Edoardo Ciscato – Matching on gender and sexual orientation

ROOM : P02

In spite of the increasing share of individuals who identify as bisexuals or report sexual attraction to both men and women, 89% of partnered bisexuals in the U.S. are in a different-sex relationship. Moreover, 14% of partnered gay men and lesbians have a partner of the opposite gender. In order to better understand these mating patterns, we estimate a multidimensional matching model of the marriage market where individuals differ by gender, sexual orientation, age, education, and race. The novelty is that the partner’s gender is endogenously chosen conditional on the agent’s sexual orientation, and is subject to trade-offs that depend on both the agents’ preferences and the pool of potential partners. We show that same-sex couples experience lower gains from live-in relationships, a difference we refer to as the “same-sex penalty”. We find a relatively large penalty for male same-sex couples, in both Germany and the U.S., although it has decreased in recent years. Moreover, we identify a smaller penalty for female same-sex couples in some regions of the U.S., but not in the Northeast, in the West, nor in Germany. Through a counterfactual experiment, we show that, absent this penalty, the share of same-sex couples in the U.S. would increase by about 50%, from 1.35% to 2.06% of all couples. Finally, we also show that a 10% increase in the men-to-women gender ratio would lead to a modest, but positive increase in the odds that a man is in a same-sex relationship (+1.8%), and to an almost symmetric decrease in the odds that a woman is in a same-sex relationship (-1.9%).