In the context of a rapidly aging population, improving the parents' health outcomes, especially in parents with poorer health, is essential for narrowing elderly health inequality. Using data from the China Health and Retirement Longitudinal Study, we took the university enrollment expansion policy as the instrumental variable and employed the two-stage least square (2SLS) and instrumental variable quantile regression (IVQR) approaches to explore the spillovers of offspring education on the elderly parents' frailty index. The results show that one additional year of offspring educational attainment was associated with a 0.017 or 4.66% decline in the parents' frailty index. These spillovers are stronger where parents are cohabiting with their children than when separating (more than 2 times higher). Moreover, there is substantial heterogeneity that is determined by the gender of parents. The spillover on mothers is greater than that on fathers. Further analysis of a cohort of parents with different frailty indexes reveals that the upward spillovers of offspring education on parents' health are non-linear and non-averaged. The spillovers may diminish as parents own health improves. These spillovers suppress the "Matthew Effect", which can lead to the further widening of health inequality.
Keywords: China; offspring education; parents’ health inequality; quantile treatment effect; spillovers; university enrollment expansion policy.