This paper examines how vocational education reforms can generate sizable crime-reduction externalities at the neighborhood level. We study the staggered rollout of Dual Vocational Training across Catalonia, starting in 2014. This new approach shifted vocational instruction toward firm-based apprenticeships, expanding workplace training from under 500 hours to over 1,000 hours and introducing subsidized paid apprenticeship contracts. Adoption differed across schools over time, generating policy-relevant quasi-experimental variation. We estimate difference-in-differences models with continuous treatment intensity at the census-tract level. The analysis links administrative records for the universe of Catalan vocational students to daily geocoded police reports (2008–2018). The results show that a 1 percentage-point increase in the dual track share reduces reported youth crime by approximately 13%. Effects build gradually and strengthen over the second and third years, consistent with persistent behavioral change rather than purely mechanical incapacitation. Reductions concentrate during daytime hours (when apprentices are typically at firms) consistent with incapacitation. Declines are strongest for economically motivated property crimes (especially theft), consistent with increased opportunity costs through earnings and improved labor-market prospects. These results imply that apprenticeship-style programs can deliver substantial non-pecuniary social returns through reduced criminal externalities and, therefore, may play a role in disrupting intergenerational disadvantage.
