The Effect of College Education on Health and Mortality: Evidence from Canada

Abstract

We investigate the returns to college attendance in Canada in terms of health and mortality reduction. To do so, we first use a dynamic health microsimulation model to document how inter-ventions which incentivize college attendance among high school graduates may impact their health trajectory, health care consumption and life expectancy. We find large returns both in terms of longevity (4.1 years additional years at age 51), reduction in the prevalence of various health conditions (10-15 percentage points reduction in diabetes and 5 percentage points for stroke) and health care consumption (27.3% reduction in lifetime hospital stays, 19.7 for specialists). We find that education impacts mortality mostly by delaying the incidence of health conditions as well as providing a survival advantage conditional on having diseases. Second, we provide qua-si-experimental evidence on the impact of college attendance on long- term health outcomes by exploiting the Canadian Veteran's Rehabilitation Act, a program targeted towards returning WW-II veterans and which incentivized college attendance. The impact on mortality are found to be larger than those estimated from the health microsimulation model (hazard ratio of 0.216 com-pared to 0.6 in the simulation model) which suggest substantial returns to college education in terms of healthy life extension which we estimate around one million canadian dollars.

Keywords: mortality, education, microsimulation, quasi-experimental, instrumental variables, veterans

JEL Classification: I14, I26