A groundbreaking study published about 25 years ago found that women musicians were more likely to secure spots in orchestras when auditions were conducted behind a screen that concealed their gender. The famous “blinding experiment” influenced hiring practices by focusing on recruiter bias, but what happens on the other side of the curtain?
The recent working paper “Discrimination, Rejection, and Job Search,” coauthored by Harvard Business School Associate Professor Katherine Coffman, explores the supply side of the job market: whether women and older job seekers are more willing to pursue a job if they can block demographic information from their resumes.
An aging population and falling birth rates portend a potential long-term labor shortage in the US that might force companies to work harder to source talent. Encouraging more women and older workers—groups that have felt the sting of sexism and ageism in the workplace—to apply could improve a company’s candidate pool.
Coffman’s study tests a simple approach—blinding—that could work even as more companies use artificial intelligence to screen applicants. The team found that blinding the process narrowed the gender and age gap by approximately 25 percent without significantly affecting young men.
“Blinding has significant, meaningful impacts on the supply of talent. A blind hiring process increases the overall size, average talent, and gender diversity of the applicant pool,” the paper states.
Coffman, the Piramal Associate Professor of Business Administration, teamed up with Anne Boring, an associate professor of economics at Erasmus University Rotterdam; Dylan Glover, an assistant professor of economics at INSEAD; and María José González-Fuentes, a doctoral student at École Normale Supérieure de Lyon and an affiliate of its Center for Economic Research and Governance, Inequality, and Conflict.
Drawing more talented women
The researchers used a simulated job market on the Prolific platform, where participants were instructed to act as recruiters reviewing applications or job applicants presenting resumes for a technical job. Job candidates were asked about their gender, age, education, and favorite academic subject (such as humanities or social sciences). Their knowledge of coding and software programming was also tested.
After that, the team measured the candidates’ willingness to participate in a blinded or non-blinded process, providing monetary incentives to participants. The results showed that talented candidates, especially women, preferred withholding information about their gender and age when applying.
“This highlights the potential of blinding to increase the supply of talented, counter-stereotypical candidates,” the paper notes.
Researchers asked applicants about their choices and concluded that self-stereotyping—when people subconsciously adopt misperceptions about them—plays a significant role in the lower application rates among women. However, female candidates, in particular, seemed to anticipate discrimination when their resumes disclosed their age and gender.
The impact of rejection
The team also examined how candidates handle rejection. Returning to the orchestra example, the paper suggests that it’s like asking if female musicians are willing to try out again after not being selected.
To find out, the team gave negative feedback to all candidates on the Prolific platform and compared their reactions. They found that rejection discourages candidates from reapplying regardless of the process, but the impact is greater in cases where job seekers applied with a blind resume.
“Rejections that could be rationalized by recruiter discrimination have significantly less impact on future application behavior than rejections resulting from a blind hiring process,” the researchers wrote. “In addition, candidates are more likely to say that the recruiter’s decision was fair and they report less disappointment when their application was rejected with a blind resume.”
Those who were rejected after revealing their gender or age in their resume said they would prefer submitting blind applications in the future. If the rejection involved a blind resume, their future preference depends on whether the candidate believes that stereotypes will help or hinder their chances of getting the job, the researchers found.
Even after being rejected with a blind resume, women and older candidates preferred concealing their demographic information. On the other hand, “men are significantly more willing to apply on non-blind applications than blind applications going forward, perhaps anticipating that gender discrimination could benefit them,” according to the research.
