Impact of retirement on mortality - Evidence from Spain

This paper aims to estimate the long-run health effects of retirement in Spain looking at mortality within five years. I use a Regression Discontinuity Design to exploit Spanish age-specific early and statutory retirement ages controlling for month of birth. Using the Muestra Continua de Vidas Laborales (MCVL) Spanish administrative data I find heterogeneity on retirement effect. Exiting the labour market at early retirement age decreases the long-run probability of death by 6.1 percentage points whereas retiring at statutory retirement age increases mortality within five years by 2.1 percentage points. The results are robust to several sensitivity tests. Moreover, I document heterogeneity at each age threshold depending on personal and professional characteristics. For instance, retiring from high physical strain occupations leads to a health-preserving effect at both eligibility ages.