This paper estimates the causal effects of civil war on years of education in the context of a school-going age cohort that is exposed to armed conflict in Côte d'Ivoire. Using year and department of birth to identify an individual's exposure to war, the difference-in-difference outcomes indicate that the average years of education for a school-going age cohort is .94 years fewer compared with an older cohort in war-affected regions. To minimize the potential bias in the estimated outcome, the authors use a set of victimization indicators to identify the true effect of war. The propensity score matching estimates do not alter the main findings. In addition, the outcomes of double-robust models minimize the specification errors in the model. Moreover, the paper finds the outcomes are robust across alternative matching methods, estimation by using subsamples, and other education outcome variables. Overall, the findings across different models suggest a drop in average years of education by a range of .2 to .9 fewer years.
Policy Research Working Papers