Abstract

PAMELA J. SMOCK University of Michigan WENDY D. MANNING Bowling Green State University* Family patterns have changed dramatically. Now nearly a majority of children in the United States are likely to spend at least part of their childhood living apart from their biological father. Social science research has indicated that father involvement-particularly economic flows to children-is crucial to children's well-being. Furthermore, policymakers are focused on child support reform as the foundation for improving outcomes for children. Yet most research on the determinants of child support relies solely on mothers' characteristics. We draw on new, matched, ex-couple data from the Panel Study of Income Dynamics to examine the importance of having information about both parents. Our findings indicate that the characteristics of nonresident parents are central to understanding levels of child support and underscore the need for data collection that includes obtaining information from both resident and nonresident parents. Key Words: child support, divorce, nonresident parents, single parents. Changes in family formation behavior over the past three decades have altered the context in which children are raised. Trends show increases in cohabitation, nonmarital fertility, and marital disruption (Bumpass, 1990; Sweet & Bumpass, 1987). The upshot of these trends is that in 1990 at least 25 million children, about 40% of all children in the U.S., did not live with their biological father (Norton & Miller, 1992) and about half of all children are expected to live in a single-parent household before adulthood (Bumpass & Sweet, 1989a) A large body of literature focuses on the effects of family structure on children's wellbeing. These studies generally show that living without both biological parents has a negative effect on children's life chances (e.g., McLanahan & Sandefur, 1994). Children living with only one parent receive less schooling, are more likely to have nonmarital births, and are more likely to grow up to be poor themselves. A good deal of this relationship stems directly from lack of income. Due to the low potential earnings of many single mothers and little or no child support, nearly a majority of these children live in poverty (Garfinkel & McLanahan, 1986; Holden & Smock, 1991). Thus, a national focus on child support reform as one way to improve the well-being of children has emerged, and many recent studies have been devoted to understanding the determinants of child support payments (e.g., Beller & Graham, 1993; Garfinkel & Robins, 1994; Hill, 1992b; King, 1994; Seltzer, 1991). Yet what is generally missing is information from and about the nonresident parent. Due to data limitations, most studies attempt to explain the receipt of child support by relying on the characteristics of the resident parent. Our research evaluates the implications of having data on only the resident parent in child support analyses and, more broadly, the implications of ignoring nonresident parents in national data collection efforts. Drawing on new, matched ex-partner data from the Panel Study of Income Dynamics (PSID), we examine two main issues. First, using descriptive statistics as well as an extension of our multivariate models, we examine whether significant differences exist in levels of child support received and paid, as reported by the resident parent and nonresident parent, respectively. Second, using bivariate tobit models, we assess the relative merits of predicting child support payments using solely the characteristics of the resident parent, compared with using the nonresident parent's or both parents' characteristics. We also examine the extent to which it is necessary to have both parents' reports of the dependent variable-child support-by assessing whether omitted variable bias is present in the data. BACKGROUND AND CURRENT INVESTIGATION Because most children in single-parent families are poor, it is not surprising that policymakers and social scientists are focusing on child support as a potential way to ensure the well-being of future generations. …

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call