While theory suggests that public expenditures on education may affect intergenerational
earnings mobility, the direction of the effect hinges on whether such outlays substitute for
or complement private human capital investments. This paper empirically evaluates the
question using census data in the Integrated Public Use Micro Sample from 1940-2000.
The results show that state-cohorts with smaller class sizes generally enjoy less
intergenerational mobility, indicating that class size reductions benefit children from
high-income families more than those from low-income families. The size of the effect is
substantial: the effect of moving from one standard deviation above to one standard
deviation below the mean class size increases earnings persistence by more than 40%.
These results are robust to controls for the average class size in the state in years the
individual was not in school, a finding which rules out many endogeneity explanations.