Divvying Up the Raise Pool

Five years ago, when I first became a department head, what worried me most about the job was assigning salaries to my fellow faculty members. I was afraid I would be unfair; I was even more worried that a disappointed colleague would chew me out, or worse, dissolve in tears.

The dean who appointed me (as one of two heads of the comparative-literature department) didn’t share my misgivings. He gave me the standard no-frills account of how to hand out raises:

In the sciences, he said, we look at grants and publications. Since you’re in the humanities, things are a little more complicated, but still pretty straightforward. Every year, he told me, just rank and count up the publications produced by each faculty member. Some departments have a point system -- one point for a book review, two for an article in a refereed journal, 12 for a book, and so on. Once you know what the raise pool is going to be, just divide that number by the total number of points earned by the department’s faculty members that year to find the dollar value of a point. Multiply that figure by the total number of points each faculty member has earned to find his or her raise.

The former heads of our department had never developed a point system, and I quickly found out why. In a discipline like ours, there are as many imponderables in assigning points as there are in assigning salaries.

Take journal articles, for example. Does a 10-page article count as much as one that is twice as long? Does an article in a top-tier journal count the same as one in a lesser-ranked publication? How should I assess a public lecture, a conference paper, a keynote address? Does a translation or a scholarly edition of a Korean text equal a book? What about a book published in Korea? Does a volume you’ve edited count as a book?

Recently one of our faculty members began writing historical novels about Greek philosophers. Do those count as books? Should a book really have the same point value as six articles? Why not seven? Or five?

And what about the faculty member who didn’t publish anything this year? Should he go without any raise at all? What if that faculty member is in the fourth year of a seven-year project with no results to show and none to come for a while? Is it fair to reward him on credit or must he wait for a windfall after publication?

After five years in the chair’s job, I find that the clamor in my head from all of those questions has died down. I’ve worked out my own solutions not only to those matters but also to a number of other conundrums that I hadn’t even imagined when I first started.

Although my experience is based on my own department and on our quirky discipline, when colleagues who are just starting out as department heads ask for my advice, I feel that I’m in a position to give them a fuller answer than my dean gave me.

First I suggest they look at two underlying assumptions of the standard faculty-salary model -- that annual raises should reflect annual productivity and that productivity and research are the same thing. As logical and widely accepted as those two assumptions are, in my experience, both lead to unfairness.

In the past five years, the raise pool allocated to our department has been different every year. When raise pools vary from year to year, a fair allocation of points does not guarantee fairness in the bottom line -- the actual allocation of raise dollars. Faculty members might publish books or articles of similar scope and importance in consecutive years and receive vastly different dollar amounts for the same work from the raise pool.

The worst-case scenario is the faculty member whose multiyear research project finally yields a book in a low-budget year. Having patiently waited for years without a significant pay increase, she is grossly underrewarded when the payoff finally comes.

Another factor that limits the effectiveness of counting publications to calculate annual raises is one familiar to every faculty member. Throughout my career, raises based on annual productivity have lagged behind increases in starting salaries for new hires. The result is salary compression. With seven years of annual raises, a productive associate professor ends up earning little more than an untried assistant professor hired last year.

Gender inequities create a similar problem. Even though women are compensated at an equal rate with men today, that cannot erase the effects of years when they were not. Our university has set aside money in the past to overcome those differences, but it has not always been enough. Sometimes the department has had to step in.

The second assumption underlying many salary models -- that productivity and research are the same thing -- is equally problematic.

Is the number of your publications an accurate measure of your research productivity? In the best of all possible worlds, every bit of competent research would culminate in an article or a book. But that has never been the case for many scholars, and given the current disturbing trends in academic publishing, it’s even less the case today.

Many books of scholarly importance lack even the minimal commercial value that academic presses today are demanding. Online publication will almost certainly relieve some of the pressure eventually, but it is of little help right now -- at least in our discipline.

As a department head, I see excellent research for which a publisher cannot be found or for which the only publisher is in a low-rated venue. That’s why, in our department, my co-chair and I have taken it upon ourselves to judge and reward both published and unpublished research.

While I share the departmental leadership with Gabriel Ruhumbika, I am more of a numbers person, so I handle much of the raise calculations. In assigning raises, I work in broad categories -- below-average productivity, average productivity, and exceptional productivity -- and divide the raise pool accordingly. It is relatively easy to assign people to those categories, and then within each one, to rank people and push the percentages up or down a bit.

If you start from a faculty member’s output and try to build a figure from there, it is very difficult to do, because it is hard to put a fair number on a particular achievement. In the end the process is a zero-sum game anyway -- my gain is your loss -- so absolute calculations turn out to be meaningless.

My co-chair and I also agree that some types of research productivity may never lead to publication, and we take those types into consideration in assessing someone’s productivity.

Teaching, for example, is productive, especially if it is the kind of teaching that a research institution is meant to foster. Like a book or refereed article, the classroom is a means of disseminating knowledge. At a nonresearch institution, instructors take what they have learned from their teachers and pass it on to their students. As instructors age, the information circulated in their classrooms may lose currency.

At a research institution, on the other hand, instructors pass on the results of their own research. They also pass on to students an understanding of how knowledge in the field is created and judged, and offer a continually changing synopsis of the state of knowledge in their fields. Teaching can, and should, be a form of research publication, and it should be rewarded from that point of view.

Student evaluations are of some use in determining quality in teaching. Students can judge the ability of a faculty member to reach them, but they are ill equipped to judge the reliability and currency of the information they receive. Yet those evaluations are often what determine institutional honors for outstanding teaching. We think collegial peer evaluations of teaching should be encouraged and taken into account in awarding raises.

Teaching is not confined to the classroom. Faculty members supervise student research at every level. They contribute to program maintenance and curriculum development. They administer sub-programs, advise students, and serve on university committees. However remote from research, all of those activities contribute to a department’s educational mission and deserve consideration when the time comes to divvy up the annual raise pool.

Early on, my co-head and I paid particular attention to salary compression and inequities. That proved to be the best time to deal with problems that required money beyond our normal annual budget. It was clear to everyone that as newly appointed leaders, we were not responsible for the inequities that we had inherited. And we were successful when we urged the dean to respond to those long-term pay problems.

In recent years, we have assigned raises by making a broadly based rolling assessment of faculty productivity. We look at a professor’s record for the past two or three years rather than just for the current year, and we take into account the raise pool in each of those years. That way we can supplement the salary of someone who happened to produce a lot in a year when the budget for raises was small.

We reward good teaching and we consider service to departmental programs as an important contribution to our overall mission.

Like our predecessors, we have not come up with a formula in which we can plug a few numbers and produce a fair raise for every faculty member. Instead, what we have devised is a process that has withstood occasional challenges from the dean’s office and, equally important, won faculty support.

In five years, we have had almost no complaints, no angry confrontations, and, best of all, no tears.