There has been intense debate over the immunological basis of schizophrenia, and the potential utility of adjunct immunotherapies. The major histocompatibility complex is consistently the most powerful region of association in genome-wide association studies (GWASs) of schizophrenia, and has been interpreted as strong genetic evidence supporting the immune hypothesis. However, global pathway analyses provide inconsistent evidence of immune involvement in schizophrenia, and it remains unclear whether genetic data support an immune etiology per se. Here we empirically test the hypothesis that variation in immune genes contributes to schizophrenia. We show that there is no genome-wide enrichment of immune loci in the largest genetic study of schizophrenia conducted to date, in contrast to six diseases of known immune origin. Among 108 regions of the genome previously associated with schizophrenia, we identify six immune candidates (DPP4, HSPD1, EGR1, CLU, ESAM, NFATC3) encoding proteins with alternative, non-immune roles in the brain. Our results suggest that genetically mediated alterations in immune function may not play a major role in schizophrenia susceptibility. Instead, there may be a role for pleiotropic effects of a small number of immune genes regulating brain development and plasticity. While our findings do not refute evidence that has accumulated in support of the immune hypothesis, they indicate a non-genetic etiology for immune processes that may be involved in schizophrenia. Whether immune alterations drive schizophrenia progression is an important question to be addressed by future research, especially in light of the growing interest in applying immunotherapies in schizophrenia.