Draft lottery data combined with Danish longitudinal administrative records show that military service can reduce criminal activity for youth offenders who enter service at ages 19-22. For this group property crime is reduced for up to five years from the beginning of service, and the effect is therefore not only a result of incapacitation while enrolled. We find no effect of service on violent crimes. We also find no effect of military service on educational attainment and unemployment, but we find negative effects of service on earnings. These results suggest that military service does not upgrade productive human capital directly, but rather impacts criminal activity through other channels, for example by changing the attitudes to criminal activity for this group.