Ride sharing and self-driving cars may transform cities, but the key to really understanding modern metropolitan areas is the steam locomotive, according to new research.

Steam locomotives and railways dramatically redefined cities by separating business and manufacturing districts from residential areas, argue University of Bristol’s Stephan Heblich, Princeton’s Stephen J. Redding, and London School of Economics’ Daniel M. Sturm.

The researchers analyzed more than a century of London’s demographics along with its commercial and residential development. They used London as a prototype of sprawling modern metropolises, in which vast numbers of people live significant distances from where they work.

They find that the development of the steam engine and rail network—more than the cotton gin, light bulb, telegraph, automobile, or airplane—may have been the single biggest factor in modern London’s growth.

When the steam locomotive was invented in the early 19th century, it more than tripled average travel speeds, from 6 mph to 21 mph. Greater London’s growth took off with the building of the city’s suburban rail network and the London Underground. Many cities around the world have since experienced the same phenomenon, and today, China’s rapidly growing urban areas are rushing to build networks of subways and light-rail systems for commuters. By 2020, China’s rapid-transit system could span 4,000 miles and cost $300 billion all told.

In 1801, the researchers report, London had a population of 1 million and spanned 5 miles east to west. A century later, London was the world’s largest city, with 6.5 million people. It measured 17 miles across—a sprawl, the researchers conclude, that is another legacy of the British capital’s extensive railway network.

Following the arrival of the railway, they find, there was a reduction in relative population growth in civil parishes close to the center of Greater London, and an increase in relative population growth in parishes further out.

A line chart plotting the change in City of London’s population, with the count of people on the y-axis and the years of 1831 to 1921 on the x-axis, including an annotation marking the Eighteen Fifties as when the railway network begins to proliferate. One line tracks the nighttime population, including people who reside within the city, which is steady around one hundred fifty thousand until 1850, after which it gradually descends to nearly zero. A second line tracks the researchers’ daytime population model which starts similarly but grows after 1850 reaching nearly four hundred thousand in the Nineteen-Tens. A third line is an actual accounting of daytime population, which does not begin until the Eighteen Sixties, but closely tracks the researchers’ model thereafter.A scatterplot charts the change in London-area parishes’ population growth rates from thirty years before the arrival of a railway station to thirty years after, with percentage points on the y-axis and the location’s distance from central London in kilometers on the x-axis. Growth rates are mostly negative for dots between zero and five on the x-axis and mostly positive, up to 2 points, for locations between five and thirty of the x-axis.

Heblich et al., 2018

“We find that much of the aggregate growth of Greater London can be explained by the new transport technology of the railway,” the researchers write. “Steam railways dramatically reduced travel times and hence permitted the first large-scale separation of workplace and residence to realize economies of scale” in business and manufacturing districts as well as services and amenities in residential areas.

The researchers explain these findings using a model of what they call the “spatial organization of economic activity within Greater London,” or peoples’ decisions about where to live and work within the city. Using data on bilateral commuting in 1921 and information on employment by residence and land values dating back to the early 19th century, they predict the impact of the construction of the railway network on employment by workplace in Greater London as early as 1831, before the construction of the first railway, in 1836.

The model successfully captures the sharp divergence between the night-time and day-time population in the City of London from the mid-19th century onward, and is also able to replicate the early commuting data, which show that at the dawn of the railway age, most people lived close to where they worked.

Removing the entire rail network, according to the model, would reduce Greater London’s population by 30 percent and “decrease commuting into the City of London from more than 370,000 in 1921 to less than 60,000,” the researchers write. By comparison, removing only the London Underground (or the “Tube”) would diminish the population by 8 percent and lower commuting into the city to just under 300,000.

More from Chicago Booth Review

Related Topics

More from Chicago Booth

Related Topics

Your Privacy
We want to demonstrate our commitment to your privacy. Please review Chicago Booth's privacy notice, which provides information explaining how and why we collect particular information when you visit our website.