Essays in child welfare and long-run economic well-being

ABSTRACT

This dissertation is composed of three papers, all using modern applied econometric methods that try to isolate causal estimates in several important settings. Each chapter examines social and formal institutions on the well-being of children.The first chapter studies the importance of parental presence for children's adulthood labor market outcomes. Through interactions with their parents, a child develops the basic foundations for cognitive and non-cognitive skills that directly shape their adulthood economic well-being. Motivated by this critical role, a large body of research has sought to uncover the consequences of disruptions to family structure and parental inputs on a child's development. Despite its importance, we know relatively little about the consequences of the most severe form of parental absence the death of a parent. Empirical evidence of the long-run causal effects of parental death has been limited since it requires sufficient longitudinal data and plausibly exogenous variation in parental mortality. To overcome these challenges in the literature, I make use of a rich dataset of over 180,000 historical mining accidents and link individual accident victims to the full-count U.S. Census. Doing so allows me to follow the sons of mining accident victims through to adulthood and study the causal effects of parental death on economic well-being. To identify the causal impact of parental death, I compare the adulthood outcomes of children of fatal mining accident victims to children whose parents suffered a serious but non-fatal mining accident. I find that, compared to children of non-fatal mining accident victims, bereaved children experienced nearly four percent lower wage income during adulthood. Further analyses reveal the most severe effects stem from those that lost their parent at an early age. Specifically, adults who were younger than primary school age when they lost their fathers had roughly 15 percent lower wages. Exploring potential channels, I show that most of the estimated earnings penalty can be attributed to differences in employment along both the intensive and extensive margins and is not due to differences in human capital accumulation. Bereaved sons were more likely to be out of work, report unemployment assistance, and work fewer weeks. Together, these employment channels can account for more than 60 percent of the estimated loss of adulthood income. The second chapter of this dissertation, coauthored with E. Jason Baron (a classmate at Florida State University) and Joseph P. Ryan (Professor of Social Work at the University of Michigan), examines the sources of over-representation of Black children in foster care systems and the causal effects of one popular reform that seeks to reduce this over-representation. The program, known as "blind removals," conceals demographic characteristics of children at-risk for foster care placement from child welfare professionals weighing whether or not to remove the child from their home. We present the first quantitative analysis of blind removals by examining a comprehensive administrative dataset of the universe of child maltreatment investigations in Michigan and presents two main findings. First, the over-representation of Black children in foster care systems is almost entirely driven by Black children being twice as likely to be investigated for child maltreatment as White children. Conditional on initial rates of investigation, White and Black children are placed in foster care at similar rates. Second, the study finds no evidence that blind removals impacted the already small racial disparities in the removal decision, but the program substantially increased the time to removal.The final chapter of this dissertation, coauthored with two classmates from Florida State University (E. Jason Baron and Cullen T. Wallace), highlights the link between educators... (PsycInfo Database Record (c) 2022 APA, all rights reserved)