Abstract Background Many studies have identified urban–rural differences in the occurrence of schizophrenia. Though unknown, the underlying causes responsible for these differences have been hypothesized to include urban–rural differences in toxic exposures, diet, infections, stress, or an artifact due to selective migration. Freeman hypothesized that persons with schizophrenia migrate towards larger cities due to development of their disorder or its prodromata. Based on this, the reason for the high frequency of schizophrenia in urban areas is not that those affected have lived in environmentally unfavorable areas, but that people with schizophrenia selectively migrate towards urban areas due to disease onset. No population-based studies accessed the extent and potential impact of this artifact of selective migration. Methods Utilizing a population-based sample of the Danish population, it was investigated if persons with schizophrenia more often migrated towards larger cities due to disease onset. The impact of selective migration on the urban–rural differences was quantified comparing a prospective and a retrospective study. Results Compared to healthy controls, persons with schizophrenia spectrum disorder migrate to a higher degree of urbanization due to disease onset (OR = 1.18 (1.14–1.23)). However, the bias in urban–rural effect sizes due to this artifact of selective migration was limited. Conclusion Persons with schizophrenia drift towards urban areas as a consequence of the disorder or its prodromata, but this drift has only limited impact on the urban–rural differences. Therefore, prospective and retrospective studies are both informative on the unknown underlying factor or factors responsible for the urban–rural differences in schizophrenia risk.