Railroads and American Economic Growth: a " Market Access " Approach *

This paper examines the historical impact of railroads on the American economy. Expansion of the railroad network and decreased trade costs may affect all counties directly or indirectly, an econometric challenge in many empirical settings. However, the total impact on each county can be summarized by changes in that county’s “market access,” a reduced-form expression derived from general equilibrium trade theory. We measure counties’ market access by constructing a network database of railroads and waterways and calculating lowest-cost county-to-county freight routes. As the railroad network expanded from 1870 to 1890, changes in market access are capitalized in agricultural land values with an estimated elasticity of 1.5. Removing all railroads in 1890 would decrease the total value of US agricultural land by 73% and GNP by 6.3%, more than double social saving estimates (Fogel 1964). Fogel’s proposed Midwestern canals would mitigate only 8% of losses from removing railroads. ∗For helpful comments and suggestions, we thank seminar participants at Berkeley, Harvard, Stanford, UC-Davis, and UC-Irvine. We are grateful to Jeremy Atack and coauthors for sharing their data. Irene Chen, Andrew Das Sarma, Manning Ding, Jan Kozak, Meredith McPhail, and Rui Wang provided excellent research assistance. During the 19th century, railroads spread throughout a growing United States as the economy rose to global prominence. Railroads became the dominant form of freight transportation and areas around railroad lines prospered. The early historical literature often presumed that railroads were indispensable to the United States’ economy or, at least, very influential for economic growth. Understanding the development of the American economy is shaped by an understanding of the impact of railroads and, more generally, the impact of market integration. In Railroads and American Economic Growth, Fogel (1964) transformed the academic literature by using a social saving methodology to focus attention on counterfactuals: in the absence of railroads, freight transportation by rivers and canals would have been only moderately more expensive along most common routes. Small differences in freight rates can cause some areas to thrive relative to others, but the aggregate economic impact may be small. This social saving methodology has been widely applied to transportation improvements and other technological innovations, though many scholars have discussed both practical and theoretical limitations of the approach (see, e.g., Lebergott 1966; Nerlove 1966; McClelland 1968; David 1969; White 1976; Fogel 1979). There is an appeal to a methodology that estimates directly the impacts of railroads, using increasingly-available county-level data and digitized railroad maps. Recent work has compared counties that received railroads to counties that did not (Haines and Margo 2008; Atack and Margo 2010; Atack et al. 2010; Atack et al. 2011), and similar methods have been used to estimate impacts of railroads in modern China (Banerjee et al. 2010) or highways in the United States (Baum-Snow 2007; Michaels 2008). These studies estimate relative impacts of transportation improvements; for example, due to displacement and complementarities, areas without railroads and areas with previous railroads are also affected when railroads are extended to new areas. This paper develops an alternative methodology for estimating the aggregate impact of railroads on the American economy. Expansion of the railroad network affects each county’s “market access,” a reduced-form expression derived from general equilibrium trade theory. County market access increases when it becomes cheaper to trade with another county, particularly when that other county has a larger population and higher trade costs with other counties. Rather than focus on changes in local railroad density, which may also be endogenous to local economic conditions, changes in market access summarize the total direct and indirect impacts on each county from changes in the national railroad network. One alternative approach is to create a computational general equilibrium model, with the explicit inclusion of multiple regions separated by a transportation technology (e.g., Williamson 1974; Herrendorf et al. 2009). In related work using a similar model, Redding and Sturm (2007) estimate the impact on population