Cohabitation VS. Marriage: Mating Strategies by Education in the Usa

In this paper I analyze the determinants of cohabitation, marriage, and divorce in the US. I first document that college graduates are more likely to marry, and less likely to cohabit and divorce, than non-college educated individuals. To account for these facts within a unified framework, I build and estimate a life-cycle model of partnership formation and dissolution where income processes differ by gender and education. I find that the main driver of education-based differences in mating strategies is that the gender wage gap is larger among college graduates. Since divorce is more costly than ending a cohabitation, marriages tend to be more stable and therefore offer women more protection from human capital depreciation during nonemployment. Consequently, marriage is a more effective means of enforcing household specialization. Since college graduates have more room for household specialization, they are more likely to choose marriage. The variance of income shocks, which affects the demand for consumption insurance, is larger among college graduates. Even if the variance of income shocks could potentially explain partnership choices, simulations suggest a small role of income volatility.