Causes and Consequences of Site Selection in Development Aid

About this event

Achieving efficient allocation of development aid is a first-order priority of international organizations, particularly in an environment with increasing funding constraints. Yet we know little about how aid is allocated and the potential scope to improve aid targeting. We systematically characterize the allocation of world bank development projects within country and study the drivers of different allocation mechanisms. Our first finding is that 30% of World Bank projects exhibit positive site-selection bias: within the same province/state, selected districts are significantly richer than non-selected ones, reflected by higher levels of nightlight intensity. Second, project managers explain around 20% of total variation in site-selection bias. Third, exploiting the quasi-random assignment of managers to countries, we find that countries receiving a manager with higher quality increase their share of projects in richer locations. In this presentation, we will discuss potential ways to measure misallocation of project to sites and testing the implications of managerial discretion in site selection for policy learning and diffusion.