The Long-Term Impact of Education on Mortality and Health: Evidence from Sweden

There is a well-documented large positive correlation between education and health and yet it remains unclear as to whether this is a causal relationship. Potential reasons for this lack of clarity include estimation using different methods, analysis of different populations and school reforms that are different in design. In this paper we assess whether the type of school reform, the instrument and therefore subgroup identified and the modelling strategy impact the estimated health returns to education. To this end we use both Regression Discontinuity and Difference in Differences applied to two Swedish school reforms that are different in design but were implemented across overlapping cohorts born between 1938 and 1954 and follow them up until 2013. We find small and insignificant impacts on overall mortality and its common causes and the results are robust to regression method, identification strategy and type of school reform. Extending the analysis to hospitalisations or self-reported health and health behaviours, we find no clear evidence of health improvements due to increased education. Based on the results we find no support for a positive causal effect of education on health.