The Causal Effects of Parental Incarceration on Child Wellbeing: Evidence from a Judge Design

• In Event: Courts, Crime, and Consequences: Using Administrative Data to Estimate Causal Effects of Court Interactions

Abstract

Parental incarceration is widely associated with adverse outcomes for children, yet few studies can directly estimate its causal effects. In this study, we estimate the causal effects of incarceration on children’s development, health, and early adulthood outcomes using a judge instrumental variables (IV) strategy and comprehensive administrative data from Wisconsin. We leverage over 300,000 criminal cases involving parents and identify children from two statewide birth cohorts. Children and families are linked to detailed records on parental criminal justice involvement, public assistance, education, health care, and child welfare, enabling us to trace detailed effects of parental incarceration. By exploiting the quasi-random assignment of judges and their differential sentencing tendencies, we identify local average treatment effects of incarceration among otherwise similar families. This strategy isolates the effect of incarceration itself, holding constant the nature of the criminal charge and other case characteristics. We estimate impacts on a wide range of outcomes—academic achievement, behavioral and physical health, teen parenthood, criminal justice involvement, and employment. We also assess heterogeneity in effects by child age, race/ethnicity, family structure, and parental co-residence, and explore mechanisms including economic changes and caregiver disruptions. These results underscore the intergenerational consequences of incarceration and have critical implications for sentencing policy, diversion programs, and supports for families involved with the justice system.

For reviewers: The project uses restricted access data. Results are embargoed until a 30 day review by the state is passed. We anticipate no issues for reporting results and expect to have public results by August.

Author

• Ezra G. Goldstein, Georgia Institute of Technology
  Presenting Author