Is retirement really bad for health? Further evidence on the health-effects of retirements in Europe

Philipp Hessel, London School of Economics

Whereas researchers of many disciplines have devoted much attention to the question how health affects individuals’ retirement-behaviours, very little conclusive evidence exists on the question how retirement itself affects health. At the same time, the hypothesis that retirement would be detrimental for individuals' health remains widespread in the public as well as scientific discourse. Besides anecdotal evidence that retirement as a so called 'stressful life-event' would negatively affect the health of individuals, much of existing research on the topic may largely overestimate these potential negative effects as only a few recent studies explicitly address the issue of ‘endogeneity’ caused by the reciprocal relationship between health and retirement. In this background, by using longitudinal data for 15 Western European countries (EU-SILC), this paper addresses the question if retirement has a significant effect on individuals’ health. More precisely, this study uses country-specific (early-) retirement ages as ‘instruments’ to adjust for retirement. The dependent variables of health are self-reported health, ADL as well as chronic conditions. In contrast to the 'simple' random- or fixed-effects models, the results of the instrumental variables models suggest that for men and women retirement has no negative effects on health. Thereby for men retirement leads to significant improvements in self-rated health as well as to fewer limitations in ADL.e.