A Weakly Balanced Dataset of Poverty in the Developing World Using Under-Five Mortality Rates, 1980–2022
摘要
Using econometric methods, this paper addresses the missing data problem of the World Bank extreme and moderate poverty rates from the World Development Indicators. An iterative interpolation model was constructed and evaluated using cross-validation methods utilizing under-five mortality rates. The result is a robust and extensive poverty dataset for 169 countries annually from 1980–2022 covering 99% of the world’s population. The data indicate that global poverty has declined dramatically throughout the world. Our database suggests that the reduction in poverty is larger than previously documented. These reductions in poverty occurred in countries that improved their institutions, governance, and human capital. Moreover, malaria continues to be an important roadblock for the poor. Our preliminary analysis using this updated data suggests that future research will be able to examine poverty in much greater detail, including, but not limited to, the nexus of inequality-poverty-growth, remittances and poverty, or institutional quality and poverty reduction.