Policy Diffusion: Theory and Its Implications

Abstract

　Various empirical studies both qualitatively and quantitatively have shown that policies diffuse across local governments through learning. Although people often regard it as a phenomenon that only good policies are survived, it is not necessarily the case. By reviewing existing theoretical findings, the present study aims to identify the conditions under which policy diffusion yields a good outcome. Based on the literature of policy experimentation, we first argue that policy diffusion through policymakers’ learning does not work when a free-riding problem across local governments is severe. Based on the literature of yardstick competition, we next argue that policy diffusion through voters’ learning fails when the electoral accountability system does not appropriately work in each local government. We finally discuss how we should design political institutions to make policy diffusion work.