The Long-Run Impact of Electoral Violence on Health and Human Capital in Kenya

This paper examines whether exposure to electoral violence during childhood or adolescence has long-lasting effects on health and human capital. Using a nationally representative population survey in Kenya, we identify who has been exposed to electoral violence during early life from 1992 to 2013 by exploiting the variation in such violence across space and time. Using coarsened matching, we find that exposure to electoral violence between the ages of 0 and 16 has a negative impact on adult height. Previous research has shown that large-scale and protracted conflicts can pass down stunting effects to descendants. Similar to these studies, we find the low-scale but recurrent electoral violence in Kenya has affected the height-for-age of children whose parents were exposed to electoral violence during their early life. This effect on the second generation is transmitted exclusively to boys. This may be because boys in Kenya and Sub-Saharan Africa are more likely to be malnourished than girls, which raises their risk of stunting. In contrast to large-scale conflict studies, we find no long-term effect of childhood exposure to electoral violence on educational attainment or consumption. Most Kenyan electoral violence has occurred during school breaks, which may explain these contrasting results.