Mare (2011) famously posited that increases in the generational overlap between grandparents and grandchildren decrease social mobility. Empirical support for this grandparent-overlap hypothesis, however, is scarce due to unresolved conceptual, methodological and empirical difficulties.  I reconceptualize grandparent overlap effects (GOEs) as multidimensional treatment effects and develop a new class of cumulative fixed-effects estimators to estimate GOEs from unique multigenerational Danish register data. I find that increasing grandparent overlap affects academic test scores and that the effect differs substantially across grandparent characteristics. This demonstrates empirically that rising life expectancy and the increase of grandparent overlap intensify the multigenerational transmission of social inequality.