Safe from Harm?

Abstract

More than 20 percent of young adult prison inmates in the United States have spent ;me in foster care, among whom a majority have lived in a congregate care (group-based) seCng. In these seCngs, youth are exposed to non-related peers, who may influence long-term social and behavioral outcomes. Using three decades of administra;ve data from Wisconsin and exogenous varia;on in the rela;ve imprisonment risk of foster care peers, we study how peer composi;on affects youths’ future criminal jus;ce system contact, educa;onal aLainment, and short-term risky behavior. We consider peer composi;on along three dimensions: (i) the share of peers placed in foster care for delinquency-related reasons, (ii) machine learning predic;ons of the future probability of imprisonment, calculated using pre-placement child and family characteris;cs, and (iii) the share of peers with formerly imprisoned parents. We find that peer characteris;cs have no effect on a youth’s likelihood of entering prison by age 20, nor on a number of other indicators of deviant behavior. These results hold across group homes, residen;al facili;es, and temporary placement seCngs, and are rela;vely robust to placement length and youth demographics. A one standard devia;on increase in predicted peer imprisonment risk is associated with a modest, marginally significant one percentage point (three percent) increase in a youth’s likelihood of dropping out of high school. Our findings have implica;ons for the recent movement to reallocate children away from congregate care.