Federal appellate judges are more likely to hire women to prestigious court clerkships after serving on panels with female colleagues, new Cornell research shows.

The findings suggest diversity at a profession’s highest levels may open doors for underrepresented groups at entry levels, potentially helping to reduce discrimination, said Eleonora Patacchini and Marco Battaglini, economics professors in the College of Arts and Sciences.

“We find a strong impact of exposure to female judicial colleagues on future hires,” Patacchini said. “It creates opportunities if judges, or company managers, may change their beliefs and then change their hiring behavior.”

Patacchini and Battaglini, the Edward H. Meyer Professor of Economics, are the co-authors with Jorgen Harris, Ph.D. ’19, an assistant professor of economics at Occidental College, of “Professional Interactions and Hiring Decisions: Evidence from the Federal Judiciary,” released in January as a working paper by the National Bureau of Economic Research.

The scholars said the NBER study exploited unique features of the federal appellate court system.

First, judges are assigned randomly to serve on typically three-person panels, and so do not choose the colleagues with whom they deliberate high-stakes cases. Second, they have broad latitude to hire from a pool of top law students that apply annually for clerkships.

“As a result, changes in the likelihood of hiring a female clerk likely reflect changes in a judge’s assessment of the likely ability of a female junior colleague,” the authors wrote.

The researchers scoured published court opinions from 2007 to 2017 for the composition of appellate panels and combined that with data about more than 200 judges and their clerks from the Judicial Yellow Book directory. About one-quarter of the observations involved female judges.

The analysis found that, on average, judges who heard more cases with female colleagues were 4 percentage points more likely to hire at least one female court clerk in the next year. The likelihood increased more among male judges (4.3 percentage points) than female judges (1.6).

The results show that exposure to an underrepresented group influenced individual hiring decisions, the authors said. The randomness of judges’ exposure made other explanations unlikely, they added.

“It changes an economically meaningful decision,” Battaglini said. “So exposure is good to the extent that we think it helps reduce discrimination.”

The researchers examined seven characteristics that might have affected judges’ interactions: gender; age; experience; political party; seniority; quality; and fraction of clerks who are women. The effect of exposure to female colleagues appeared strongest for judges who were male, who had few women clerks, or who were less experienced, though these differences were not statistically significant due to smaller sample sizes.

That evidence was consistent with the researchers’ theory that increased exposure to female colleagues on judicial panels had shifted the attitudes of judges who had less prior exposure to female colleagues.

“Most of these judges started their careers when women were less prevalent in the legal profession,” Patacchini said. “They might well update their beliefs about women’s professional capabilities.”

Such reassessments likely occurs in other professional contexts, the researchers said, pointing to the value of policies promoting diversity in a profession’s upper rungs.

“Policies aimed at increasing the diversity in the leadership of a profession, such as affirmative action policies or policies requiring that a certain number of board seats be filled by women,” they wrote, “may have benefits beyond their immediate beneficiaries.”