Abstract: This paper investigates the effects of school starting age on crime while relying on variation in school starting age induced by administrative rules; we exploit that Danish children typically start first grade in the year they turn seven, which gives rise to a discontinuity in children's school starting age. Analyses are carried out using register-based Danish data. We find that higher age at school start lowers the propensity to commit crime. Importantly, we find that the individuals who benefit the most from being old-for-grade are those with high latent abilities whereas those with low latent abilities seem to be unaffected or even worse off by being old-for-grade in school. Moreover, for girls much of the crime reduction stems from an incapacitating effect of schooling as the effect fades in the long run. For boys, on the contrary, we find the reduction in the crime rate to be permanent. Finally, we show how these results are caused by absolute school starting age and not one's age relative to the age of one's peers.