Chapter 9 Effects of Education on Adult Health in Sweden: Results from a Natural Experiment

Understanding health determinants and exactly how they affect health is an important social policy question. Empirical tests in the health literature typically find that the number of years of formal schooling completed is the most important correlate of good health. However, there is less consensus as to whether this correlation reflects a causal relationship between more schooling and better health. This chapter capitalizes on a unique social experiment: the 1950 Swedish comprehensive school reform, which was implemented in stages and by municipal areas, through which people born between 1945 and 1955 went through two different school systems (one of which required at least one more year of schooling). It uses an instrumental variables technique to estimate formal schooling's causal effect on adult health in Sweden. The instrumental variable for degree of education (schooling) generated from compulsory school reform yields a consistent estimate of education's causal impact on health, as measured by an bad health index and of body mass index in the healthy range. The additional schooling generated by Sweden's compulsory school reform produces improved adult health (controlling for cohort and county effects, family background characteristics, and individual income).