(When) Does Appearance Matter? Evidence from a Randomized Controlled Trial

While there is evidence about labor market discrimination based on race, religion, and gender, we know little about whether physical appearance leads to discrimination in labor market outcomes. We deploy a randomized experiment on 1,000 respondents in India between April and May 2020 to investigate how ``looks'' affect labor market outcomes. Using machine learning and computer face rendering techniques, we simulate North, South, and North-eastern Indian faces and randomize two waves of treatments where we attach faces with otherwise identical CVs. However, the second wave also contains a vignette where we provide respondents with information about the incidence of discrimination. We focus on North-eastern Indian faces that resemble those in China to focus on the potential for discrimination that has historically been present in anecdotal accounts. While we do not find evidence of discrimination against North-eastern faces in the first wave, we do so in the second among those who do not see the vignette. We provide suggestive evidence that these results are consistent with the presence of homophily under limited attention