SEMINAR 12 November 2012: Kerstin Enflo, Lund University of Economics and Management

Locomotives of Local Growth: The Swedish Railroad Network as a Quasi-Natural Experiment, 1856-1890

Abstract

We treat the Swedish railroad network as a quasi-natural experiment to analyze the effect of infrastructure improvement on urban growth, industrialization and productivity. Among all Swedish cities, we find that cities that gained access to the railroad network on average grew 2 percentage points faster than those that did not. This accounts for half of the total growth of these cities between 1856 and 1890. In addition, cities that gained access were on average twice as industrial in 1870. In cities that gained access to the railroad network production was more likely to be organized in factories rather than along artisanal lines, factories were bigger, and productivity was higher.

The Swedish railroad network, initiated by the state in 1856, is interesting since the lines were not drawn to connect the economically most important cities. Contemporaries at the time criticized the plan for its “horror of waterways and cities” since they found that it retarded rather than favored many old cities during the early stage of the construction. We show that cities that gained access to the railroad (our treatment group) shared a common growth trend with cities that did not (our control group) prior to the construction of the network, supporting the notion that the plan was laid out in a quasi-randomized fashion. We control for endogeneity by omitting major terminal cities and we find no spurious growth effects for cities that were intended to gain access in alternative plans that were never constructed.