Article: Silver, Ian A., Meckes, Jessica, and Dole, Jenna L. 2026. “Exploring the Influence of Criminal Legal System Involvement Before 18 on Residential Mobility.” Journal of Criminal Justice, 106:102734. https://doi.org/10.1016/j.jcrimjus.2026.102734
1. Background (PDF)
Moving to a new home or a new community is often seen as a burden, but for criminal legal system (CLS) involved individuals it can offer an opportunity to start new. Prior research suggests that formerly incarcerated adults often move frequently but rarely escape high-poverty neighborhoods when they relocate. Far less is known about how CLS contact during adolescence shapes residential mobility during adulthood. This study is the first national study to examine whether being arrested, incarcerated in a juvenile facility, or incarcerated in an adult facility before age 18 affects both within-state moves (which is perceived to preserve access to family, jobs, and familiar services) and out-of-state moves (which is perceived to offer more separation from past environments but demand more resources). Both within-state and out-of-state moves could offer new opportunities for individuals who were involved in the CLS before 18.
2. Results
Using the National Longitudinal Survey of Youth 1997 (NLSY97) and a quasi-experimental design, the study assessed the effects of being arrested, incarcerated in a juvenile facility, or incarcerated in an adult facility before age 18 on within-state moves and out-of-state moves. The findings suggested that:
- Being arrested before 18 slightly reduced the likelihood of ever moving within-state, but modestly increased the number of within-state moves and the number of out-of-state moves compared to youth with no CLS contact.
- Being incarcerated in a juvenile facility before 18 reduced the likelihood, frequency, and increased the time to a first within-state move. Also reduced the frequency of out-of-state moves and delayed the first out-of-state move by about 13%. Overall, this group showed the most consistent pattern of reduced and delayed mobility.
- Being incarcerated in an adult facility before 18 increased the likelihood and frequency of within-state moves, but reduced the likelihood and frequency of out-of-state moves, and delayed a first out-of-state move by 58% compared to youth with no CLS contact. This group appeared to move more often, but stayed within the same state.
CLS involvement before 18 was generally associated with an individual’s first move occurring later in life (increased number of years until first move).
3. Implications
These findings suggest that individuals exposed to the CLS before 18 might need housing navigation to help establish themselves within a community that promotes prosocial behavior. This effort might have to exist for a number of years after system involvement, as the effects of CLS involvement could extend well into adulthood. additional efforts could include: 1) Housing vouchers or transitional housing assistance targeted at formerly CLS-involved youth, 2) Brief (1–2 day) housing/relocation seminars integrated into existing reentry or vocational programming, and 3) Employment support, since limited job prospects are a plausible reason CLS-involved youth stay in place even when relocation might benefit them.
A caution for interpretation: more mobility isn’t automatically good, and less mobility isn’t automatically bad. Frequent within-state moves among adult-facility youth may reflect instability rather than opportunity. Researchers should try to understand why individuals are or are not moving after CLS involvement before 18.
4. Data and Methods
The study used the NLSY97, a nationally representative U.S. birth cohort of 8,984 individuals born between 1980 and 1984. The analytical sample included 8,961 individuals who were 12–17 years old at the 1997 baseline. CLS involvement before 18 (arrest, juvenile facility incarceration, adult facility incarceration, or no contact) was based on monthly arrest/incarceration summaries from 1992–2021 and juvenile sentencing data from 1997–2002.
Residential mobility was tracked from ages 18–41 using six outcomes: whether a within-state move occurred, the number of within-state moves, and years until the first within-state move; and the same three measures for out-of-state moves. A within-state move meant relocating to a different county within the same state; an out-of-state move meant relocating to a different state (or a foreign country).
Gradient boosted inverse probability weights (an Average Treatment Effect approach) balanced 20+ covariates across the four CLS groups. These weights were applied in logistic regression models (ever moved), Poisson regression models (number of moves), and lognormal survival models (time to first move). Missing data were addressed using random forest imputation, and results were replicated across negative binomial models and five separate imputed datasets for robustness.
5. Conclusion
This study is the first national investigation of how youth CLS involvement shapes both the likelihood and timing of adult residential mobility and the first to separate within-state from out-of-state moves. The results show that incarceration before 18, particularly in juvenile facilities, is associated with reduced and delayed mobility of nearly every kind, while incarceration in adult facilities produces a more complex pattern: more local movement, but far less ability to leave the state. Together, these findings suggest that youth CLS involvement can constrain the kind of geographic mobility that supports fresh starts, reinforcing the case for reentry-focused housing and employment supports that extend well beyond the immediate post-release period.
Disclosure: This research brief was prepared by Claude and reviewed/edited by Ian A. Silver
ianasilver.com 