Health Economics Seminars (EUR)

Carlos Riumallo-Herl (Harvard University, United States)
Tuesday, 20 June 2017


This study explores the effect of pension fund privatization during 1981 Chilean pension reform on retirement wealth and longevity. Starting in 1981, the Chilean pension system mandated new labor force entrants as well as individuals previously not covered by social security to choose among 12 pre-selected pension funds. Given that very limited information was provided regarding fund portfolios or strategies ex-ante, and that funds ended up performing differently ex-post, initial fund selection created quasi-random variation in pension wealth. Using administrative data from 1981 to 2013, I document wealth differentials of up to 14% between contributors to the best and worst pension funds, with large reductions in wealth following major recessions.

Using initial fund choice and the timing of the reform as an instrumental variable for pension wealth, I assess the impact of pension wealth on mortality. I show that a 10% increase in pension wealth reduces old age mortality risk by approximately 4.9%.

This translates into a gain of 0.30 years in life expectancy for males, and 0.33 years for females, implying a 1.9% and 1.4% larger median life expectancy at age 65. Furthermore, the effects of pension wealth differ by gender and income group, with a 10% increase in pension wealth leading to a larger proportional increase in life expectancy for those in the lowest wealth quintiles. Using data from household surveys I find that food, health, and services are most responsive to pension income when compared to other areas of consumption and could represent the mechanisms through which pension wealth influences longevity in Chile.