Effect of Selective Mortality on the Education-Mortality Relationship across the Life Course

Anna Zajacova, Princeton University

Two opposing hypotheses have been proposed to explain the relationship between education and health across age: cumulative advantage and age-as-leveler. Empirical studies generally found a pattern consistent with the latter perspective, showing a diminishing strength of the education-health relationship in later life. Some researchers have suggested that the observed trend does not necessarily support the convergence hypothesis but that it is the result of selective mortality in the population where the true effect is cumulative. The goal of this paper is to show that the observed curvilinear pattern between education and mortality across age can be caused entirely by mortality selection due to unobserved heterogeneity in population frailty. We use simple macrosimulation models to show that in a population characterized by heterogeneity in mortality hazard, a true linearly increasing effect of education on mortality will appear to have the inverted U-shape curve observed in many empirical studies.