Association for Public Policy Analysis & Management

Panel: Poverty, Job Loss, Family Structure, and Child Maltreatment: Evidence from New Data and Methods

(Family and Child Policy)

Policymakers and researchers have long struggled to identify the etiology of child maltreatment and to develop effective prevention programs and policies. Prior research has frequently identified poverty, single parenthood, job loss, mental health, and family structure, among others, as primary predictors of child maltreatment. Despite the large numbers of children who experience some form of maltreatment every year in the United States, little causal empirical research exists that identifies its antecedents. This panel takes up the question of whether traditional indicators of maltreatment, including poverty, job loss, and family structure, are indeed predictive of child maltreatment. The first paper draws on a unique randomized evaluation of an income support intervention intended to prevent child maltreatment and CPS involvement. This work represents an important step forward in a nascent literature examining the causal effect of income on child maltreatment (see e.g., Berger, Slack, & Noyes, 2017). The paper distinguishes between the effect of both public and private income transfers and may offer important insight into the role of public policy in preventing child maltreatment. Similarly, the second paper draws on a growing body of research about the exogenous effect of job losses on child maltreatment and intimate partner violence (Gassman-Pines et al., 2015; Schneider et al., 2016). Linking data from a variety of sources, the author offers insights into how job losses for men and women and by racial/ethnic background may be differentially associated with child maltreatment and intimate partner violence. Importantly, the paper also examines how traditional responses to job loss – Unemployment Insurance and Trade Adjustment Assistance – respond in a variety of contexts. The third paper re-examines three often-cited findings in the child maltreatment literature by drawing on a large longitudinal birth cohort study – (1) that first births are at increased risk for child maltreatment; (2) that having many children in a household is associated with an increased risk for maltreatment; and (3) that these associations are particularly prevalent among low-income, unmarried, and younger mothers. The author finds that while first birth is associated with the risk for child maltreatment, the number of children in the household is associated with a decrease in the risk for maltreatment, and that neither poverty nor marital status moderate the associations. A number of interventions have been developed targeted at new mothers and results from this work may help to inform policymakers in program design. Together, these three studies utilize unique experimental data, combine data from disparate sources, and apply longitudinal data in previously unused ways to answer important and policy-relevant questions about how best to prevent child maltreatment. Furthermore, each of the studies represents an important methodological advance, through randomized control trial, exogenous effects, and individual fixed effects methods, in the study of child maltreatment.