WP61 2012/03 "The Economics and some simple Econometrics of randomized R&D-Program Evaluation"

Despite a sound rationale for government intervention in the field of R&D, the question on the achievement of innovation policy objectives remains an empirical issue which has to be addressed in evaluation exercises.  However, evaluations of R&D interventions do not necessarily provide the rigorous results needed to inform policy makers on which of the tough choices they should make. The paper presents economic and econometric evidence that supports a (selective) use of randomization in the design of innovation policies because it enables more credible estimations of causal policy effects and can improve outcomes of policy intervention due to a more “creative” project selection compared to traditional methods such as peer review.