This paper analyses four historical time series on homicide, assault, robbery and theft from Finnish and Swedish vital and conviction (judicial) statistics. It is suggested that long-term criminal justice statistics – at least in Finland and Sweden – are highly usable and that they add relevant basic knowledge to important current criminological and crime policy issues. It is concluded that law reforms are greatly influential in terms of changes in the sanctioning system (strong evidence of policy impact); but law reforms do not contribute to lasting changes in the development of crime as measured by conviction statistics (no substantiated evidence of crime impact). In the long run, crime seems to persistently evade the manifold legislative efforts to ‘combat’ it.