Reporting from Workstation 9

I was reading a post on GM's blog on whether Levitt was ruining economics, and felt glad that there were many out there who share this concern, including M himself.
I have not been a big fan of empirical work. Primarily because it's not trustable.

Essentially, you already have an idea of what result you want to achieve. You then carefully work at it. Econometrics is a dicey subject. As far as theoretical econometrics goes, I find it far too challenging for someone with my poor math grounding.

Hence when I was choosing my thesis topic, I could go for a tough topic. I was working on a Dynamic Stochastic General Equilibrium Model, for Canada. DSGE models are used in most of the first world Central Banks. But beyond computer simulations, come it matrices of magnitude 30 by 30, convergence conditions etc, etc. At the point when I realized that the amount of math was slowly diluting any economic intuition I might gain, I thought maybe I must research a topic that I can understand and contribute to.

Peter Phillips told us in class, that research can be done two ways. You could sit in your room, with your legs on top of your desks and think of other ways of looking at a concept:-think out of the box, or like Mr Levitt, you can sit at your desk and run regressions day and night. It's a choice we had to make.

But M says that eventually papers with ideas get cited more often and hence, there will always be an incentive in the intellectual community to gain the respect of other academics and not popular fame. Hence all is not bad. Mr Levitt has indeed gotten more people interested in applied economics, however the beauty of economics goes far beyond regressions..

Labels: Economics