Friday, May 11, 2007

Free Lunch Thinking at Harvard

"There is no such thing as a free lunch" is as good a slogan as any I know. Whenever I hear someone propose a reform and suggest that it does not involve any costs, my first reaction is to think that they aren't being honest. Life is full of hard tradeoffs. If you think otherwise, you probably haven't thought enough.

I see some students being lulled into free lunch thinking when discussing raising wages of low-income workers at Harvard. Even putting aside any possible adverse side effects of above-equilibrium wages, advocates of higher wages need to confront the real question of limited resources. If Harvard is to raise wages above the going market levels, the money to pay those wages has to come from somewhere, such as higher tuition, less financial aid, or fewer faculty. Saying "Harvard has a large endowment" is not an answer--the endowment represents future spending on items the university values.

Harvard administrators are lulled into free lunch thinking when discussing teaching quality. An article in yesterday's NY Times offers this insight into the issue:

The aim of the report is not to de-emphasize research in any way, but to bring about a greater institutional focus on teaching, Professor Ulrich said. “This is not a report that says we’re going to hire teachers who are not also scholars,” she said. “We want both.”
But better teaching is not a free lunch. If the university puts more weight on teaching in its hiring and promotion decisions, it must put less weight on research. Sure, there are some scholars who are great at both activities, but those are not the marginal hires.

You will know the university is serious about increasing teaching quality when it says it is willing to hire scholars with less distinguished research records in order to get better teachers, or when it says it will spend more money to compete more vigorously with Princeton and Yale for the best professors (and tells you where that money will come from). But unless the university says it is willing to confront the hard tradeoffs, it is just posing.


Update: Some commenters point out that sometimes the world does offer Pareto improvements, such as the gains from trade in a Ricardian trade model, and that these are free lunches. True enough. Slogans are more rules of thumb than theorems. Experience suggests, however, that purported free lunches are vastly more numerous than actual free lunches.

I should note that purported free lunches are found on both sides of the political divide. Politicians on the right like to claim that their proposed tax cuts will completely pay for themselves with extra economic growth. Politicians on the left like to claim that their proposed spending programs are investments that will yield returns so great that the spending becomes self-financing. Either outcome is possible in theory, but skepticism is usually the best response in practice.