Measurement Errors and the Convergence Hypothesis

Abstract

Specialisation has benefits, but these benefits can be more than fully offset by coordination failures. I think we economists have carried intellectual specialisation too far. Some of us think about how economic systems ought to work, some of us collect data, some analyse data and others think about how we should analyse data. We have honed our skills at these separate tasks to razor-sharpness, but there is very little overall coordination of these disparate specialities. It is as if to build our intellectual edifice, we hired the very best carpenter, the very best plumber and the very best electrician, but did not allow any coordination. What kind of house would be built?

Support from NSF grant SES 8910950 and comments of Dalia Marin are gratefully acknowledged. Able computational assistance was provided by Rodrigo Fuentes and Josef Perktold.