Is electricity a vehicle of children and women empowerment in poor rural areas of the developing world? In this paper we approach the question through the human capital and labor market transmission channels using a rich and detailed micro-level household survey data from Rwanda. Our analysis considers effects occurring directly at the household level and indirectly at the village economy level. We address endogeneity raised by the non-random distribution of grid access across villages and household selection into treatment within villages using a combination of inverse probability weighting methods and instrumental variable strategy. Reweighting allows us to carefully choose the control group in a multilevel treatment context in order to uncover the direct and indirect effect of electrification, while the instrument deal with the unobserved selection bias at the village or household level. As a whole, we find little impact of electricity on children employment. A distinction by gender shows however that spillover effects bring boys into wage employment, while the direct household electrification exerts an effect in the opposite direction. Additional time devoted to work does not result in any negative impact on schooling. Women participation to paid employment is not affected by household connection or village access to the grid. Instead, women that are already likely to participate to paid employment, work longer hours as wage worker and in independent businesses. Looking at inequalities between spouses, women position relative to their male spouse is improved in terms of employment participation due to their greater involvement into family businesses.
