Remarriage Less Likely Among Divorced Dads Paying Child Support

The financial benefits to children in single-parent families of improved enforcement may be substantially or completely offset by the negative effects of enforcement that operate indirectly through diminished remarriage.

Sometimes, even the best intended social policies have negative side effects. A recent National Bureau of Economic Research study shows that child support obligations deter remarriage among nonresident low-income fathers (that is, divorced fathers who do not live with their children). Thus, according to NBER Research Associate David Bloom and co-authors Cecilia Conrad and Cynthia Miller, the benefits to children of stricter child support enforcement are diminished by the negative effects of child support on remarriage. This occurs because a substantial share of nonresident low-income fathers who would have remarried and supported women with children no longer do so. Indeed, the authors conclude, "the financial benefits to children in single-parent families of improved enforcement may be substantially or completely offset by the negative effects of enforcement that operate indirectly through diminished remarriage."

In Child Support and Fathers' Remarriage and Fertility (NBER Working Paper No. 5781), the authors report estimates from two independent samples of the effect of a typical child support payment on a man's yearly probability of remarriage. Based on income, the average amount of support paid, and the amount that would be paid under a perfectly enforced system using the Wisconsin child support guidelines, Bloom and his coauthors estimate that child support payments may reduce a man's yearly probability of remarriage by up to 15 to 20 percent. In terms of enforcement in general rather than payments specifically, a "10 percent increase in enforcement, for example, can be expected to reduce the yearly probability of remarriage for low-income fathers by up to 10 percent."

The data used in this study come from two sources. The Survey of Income and Program Participation, a nationally representative sample of households interviewed every four months, includes marital and fertility history and details of household relationships. The authors restrict their sample to divorces that occurred between 1981 and 1989.

They also use data from the 1979-92 National Longitudinal Survey of Youth, which covered approximately 13,000 men and women aged 14 to 21 in 1979. Information from the 1990 and 1991 marital and fertility history questions allows the authors to construct a sample of men who divorced between 1981 and 1989; a subset of these divorces occurring from 1982-7 includes information about child support paid.