WP 2013-10 Impact Evaluation and Welfare Evaluation
Published in 2014 in The European Journal of Development Research, Volume 26, Issue 1.
ABSTRACT: The expression “impact evaluation” means different things to different people, but to most economists now it means the use of randomized controlled trials (RCTs) or “quasi-experiments.” I want to focus on that side of the research and argue that, even when statistical circumstances are ideal, it constitutes at best an intermediate input into the welfare evaluation of interventions. That intermediate input is valuable, but should not be confused with the final product, a proper cost-benefit analysis. I summarize arguments here that are developed at greater length.