Children of married couples in the United States often have better health or academic outcomes than those of unmarried or cohabiting couples. It has been suggested in the economics and sociology literature that this correlation results from a causal difference in the ability of married versus cohabiting parents to commit to joint investments in children. In this paper, we quantify the effects of family relationship instability on the outcomes of children. We create a  model of relationship and labor market choice, and estimate it using measurements on children's outcomes drawn from the Fragile Families panel. Our counterfactuals consider the effects on children of increasing the desirability and stability of marriage, and we find that even large increases in marriage rates (from 25% to 50% in our sample) are associated with only small increases in child outcomes, with an average effect size that moves a child up 2 percentage points in the quality distribution relative to the baseline. This suggests that observed differences between child outcomes across relationship statuses are driven by selection on unobserved child quality, and that policies targeting parental relationships may not have large returns.

by invitation: for information or to receive the invitation link contact patrizia.pellizzari@unibocconi.it