Spatial Inequality: Overcoming Neighborhood Effects in Africa

Spatial inequality in global economic development has left Africa with the least progress in improving living standards among developing regions of the world. Moreover, there are strong neighborhood effects within Africa. This paper revisits the explanation of unequal growth across countries in an African context. We argue that some of the lingering disagreements over the channels through which institutions and geography may explain differences in income per capita across countries could be resolved by accounting for neighborhood effects often overlooked in past analyses. Through simultaneous equations we test how trade, urbanization, and agricultural productivity are affected by a country’s policies and factor endowments and the degree to which each aspect of economic development is affected by spillovers from neighboring countries. We use both limited and full information estimators, based partly on a generalized moments estimator for spatial autoregressive coefficients, which allow for spatial error correlation, correlation across equations, and the presence of spatially lagged dependent variables. With this specification, after controlling for spatial proximity, both institutions and geography variables exert, through trade and urbanization, an independent effect on income.