Smart Café Cities: Testing Human Capital Externalities in the Boston Metropolitan Area

Existing studies have explored either only one or two of the mechanisms that human capital externalities percolate at only macrogeographic levels. This paper, by using the 1990 Massachusetts census data, tests four mechanisms at the microgeographic levels in the Boston metropolitan area labor market. We propose that individual workers can learn from their occupational and industrial peers in the same local labor market through four channels: depth of human capital stock, Marshallian labor market externalities, Jacobs labor market externalities, and thickness of the local labor market. We find that all types of human capital externalities are significant across census tracts and blocks. Marshallian labor market externalities and the effect of labor market thickness in terms of industry employment density are significant at the block level. The mechanisms of knowledge spillovers vary across industries and occupations. Different types of externalities attenuate at different speeds over geographic distances. The effect of labor market thickness -- in terms of industry employment density -- decays rapidly beyond 1.5 miles away from block centroid; the effect of human capital depth decays rapidly beyond three miles; while Jacobs externalities decay very slowly, indicating a certain degree of urbanization economies. We conclude that knowledge spillovers are very localized within microgeographic scope in cities that we call, "Smart Cafe Cities."