This report has been prepared in thecontext of a major global effort in the past eight years tobetter measure results in development assistance. The agendafor this effort was articulated and refined in a series ofinternational conferences, beginning with the internationalconference on financing for development in Monterrey in 2002and continuing through the Accra agenda for action in 2008.Cost-benefit analysis entails measuring results, valuingresults, and comparing results with costs, and hence ishighly relevant to the results agenda. Cost-benefit analysiscan provide a comprehensive picture of the net impact ofprojects and help direct funds to where their developmenteffectiveness is highest. This study draws two broadconclusions. First, the bank needs to revisit its policy forcost-benefit analysis in a way that recognizes thelegitimate difficulties in quantifying benefits whilepreserving a high degree of rigor in justifying projects.Second, the bank needs to ensure that cost-benefit analysisis done with quality, rigor, and objectivity: poor data andanalysis misinform, and do not improve, results. Reforms arerequired to project-appraisal procedures to ensureobjectivity, improve both the analysis and the use ofevidence at appraisal, and ensure effective use of costbenefit analysis in decision making.