Do subjective expectations of survival affect retirement intentions?

Hanna van Solinge, Netherlands Interdisciplinary Demographic Institute (NIDI)

According to economic models of lifetime consumption, life-expectancy is important for long-term planning such as savings and investment. At the aggregate level, studies have found a general positive correlation between increases in average life-expectancy at birth and savings and investments in schooling. Economic life cycle models postulate that individuals expecting to be long-lived will retire at a later age than those expecting to die early as they will need greater wealth to finance more years of retirement. This relationship between subjective life expectancy and saving behavior has been documented in a variety of studies. Saving for retirement and retirement decision making may be related, but not necessarily the same decision. This study investigates the relationship between subjective life expectancy and the propensity to retire early (preferences to take early retirement). Two questions are addressed in this study: 1) What are the correlates of subjective life expectancy? 2) Do individuals who perceive shorter individual time horizons are more inclined to retire early than person’s who expect to life longer? Using multi-actor panel data from the NIDI Older Worker and Retirement Survey 2001-2006 (N=1.675 older workers and spouses), regression models are estimated to examine the effect of subjective life expectancy on retirement decision making. Our measure for subjective life expectancy is based on the individual’s self-rated probability/chance they would live to 75 or 95. Preliminary results reveal that individuals seem to base their subjective life expectancy largely on their relatives’ longevity. In explaining retirement intentions, subjective life expectancy has effects independent of economic and health effects. The results would be useful additions to models that forecast labor force participation by older workers.