The effects of institutional gaps between marriage and cohabitation

Paula Calvo
PhD Yale University

Abstract: This paper examines the effects of institutional differences between marriage and non-marital cohabitation on household formation, individual’s welfare, and child human capital. I first show that, conditional on observable characteristics, cohabiting couples have, on average, higher separation rates, higher female labor supply, and worse cognitive outcomes among their children, relative to married couples. To explain these empirical findings, I model the individuals’ life-cycle problem within an equilibrium marriage market framework that features the choice between marriage and cohabitation. I estimate the model using U.S. household data. The results indicate that low-educated cohabiting women receive a lower share of the household’s resources than low-educated married women. Moreover, consistent with the empirical findings, their children accumulate less human capital compared to those born to low-educated married women, explained by lower maternal time investments and higher separation rates between cohabiting couples. In counterfactual analysis, I equalize child custody laws for unmarried and married parents upon separation. I find that this policy would improve the welfare of low-educated cohabiting women and the outcomes of their children. Accounting for marriage market equilibrium effects is critical for this result: Under the baseline equilibrium, this policy change would reduce the welfare of low-educated cohabiting women (by reducing their access to children upon separation). However, in the new marriage market equilibrium, these women would be compensated with a higher share of the household’s resources, which induces them to form cohabiting relationships. This policy also contributes to closing the human capital gap between children born to low-educated cohabiting and married women