The literature shows that regional disparities in growth and poverty are often relatively high, that these regional disparities do not necessarily disappear as the economies grow and develop and that these disparities are itself in many cases an important driver of the overall performance of an economy. In this paper we make use of the advantage of a multilevel random coefficient model to explain spatial disparities in incomes among Burkinab`e households. Our findings show that it is not a geographical concentration of people with poor endowments that make areas poor in Burkina Faso. Household income disparities are largely driven by differences in neighborhood endowments and to a smaller extent by provincial or regional characteristics. We conclude that the policy should target small scale geographical units, such as villages. Providing infrastructure, enhancing the functioning of labor markets and fostering demand for education can compensate for climatical disadvantages.
spatial inequality poverty multilevel modeling decomposition Sub-Saharan Africa