While we know a lot about how countries become prosperous, we have only begun to understand how aid contributes to economic growth and poverty reduction. The development record is mixed and no robust association between the volume of aid and development performance has been discovered. The limits of cross-country regressions have become clear: they do not throw much light on the reality of aid. But the novel mix of qualitative and quantitative methods fashioned by independent evaluators constitutes a serviceable approach to the assessment of aid effectiveness both at project level and at country level. In particular, a new brand of country assistance evaluations (CAEs) has demonstrated that success at project level matters even if it does not automatically translate into success at country level?the ?micro-macro paradox?. Evaluations confirm that well-managed aid, using instruments that are tailored to specific country contexts, works. They show that budget support mechanisms and programme aid instruments have a role to play in certain circumstances while projects are the aid vehicles of choice in others. The popular notion that development effectiveness can be ensured through the targeting of aid towards countries classified as good performers by idealized sets of indicators has been discredited. Recent policy research suggests that, despite the risks ...
