This paper is concerned with the differing modalities of pro-poor policy formation and implementation. Debates are increasingly pulled in these two distinct directions: on the one hand, towards the importance of expert-led or evidence-based/informed policy (i.e. research evidence based) and on the other hand, the need for citizen-led (i.e. participatory) policy processes. This paper argues neither modality is necessarily pro-poor. We explore the conditions under which each might be pro-poor.