The Retirement-Health Puzzle: A Sigh of Relief at Retirement?

Abstract

Objectives: While the health effects of retirement have been well studied, existing findings remain inconclusive, and the mechanisms underlying the linkage between retirement and health are unclear. Thus, this study aimed to evaluate the effects of retirement on health and its potential mediators.

Methods: Using a national household survey conducted annually from 2004 to 2019 in Japan (the Japan Household Panel Survey), we evaluated the effects of retirement among Japanese men aged 50 or older on their health, in addition to other outcomes that could be attributed to health changes associated with retirement (i.e., health behaviors, psychological well-being, time use for unpaid activities, and leisure activities). As outcomes are not measured every year, we analyzed 5,794-10,682 person-year observations for 975-1,469 unique individuals. To address the potential endogeneity of retirement, we adopted an instrumental variable fixed-effects approach based on policy changes in eligibility ages for employee pensions.

Results: We found that retirement improved psychological well-being, exercise habits, and time spent on unpaid work. The psychological benefits of retirement were no longer observed for longer durations after retirement, whereas healthy habits and unpaid activities continued. Moreover, health-related improvements after retirement occurred mostly in the higher-income group.

Discussion: Enhancement in personal quality of life owing to increased leisure time and stress reduction from work in addition to lifestyle changes may be key to understanding the health benefits of retirement. Considering the mechanisms behind retirement-health relationships and potential heterogeneous effects is essential for healthy postretirement lives when increasing the retirement age.

Keywords: Causal inference; Employment; Health disparities; Health outcomes; Retirement.