Economist's Notebook: Specialization vs. Generalization and the Research University Model

This little article in The New York Times on some research by David Figlio and others caught my eye: it essentially claims that non-tenured instructors do a better job teaching introductory-level university courses than do tenured and tenure-track faculty.  There could be many reasons for this and as an economist I am obliged to talk about incentives first.  Instructors have to be more worried about performance in the classroom than to tenured faculty - although I should throw in a note here about the increasing use of merit-pay and promotion to improve the classroom performance of tenured faculty.

Other reasons could be that research takes time and energy away from a professors classroom performance, or that instructors do not teach as difficult a class, who knows?  What interests me is that the mostly likely reason in my mind is due to specialization: instructors that focus on intro classes become extremely good at it thanks to the time and energy they can devote.  They also have time to experiment with new classroom techniques and teach these classes more frequently so there is no depreciation in their skills.

This, of course, points out that the traditional role of a professor at a research university is not how an economist would necessarily design things.  The principle of comparative advantage suggests that the efficient distribution of tasks would have those with a relative advantage in research do mostly research, those with a relative advantage in teaching do more teaching and so on.  The ideal of the tenured research faculty is one who devotes about half their time to research and half their tie to teaching - the very thing David Ricardo suggested was a bad idea!

But this makes two big assumptions: one, that the goal is efficiency; and two, that there are no spillovers - that having to do research does not help teaching and vice versa.  Even if you accept the efficiency goal, I believe strongly that in some classes this latter statement is false.  Perhaps not as much in intro classes though, which would help explain the Figlio, et. al., result.  But I for one have become a better economist from having to teach and continue to think deeply about very basic economic principles and I think my research experience definitely makes me a better teacher of classes that overlap my research areas (and in a big research university the ideal is that professors teach within their research areas) but also in the more intro and intermediate level classes as well.

Despite this, it does make me wonder whether the new model, that almost all universities seem to have gone to, where each department includes a team of instructors working along side the research faculty, isn't perhaps a better model.   Despite it being the subject of much derision, perhaps it has been better all along?