorter, unpublished work that presents interesting findings, raises important questions about published work, or simply points people toward interesting work. By no means should the weight on journal articles drop toward zero. But I want more Tyler Cowens and Bernie Saffrans. And more David Colanders.

The secon is a post in which Kling has reacts to an article by Robert Solow on Model Building. (Read Solow's article here.)

In the late 1980's, the controversies in macroeconomics more or less ended in a stalemate, because each side had learned how to make its model fit the data. Models with totally opposite microfoundations and policy implications could be shown nonetheless to be "observationally equivalent."

This highlights a particular danger in model building for many areas of study, ranging from economics and climate change.

Read both posts. They contain far more than what I've highlighted here.