What Can Experiments Tell Us About How to Improve Government Performance?

Abstract In recent years, experimental methods have been both highly celebrated, and roundly criticized, as a means of addressing core questions in the social sciences. They have received particular attention in the analysis of development interventions. This paper focuses on two key questions: (1) what have been the main contributions of RCTs to the study of government performance? and (2) what could be the contributions, and relatedly the limits? It draws inter alia on a new systematic review of experimental and quasi-experimental studies on governance to consider both the contributions and limits of RCTs in the extant literature. A final section introduces the studies included in this symposium in light of this discussion. Collectively, the studies push beyond polarized debates over experimental methods towards a new middle ground, considering both how experimental work can better address identified weaknesses and how experimental and non-experimental techniques can be combined most fruitfully.