LAUSR

ABSTRACT This article investigates whether and to what extent retirement changes health behaviour. For identification we use an instrumental variable approach that exploits exogenous variations in the early and normal retirement ages within and across 10 European countries. Our results reveal that among those who abstained from alcohol and vigorous or moderate exercise at baseline, retirement increased those activities. Non-smokers did not increase smoking upon retirement. Retirement led to less smoking for those who smoked before retiring. It also brought about an increase in vigorous exercise for those who had the behaviour at baseline. These results further vary by a person’s job type, but less so with respect to gender or geographic region. Overall, our findings provide new empirical evidence on the causal link between retirement and health behaviours and how such link relates to four sources of individual heterogeneity: gender, European geographic region, job type and baseline health behaviour.