Are Pilot Experiments Random? Evidence from China’s Social Security Expansion

Abstract

This paper provides new evidence of how political regimes and social connections determine participation in pilot experiments. Using data from China’s national pilot experiments of New Rural Pension Scheme (NRPS), we show that county-level charac- teristics and political regime have explanation power in the early stage of county pilot decision. Local party secretaries connected to Central Committee (CC) members, a po- litical body that comprises the top leaders of the Communist Party, are discouraged to enroll. In contrast, counties that are send-down movement destinations of CC members during Cultural Revolution (1966-1976) are more likely to enroll. We also find counties with lower cost, higher benefits, and more government revenue are more likely to enroll. Local government leaders may face trade-off between economic performance and social welfare, and connected officials have more incentives to allocate resources in economic growth rather than social welfare. These findings are characterized by a theoretical model that describes how connections change local leaders’ allocation of government resources under a multitask problem