This paper examines the drivers of income inequality within and across countries using relevant measures of inequality and an estimation technique that jointly accounts for both model and estimation uncertainties. The estimations are applied to a global sample and to three categories of vulnerable developing countries: Africa, least developed countries (LDCs), and landlocked developing countries (LLDCs).

We find that multiple factors contribute to income inequality within and across countries but that there are significant differences in the key drivers globally and in Africa, LDCs and LLDCs. We also find strong support for the Kuznets hypothesis in the global and the developing countries samples but not in the Africa, LDC and LLDC samples. These differences underscore the need for policymakers to account for country-heterogeneity in the design of policies to combat inequality within and across countries.