Citation

Ikeda SS, Zhang Y. Econ. Bull. 2016; 36(2): 640-664.

Copyright

(Copyright © 2016, AccessEcon, LLC [Beware: the publisher website links to malware])

DOI

unavailable

PMID

unavailable

Abstract

We construct a regional panel dataset spanning the period 1984-2012 and covering about 1,000 municipalities in Japan and applied a careful treatment of the regional fixed effects, time effects, persistence of suicide risk, and potential endogeneity of economic variables in a dynamic panel regression framework. The results indicate that (a) the removal of regional fixed effects and time effects is necessary but not sufficient for a robust analysis of suicide risk in Japan, (b) the balance of lagged and current savings and lagged real-estate liabilities are anti-risk factors for Japanese males, whereas widowed status and non-estate liabilities are risk factors for them, and (c) the balance of payments for life insurance premiums and the unemployment rate for the same gender are weak risk factors for males but significant anti-risk factors for females.

Language: en