Theses Doctoral

Social Policy and Family Well-Being: Essays in Applied Microeconomics

Rossin-Slater, Maya

In my dissertation, I study how individuals respond to changes in their options and constraints as a result of government policies and their local environments. I focus on issues in maternal and child well-being, as well as family structure and behavior, and draw implications for addressing the needs of disadvantaged populations in the United States. I use quasi-experimental empirical strategies with large and varied data sets to provide credible causal estimates. I believe that the results from my research can shed some light on the causes and consequences of disadvantage in the United States, contribute to cost-benefit analyses of some of the largest social welfare programs, and help inform decisions about public spending. The focus on maternal and early childhood well-being is motivated by increasing support for the notion that fetal and infant health are predictive of individuals' later-life outcomes (Almond and Currie, 2011a,b). This evidence highlights the potential value in programs and policies aimed at pregnant women and new mothers. Indeed, successful programs that improve the welfare of disadvantaged women during pregnancy and post-partum may play an important role in ameliorating inequalities at birth, and thereby potentially mitigating the intergenerational transmission of low socio-economic status. In the first essay, titled "The Effects of Maternity Leave on Children's Birth and Infant Health Outcomes in the United States" (published in the Journal of Health Economics, March 2011), I provide the first quasi-experimental analysis of the effects of the unpaid maternity leave provisions of the 1993 Family and Medical Leave Act (FMLA) on children's birth and infant health outcomes in the United States. My identification strategy uses variation in pre-FMLA maternity leave policies across states and variation in which firms are covered by FMLA provisions. Using Vital Statistics data and difference-in-difference-in-difference methodology, I find that maternity leave led to small increases in birth weight, decreases in the likelihood of a premature birth, and substantial decreases in infant mortality for children of college-educated and married mothers. The fact that I only find positive impacts on the health of children of college-educated and married women, while children of less-advantaged women experience no health benefits, suggests that unpaid parental leave policies may exacerbate disparities in child health as they only benefit the parents who can afford to use them. In the second essay, "Engaging Absent Fathers: Lessons from Paternity Establishment Programs," I examine behavior among parents who have children out-of-wedlock. Single-mother households are disproportionately disadvantaged, and children raised in two-parent households fare better along numerous measures of well-being. These facts motivate the implementation of policies that encourage father involvement among unmarried parents. I conduct the first comprehensive causal analysis of one of the largest U.S. policies that aims to engage unmarried fathers with their families, In-Hospital Voluntary Paternity Establishment (IHVPE), and place my findings in the context of a conceptual framework rooted in family economics theory (Edlund, 2011; Browning, Chiappori, and Weiss, forthcoming). The program significantly reduces the costs of formal paternity estabishment, which is the only available legal contract that assigns partial parental rights and obligations to unmarried fathers. Using data from a multitude of sources and variation in the timing of IHVPE initiation across states, I show that IHVPE achieves its stated goal of substantially increasing paternity establishment rates. However, I show that IHVPE also affects another margin of parental behavior. I find a \emph{negative} effect on parental marriage -- specifically, for each additional paternity established as a result of IHVPE, there are 0.13 fewer parental marriages occurring post-childbirth. Accounting for the decline in parental marriage, I find that the net effects on some measures of father involvement are negative, while overall child well-being is largely unaffected. Why might paternity establishment serve as a substitute to marriage for some parents? To explain this finding, I offer a simple conceptual framework, in which parents trade-off their utility from access to children with their match quality. Paternity establishment offers an "intermediate" parental relationship option between the "extremes" of no formal relationship and marriage. When the cost of establishing paternity is lowered, parents who would have previously maintained no formal relationship and parents who would have previously been married are more likely to choose the intermediate contract. If fathers are more involved with their children when they have greater parental rights (Weiss and Willis, 1985; Edlund, 2011), then the net effect on father involvement is ambiguous, and can be negative if the increase in involvement among switchers out of no relationship is lower than the decrease in involvement among switchers out of marriage. My results suggest that the trade-off between access to children and match quality is empirically relevant for parents who have children out-of-wedlock, and policies based on the notion that more father involvement is essential to child and family well-being must account for the parents' agency in choosing their partners. A paternity establishment program that intends to engage absent fathers and increase father involvement can actually have the opposite effects by discouraging some parents from marriage and reducing the support provided by otherwise married fathers. Finally, in the third essay, titled "WIC in Your Neighborhood: New Evidence on the Impacts of Geographic Access to Clinics" (accepted at the Journal of Public Economics), I examine how geographic proximity to Special Supplemental Nutrition Program for Women, Infants, and Children (WIC) clinics affects food benefit take-up, pregnancy behaviors, and birth outcomes. WIC is the major U.S. program with a goal of enhancing the health and nutrition of low-income pregnant women and children. Rigorous evaluation of the program is necessary both for policy-making purposes and for providing new estimates of the determinants of fetal and infant health. Although there are several studies that examine the relationship between WIC and birth outcomes (e.g. Bitler and Currie, 2005; Joyce et al., 2005; Joyce et al., 2008; Figlio et al., 2009; Hoynes et al., 2011), much less attention has been paid to the determinants of WIC benefit take-up. Moreover, consensus on the effectiveness of WIC has not been reached: the existing literature suffers from problems due to omitted variables bias, lack of data on important variables such as benefit take-up and breastfeeding, and other econometric and measurement issues. I employ a novel empirical approach on data from birth and administrative records over 2005-2009 that uses within-zip-code variation in WIC clinic presence together with maternal fixed effects, and accounts for the potential endogeneity of mobility, gestational-age bias, and measurement error in gestation. I find that access to WIC increases food benefit take-up, pregnancy weight gain, birth weight, and the probability of breastfeeding initiation at the time of hospital discharge. The estimated effects are strongest for mothers with a high school education or less, who are most likely eligible for WIC services.



  • thumnail for RossinSlater_columbia_0054D_11239.pdf RossinSlater_columbia_0054D_11239.pdf application/pdf 2.95 MB Download File

More About This Work

Academic Units
Thesis Advisors
Kopczuk, Wojciech
Ph.D., Columbia University
Published Here
April 30, 2013