In light of findings unveiled at the September 16 faculty meeting concerning two issues of faculty salary discrepancies, an ad hoc committee has been formed to examine apparent gender discrepancies. In addition, the Faculty Compensation Committee is continuing talks with the administration about what should be done about a group of assistant professors whose incomes are significantly lower than those predicted by a statistical model.
Chair of the Faculty Compensation Committee Richard De Veaux, associate professor of mathematics, presented the committee’s findings at the faculty meeting. The committee created a statistical model to predict faculty salaries that took into account such factors as department, year of Ph.D. and tenure.
According to committee members there are two major concerns: first, approximately 20 professors, only one of whom is female, have salaries significantly higher than the model predicts, raising questions about the distribution of merit-based raises. The second concern is that a group of associate professors have salaries that fall $4000 to $6000 below what their salaries should be as predicted by the model.
At Wednesday’s faculty meeting, the faculty compensation committee asked that the administration assemble a group to examine the reasons for the higher salaries of the 20 faculty members whose incomes exceeded the predicted average.
Provost Stuart Crampton explained that an ad hoc committee of former and current provosts and deans of the faculty has been formed for this purpose, in the hope that those who may have determined the raises in the past will be able to give a historical perspective on the raises.
“The process so far has been a good one,” Crampton said. “We try to be fair with salaries. We hadn’t looked at them in such a way as to see this particular problem. It’s important to know whether the salaries reflect our values. It may be consistent with our values to give higher salaries to those perceived to be better teachers, more productive scholars.”
Professor of English Suzanne Graver, a member of the faculty compensation committee, explained the work facing the new ad hoc committee.
“What we’re doing is trying to shed some light on mechanisms by which salaries are determined,” Graver said. “We’ll look at the salaries of the top 20 whose salaries are higher than the model predicts. . .About 12 are significantly higher [than predicted].”
“We can only account for some factors that might have influenced those in the top group,” Graver explained. “This is less to explain or justify, [it’s] just to describe what factors entered into it and see discussion take place on what can enter into determining pay.”
Assistant to the President Nancy McIntire said the committee will attempt to determine what factors the seemingly over-compensated group has in common.
Crampton said the committee reached few resolutions at its first meeting last Thursday.
“I think we’re going to know a lot more in a month,” he said. “We met yesterday and realized we needed more information â€“ enough to get a sense of what things may have been important. It was a good hour, but only an hour.”
The committee will not make any policy changes or attempt to resolve the salary discrepancies. “This is fundamentally a prelude to whatever else will be done,” Graver said.
De Veaux realizes the advantage of some merit-based salary increases, calling merit and high residual salaries the touchier part of the problem.
“I think there has to be some room for being rewarded for performance. I’d be sad if there were no economic incentive for performance,” De Veaux said.
Professor of German and Russian Bruce Kieffer agreed, but expressed concern that the 20 professors whose salaries are above average offset the full professor’s salary average so that the lower group is in effect subsidizing the salaries of higher paid professors.
Kieffer did not dismiss the idea of merit-based salary increases, but made the distinction between legitimate and questionable reasons for these raises. For example, the College often gives raises to keep valuable faculty members when they have high salary offers from other universities. “What troubles some people is the question of whether the College is meeting or trying to pre-empt a higher offer,” he said.
De Veaux commented on Provost Crampton’s involvement with the issue. “I think he’s wrestling with how to do it. I think he’s committed to fixing it. [The discrepancy] is not going to go away; In fact it will just compound,” he said.
“It’s potentially a very significant problem, but it’s also potentially not really a problem at all,” Kieffer said.
Several faculty and administration members expressed concern over the resolution of the discrepancies.
“One has to admit that it’s a very complicated situation,” Kieffer said. “Discussions about salary at Williams have always been enveloped in a faith in equity. That there are other discrepancies has not really been known or appreciated before.”
President of the College Hank Payne emphasized the importance of reaching some concrete conclusions.
