Studies have shown the impact of a population’s age structure on the crime rate. Germany is — like many other industrialized countries — facing an ageing of its population. This trend will continue in the future: Until the year 2020 the share of younger people aged 14 to 24 years will decrease from 12.3 % to 10.7 % and the share of elderly persons aged 60 years and older will increase from 25.9 % to 30.1 %. Crime is, however, not only influenced by age, other factors also play an important role. Research has shown that the level of social disorganization is especially related to the crime rate. The aim of the present contribution is to explain the crime trends between 1995 and 2010 using multivariate panel estimators that take into account the demographic changes and social disorganization. These models are in a second step used to forecast the crime trends until the year 2020. The data base consists of pooled time-series at the county level from four German states (Bavaria, Brandenburg, Lower Saxony and Saxony-Anhalt). The results show that the age structure plays only a limited role in explaining the past crime trends. The most important factor is residential instability. The forecasts expect a decline of the number of registered offences till 2020. However, the decline will be faster in the eastern states than in the western states and some offences are expected to increase in the future.