This paper estimates the effects of privatization on worker separations and wages using retrospective data from a national probability sample of Ukrainian households. Detailed worker characteristics are used to control for compositional differences and to assess types of observable 'winners' and 'losers' from privatization. Preprivatization worker-firm matches are used to control for unobservables in worker and firm selection. The results imply that privatization reduces wages by 5 percent and cuts the layoff probability in half. Outside investor ownership reduces separations but leaves wages unaffected. Winners from privatization tend to be higher-skilled employees of larger firms, but there is no discernable relationship with gender, education, or experience.