Is Foster Youths' Juvenile Justice System Involvement Context-Dependent?: A Life Course Approach to the Differential Experiences of Older Foster Youth

Background: Foster youth are at increased risk of entering the justice system during their transition to adulthood. However, little research has examined how state systems can provide youth with structural support to prevent juvenile justice involvement (JJI). California adopted federally-funded extended foster care (EFC) in 2010 and began offering those services in 2012. From the life-course perspective (Elder, 1994), this policy change marks a critical shift in the opportunity structure (i.e., extending systemic supports beyond age 18) for transition-age foster youth, and thus presents an opportunity to evaluate whether a socio-historical event influenced youths’ life experiences. To improve our understanding of contextual factors on crime, this study seeks to address two questions: Is EFC policy implementation associated with changes in the risk that youth would experience JJI? If so, does the policy change affect JJI differently for foster youth closer to and further from the age of majority (17-year-olds vs. 16-year-olds)?

Methods: Analyses draw from 2006-2016 California child welfare data on individuals in care between their 16th and 18th birthdays (n=69,140). Other juvenile court statistics data were used to account for California juvenile delinquency rates. Differences-in-differences analysis was used to examine JJI between 16-year-olds and 17-year-olds during the periods before and after California implemented EFC. In this analysis, Cox models examined whether the policy had differential effects on the timing of JJI by age group. The outcome variable was time until JJI for youth who moved from child welfare-supervised placements to probation-supervised placements for the first time prior to their 18th birthday (or censoring due to other exits from care). Control variables included demographic characteristics, maltreatment and foster care history, and California delinquency rates. The independent variables included two grouping measures: policy period was dichotomously measured using “pre-EFC” (2006-2011) and “post-EFC” (2012-2016); age-cohort membership was dichotomously measured using “16.00-16.99” and “17.00-17.99”.

Results: Foster youth in this study exhibited higher rates of JJI than the state youth population (31.1 vs. 15.4 per 1,000 youth). JJI was lower after the EFC policy than before in both age groups. Regression results found a significant association of JJI with the EFC policy and age-cohort. The estimated hazard ratios (HRs) of CJI were significantly lower for youths in the post-policy period than in the pre-policy period (HR=0.46, p<.001) and for 17-year-olds than 16-year-olds (HR=0.77, p<.001). The DID estimator was not significant, suggesting there was no evidence that the policy affected 17-year-olds differently than 16-year-olds.

Conclusions/Implications: This is the first study examining effects of the EFC policy on JJI for minors. We did not find evidence of a differential policy effect by age, but found that the policy may have a general impact on JJI. This study confirms that structural supports and social context are key contributors to JJI of older foster youth. Paying attention to important social contexts surrounding foster youth may help inform strategies (e.g., policy development, targeted service delivery) that can facilitate healthy transitions to adulthood for foster youth. Future research examining the mechanisms by which EFC influences JJI seems warranted.