2014-13 Self-identified sexual orientation and the lesbian earnings differential

Two decades of research on differences in labor market outcomes by sexual orientation has concluded that lesbian workers earn more than heterosexual women. This research, however, is largely based upon data that do not ask respondents about their own sexual orientation. We are the first to use nationally representative data that includes selfidentified sexual orientation. We find evidence of a sizable wage penalty for self-identified lesbians in 2008 and 2010. We show that using common behavioral proxies for sexual orientation overstates of the earnings of lesbians in some years, but understates it in other years. Using the different methods of identification leads to different conclusions about the way the recession affected lesbians: It took much longer for the wages of self-identified lesbians to recover.