Changes in compulsory schooling and the causal effect of education on health: evidence from Germany.

In this paper we investigate the causal effect of years of schooling on health and health-related behavior in West Germany. We apply an instrumental variables approach using as natural experiments several changes in compulsory schooling laws between 1949 and 1969. These law changes generate exogenous variation in years of schooling both across states and over time. We find evidence for a strong and significant causal effect of years of schooling on long-term illness for men but not for women. Moreover, we provide somewhat weaker evidence of a causal effect of education on the likelihood of having weight problems for both sexes. On the other hand, we find little evidence for a causal effect of education on smoking behavior. Overall, our estimates suggest significant non-monetary returns to education with respect to health outcomes but not necessarily with respect to health-related behavior.