Intergenerational Risks of Criminal Involvement Submitted for National Poverty Center Research Grant Proposal

Project Summary: The enormous increase in incarceration led to a parallel, but far less well-documented, increase in the proportion of children who grew up with a parent incarcerated at some point during their childhood. Moreover, the concentration of these incarceration trends among less educated African-Americans has resulted in a larger gulf between the early-life experiences of white and black children, which may have profound effects on their later-life chances. The implications for child well-being of policy-induced increases in the incidence of parental incarceration are not well understood. The proposed study seeks to provide evidence on three related questions: 1) intergenerational correlations in the likelihood of criminal involvement (arrest, conviction, incarceration) 2) prevalence of parental incarceration by race/socioeconomic status, and effects of parental incarceration on child outcomes, including youth crime 3) assess importance of neighborhood and family background in influencing deviant behavior in adolescence and criminal involvement in adulthood; including analysis of intertemporal correlations of risky behaviors in adolescence and adulthood. This project proposes to provide some of the first nationally-representative estimates of the intergenerational correlations in the likelihood of criminal involvement (arrest, conviction, incarceration)—and effects of parental incarceration on child outcomes, including youth crime. Using longitudinal data of a nationally-representative cohort of children born between 1951-1975 and followed through 2003, I estimate sibling and childhood neighbor correlations in adult risk preferences, adult criminal involvement, deviant behavior, risky health behaviors—age of onset of cigarette smoking initiation in adolescence and smoking behavior in adulthood—to assess the relative importance of neighborhood and family background on the early formation of these behaviors. I begin by using the restricted-use, geocoded version (at the neighborhood census block/tract level) of the Panel Study of Income Dynamics (PSID) and its Child Development Supplement (CDS), to produce nationally-representative estimates over the life cycle of the cumulative risk of being suspended/expelled from school, likelihood of being placed in a reform school, probability of ever being charged or convicted of a crime, and the risk of death or imprisonment. These descriptive statistics are presented separately by gender, race, education, and parental socioeconomic status. Using the PSID-CDS, I will provide evidence on a series of important descriptive questions regarding how often white, black, and Hispanic children experience parental incarceration, how the risk has changed over the past 25 years (recent birth cohorts versus older birth cohorts from other data sources), and how this risk varies within racial/ethnic groups. I will present evidence …