Can pensions save lives? Evidence from the introduction of old-age assistance in the UK

I study the impact of old-age assistance on mortality using the introduction of public pensions in the UK in 1909 as a quasi-natural experiment. Exploiting the newly created pension eligibility age through a difference-in-difference as well as an event-time design, I show that elderly mortality in England and Wales declined after the pension was introduced. The estimated mortality decline is economically relevant, more pronounced in counties with a higher share of pensioners and is driven by fewer deaths from infectious as well as non-infectious diseases. An analysis of full-count individual-level census data points to a reduction in residential crowding and retirement, especially from occupations associated with high mortality rates, as likely channels.