Objective This study aimed to explore the relationship between psychosocial job stressors and suicidal behavior (fatal and non-fatal) among Swedish men while controlling for potential confounders. Methods Population-based Swedish longitudinal cohort study of male conscripts without previous self-harm (N=1 483 310) enlisting 1968–2002. Conscription examinations included measures of IQ, stress resilience and psychiatric diagnoses. Job demand–control (JDC) exposure was assessed using the Swedish Job Exposure Matrix linked to specific occupations. Suicidal behavior among men aged 30–64 was identified in the National Hospital Register (non-fatal self-harm) and Swedish Cause of Death Register (suicide) during follow-up 2002–2014. Cox regression models were used to estimate associations between JDC category and suicidal behavior. Results In fully adjusted models, passive jobs (low demand-low control) showed the highest risk of suicidal behavior [hazard ratio (HR) 1.33, 95% confidence interval (CI) 1.25–1.43] compared to those with low strain (low demand-high control), followed by high strain (high demand-low control) (HR 1.12, 95% Cl 1.03–1.22). A lower risk of suicidal behavior was found in the active category, where levels of both demand and control are high (HR 0.64, 95% Cl 0.60–0.70). Separate analyses for suicide as outcome revealed a lower risk of suicide in persons with active jobs (high demands-high control). The passive category showed a higher risk for suicide, but the association did not remain after adjustment for stress resilience and IQ. Conclusions These results show that psychosocial job stressors among men are associated with risk for suicidal behavior. Improving job control has the potential to decrease suicidal behavior for this group.