Friday, September 7, 2007

Incentive pay for professors?

When Adam Smith lectured in Scotland, he collected admission from those who wanted to hear him. This kind of market pressure ensured that he would deliver content that students were willing to pay for.

These days, we rely on different mechanisms to align the incentives of faculty with the goals of students. For example, the principal-agent conflict at Dartmouth between the Board of Directors and the faculty can be viewed as an effort to control the "normal" incentive conflict between producers (faculty) and consumers (students). However, this control mechanism may not be working as well as some students would like (Do business Schools practice what they preach?)

James Miller, an economist at Smith College, has a modest proposal to better align the incentives of faculty with the goals of students.

... [give] each graduating senior $1,000 to distribute among their faculty. Colleges should have graduates use a computer program to distribute their allocations anonymously.

My proposal would have multiple benefits. It would reduce the tension between tenure and merit pay. Tenure is supposed to insulate professors from retaliation for expressing unpopular views in their scholarship. Many colleges, however, believe that tenured professors don’t have sufficient incentives to work hard, so colleges implement a merit pay system to reward excellence. ... And because the proposal imposes almost no additional administrative costs on anyone, many deans and department heads might prefer it to a traditional merit pay system.

The benefit of this proposal is similar to the benefit of vouchers. Consumer choice disciplines the producers, as vouchers can be used at any producer.

Thanks to colleague Bruce Barry for the heads up.

3 comments:

  1. Won't the majority of students just give the money to the Professor who is the easiest grader? I think you're too heavily assuming the money would go to the "best" Professor, and that students will allocate funds based on where they learned the most.

    I'd assume money will be allocated to the easiest grader/s, and then to the "coolest" Professor (most entertaining, most fun, etc).

    ReplyDelete
  2. Enthusiasm for this approach reflects a misunderstading of the nature of teaching. Student happiness should not be the primary metric for teaching quality - not because their happiness doesn't matter, but because they are not in a position to be qualified to judge teaching quality, especially right after the course. And as always, the "customer" metaphor is flawed. It is fine for a university to treat students as "customers" in related to the administrative services provided to them (food services, placement services, and so forth) but that is not the way to imagine the teacher-student relationship when the goal if intellectual advancement.

    These philosophical matters aside, it is silly to give a graduating student money to spread around to professors because they will not just reward (as the other commenter suggests) the professors who are easiest or most entertaining; many also are likely to be influenced by a recency effect, rewarding good professors they have had more recently over those they had earlier in their time at the university.

    ReplyDelete
  3. I think you would be surprised by that empirical evidence likely does not back up the"easiest" comment. While I have not seen a study on this, my experience suggests the easy professors seldom win teaching awards voted on by the students. They, instead goto the most dynamic teachers...

    Also, no one is advocating making ALL the compensation variable in this way, just some of it. Why not make a small component of someone's salary contingent on the perception of customers?

    ReplyDelete