Development Effects of Electrification: Evidence from the Geologic Placement of Hydropower Plants in Brazil

Expansions of electricity grids reflect both cost considerations (where is it cheapest to generate electricity?) and demand-side concerns (where are firms and people located, and where is demand for power likely to grow most?). Demand evolves simultaneously with power generation, and complicates efforts to estimate the effects of electrification. This paper attempts to isolate the portion of the variation in grid expansions in Brazil that is attributable to “exogenous” engineering cost considerations to estimate the development effects of electrification between 1950 and 2000. Brazil relies almost exclusively on hydropower, and hydro-power generation requires intercepting water at high velocity. A portion of the spatial variation in the expansion of the electricity grid in Brazil during this period is therefore driven by river gradients suitable for hydro-power generation. We predict hydropower plant placement based on geologic characteristics (river gradient and water flow) of locations throughout Brazil and then develop a cost-minimizing “engineering model” to predict the expansions of transmission lines from each of those predicted hypothetical stations every decade. The model generates maps of hypothetical electricity grids for Brazil in each decade which show what the grid would have looked like had infrastructure investments been made based solely on geologic cost considerations, ignoring all demand-side concerns. We use these modeled hypothetical maps to instrument for the actual variation in electrification across Brazil over time. We begin by showing that the very strong cross-sectional correlation between electrification and population density is largely driven by the correlation between population density and water availability. In fixed effect regressions, the effect of grid expansions on subsequent changes in population density is much smaller, and fixed effects instrumental variables estimates show that even this smaller effect is likely a result of electricity grid expansions following population projections, rather than electrification inducing an increase in density. Electricity is estimated to increase GDP per capita under the fixed effects IV. Since the population density results suggest that other ‘general equilibrium’ effects such as inmigration to electrified areas or selective in-migration by skilled workers and firms are not taking place, we interpret the positive GDP result as an indication of a true causal effect of electricity on some aspect of productivity. *We thank the University of Colorado NICHD Population Center, Corporación Andina de Fomento, Center for Advancement in Research and Teaching in the Social Sciences at the University of Colorado and the Macmillan Center at Yale University for the financial support that made this data collection possible. We also thank Daniel Ortega and seminar participants at Corporación Andina de Fomento, UC Energy Institute and Yale School of Management for comments, and Vanessa Empinotti and Steven Li for excellent research assistance. 2