Stigma and Labor Market Supply

Study Context

Our proposed labor market experiment explores how the stigma of living in a favela affects decisions to apply to jobs and its impact on the matching process. Stigma is a mark associated with negative characteristics, which leads to discrimination. Stigma can be especially damaging in the labor market, since it can psychologically affect labor suppliers. For those living in favelas, one’s home address may be associated with stigma. We outline a field experiment in which we manipulate stigma (address) visibility to estimate its effects on job application decisions and interview performance.

Study Design

This study will involve focus groups and a pilot experiment in Brazil, where 6% of the population live in favelas: urban slums that are usually ruled by organized crime and adjacent to richer neighborhoods. In the first stage, the researchers will manipulate home address visibility in a job application decision manipulating visibility amounts to varying the anticipation of stigmatization. The second stage of the experiment focuses on understanding the impact of the stigma in the job interview performance. A second randomization will vary the anticipated stigmatization: this time, applicants will be randomly informed about whether the interviewer has information about their home address or not. Interviewers will be blind to the experimental design and will not be allowed to discuss home addresses during the interview.

Results and Policy Lessons

In the first stage, the research team found that removing the need to declare a home address from the application process does not increase job application rates. In a second ongoing stage, the researchers randomly tell job-seekers the results of an audit study where they found no discrimination against favela residents in callback rates. Current estimates suggest that job-seekers updated their beliefs regarding discrimination, but that does not translate into much higher job application rates. Finally, the researchers randomized whether job interviewees believed their interviewer knew their address or not, and their current results suggest that candidates did worse in the interviews when their addresses were known.

There seems to be interesting heterogeneity in treatment effects. For instance, white favela jobseekers apply more frequently when they are told their addresses are not needed for applying, and people of color and women apply more often when they learn the results of the researchers’ audit study showing no discrimination based on address.

Overall, results suggest that anticipated discrimination is deeply ingrained for many individuals, and policies that allow people to hide their addresses may not be effective across all demographics. At the same time, providing information on actual discrimination rates can lead women and people of color to apply more often. Finally, the researchers show evidence that anticipated discrimination can be damaging in job interviews, when candidates are under pressure.