Although the practice of military conscription was widespread during most of the past century, credible evidence on the effects of mandatory service is limited. Angrist (1990) showed that the Vietnam-era draft in the U.S. lowered the early-career wages of conscripts, a finding he attributed to the low value of military experience. More recent studies have found a mixed pattern of effects, with both negative (the Netherlands) and positive (in Sweden) earnings impacts. Even among Vietnam era draftees, Angrist and Chen (2011) find that the net effect on earnings by age 50 is close to zero. We provide new evidence on the long-term impacts of peacetime conscription in a low education labor market, using longitudinal data for Portuguese men born in 1967. These men were inducted at a relatively late age (21), allowing us to use pre- conscription wages as a control for potential ability differences between conscripts and non- conscripts. Our estimates of the average impact of military service for men who had entered the labor market by age 21 are slightly positive (1-2 percent) but not significantly different from zero throughout the period from 2 to 20 years after their service. These small average effects arise from a significantly positive later-life impact for men with only primary education, coupled with a zero-effect for men with higher education. The positive impacts for less-educated men suggest that mandatory service can be a valuable experience for poorly-educated men who might otherwise spend their careers in low-level jobs.
military conscription longitudinal earnings quasi-differences sensitivity analysis