In this paper we estimate the impact of higher early preschool access on offending behavior later in time, exploiting a nationwide public preschool expansion that took place in Spain over the 1990s. This is the first study providing evidence on the relationship between preschool and crime using a large-scale, universal-access program rather than targeted initiatives reserved for disadvantaged children. Our identification strategy relies on the staggered implementation of the Spanish preschool expansion across regions and birth cohorts. We link the education data to a unique administrative crime dataset recording the offenses committed in the region of Catalonia over the period 2009-2014. On average, a 1 percentage point increase in preschool access among the age 0-3 population yields 1.6% fewer crime actions during youth and young adulthood. Larger effects on impulsive crime categories provide suggestive evidence for non-cognitive-skills improvements being important drivers of our results. (c) 2021 Elsevier B.V. All rights reserved.