Predicting Income Distributions from Almost Nothing

This paper develops a method to predict comparable income and consumption distributions for all countries in the world from a simple regression with a handful of country-level variables. To fit the model, the analysis uses more than 2,000 distributions from household surveys covering 168 countries from the World Bank’s Poverty and Inequality Platform. More than 1,000 economic, demographic, and remote sensing predictors from multiple databases are used to test the models. A model is selected that balances out-of-sample accuracy, simplicity, and the share of countries for which the method can be applied. The paper finds that a simple model relying on gross domestic product per capita, under-5 mortality rate, life expectancy, and rural population share gives almost the same accuracy as a complex machine learning model using 1,000 indicators jointly. The method allows for easy distributional analysis in countries with extreme data deprivation where survey data are unavailable or severely outdated, several of which are likely among the poorest countries in the world.