Military service, along with marriage and employment, is often cited as a positive turning point in the lives of criminal offenders. Despite the findings of studies based on WWII and Korean War veterans, the evidence that military service in other eras (e.g. the Vietnam War) has had enduring effects on crime is mixed. This could indicate that the impact of military service varies over time in response to changes in recruitment, the experience of service and reintegration, or the support veterans receive (e.g., the GI Bill). Alternatively, these discrepancies could reflect methodological differences, including inconsistencies in how researchers have dealt with the selection biases associated with military recruitment. To investigate these issues, I analyse historical census data. Taking advantage of between-cohort variation in the probability of enlistment, I evaluate the impact of service in WWII, the Korean War, and the Vietnam War on the risk of incarceration at key stages of the life-course. Results indicate that military service had modest, negative effects on imprisonment among men who served in WWII, but little impact on veterans of other wars. I discuss the implications of these findings for life-course criminological theory and research and our understanding of the life-course consequences of armed conflict.