This paper investigates the impact of same-sex legalization and same-sex relationship recognition on labor outcomes such as wages and specialization for lesbian women. Using 1990 U.S. Census data and 2013 American Community Survey (ACS), I exploit the temporal and spatial variation of legalizing same-sex marriage using a triple difference-in-differences-in-differences estimator, and employing an Oaxaca-Blinder wage and specialization decomposition. Results demonstrate that unmarried lesbians who are primary partners experience a wage premium of 15.2% and unmarried lesbians who are secondary partners experience a more modest wage premium of 3.7%. Married lesbians who are secondary partners experience a wage penalty of 10.7%. For primary partners, wage premiums are 19.3% under domestic partnerships, and 15.9% under civil unions. No such effects are detected under same-sex marriage for the general lesbian population, suggesting that the primary partner marriage premium is being offset by the secondary partner wage penalty. Marriage, more than any other legal institution, signals greater commitment and in turn facilitates specialization. There is little evidence of marriage market selection and employer discrimination related mechanisms.