Social Security as Viewed through Micro-data

Part 4: Explaining Retirement Behavior

SHIMIZUTANI Satoshi Consulting Fellow, RIETI

One of the main objectives of the Japanese Study of Aging and Retirement (JSTAR) is to analyze retirement behavior. The average retirement age (average effective age of retirement) in Japan is one of the highest among advanced countries for both genders. In contrast, there has been a growing tendency in Europe to opt for early retirement following a significant easing in the eligibility criteria for disability pension benefits, which is one of the major factors behind the deteriorating fiscal positions in many European countries. Therefore, late retirement in Japan is drawing much attention.

Retirement should be taken as a long-course event and thus can be analyzed only by looking at panel data. Retirees may re-enter the labor market. In the case of married persons, retirement decisions may be made jointly by couples. Extensive research has been undertaken overseas to analyze such and various factors characterizing retirement behavior, resulting in a significant accumulation of knowledge and expertise. In this regard, Japan has long lagged behind as relevant data were not available until recently.

In a joint research conducted with Professor Hidehiko Ichimura using data collected in the first and second wave of the JSTAR survey, we found that Japanese people's retirement decisions are affected by health conditions, namely, changes in everyday physical functions, mental health (depression), and decline in cognitive ability (assessed by actually measuring each respondent's ability to recall words). As such, there is no doubt that health conditions affect retirement decisions. At the same time, however, we also found that, without objective and multifaceted data on respondents' health conditions, we cannot definitely conclude that health conditions are the determining factor in making retirement decisions. Furthermore, a significant correlation between retirement decisions and the employment status of spouses was observed among married couples in Japan, as is the case in other countries, and there is a difference in the pattern of correlation between men and women. We also learned that those who have been employed rarely go into business for themselves after retirement, and, once retired, very few of them re-enter the workforce.

Understanding the whole process leading to retirement has significant policy implications. With its population on the verge of decline, Japan is facing an urgent need to promote vigorously the employment of elderly people. Indeed, we often hear the term "shogai geneki (staying active for life)." In reality, however, not all elderly people can continue to work as they are often faced with significant health and other constraints. Meanwhile, using cross-nationally comparable data, Professor Robert Willis of the University of Michigan and Susann Rohwedder, senior economist at the RAND Corporation, showed that early retirement has a negative impact on the cognitive ability of people in their early 60s, and that the effect is quite significant. As factors behind this, they pointed out that the work environment is cognitively more stimulating than the home environment and that people preparing for early retirement have a weaker incentive to invest continuously in human capital.

Finely tuned employment matching based on the health and cognitive conditions of individual elderly persons is indispensable to the proper functioning of the elderly labor market. Individual differences among elderly job candidates are far greater than those among new graduates. Thus, in order to address job mismatches for the elderly, it is necessary to compare meticulously and align individual workers' skills and abilities to the specific demands and needs of the labor market by utilizing micro-data.

>> Original text in Japanese

* Translated by RIETI from the original Japanese "Yasashii Keizaigaku" column in the September 15, 2011 issue of Nihon Keizai Shimbun.

September 15, 2011