Author: Yohei Kobayashi, Consulting Fellow, RIETI
Initiatives related to EBPM (evidence-based policy making) have recently spread among national and local government bodies in Japan. The main differences between EBPM and previous policy making have been (1) the use of evidence (scientific basis that demonstrates the causal effects of the policy) and (2) the generation of evidence. The use of evidence means the selection of measures that contribute to the attainment of a policy goal based on the existing causal evidence. The generation of evidence refers to examining the effects of a policy for which the causal effects have not been proven.
If causal evidence already exists, we can simply use EBPM for policy making. However, the reality is that there are many cases in which such causal evidence has not been extracted through research because new policy challenges arose, as well as cases in which there are uncertainties regarding whether the same effects can be expected in Japan for policies that were implemented overseas, and for this reason there are not many cases in which it is possible to make political decisions based on the existing evidence.
For this reason, it is important to both generate and accumulate evidence. While there are many methods for generating evidence, one useful tool is a randomized controlled trial (RCT). RCT is a method used to measure the causal effects of a policy by randomly assigning subjects to groups that are eligible or not eligible for a policy. However, several barriers can be faced if we try to use RCT in the field of policy. Frequently encountered barriers are as follows.