The rise of working from home (WFH) has been though to amplify existing labor market inequalities because it often disadvantages certain social groups in terms of prevalence and wage returns. The Inter-American Development Bank (IDB) sponsored a research partnership between Universidade Estadual de Campinas (UNICAMP) and Columbia University to investigate differences in earnings penalties associated with WFH between groups of gender and race before and during the COVID-19 pandemic in Brazil. The study explains how and why the earnings penalty associated with WFH diminished for White and Black men during the pandemic while remaining high for White and Black women.

The preliminary version of the study was published at the IDB Working Paper Series (click here for full access) by Alexandre Gori Maia (UNICAMP) and Yao Lu (Columbia). The study used a unique longitudinal dataset that is representative of the Brazilian population to estimate the changes in earnings of those who shifted from WFO (working from the office) to WFH and vice-versa before and during the pandemic. Fixed-effect estimators removed the potential selection biases from time-invariant characteristics, such as less productive workers being more likely to WFH. The study also conducted additional tests to assess the robustness of the estimates against time-variant omitted variables (e.g., poor performance in the office) and attrition (e.g., individuals WFH less likely to experience job loss).

Findings highlight how, during the pandemic, the earnings penalty associated with WFH diminished for White and Black men while remaining high for White and Black women. Three mechanisms help to explain these changes in the earnings penalty associated with WFH: i) The equilibrium between supply and demand, as more women than men transitioned to WFH during the pandemic; ii) Labor productivity, as the effective working hours of Black women were more affected by WFH; iii) Visibility and promotion, as White women became less likely than White men to be promoted when WFH during the pandemic.

The main policy implication of this study is that measures to stimulate remote work may inadvertently have detrimental impacts on gender and racial pay equality when they are not accompanied by changes in the social and labor norms responsible for the earnings penalty of WFH. Reducing the asymmetry in the distribution of women and men working remotely may be a way to reduce the gender earnings penalty gap of WFH. Another way to reduce the earnings penalty of WFH is developing technologies that make WFH more accessible for some occupations, as the racial gap in the earnings penalty of WFH disappeared during the pandemic in occupations that can feasibly be done from home using IT technologies.