The role of educational attainment and migration on human capital and income growth in the United States, 1840--2000

This primary contribution of this work is to provide new data on annual average years of schooling and per capita incomes at the state level of the United States from 1840--2000. The New England, Middle Atlantic, East North Central, West North Central, and Pacific regions have been educational leaders over the entire period, while the South Atlantic, East South Central, and West South Central regions have been educational laggards. Estimates of the returns to schooling indicate that a year of schooling increased income by between 8 and 15 percent, with a point estimate close to 9 percent.;The relative importance of human capital and total factor productivity in explaining output growth across states is analyzed. Under alternative assumptions concerning the stock of physical capital across states, between 49% and 68% of output growth is explained by aggregate input growth. While explaining a large fraction of the levels of output growth, aggregate input growth explains only a small fraction of the variance of output growth across states.;The effect of migration on average educational attainment levels across states is examined. We decompose migration into two components: quasi-random migration (migrants are assumed to be randomly selected from the educational distribution of the state from which they leave) and selective migration (migrants are not randomly chosen). We calculate a measure of quasi-random migration from 1840--2000, and a measure of selective migration from 1940--2000. Quasi-random migration has a dramatic effect on the income levels of states and regions prior to 1900 and reduces the standard deviation of educational attainment levels across states, especially from 1980--2000. Selective migration has little effect on the income levels of census regions. The vast bulk of the effect on educational attainment levels can be accounted for by the quasi-random component.