In this paper we assess the impact of rural electrification on child labor in Nigeria using panel data provided by the General Household Survey. This relationship is theoretically ambiguous and the few existing empirical results do not converge. Given unreliability of the power grid and heterogenous equipment rates, electrification cannot be only captured using access to the grid.
We investigate in particular how child labor varies depending on the nature of electricity supply and the electrical appliances used in the household. When controlling for a large set of individual characteristics and for selection on unobservables, we find that the employment probability of children from electrified households is lower than that of children living in non-electrified households only when the household combines grid access and a generator as sources of electricity. This combination allows households to be able to use appliances that allow them to save time and reallocate it among their members. While this channel is the preferred one to explain the decline in child labor, it is difficult to relate it statistically to time-saving appliances given their low use among rural households in Nigeria. In addition, we find that girls from households using entertainment appliances (mostly television) are more likely to work than other girls, a result which is robust to both selection on unobservables and heterogenous effects of the treatment.
