Sunday, August 2, 2009

Ed Glaeser on education

This piece on teachers by Ed Glaeser has been sitting in my queue for a while. I think it falls short of Ed's usual standard on a number of dimensions. First, it is easy to say we should have better teachers and much harder to actually have them for two reasons. First, the literature makes pretty clear that we do not (yet?) know what observable characteristics make for a good teacher. Thus, at the moment, there is no clear way to change the hiring process to bring in more individuals who will be good teachers. Ed is more coy about this fact than he should be. Second, it takes some time to sort out who is a good teacher and who is not through classroom observation. By the time it is sorted out, teachers in most public schools will have tenure and essentially be impossible to fire.

This line of argument suggests that what is needed is a change in the industrial organization of schools, away from a predominately government / union near monopoly model toward a model that includes a more flexible labor market. Oddly, Ed does not mention this but instead suggests throwing good money after bad by putting more money into the current system. That makes little sense given that in the current system teacher salaries are essentially a constant plus a fixed amount times number of years in service plus another constant for having a master's degree. In such a system one cannot raise the salaries of the good teachers without similarly raising the salaries of all the mediocre ones.

Finally, I have always thought that (former Chicago economics professor) Sherwin Rosen's model of hierarchies suggested that you do not really want the "best and the brightest" in teaching. Teachers influence a small number of people at a time a little bit. That is important, but not important enough that, from a social standpoint, you want to take someone who would otherwise develop some great new product, or start a new firm, away from that and put them into teaching.