Railroads and Growth in Prussia

We study the effect of railroad access on urban population growth. Using GIS techniques, we match triennial population data for roughly 1,000 cities in nineteenth-century Prussia to georeferenced maps of the German railroad network. We find positive short- and long-term effects of having a station on urban growth for different periods during 1840--1871. Causal effects of (potentially endogenous) railroad access on city growth are identified using propensity score matching, instrumental variables, and fixed-effects estimation techniques. Our instrument identifies exogenous variation in railroad access by constructing straight-line corridors between nodes. Counterfactual models using pre-railroad growth yield no evidence to support the hypothesis that railroads appeared as a consequence of a previous growth spurt.