This paper proposes using satellite data on night lights as a proxy for economic growth, particularly in countries with poor national income accounts data. The authors develop a statistical framework to combine night light data with existing income growth measures, assuming that measurement errors in both are uncorrelated. For countries with good national income accounts, night light data provide marginal value in estimating true income growth, while for countries with poor data, the optimal estimate of true income growth is a composite with equal weights given to both night light and income data. The study finds that real incomes in non-coastal areas of Sub-Saharan Africa grow faster than those in coastal areas, and non-malarial areas grow faster than malarial ones. Night light data also allow for the measurement of income growth at sub- and supranational levels, providing a valuable tool for economic analysis and policy-making. The paper discusses the limitations of GDP data, the advantages of night light data, and the statistical methods used to combine these two sources of information.This paper proposes using satellite data on night lights as a proxy for economic growth, particularly in countries with poor national income accounts data. The authors develop a statistical framework to combine night light data with existing income growth measures, assuming that measurement errors in both are uncorrelated. For countries with good national income accounts, night light data provide marginal value in estimating true income growth, while for countries with poor data, the optimal estimate of true income growth is a composite with equal weights given to both night light and income data. The study finds that real incomes in non-coastal areas of Sub-Saharan Africa grow faster than those in coastal areas, and non-malarial areas grow faster than malarial ones. Night light data also allow for the measurement of income growth at sub- and supranational levels, providing a valuable tool for economic analysis and policy-making. The paper discusses the limitations of GDP data, the advantages of night light data, and the statistical methods used to combine these two sources of information.