Schooling Externalities, Technology and Productivity: Theory and Evidence from U.S. States

The recent literature on local schooling externalities in the U.S. is rather mixed: positive external effects of average education levels are hardly to be found but, in contrast, positive externalities from the share of college graduates can often be identified. This paper proposes a simple model to reconcile this mixed evidence. The key idea is that advanced technologies are complementary to highly educated workers, as opposed to traditional technologies which are complementary to less educated workers. Our calibrated model predicts that workers with high school education or less are employed in the traditional sector, while more educated workers are employed in the advanced sector. As the advanced sector is associated with the production of differentiated goods and services this generates a positive pecuniary externality (positive TFP effect) of college educated workers. By contrast, as no externalities are associated with the traditional technology, high school education only increases private returns. The model predictions are tested using data on U.S. states. We use compulsory attendance and child labor laws, push-driven immigration of highly educated workers and the location of Land Grant colleges as instruments for schooling attainments of workers in different states. The empirical estimates confirm that an increase in college education, but not an increase in high school education, had significant positive production externalities in U.S. states during the period 1960-2000.