The Development Effects of Electrification in Brazil

Project Summary
Governments and donors choose whether to allocate extra dollars of public funds to health or education programs, to build roads, or to extend coverage  of electricity or other public services.  Social scientists are expected to inform policy-makers about the returns to public investments through conducting small-scale randomized experiments.  In light of this responsibility, this project focuses on the construction and expansion (through generation plants and transmission lines) of the electric grid that was created in Brazil between 1960 and 2000.

Expansions of electric 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?). The latter complicates efforts to estimate the effects of electrification. Grid placement reflects demand side concerns, cost considerations, and geography (where hydro-power generation is able to intercept water at a high velocity).  This research isolates the portion of the variation in grid expansions in Brazil that is attributable to “exogenous” cost considerations by building an engineering model that takes as inputs only geography (river gradient, water flow and jungle), and predicts the placement of hydropower plants and transmission lines in order to generate maps of hypothetical electricity grids for Brazil that show how the grid would have evolved 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 during 1960-2000, and estimate electricity’s effect on a broad range of development outcomes.

We find that electrification of a county raises the value of the housing stock by over a full standard deviation in both urban and rural areas. It increases the employment rate by 15 percentage points, raises average incomes by a full standard deviation (50% of the mean income across Brazil), reduces the poverty head count ratio by 36 percentage points, and overall, raises the county’s UNDP Human Development Index score by half a standard deviation. We find negligible effects of electricity on population density, in-migration and mortality, which is suggestive that the positive developmental outcomes are not entirely driven by other ‘general equilibrium’ effects such as immigration to electrified areas or selective in-migration by skilled workers and firms. We thus interpret the positive effects as indications of a true causal effect of electricity on some aspect of productivity.

We also find that electrification allows farmers to irrigate, which leads to higher productivity, especially in areas with volatile rainfall patterns. Farming expands into frontier lands, and cultivation intensifies in existing land through investments in mechanization, fertilizer and pesticides. Crop intensification and substitution away from land-intensive cattle grazing lead to greater protection of forests and native vegetation within farms. Expansion of agriculture and this substitution have opposing effects on forest protection, and overall we estimate a net decrease in deforestation associated with electrification.<

Maps of Transmission with Distribution:

1960s

1970s

1980s

Decreasing Deforestation through Increased Productivity in Agriculture & Ranching
We plan to conduct a randomized controlled trial in collaboration with Amazon Environmental Research Institute’s Sustainable Settlements Project (IPAM-PAS) and the Climate Policy Initiative to evaluate the effects of  providing credit and extension training to rural farmers and ranchers in the Amazon.  Finding effective ways to increase agricultural and ranching output without increasing the amount of land employed by these individuals is a promising way to decrease the rate of deforestation in Brazil.

Research Partners:
Juliano Assunção (PUC-Rio), Tania Barham (U. Colorado at Boulder), Desirée Lopes (Yale University),
Molly Lipscomb (U. Notre Dame), Dimitri Szerman (PUC-Rio)

Research Papers
M. Lipscomb, A. M. Mobarak, T. Barham. “Development Effects of Electrification: Evidence from the Topographic Placement of Hydropower Plants in Brazil,” American Economic Journal: Applied Economics, 5(2): 200–231, April 2013Development Effects of Electrification: Evidence from the Topographic Placement of Hydropower Plants in Brazil

Abstract: We estimate the development effects of electrification across Brazil over the period 1960-2000. We simulate a time series of hypothetical electricity grids for Brazil for the period 1960-2000 that show how the grid would have evolved had infrastructure investments been made based solely on geography-based cost considerations. Using the model as an instrument, we document large positive effects of electrification on development that are underestimated when one fails to account for endogenous targeting. Broad-based improvement in labor productivity across sectors and regions rather than general equilibrium re-sorting appears to be the likely mechanism by which these development gains are realized.

J. Assunção, M. Lipscomb, A. M. Mobarak, D. Szerman. “Electrification, Agricultural Productivity and Deforestation in Brazil,” Working Paper

Abstract: We study the effects of the impressive growth in electrification in Brazil between 1960 and 2000 on changes in the structure of rural agricultural production, including adoption of irrigation, increases in agricultural productivity and investments, changes in land use, and the allocation of effort between crop cultivation and cattle grazing. The instrumental variables based identification strategy simulates a time series of hypothetical electricity grids that show how the grid would have evolved had hydro-power dams (which require intercepting water at high velocity) been allocated solely on geography-based cost considerations. Electrification allows farmers to irrigate, which leads to higher productivity, especially in areas with volatile rainfall patterns. Farming expands into frontier lands, and cultivation intensifies in existing land through investments in mechanization, fertilizer and pesticides. Crop intensification and substitution away from land-intensive cattle grazing lead to greater protection of forests and native vegetation within farms. Expansion of agriculture and this substitution have opposing effects on forest protection, and overall we estimate a net decrease in deforestation associated with electrification.

Documentation
Development Effects of Electrification Slides (Download)
Electrification Project Power Maps (Download)