Impact Evaluations : a tool for accountability ? Lessons from experience at AFD

This paper relates the Agence Française de Développement's experience with respect to impact evaluations. Our purpose is to assess the extent to which such studies, when designed before an actual program implementation can provide the type of summative evidence that donors still seek for when promoting such studies. Specifically, we rely on three large scale randomized control trials, and scrutinize their capacity to answer questions related to the underlying program’s impact ‘on whom’, ‘on what’ and ‘of what’. We conclude that experimental studies should be promoted to answer the type of ‘tunnel’ questions characterized by a limited number of wellspecified homogeneous inputs, a tried and tested process, a short and external events-proof causal chain, a rapid and stable appropriation by beneficiaries, a large and stable participation, and a set of measurable outcomes in the short run. While a number of such issues exist and are very much worth studying experimentally to inform future development policies, few development interventions satisfy them, and summative use is thus limited.