Health Consequences of Transitioning to Retirement and Social Participation: Results based on JSTAR panel data

Author Name HASHIMOTO Hideki  (University of Tokyo)
Creation Date/NO. September 2013 13-E-078
Research Project Toward a Comprehensive Resolution of the Social Security Problem: A new economics of aging
Download / Links


Despite an extensive amount of published economic, psychological, and public health research, a consensual view on the causal relationship between retirement and health remains to be articulated. This lack of consensus is arguably due to the diversity in the transitional process from employment to full retirement, the usage of various characteristics of outcome measures, social and economic conditions affecting the retirement decision, and the impact of crowding-out by activities not related to formal work (e.g., in the family and community network). We used panel data from the Japanese Study of Aging and Retirement (JSTAR) to scrutinize the complex relationships among employment status transition; physical, functional, and cognitive aspects of health measures; and types of social participation. We confirmed that transitioning from employment to retirement is a diverse and gradual process with distinct gender-related aspects. Social participation is significantly related to exiting formal work situations for men, but not for women. There were distinct patterns of health transition across employment status transition, by types of health measures, and by reasons for retirement. Regression analyses were conducted to identify the effect of retirement, as leave from paid work, on health conditions. Variables included in the analyses accounted for social participation, stress received from the former job, and reasons for retirement. The results which included propensity weighting reveal that psychological distress and cognitive function decline after retirement for men, but not for women. Retirement from jobs with high stress was followed by an improvement in health, especially among men. Additional results indicate that retirement is accompanied by increased social participation. Social participation ameliorates psychological distress and cognitive decline among men, but not among women. Limitations in the instrumental activities of daily life as well as in grip strength are not considerably affected by retirement. Among women, retirement to engage in family care significantly and heavily affected the level of psychological distress. These results indicate that the theories on which aspects of health status determine—and are determined by—the mode of employment status transition should be improved. Policies on work and health in the elderly population should not seek a one-size-fits-all solution, but should target different segments in terms of work characteristics, economic and social needs, and gender roles in the household.