Beyond Evaluations: Assessing the Local Performance of World Bank Projects in Africa

Ex-post evaluations of development projects are commonly used as a measure of international organizations’ (IO) performance. More recently, however, studies have started to criticize this approach. They demonstrate that evaluation ratings can be institutionally biased, not standardized, subjective, and highly influenced by project management criteria. Building on these insights, this paper examines to what extent evaluations reflect the actual achievements on the ground. Employing a methodology that combines geo-coded World Bank project data with survey data and sub-national development indicators, we assess whether projects deemed successful exert a more substantial impact on the local economic development and perceived institutional quality in Africa than projects evaluated as unsuccessful. This approach enables us to test our central assumption that there are no significant differences in development outcomes between projects evaluated as distinctly successful or unsuccessful. By challenging the current reliance on evaluation ratings as definitive performance indicators, this research contributes to the literature on the politics of evaluations and IO functioning.