We examine the location and growth of the U.S. population using county-level census data from 1840 and 1990. Counties are described by natural and produced characteristics they possessed in 1840. Natural characteristics include climate, mineral resources and access to natural transportation networks. Produced characteristics include industry mix, educational infrastructure, literacy rates, and access to man-made transportation systems. We investigate how natural characteristics influenced settlement patterns in 1840, and how natural and produced characteristics influenced population growth over the subsequent 150 years. We find that natural characteristics heavily influenced where populations located in 1840. We also that educational infrastructure, literacy rates, industry mix, and access to transportation networks had a significant influence on growth. There is little evidence of population convergence among our full sample of counties; such evidence appears only when the most-heavily-populated counties in 1840 are excluded from the sample. Moreover, when counties located on the western frontier are excluded from the full sample, on the assumption that they were relatively far from their steady state populations, there is evidence of population divergence.We thank Van Beck Hall for help identifying data sources; Carol Kraker and Steve Lehrer for research assistance; and

No abstract

New water purification technologies led to large mortality declines by helping eliminate typhoid fever and other waterborne diseases. We examine how this affected human capital formation using early-life typhoid fatality rates to proxy for water quality. We merge city-level data to individuals linked between the 1900 and 1940 Censuses. Eliminating early-life exposure to typhoid fever increased later-life earnings by one percent and educational attainment by one month. Instrumenting for typhoid fever using typhoid rates from cities that lie upstream produces results nine times larger. The increase in earnings from eliminating typhoid fever more than offset the cost of elimination.

This paper examines the effect of water-borne lead exposure on infant mortality in American cities over the period . Infants are highly sensitive to lead, and more broadly are a marker for current environmental conditions. The effects of lead on infant mortality are identified by variation across cities in water acidity and the types of service pipes that the water ran through -lead, iron, or concrete -which together determined the extent of lead exposure. Estimates that restrict the sample to cities with lead pipes and panel estimates provide further support for the causal link between water-borne lead and infant mortality. The magnitudes of the effects were large. In 1900, a decline in exposure equivalent to an increase in pH from 6.675 (25th percentile) to 7.3 (50th percentile) in cities with lead-only pipes would have been associated with a decrease in infant mortality of 7 to 33 percent or at least 12 fewer infant deaths per 1,000 live births. This paper examines the effect of water-borne lead exposure on infant mortality in American cities over the period 1900-1920. Infants are highly sensitive to lead, and more broadly are a marker for current environmental conditions. The effects of lead on infant mortality are identified by variation across cities in water acidity and the types of service pipes that the water ran through -lead, iron, or concrete -which together determined the extent of lead exposure.Estimates that restrict the sample to cities with lead pipes and panel estimates provide further support for the causal link between water-borne lead and infant mortality. The magnitudes of the effects were large. In 1900, a decline in exposure equivalent to an increase in pH from 6.675 (25th percentile) to 7.3 (50th percentile) in cities with lead-only pipes would have been associated with a decrease in infant mortality of 7 to 33 percent or at least 12 fewer infant deaths per 1,000 live births.

No abstract