This paper presents an analysis of the impact of political regimes and type of military recruitment on the probability of the occurrence of international conflicts. In the last few years, the (re) introduction of military conscription has been a focus of public debate, but empirical analysis of the issue remains limited. We argue that democratic nations with conscription-based military recruitment in place are less likely to initiate international conflicts than those with voluntary recruitment because public opinion will estimate a higher probability of direct involvement in disputes, causing political leaders to refrain from conflicts, even though stable military resources are in place. On the other hand, authoritarian nations with conscription-based recruitment systems are more likely to engage in conflicts than those with voluntary recruitment systems because political leaders are not accountable to the people, even though the cost of war is calculated in the same manner as that in democratic nations. To test this reasoning, we use directed-dyadic data from 1816 to 2005. Our analysis strongly supports our theoretical expectations.