University of Texas at Austin Professor of Economics Daniel Hamermesh lectured on whether professors’ looks impact their course evaluations and the affects of gender in the American Economic Association’s elections Monday.

Hamermesh’s first paper, “Beauty in the Classroom: Professors’ Pulchritude and Putative Pedagogical Productivity,” examined the correlation between students’ ratings of professors physical attractiveness and those professors’ course evaluations. Six undergraduate students rated the attractiveness of 94 professors from the University of Texas on a scale of 1-10. Some factors are known to correlate with lower course evaluations, such as if the instructor is foreign or the class is lower division or large.

Controlling for these factors, Hamermesh found that beauty had a significant effect on professors’ course ratings. Additionally, looks mattered more in lower division classes than upper division classes and more for men than women.

“Wage effects, promotion possibilities — in everything, looks matter more for men than women,” Hamermesh said. “Some people get upset by this, but I keep finding it, so I believe it.”

For Hamermesh, who stated, “I know that better looking people are absolutely paid more,” the question is not whether or not beauty has an effect, but why.

“Is beauty rewarded with an evaluation? Or are you rewarding it because it makes you learn better?” Hamermesh asked. Discrimination is defined as a difference in monetary outcomes that doesn’t stem from productivity difference, so the question of whether or not there is a productivity difference related to physical attractiveness is an important one. However, Hamermesh said he doesn’t know the answer.

His second paper, “Discrimination by Gender: Voting in a Professional Society,” expanded on this difficulty in measuring productivity. In it, Hamermesh examined the American Economic Association’s electoral data from 1935 to the present. He found that although only one-eighth of the candidates were women, they had a much higher chance of getting elected than their male counterparts. Out of all the factors Hamermesh measured – including race, academic affiliation, and whether one is or will be a Nobel Prize winner – only gender and whether one had held a high-level position in the government showed significant effects. Compared to the expected 50 percent, female candidates had a 74 percent chance of winning elections.

This might seem like an example of reverse discrimination, but Hamermesh cautioned against any simplistic interpretation of the results. Again, the possibility for productivity difference must be taken into account.

In this study, Hamermesh measured productivity by the number of times a candidate had been cited in the years directly before the election. This measure had no significant effect on a candidate’s probability of winning the election.

Productivity, however, is an incredibly difficult thing to measure.

“The inability to control for productivity differences in empirical studies of possibly discriminatory outcomes is hardly unique to this study, to studies of electoral outcomes generally, or even to studies of economic outcomes,” Hamermesh said.

Hamermesh has written extensively on labor economics. His published work includes the textbook Economics is Everywhere, which is taught at Oberlin.