Yusuke Adachi (p)

Patricia Melo

We provide a theory and evidence that highway construction in metropolitan areas has a causal effect on residents and employment decentralization. We develop a theoretical model that illustrates the effect of highway construction on workers' residence and workplace choices. The analysis of the model indicates that highway construction reduces not only the population, but also employment in the center of metropolitan areas when this area has a large population. To examine the validity of our theoretical model, we use partial identification with data on central cities in the United States from 1950 to 1990. The results imply that our theoretical model is evidenced by all cities of metropolitan areas, though data on populated cities in 1950 do not support previous land use models.

Richard Franke (p), Sebastian Braun

Patricia Melo

This paper provides a comprehensive assessment of the effect of railways on the spatial economic development of a German economy, the Kingdom of Württemberg, during the Industrial Revolution. Our identification strategy compares the economic development of `winning' municipalities that were connected to the railway in 1845-54 to the development of `losing' municipalities that were the runners-up choice for a given railway line between two major towns. Estimates from both differences-in-differences and inverse-probability weighted models suggest that railway access increased annual population growth by 0.4 percentage points over more than half a century. Railways also increased wages, income and housing values, in line with predictions of economic geography models of transport infrastructure improvements, reduced the gender wage gap, and accelerated the transition away from agriculture. We find little evidence that these effects are driven by localised displacement effects.