Previous studies have shown that wars restrict the acquisition of human capital. We
expand this literature by exploring how the stock of human capital is affected by wars.
Applying a “missing people” approach to the 1994 Rwanda genocide we find that the size
of the educated cohort --those with completed primary school or more-- shrunk by 39
percent relative to their less educated counterparts and compared to its size in 1991. This
excess missing rate for the educated is not due to outmigration and contrasts with the
patterns observed in other countries of the region and also with the pre-genocide trends in
Rwanda. When the spatial variation in the intensity of the genocide is exploited there is
no evidence of statistically significant differences, suggesting an aggregate effect from
the genocide. We discuss how this reduction affects labor markets post-conflict and the
returns to education.