On Tenure

Steven Levitt (author of Freakonomics) had an interesting post on his blog yesterday regarding tenure. He essentially calls for the entire concept of tenure to be abolished.

For those unaware (I'm sorry if I'm insulting anyone's intelligence but I want to make sure everyone understands what the hell we're talking about), academics are up for tenure review between three and five years after they start teaching at a particular institution. If you get tenure, it means you can never be fired from your job for performance-related reasons (sexual assault will still get you fired, if not thrown in jail). If you do not, you're essentially fired on the spot. So say I finish my PhD in 2011 (here's hoping) and get a assistant professorship at Random State University. Around 2015, based on my productivity/brilliance/teaching/hygeine, the tenure review board at RSU will decide my future by either granting me tenure, in which case I'm set for life at RSU (I only leave if I want to) or by not granting me tenure, in which case I'm out of a job (there's no grace period - as soon as you're turned down for tenure, you're done. This is why many people start looking for another job a year before they're up for tenure). The historical reason for the concept of tenure which, admittedly some find quite strange ("How can you not be fired? What is this, France?") is guaranteeing academic freedom. In other words, professors are given tenure so that they never have to fear that anything they write or research will cause them to get fired, thus ensuring that an academic will do the work he/she considers most valuable/interesting, not the work considered most safe/uncontroversial. That's the idea of tenure in a nutshell.

I'm reproducing Levitt's post in full because he raises some issues that are worthy of discussion.

If there was ever a time when it made sense for economics professors to be given tenure, that time has surely passed. The same is likely true of other university disciplines, and probably even more true for high-school and elementary school teachers.

What does tenure do? It distorts people’s effort so that they face strong incentives early in their career (and presumably work very hard early on as a consequence) and very weak incentives forever after (and presumably work much less hard on average as a consequence).

One could imagine some models in which this incentive structure makes sense. For instance, if one needs to learn a lot of information to become competent, but once one has the knowledge it does not fade and effort is not very important. That model may be a good description of learning to ride a bike, but it is a terrible model of academics.

From a social standpoint, it seems like a bad idea to make incentives so weak after tenure. Schools get stuck with employees who are doing nothing (at least not doing what they are presumably being paid to do). It also is probably a bad idea to give such strong incentives pre-tenure — even without tenure young faculty have lots of reasons to work hard to build a good career.

The idea that tenure protects scholars who are doing politically unpopular work strikes me as ludicrous. While I can imagine a situation where this issue might rarely arise, I am hard pressed to think of actual cases where it has been relevant. Tenure does an outstanding job of protecting scholars who do no work or terrible work, but is there anything in economics which is high quality but so controversial it would lead to a scholar being fired? Anyway, that is what markets are for. If one institution fires an academic primarily because they don’t like his or her politics or approach, there will be other schools happy to make the hire. There are, for instance, cases in recent years in economics where scholars have made up data, embezzled funds, etc. but still have found good jobs afterwards.

One hidden benefit of tenure is that it works as a commitment device to get departments to fire mediocre people. The cost of not firing at a tenure review is higher with tenure in place than it is without it. If it is painful to fire people, without tenure the path of least resistance may be to always say you will fire the person the next year, but never do it.

Imagine a setting where you care about performance (e.g. a professional football team, or a currency trader). You wouldn’t think of granting tenure. So why do it in academics?

The best case scenario would be if all schools could coordinate on dumping tenure simultaneously. Maybe departments would give the deadwood a year or two to prove they deserved a slot before firing them. Some non-producers would leave or be fired. The rest of the tenure-age economists would start working harder. My guess is that salaries and job mobility would not change that much.

Absent all schools moving together to get rid of tenure, what if one school chose to unilaterally revoke tenure. It seems to me that it might work out just fine for that school. It would have to pay the faculty a little extra to stay in a department without an insurance policy in the form of tenure. Importantly, though, the value of tenure is inversely related to how good you are. If you are way over the bar, you face almost no risk if tenure is abolished. So the really good people would require very small salary increases to compensate for no tenure, whereas the really bad, unproductive economists would need a much bigger subsidy to remain in a department with tenure gone. This works out fantastically well for the university because all the bad people end up leaving, the good people stay, and other good people from different institutions want to come to take advantage of the salary increase at the tenure-less school. If the U of C told me that they were going to revoke my tenure, but add $15,000 to my salary, I would be happy to take that trade. I’m sure many others would as well. By dumping one unproductive, previously tenured faculty member, the University could compensate ten others with the savings.

It must not be that simple because few schools have tried, and my sense is that those that took a stab at it capitulated quickly and reinstated tenure. What am I missing?

