Productivity differences: the importance of intra-state black-white schooling differences across the United States, 1840-2000

Using newly created data containing real output per worker, real physical capital per worker, and human capital per worker for US states from 1840 to 2000, Turner et. al (2007) analyze the growth rates of aggregate inputs and total factor productivity (TFP). We continue this line of work by documenting the importance of TFP differences in explaining cross sectional variation in the levels of (log) output. We construct plausible upper bounds on the fraction of the variance in output levels that can be explained by TFP and inputs. Similar to the growth rate analysis, we find that TFP can, on average, explain nearly 90% of output variance while inputs can explain up to only 50% of output variance. We then consider the possibility that one major institutional difference across states, the extent to which blacks were denied access to formal education, might explain TFP differences across states. To this end, we generate and present a years of schooling measures, by race, at the state level from 1840 to 2000. While directly exploiting this series has very little impact on the upper bound of the fraction of output variation that can be explained by inputs, we do find that the size of the gap between white and black years of schooling is negatively related to TFP in the period from 1840 to 1950. We also consider the extent to which time-varying rates of return on education alters the upper bound on the fraction of output variation that can be explained by inputs, finding that time-varying rates have little impact. Finally, we find some evidence for external effects of higher education and physical capital.