Target at the Right Level: Aid, Spillovers and Growth in Sub-Saharan Africa

Abstract

This article uses spatial analytical skills to investigate aid effectiveness and aid spillovers at the sub-national level in Sub-Saharan Africa over the period 1995-2013. The previous literature examines aid-growth relationship and gets mixed results. One reason of the conflicting conclusions is that it suffers from the so-called “aggregation bias”. By using geocoded aid dataset and night lights data as proxy for economic activities, I am able to reveal more detailed pictures of aid effectiveness at different aggregate levels. Overcoming the potential simultaneity problem, I find that aid targeted at the local level tends to promote local economic growth, while aid received at more aggregate levels depresses local economic activities; there exist positive aid spillovers of aid across adjacent neighbors at the local level. One possibility is that more specifically targeted aid tends to be less fungible compared to “general” aid, while aid generally given to a more aggregated level is more likely to be misappropriated for other purposes, thus creating rent-seeking opportunities to cause corruption and hurt institutional environment. Aid at the local level promotes total economic flourish and slows down population growth, while aid at more aggregate levels depresses total economic activities but stimulates population growth. Aid directly received at all levels exhibits diminishing returns, which is consistent with the theory that aid directly stimulates investment and adds capital accumulation. While aid spillovers show weak increasing returns, which suggests the spillover effects partly function through technology and knowledge dissemination. As to the conditional aid effectiveness, no systematic story is found that aid is effective conditional on policy or institutions, probably due to data limitations that local policy and institutions data are unavailable in Sub-Saharan Africa. The findings have very profound policy implications that to promote local economic growth, we should focus more on specifically targeted and less-fungible aid projects rather than aid generally given to governments at more aggregate levels; also we should reduce barriers to resource movements and knowledge dissemination within the country to promote positive spillover effects.