Numerous studies have shown that childhood interventions can foster improved outcomes in youth and adulthood. Less well understood is precisely how – that is, through which developmental pathways – the interventions produce the positive outcomes. This study seeks to assess various mechanisms by which a randomized childhood intervention, called Fast Track, reduced young adult delinquency and arrests. Decomposition of the overall treatment effects allows a comparison of program components that aim to promote different personal competencies, thus illuminating which kinds of skill enhancements most effectively lessen the likelihood of criminal behavior. First, this paper identifies a key set of cognitive, self-regulatory, and social capabilities that the Fast Track program affected. It then statistically dissects Fast Track’s impacts on adult and juvenile arrests and delinquent behavior, separating treatment effects into: 1) those plausibly driven by increases in each type of participant capability; and 2) those driven by other mechanisms. Although the analysis leaves a majority of effects unexplained, it indicates that improvements in a few capabilities account for a significant portion of Fast Track’s impact on arrests and delinquency. In particular, the results suggest the efficacy, for crime prevention, of early interventions that target language and reading skills, emotional regulation, and prosocial behavior.