Search Results

This paper examines the impact of child support on one important set of indicators of child well-being: health outcomes. We expect that, all else equal, higher levels of child support payments improve a child's health status, assuming that health is a normal good. This effect could work either directly through greater out-of-pocket spending on health inputs (i.e. food, medical care or medication) or indirectly through the purchase of health insurance coverage. Based upon research on other outcomes related to child support, we also expect that higher levels of child support will be positively correlated with a child's health status, even after controlling for total family income. The literature provides several possible explanations for this finding, including unobserved heterogeneity, decreased child receipt of public benefits, an improved relationship between the parents and strategic bargaining between the parents.

The data for the project comes from a sample of single, custodial mothers in the 1979 National Longitudinal Survey of Youth (NLSY). NLSY79 respondents, who are between the ages of 41 and 48 in the latest interview wave, have reported data on their families over the course of the panel, including children as they were born. The matched Child and Young Adult (CYA) file provides individual-level panel data on those children, including health insurance coverage, utilization of non-emergency medical care, and several indicators of health status.

Using this data, we estimate a set of models in which a child's health outcomes are a function of the annual child support received by his or her custodial parent, controlling for a full set of demographic characteristics of child and parent. Actual child support awards are likely to be endogenous because most state child-support guidelines include supplemental awards for children with extraordinary health care needs. In order to address this endogeneity problem, we instrument for actual child support using the set of parameters in each state's child support guideline formula, which change at least once for most states during the 1990 to 2006 period of our analysis. Our preliminary results suggest that higher levels of child support receipt increase health insurance coverage rates, and may also result in more annual doctor check-ups and fewer children who are underweight. This holds true even after controlling for total family income. Our next step will be to add family-level fixed effects and control variables such as public program participation to the model in order to try to identify the mechanism (from the list of possible explanations from the literature) by which child support improves health above and beyond its income effect.