The Dynamics of Referral Hiring and Racial Inequality: Evidence from Brazil
We study how referral hiring contributes to racial inequality in firm-level labor demand over the firm's life cycle using data from Brazil. We consider a search model where referral networks are segregated, firms are more informed about the match quality of referred candidates, and some referrals are made by non-referred employees. Consistent with the model, we find that firms are more likely to hire candidates and less likely to dismiss employees of the same race as the founder, but these differences diminish as firms' cumulative hires increase. Referral hiring helps to explain racial differences in dismissals, seniority, and employer size.
We thank Joaquín Fuenzalida for excellent research assistance. We thank Kathryn Anderson, Sydnee Caldwell, Jonathan Leonard, David Levine, Ricardo Perez-Truglia, Isaac Sorkin, Guo Xu, and seminar participants at the 2020 Southern Economic Association meetings, the OZ Virtual Seminar, Notre Dame, Cornell, UC Berkeley, Claremont McKenna College, Rutgers, Indiana, LSE, West Point, Chicago Harris, UCSB, the Southeastern Micro Labor Workshop, Inter-American Development Bank, Yale, and NBER Personnel for comments. Miller gratefully acknowledges financial support from the Upjohn Institute and the UC Berkeley Institute for Research on Labor and Employment. The views expressed herein are those of the authors and do not necessarily reflect the views of the National Bureau of Economic Research.