Two important determinants of crime rates in economic theories of crime are deterrent effects and legal income opportunities. For at least two reasons, youth crimes do not fit into this picture since: 1) most youths do not work 2) deterrent effects, in terms of punishment, are non existent or reduced for youths. Understanding the processes behind youth crimes is important for the reduction of crime both in the short and the long run. This paper explores the role of public spending on youths and crime rates. Using a panel of 261 Swedish municipalities the effects on four crime rates of leisure related municipality expenditure and municipality expenditure on upper secondary school are explored using non-linear fixed effect models. The main findings of the paper are: 1) there exists statistically significant effects of the two types of spending on crime rates 2) there is a trade off between fighting different crimes, i.e. spending that reduces one type of crime might increase another type 3) the effects differ, both in direction and magnitude, between different types of municipalities, e.g. rural vs. urban 4) the effects are not constant, but rather increasing/decreasing in spending level.