We investigate the effect of skin tone on employment probabilities in a longitudinal data set. Using an objective measure of skin tone from a light-spectrometer and a self-reported measure of race we find that over time the effect of skin tone on employment has diminished. These results hold both across the white and African-American samples as well as within the African-American sample itself with regard to skin tone. Further investigation indicates that all of the gains can be attributed to African-American women; there are no changes in the employment probabilities for African-American men in the 15 year panel data. We find that the expansion of employment for women is concentrated in the services occupations.
race gender employment discrimination skin tone panel data