Universal Pre-K Programs and Child Maltreatment

• In Event: When Policy Meets Practice: Policies that Influence the Child Welfare Trajectory

Abstract

Background
Universal prekindergarten (UPK) expansions have become an increasingly common state policy aimed at improving early childhood education access and supporting parental labor supply, yet little is known about their unintended effects on child welfare systems. Moreover, the impact of these policies on child maltreatment is ex ante ambiguous. UPK could support employment among families, which could increase financial resources and reduce maltreatment. A second plausible mechanism is mandated reporting: increased exposure of young children to educators may raise detection and reporting of maltreatment without necessarily changing underlying incidence. Existing evidence on early childhood programs and maltreatment is limited and largely descriptive, leaving open questions about causal impacts, age specificity, and policy dynamics over time.

Data and Methods
We combine state–year child maltreatment data from the National Child Abuse and Neglect Data System with a newly compiled panel of state prekindergarten enrollment from the National Institute for Early Education Research. Using these data, we construct both a binary indicator of UPK adoption—defined as the first year in which at least 50% of four-year-olds in the state are enrolled—and a continuous measure of enrollment intensity—defined as the share of four-year-old children enrolled per state. Our analytic sample spans all U.S. states and the District of Columbia from 2005 to 2023 and includes maltreatment outcomes disaggregated by age group and maltreatment type. We estimate difference-in-differences and triple-difference regressions that compare changes for four-year-olds relative to other age groups within states before and after UPK adoption, controlling for state, year, and age-group fixed effects. We assess robustness using alternative weighting schemes, state-specific time trends, leave-one-out analyses, and placebo age groups.

Results
UPK adoption is associated with a sharp and persistent increase in reported maltreatment among four-year-olds, with no corresponding changes in overall victimization or substantiated maltreatment. Event-study estimates show no evidence of differential pre-trends and a clear post-adoption divergence that grows over time. In cumulative specifications, maltreatment reports increase by approximately 10–15 cases per 1,000 four-year-olds in later post-adoption years. These effects are robust across alternative weighting choices, remain stable when allowing for state-specific linear trends, and are not driven by any single treated state. Placebo analyses using older children (ages seven through ten) show no comparable post-adoption patterns. Models using continuous enrollment intensity indicate a dose–response relationship: states with higher four-year-old enrollment experience larger increases in neglect reporting.

Discussion
Our findings indicate that UPK expansions are associated with increased neglect reporting among four-year-olds, consistent with enhanced detection through mandated reporters rather than increases in maltreatment severity. This pattern likely reflects greater institutional contact and earlier identification of unmet needs rather than worsening parental behavior. While increased reporting may facilitate access to services and supports for some families, it also expands interactions with child welfare systems, highlighting an important institutional linkage between early childhood education policy and child protective services. More broadly, evaluations of early childhood policies should account for how program expansion reshapes reporting environments, service access, and family support systems, not solely downstream educational outcomes.

Authors

• Eric Geng Zhou, Icahn School of Medicine at Mount Sinai
  Presenting Author

• Catherine Maclean, George Mason University
  Non-Presenting Co-Author