All roads lead to Rome ... and to sprawl? Evidence from European cities

Abstract

I investigate the effect of highways on residential sprawl in European cities between 1990 and 2012. I find that a 10% increase in the stock of highways (km) causes a 0.4% growth in the residential land area, a 1.7% growth in the number of residential lots, and a 0.7% growth in the percentage of undeveloped land surrounding residential land over 20 years. At the regional level, only the effect on residential area is smaller in Northwestern cities than in Mediterranean and Eastern LUZs. I also explore the impact on population growth a la Duranton and Turner (2012) and find significant positive effects. Jointly, land and population results show a negative effect of highways on the intensity of use of land. As a whole, these results confirm that highways expand cities with more fragmented residential developments surrounded by undeveloped land and reducing the overall city density.