Three Essays on Antipoverty Programs and Reductions in Child Maltreatment



Journal Title

Journal ISSN

Volume Title



Millions of children are reported maltreated in the United States each year. In addition to the costs imposed on victims, maltreatment also imposes costs on society at large, including short and long-term medical care, reductions in education and workforce productivity, and increased criminality, among other costs. Finding ways to reduce maltreatment risk is a critical concern of public policy. Because poverty is a primary risk factor for child maltreatment risk, researchers have considered whether antipoverty programs might reduce child maltreatment. This dissertation consists of three papers discussing and assessing the effects of antipoverty programs on child maltreatment. The first paper discusses theories that explain the relationship between poverty and child maltreatment generally – primarily family stress and family investment models – and then considers literature on the relationship between child maltreatment and several antipoverty programs and policies including the Earned Income Tax Credit, the Supplemental Nutrition Assistance Program (SNAP), Medicaid, Temporary Assistance for Needy Families (and its predecessor, Aid to Families with Dependent Children), and the minimum wage. Findings from the literature suggest strong theoretical reasons to expect that antipoverty programs should reduce maltreatment risk, that there are correlations between antipoverty program increases and reductions in child maltreatment, and some more recent causal studies demonstrate policies can reduce maltreatment risk. The second and third papers each consider a different antipoverty program (Medicaid and SNAP, respectively) and use variation in policy decisions at the state-level to assess effects on child maltreatment outcomes. Paper two considers Medicaid’s effects on child maltreatment. First, the paper discusses why Medicaid might reduce maltreatment risk, both via the proposed theoretical models relating to socioeconomic status and via changes in health care utilization for both adults and children. Because the Patient Protection and Affordable Care Act’s (ACA) Medicaid expansions, which were originally mandatory, were rendered optional by the U.S. Supreme Court’s decision in National Federation of Independent Business v. Sebelius, state selection to expand or not expand Medicaid functions as a sort of natural quasi-experiment. This paper exploits variations in state selections to expand or not expand Medicaid to identify the causal effect of Medicaid expansion on child maltreatment outcomes. Prior to the ACA, a number of states had state-funded programs expanding Medicaid similar to ACA Medicaid expansions. Some states also chose to expand Medicaid early or to partially expand their programs. Due to these and other variations, precisely defining which policy changes constitute “Medicaid expansion” can be complicated; this paper considers several different definitions of expansion to examine whether inclusion or exclusion of some states affects results. The paper finds that January 2014 Medicaid expansions led to reductions in child neglect, but the robustness of that result is sensitive to which states are included in the sample in terms of when they expanded and the generosity of their prior Medicaid coverage. The third paper considers the relationship between SNAP and child maltreatment. The paper first considers theoretical reasons why SNAP might affect child maltreatment risk, including family stress and family investment models, and then considers additional factors relating specifically to food insecurity. Then the paper turns to empirically assessing whether SNAP leads to reductions in child maltreatment. It exploits variation in state decisions regarding broad-based categorical eligibility (BBCE) in SNAP, a policy which increases the number of people eligible for SNAP and can also simplify application processes. State selection of BBCE leads to reductions in neglect and sexual abuse, and some findings indicate BBCE may also reduce physical abuse and medical neglect, though those findings are sensitive to model specification. This dissertation discusses several theoretical reasons why antipoverty programs should reduce maltreatment risk, assesses literature on several programs, and empirically assesses the causal effect of two programs – Medicaid and SNAP – on child maltreatment outcomes. Findings suggest Medicaid expansions may reduce neglect, though the results are sensitive to how Medicaid expansion is defined, and that SNAP broad-based categorical eligibility reduces neglect and sexual abuse. Results support the proposed theoretical models and more generally support the idea that antipoverty policies and programs may reduce child maltreatment.



Antipoverty programs, Child maltreatment, Medicaid, Supplemental Nutrition Assistance Program