LAUSR

Cash transfer programs (CTs) have been rigorously evaluated since their inception in the 1990s. However, to date, there has been no study critically examining the utility of the methodological approaches used to evaluate CTs. This article reviews the approaches used to evaluate CTs in sub-Saharan Africa (SSA) to provide recommendations for improving future CTs evaluations. We conducted searches for CTs evaluation studies in SSA in the peer-reviewed and grey literature using electronic databases, hand searching of selected journals, organisational websites, Google Scholar and Scirus Internet search engines. The review included 53 evaluation studies which were largely outcome-focused evaluations (95%; n = 50). The studies were undertaken within 24 CT programs comprising 11 unconditional CTs (UCTs), eight conditional CTs (CCTs) and five combined UCTs and CCTs. The review finds that while there is evidence of CTs impacts on a broad range of outcomes, the current evaluation approaches have primarily been experimental designs and have largely failed to provide explanations for mechanisms of change. To improve CTs policy and practice, there is the need to consider theory-based evaluation approaches such as realist evaluation that provide insights about the contexts and mechanisms through which programs generate outcomes in different circumstances.