Essays on the Evaluation of Public Policies

Abstract: This thesis consists of four self-contained papers.Essay 1: This paper uses the synthetic control (SC) method to examine how the establishment of Nuclear Power Facilities (NPFs) in Japan in the 1970s and 1980s has affected local per capita income levels in the municipalities in which they were located (NPF municipalities). Eight quantitative case studies using the SC method clarify that the effects of NPF establishment on per capita taxable income levels are highly heterogeneous. The estimated effects are often economically meaningful and in some cases huge: the income level was 11% higher on average and 62% higher in one municipality in 2002 when compared with counterfactual units. On the other hand a few of the NPF municipalities have received only weak or negligible effects from NPF establishment. The post-estimation comparisons of employment between the NPF municipalities and the SC units suggest that the size of the direct labor demand shocks and subsequent indirect employment effects on nontradable service sectors have contributed to the increase in per capita income levels.Essay 2: In a regression kink (RK) design with a finite sample, a confounding smooth nonlinear relationship between an assignment variable and an outcome variable around a threshold can be spuriously picked up as a kink and result in a biased estimate. In order to investigate how well RK designs handle such confounding nonlinearity, I firstly implement Monte Carlo simulations and then study the effect of fiscal equalization grants on local expenditure in Japan using an RK design. Results in both the Monte Carlo simulations and the empirical application suggest that RK estimation without covariates can be easily biased, and this problem can be mitigated by adding basic covariates to the regressors. On the other hand, a smaller bandwidth or a higher order polynomial, even a quadratic polynomial, tends to result in imprecise estimates although they may be able to reduce estimation bias. In sum, RK estimation with a confounding nonlinearity often suffers from bias or imprecision and estimates are credible only when relevant covariates are controlled for.Essay 3: This paper investigates the effects of fiscal equalization grants on total expenditure and disaggregated expenditures by exploiting two different formula-based exogenous variations in grants. Examining the institutional settings of the Japanese fiscal equalization scheme and estimating local average grant effects with a regression kink design and an instrumental variable approach, I demonstrate that there exist heterogeneous grant effects for two groups of municipalities with different fiscal conditions. That is, estimated grant effects on total expenditure are approximately one-to-one for municipalities around the threshold of grant eligibility, but much more than one-to-one for municipalities that are heavily dependent on fiscal equalization grants. In addition, grant effects on disaggregated expenditures are dispersed across different expenditure items in the former type of municipality but concentrated on construction expenditures in the latter type. I then discuss that the observed grant effect heterogeneity is a consequence of the institutional settings of the Japanese fiscal equalization scheme.Essay 4 (with Reo Takaku): We evaluate the impact of patient cost sharing on the use of dentures and subjective chewing ability exploiting a sharp reduction in the coinsurance rate, the percentage of costs born by the user, from 30% to 10% at the age of 70 with a regression discontinuity design. Using data from the Japanese Study of Aging and Retirement (JSTAR), we find that the utilization rate of dentures increases from approximately 50% to 63% around the threshold, implying that the extensive margin elasticity of denture usage with respect to the coinsurance rate is about -0.41. In addition, we find this jump is almost entirely due to the change in the rate among women. On the other hand, we do not find a significant improvement in self-reported chewing ability. Our empirical findings are also confirmed by complementary analysis with randomization tests and placebo randomization tests.