UC San Diego

This dissertation investigates how individuals, municipalities, and firms respond to incentives created by public policies and provides empirical evidence on their efficacy.

Chapter 1 investigates the effects of increasing the eligibility age for public pension on workers’ retirement decisions, focusing on recent Japanese public pension reforms. In Japan, the pensionable age for Employees’ Pension Insurance benefits gradually increased from 60 to 65 for males over the course of a decade. Using individual-level restricted-use data and a regression discontinuity design, I find that raising the pensionable age for flat-rate benefits by one year increases male employment at the critical ages by about 7-8 percentage points. Individual labor supply responses at the critical ages are heterogeneous across closeness to the implementation date

due to anticipatory responses.

Chapter 2 studies the effect of allocating central administrators on local government units. During the 2000s, Japanese central administrators were actively transferred from the central government to mentor and monitor local governments. Exploiting the timing of hosting transfers and rich administrative data, I find that municipalities with transferred central administrators in fact persistently improved fiscal discipline by shrinking expenditure and lowering debt. Voters seem to reward the incumbent mayor in the local election for better administration and fiscal conditions. Heterogeneity analyses reveal, though, that transferred administrators temporarily increase local expenditure and categorical grants in fields closely related to their respective departments.

Chapter 3 investigates the effect of raising the mandatory retirement age and introducing a continued reemployment system on older workers and young job-seekers. In 2006, Japanese companies were required to raise the mandatory retirement age from age 60 to at least age 63 or to introduce a continued employment system that creates flexible positions for older workers to continue at the same company. Relying on quasi-experimental variation in exposure to the policy change according to pre-reform norms by industry, geography, and firm size, I find that the reform was effective in terms of decreasing the job separation rate of older workers. It also decreased the job finding rate of young people for firms that were more affected by the policy change.