This paper provides evidence of the relation between the risk of sexual harassment and wages. While one approach to detecting the effect on wages of sexual harassment would be to estimate wage equations controlling for whether an individual reports that he or she had been sexually harassed, sexual harassment on the job is unlikely to be exogenous with respect to wages, and it is difficult to identify appropriate variables that would allow instrumental variables estimation. In addition, there are almost no data reporting information on sexual harassment as well as wages and other determinants of wages. To avoid these problems, I adopt the conventional hedonic wage methodology used to estimate compensating wage differentials for risk of injury or death. Specifically, using data on individual charges filed with the EEOC (which I obtained under the Freedom of Information Act), I calculate gender-specific estimates of the risk of sexual harassment by industry and age group.