"Human capital is now widely recognized to confer numerous benefits, including higher incomes, lower incidence of unemployment, and better health, to those who invest in it. Yet, recent evidence suggests that it also produces larger, social (external) benefits, such as greater aggregate income and productivity as well as lower rates of crime and political corruption. This paper considers whether human capital also delivers external benefits via reduced mortality. That is, after conditioning on various individual-specific characteristics including income and education, do we observe lower rates of mortality in economies with higher average levels of education among the total population? Evidence from a sample of more than 200 U.S. metropolitan areas over the decade of the 1990s suggests that there are significant human capital externalities on health. After conditioning on a variety of city-specific characteristics, the findings suggest that a 5 percentage point decrease in the fraction of college graduates in the population corresponds to a 14 to 36 percent increase in the probability of death, on average. Although I am unable to identify the precise mechanism by which this relationship operates, it is certainly consistent with the idea that interactions with highly educated individuals - who tend to exhibit relatively healthy behaviors - encourage others to adopt similar behaviors. Evidence of a significant inverse relationship between aggregate human capital and smoking, conditional on personal characteristics, in a sample of 201 U.S. metropolitan areas is also consistent with this hypothesis"--Federal Reserve Bank of St. Louis web site.