This paper by Enrico Moretti examines the impact of workers' education on productivity in US cities, focusing on the spillover effects of human capital. Using a unique firm-worker matched dataset from the Census of Manufacturers and the Census of Population, Moretti estimates plant-level production functions to assess the productivity gains associated with increases in the share of college graduates in a city. The study finds that plants located in cities with higher levels of human capital experience significantly higher output compared to similar plants in cities with lower levels of human capital. The spillover effects are particularly pronounced in high-tech plants, while they are negligible in low-tech plants. The estimated productivity differences between cities with high and low levels of human capital align well with observed wage differences, suggesting that the productivity gains from human capital spillovers are offset by increased labor costs. The paper also includes several specification tests to validate the results, such as examining the impact of physical capital and the interaction between plants within the same city. Overall, the findings suggest that human capital spillovers have a significant but limited impact on economic growth.This paper by Enrico Moretti examines the impact of workers' education on productivity in US cities, focusing on the spillover effects of human capital. Using a unique firm-worker matched dataset from the Census of Manufacturers and the Census of Population, Moretti estimates plant-level production functions to assess the productivity gains associated with increases in the share of college graduates in a city. The study finds that plants located in cities with higher levels of human capital experience significantly higher output compared to similar plants in cities with lower levels of human capital. The spillover effects are particularly pronounced in high-tech plants, while they are negligible in low-tech plants. The estimated productivity differences between cities with high and low levels of human capital align well with observed wage differences, suggesting that the productivity gains from human capital spillovers are offset by increased labor costs. The paper also includes several specification tests to validate the results, such as examining the impact of physical capital and the interaction between plants within the same city. Overall, the findings suggest that human capital spillovers have a significant but limited impact on economic growth.