Conscription and educational outcomes.

Peacetime military service has both positive and negative effects on human capital. While it depreciates academic skills, it also enhances non-cognitive ones. The net effect of conscription is hard to identify due to issues of self-selection, endogenous timing and omitted variables bias. We exploit the compulsory service of men in the Republic of Cyprus preceding university enrolment to deal with the first two problems. After controlling for prior academic performance and other relevant controls in a selection on observables model, we find that duration of service has a positive and significant effect on men's subsequent academic performance as measured by grade point average. Two exogenous reforms-one at the extensive margin and one at the intensive margin of military service-allow us to deal with omitted variables bias. We estimate difference-in-differences models, where female students act as a control group, and show that an increase (reduction) in the average length of army service has a positive (negative) and significant effect on men's academic performance.

Supplementary information: The online version contains supplementary material available at 10.1007/s00148-023-00944-2.