Results from single-donor analyses of project aid success seem to generalize pretty well across donors

Abstract

Much research on foreign aid presents claims that apply to aid in general but tests these claims using data from one or a small number of donors. This makes it difficult to know if we have learned something about aid, or merely something about one donor. For example, the literature on project aid success has found that per capita GDP growth rates or Freedom House scores in recipient countries correlate with project success. However, these claims have been tested against data from only multilateral donors and often against data from only the World Bank. I re-examine these analyses using a dataset of harmonized project outcome scores for seven diverse donors. Most donors seem to be similarly influenced by recipient-level and project-level factors, though a few notable exceptions exist. Analyses of project aid success that focus on single donors may be able to produce knowledge about aid in general.