For starters, I would say that Levitt is right about two things, both to do with the question of incentives. First, he's completely right in his point that making incentives so strong in the beginning of an academic's career is almost superfluous. When someone gets hired right out of grad school, they're eager to impress anyway; they don't need the pressure of tenure to make them productive. Think about it: in any other job, when you first get hired, aren't you most cognizant of the need to show people you're worthy in the first few years of the job? Stop nodding, it was a rhetorical question.

Second, he's completely right that tenure does make some professors lazy after they get it. This is a fact and everyone knows it. There's really nothing else to add here.

Levitt, however, jumps the gun when he says tenure should be abolished wholesale. I say this for a few reasons.

First, Levitt, as I'm sure you know, is an economics professor. If he can't imagine research "in economics which is high quality but so controversial it would lead to a scholar being fired", it's beacuase he's talking about economics. He would not, I guarantee you, be saying this if he were a political scientist. I'm not making any normative claims here on why political science is cooler than economics (which, incidentally, it is); all I'm saying is that there is a greater likelihood that a research program in political science is controversial than one in economics being so. I can't imagine any reasonable person quibbling with that assertion, so let's move on.

Second, the controversiality of research is not the only reason scholars might be persuaded to follow different research agendas if tenure was abolished. It is also a question of time. If tenure was abolished, academics would always feel the whole "publish or perish" pressure. If you want to publish fairly regularly in order to preclude the possibility of getting fired, you will never pursue research that takes a long, long time. Some projects can take up to ten years. Some of these projects are incredibly valuable to both the academy and the general public. We risk losing these works if tenure is abolished because scholars will be too scared to ever contemplate doing research that takes such a long time; they'd rather do short and sweet research that gets them in a journal every four years.

Third, I think Levitt ignores just what a world without tenure would look like. As it stands right now, old professors choose whether or not a young professor is going to be allowed to carry on at a particular university. Now, if tenure is abolished, can you guess what will happen? I can. Old professors will never hire brilliant young professors because they'll be afraid that said young professor will be so good that they (the old professors) won't be needed. Only when a young professor is not a threat to the old professors will they make the best choice for the university. Hiring committees should not, for the sake of the university, ever be motivated by anything close to fear; if they are, everyone loses. And abolishing tenure would do just that: bring fear into the calculus.

Fourth, tenure compensates for cash. Academia is one of the lowest paying professions when you compare salary-to-years-in-school ratios. Consider the fact that I will have at least five, if not six, more years of education than my friends on Wall Street (well, friend). I would have read the equivalent of about 1000 more books than him by the time we're 28. I would have written a friggin' dissertation. That's a whole lot of education. What are the rewards? Well, financially, they're not great. Most assistant professors (the post you get straight out of grad school) will make between $50,000 and $80,000, depending on (a) what university they're teaching at; (b) the city the university is in; (c) how good their dissertation was; (d) which other universities, if any, have given him/her an offer; and (e) which university they got their PhD from. An associate professor (up one step on the ladder) makes between 65k and 95k. Professors (top rung) make between 75k and 110k. These are very general numbers, some are lower and some are higher. The point to be made, however, is that for all the years slogging through school, academics aren't particularly well paid, especially if you compare them to lesser educated people (sorry Nikhil) who make substantially more. Tenure (and summers off, no doubt) is a way of evening the score. The bargain is: we make less, but we're secure. You make more, but you're dreading the next recession. Now, some academics (like Levitt above) are perfectly willing to give up that security for more money (as he says, he'll give up his tenure for 15k more). But I doubt that all of them, or even a majority of them, are willing to do so.

All that said, I myself am uncomfortable with the way tenure is structured today. I don't think it should be abolished but I do think it can, and should, be modified. I would make two broad changes. One, I would change when faculty are first up for tenure. The first few years of an assistant professor's life are stressful enough; they really don't need this sword of Damocles hanging over their head. If faculty are up for tenure about six or seven years after getting hired, as opposed to the three or four right now, I think they'd be more likely to do better quality research (one consequence of having it after three years is that people want to get their name in a journal as much as possible, so will write and do less than perfect work. They basically sacrifice quality for quantity).

Two, I wouldn't leave tenure as a life-long promise of employment. I would make it subject to revision every, say, eight or twelve years. I think this would solve the problem that Levitt and I agree is a big problem: the lazy prof who's done nothing for 25 years but is still cashing the university's checks (I should say there's not a great number of these so-called lazy profs, but there are enough of them to ensure that the system is less than perfect). The exact number of years that elapse before another tenure review is not important, what is important is that such a review take place periodically to ensure motivation among faculty.