Nighttime Lights and Power Infrastructure Investment: Evidence from Angola

An increasing number of papers in the literature use satellite data on nighttime lights as a proxy for economic activities, such as GDP or GDP growth. They implicitly assume that the relationship between GDP and nighttime lights works through the demand side, and there is no constraint on the supply of electricity. This paper first points out a paradox in using this method: the countries for which the method is needed the most, i.e. the countries with poor statistical capacity, are just the countries, for which the assumption of the method is satisfied the least, i.e. the countries with a large power infrastructure deficit. Motivated by this, we collected the data on power infrastructure investment in Angola, a country with a large power infrastructure funding gap. Indeed, we find that in the case of Angola the stable relationship between GDP growth and lights growth assumed in the literature is broken. Instead,increase in lights strongly co-moved with increase in power infrastructure investment. The strong link between lights and investment enables us to develop a new method of quantitatively evaluating value-for-money for infrastructure investments, which directly estimates the cost-effectiveness of transforming investment to welfare, as measured by lights. We estimate the overall cost-effectiveness, and the cost-effectiveness of different financing methods in the case of Angola.