The presence of crime is often considered as one of the most important social variables to be negatively associated with economic development. As is well known, Italy exhibits an economic divide between its central‐northern and southern regions. This paper tries to ascertain whether there is also a divide in the diffusion of crime. To this purpose, Italian provinces are classified into groups by using a new methodology that combines cluster analysis with time series cross‐section data. The statistical searches for groups are carried out by using an unbalanced time series cross‐section data‐set made up of yearly observations. The results of the analysis show that crime is not inextricably tied to certain geographical locations, and not others, as is usually believed.