The role of cognitively stimulating activities in explaining the association between retirement timing and cognitive functioning in old age

Background: In light of the rising life expectancy and financial pressure on old-age pension systems in industrialized economies, many European countries – including Switzerland – are currently reforming related policies. Some of the countries have increased or plan to increase the eligibility age for old-age pension. In this context, we examine the association between the prolongation of working life and cognitive health in late-life based on data from Sweden, where retirement age is more flexible than in most other countries. The aim of our study is to analyze whether the level of job-related cognitive stimulation affects the association between retirement age and cognitive functioning in old age. Methods: We use a representative dataset from Sweden, linking survey data from the Level of Living Survey (LNU), the Swedish Panel Study of Living conditions of the Oldest Old (SWEOLD) and annual tax register data. Using a propensity score matching approach, we create a quasi-experimental study design comparing individuals who extend their working life beyond age 65 with those who retire earlier. In addition, we use ordered logistic regression analysis to examine the effect of working longer by occupational groups. Results: We observe that on average, individuals who retire at age 65 or later do not have a higher level of cognitive functioning around age 80 than those who retired earlier. We observe higher levels of cognitive functioning around age 80 for individuals who retire at age 65 or later as compared to those who retire earlier. In addition, our analysis shows that if we control only for socio-demographic and occupational characteristics, individuals who worked in more complex and non-manual occupations have a better cognitive functioning in old age. However, if we include leisure activities that individuals attend during their retirement into the analysis, the effect of occupational characteristics becomes less strong and loses statistical significance whereas the effect for leisure activities is significant. Conclusion: Using longitudinal data and a causal approach that allows for tapping into mechanisms underlying the association between retirement and cognition, our results suggest that engaging in stimulating activities contributes to the maintenance of cognitive abilities in old-age. This finding allows informing the policy debate about the transition from employment to retirement