Confinement as a two-stage turning point: Do changes in identity or social structure predict subsequent changes in criminal activity?^*

Scholars frequently characterize incarceration as a possible turning point in criminal activity. This implies a two-stage process: 1) change in life-course mechanisms around confinement and reentry result in 2) subsequent change in criminal activity relative to preconfinement. Following this model, we examine change in criminal activity, criminal identity, and social/structural challenges using data from the Prison Project, a cohort of adult males with short-term confinement in the Netherlands in 2010–2011. Results of a novel test for within-individual change in arrests from preconfinement to post-reentry show that most individuals are stable—yet there is a substantial group who go down meaningfully and a much smaller group who go up. Even though changes in criminal identity from the intervening period do not predict these change groups, increases in social/structural challenges predict those who go up in criminal activity. We build from prior work on desistance and reentry, contrasting our findings and highlighting the unique insight gained from, as well as challenges of, measuring individual change within our two-stage turning point model. Although life-course mechanisms often correspond with changes in criminal activity concurrently, identifying individual changes that are predictors of subsequent shifts in criminal offending remains elusive.