“I think it’s very important that there be a sense of trust and integrity in salary structure,” he said. “This is something we can [address] soberly and sensitively over the next few months and should be able to resolve. We’ve got the right faculty governance organs looking at the right questions now. These are serious questions, but I think we have the good will to resolve it.”
Payne was not particularly concerned that the issue will create tension among the faculty.
“I like to think we’re all working in good faith,” he said. “This is not blatant discrimination or willful manipulation of the system.”
Explaining “the dip”
Faculty income at Williams has two components: salary and benefits. According to Associate Professor of Biology Heather Williams, the College discovered three years ago that it had been incorrectly calculating the mortgage benefit for professors, counting it twice and creating the illusion that the incomes for this group of professors were meeting the salary goal.
The College attempted to fix the problem in 1995 by boosting the salaries of those employees with mortgages, most of whom were full professors. This percentage raise applied to all who were full professors in 1995, but neglected to account for those who have become full professors since then.
Crampton elaborated on the mistake.
“There was a correction in the way this is calculated,” he said. “Because of the way the mortgage benefit was miscalculated, we thought we were meeting the salary goals of full professors,” he said, explaining that once the error was discovered, the full professors’ salaries were increased. However, the full professors were given a higher percent salary raise than the associate professors. Crampton said if no further adjustment is made, the discrepancy will increase.
“The most important thing which was pointed out by the subcommittee is that if somebody’s salary is lower for reasons having nothing to do with how good a teacher they are, that difference will compound over time,” Crampton said.
Although the Williams faculty is 40 percent female, 70 percent of associate professors are women. The subcommittee on gender and salary equity, led by Williams, pointed out that if the trend is allowed to continue, female full professors will have significantly lower average salaries than male full professors.
“This is certainly not gender discrimination now, but allowing it to pe
rsist would be dubious,” Williams said.
Professor Williams blames the slow discovery of the mistake on the administration’s reluctance to release faculty salary figures.
“The reason [the discrepancy] was not uncovered until now is because the administration has historically refused to release information on salaries,” she said.
Crampton said the administration has been concerned about breaching faculty confidentiality. However, under a new agreement, faculty salary information will be released for review once every three years, but no names will be attached to the data.
Professor Williams also explained that very few Williams college faculty members belong to the American Association of University Professors (AAUP). According to Williams, most prefer that the faculty is not unionized because it allows for a more intimate and civil relationship between faculty and administration.
“In my ideal world we don’t need [unions],” De Veaux said. “But it’s nice that they’re there. I don’t want an us and them [relationship with the administration].”
De Veaux explained that the discrepancy will not disappear as associate professors become full professors. “If corrections aren’t made, the full professors will become two groups in the future: those that got raises three to four years ago and those who didn’t,” De Veaux said.
De Veaux and the faculty compensation committee will continue discussions with the administration to try and resolve the discrepancy among associate professors. He had several suggestions about how to ameliorate the problem.
“There are many ways to correct [the problem],” he said. “It will involve some aspect of getting associate professor increases in line with assistant and full [professors]. Another possibility is to increase the permanent raise one gets between associate and full [professorship].”
According to De Veaux, there are two major possibilities for a solution. One is to adjust the salaries of all associate professors now, so that a percentage increase upon promotion will put their salaries in line with those of current full professors. Another possibility is to give associate professors a block salary increase upon promotion, so that their salaries match the full professors’ salaries.
Williams agreed that a solution should include a block salary increase and compensation for salary and retirement funds lost because of the discrepancy. “To me it seems really simple,” she said. “The solution is easy to see.”
Williams and De Veaux worry that the issue of retroactive compensation will be a contentious one.
“They ought to straighten out the salaries of those who missed it, and that’s going to involve back pay,” De Veaux said. “I think they ought to have some retroactive compensation, at least a retroactive bump. We’re working on making it equitable in terms of years of service by rank.”
However, Kieffer doubts that back pay will be a part of the solution.
“The College never does that,” he said. “The College does what it calls adjustments. If you gave those people back pay you might see mathematical consequences that would create another problem.”