Abstract. We develop a dynamic discrete choice model of schooling and career decisions to explore the effect of compulsory military service on men's earnings and education in Iran and fit the model to actual data from household surveys. We show that, on average, conscription has a significant negative impact on men's lifetime earnings, but the earnings gap disappears in older ages. The counterfactual experiments suggest that experience loss due to compulsory military service is the main mechanism explaining the earnings gap over the lifetime. In contrast, the earnings loss during conscription has a short-term effect on the lifetime earnings in younger ages, but the experience mechanism dominates its effect in the 30s. The third channel through which conscription affects earnings is changing schooling decisions, but its overall effect is lower than the two other mechanisms. However, unlike experience and earning losses, the schooling channel's impact is dissimilar among different individual types. While pursuing higher education to defer conscription reduces the lifetime earnings of one type, in another type, such a draft avoidance behavior leads to higher lifetime earnings.