Two papers in the latest issue of World Development look, from a first glance, like they need to have a chat
Why there Should be No Political Foreign Aid Curse
Ceren Altincekic, David H. Bearce
This paper considers the causality underlying the so-called political aid curse, which proposes that foreign aid, like oil, should hinder democracy. Using a theoretical model which identifies repression and appeasement as the primary alternatives to democratization, it argues that aid revenue should not produce a political curse because it is less fungible, more conditional, and less constant than state oil revenue, making it difficult for recipient governments to use their aid to fund either repression or appeasement. Using several different measures associated with repression and appeasement, the statistical results show that aid cannot be associated with any of these dependent variables.
I analyze a unique dataset of sub-national resource allocations in Kenya from 1989 to 1995 and show that project aid and local funds were disproportionately directed to the president’s political base. Per-donor analyses of aid flows show that bilateral donors and the African Development Bank were most likely to skew their aid to the president’s base. Kenya’s autocratic leader was able to exercise strong political influence over the location of many aid projects, even under unfavorable circumstances. While disbursing aid as projects may have ensured better accounting of funds, it did little to prevent aid from becoming patronage.
For what it’s worth, I don’t think the findings are as at odds as they seem. As best I can tell, most of the time aid is relatively impotent in the face of domestic political economy factors. It might be able to help (or hinder) a bit, but my experience based on Solomon Islands is that it is very hard for aid actors to transform a domestic political economy: for better or for worse. Although I agree with Bearce that they probably do have enough power to offset the political resource curse problem (and maybe a bit more), if they act wisely and with recipient population welfare as their first motivation. And this is the likely second explanation of the difference: look at the time frame of Brigg’s dataset. The tail end of the cold war and just after (which given some path dependency more or less means ‘in’).