High crime rates among second-generation immigrants are usually attributed to the ethnic group’s weak socioeconomic position in the host society. The causes of crime can, however, also be sought in their native countries or regions. Owing to a lack of empirical data, this has rarely been tested. The Netherlands is an exception: small-scale ethnographic case studies among young Moroccan men in Dutch cities suggest that their regional background and culture, particularly if they are from the less developed Rif Mountains area, may explain their high crime rates. In this article we examine whether this applies to the results of a quantitative study on all the Moroccan male juveniles in the Netherlands. At the individual level, our unique dataset is a combination of their native regions, criminal records (ever suspected of a crime) and demographic and socioeconomic characteristics. Logistic regression analysis shows that current socioeconomic position is a strong predictor of ever having been suspected of a criminal offence, and the impact of geographical descent, directly or indirectly via socioeconomic position in the Netherlands, is negligible. Accordingly, our findings on Moroccans in the Netherlands do not warrant our questioning the common explanation of the immigrant–crime connection in criminology and suggest further research to determine the specific host society’s features that explain the overrepresentation of ethnic groups in crime statistics.