Racial Earnings Disparities with Endogenous Labor Market Participation

This paper examines racial disparity in wage income earnings in the USA across different quantiles, after accounting for the endogeneity of labor market engagement which is captured using the continuous variable: annual number of weeks worked. Controlling for this endogeneity is important because changes in income earned would affect the labor supply, while on the labor demand side any pressures that cause racial disparity is likely to affect both hiring and pay decisions. State-sector level averages for weather-related work absences are used as instruments for number of weeks worked at the individual level and robustness of the results are verified. The instrumental variable conditional and generalized quantile regression techniques are used for analyzing the disparity. The results find that the racial disparity gets larger at the upper quantiles even when controlling for other covariates ensure that the individuals are similar in important attributes.