Understanding how electricity demand is likely to rise once households gain access to it is important to policy makers and planners alike. Current approaches to estimate the latent demand of unelectrified populations assume constant elasticities of demand. Here we use a simulation-based structural estimation approach employing micro-data from household surveys for three developing nations to estimate responsiveness of electricity demand to income considering changes both on the intensive and extensive margin. We find significant heterogeneity in household response to income changes, which suggest that assuming a non-varying elasticity can result in biased estimates of demand. We use our estimated model along with information on the current and projected future distribution of income and population in each nation to estimate the total latent electricity demand of achieving universal access to electricity services. Our results confirm that neglecting heterogeneity in individual behavior and responses can result in biased demand estimates.
Presented in Session 121. Population Processes, the Environment, and Energy