From ancient Rome to modern Atlanta, the shape of cities has been defined by the technologies that allow commuters to get to work in about 30 minutes.
In 1994, Cesare Marchetti, an Italian physicist, described an idea that has come to be known as the Marchetti Constant. In general, he declared, people have always been willing to commute for about a half-hour, one way, from their homes each day.
This principle has profound implications for urban life. The value of land is governed by its accessibility—which is to say, by the reasonable speed of transport to reach it.
Even if there is a vast amount of land available in the country, that land has no value in an urban context, unless transportation makes it quickly accessible to the urban core. And that pattern has repeated itself, again and again, as new mobility modes have appeared. This means that the physical size of cities is a function of the speed of the transportation technologies that are available. And, as speed increases, cities can occupy more land, bringing down the price of land, and therefore of housing, in newly accessed territory.
Modern Atlanta may bear little resemblance to the cities of past millennia, but its current residents share something fundamental with urbanites of the distant past: The average one-way commute time in American metropolitan areas today is about 26 minutes. That figure varies from city to city, and from person to person: Some places have significant numbers of workers who enjoy or endure particularly short or long commutes; some people are willing to travel for much longer than 30 minutes. But the endurance of the Marchetti Constant has profound implications for urban life. It means that the average speed of our transportation technologies does more than anything to shape the physical structure of our cities.
To see how, let’s travel back in time by more than 2,000 years, and move towards the present. (This journey will take less than 30 minutes.)
The city on foot: 800 BC-1700 AD
Until the Industrial Revolution, there was pretty much only one way for most people on dry land to get around: on foot. With services concentrated in the center of cities, the radius of development from the heart of the city was limited to not much more than one mile—about the distance a person can walk in 30 minutes.
Sure enough, most cities from the ancients to the Industrial Revolution did not grow much bigger than a two-mile diameter. Their core areas were often even smaller, though some of the poor lived in settlements outside the city gates. Ancient Rome packed as many as a million people into an area a little more than two miles in diameter. Medieval Paris stretched about two miles from the Bastille to the Louvre, Vienna’s Innere Stadt measures only one mile in diameter, and the historic City of London is nicknamed the “Square Mile” for a reason. Beijing’s walls enclosed an inner city about three miles in diameter; into the 20th century, that still made up most of the developed area.
So as these old cities grew, they became denser. The narrow warrens of surviving European medieval cities, like Florence, Prague, or a few parts of Paris untouched by Haussmann, give one a good idea of this process. The arrival of horse-drawn mass transportation—the first fixed-route buses were created by mathematician Blaise Pascal in Paris in 1662—didn’t change this pattern; hooves didn’t get people around much faster than feet did.
For most residents, the cities of this era were awful places to live, with extreme crowding, dismal sanitation, and routine outbreaks of vicious disease. Tenement apartments may seem livable when occupied by a couple, with an air conditioner in the window; they weren’t so pleasant when occupied eight to a room. The idea of escaping the city to the bucolic countryside is therefore as old as the city itself. Aristocrats kept country estates and villas for thousands of years; Vesuvius buried some examples under its rain of ash.
But even for the wealthiest, long-distance travel was an uncomfortable exercise, involving days on a horse or in a carriage on a rutty road. When Louis XIV centralized the French nobility in Versailles, he could be sure that their visits to their domains would be infrequent, which served his goal of limiting their power.
For the aristocracy of the pre-industrial age, commuting from a rural estate to a job in the city wasn’t possible on a daily basis—but such people did not tend to have what modern people would consider to be normal jobs. By the 19th century, the mercantile elite could not afford to be too distant from their urban enterprises. Invention quickly allowed the privileged few an escape.
The city on rails: 1840s-1950s
In 1830, British civil engineer George Stephenson opened the world’s first steam-powered public railway between Manchester and Liverpool. Soon, railways focused as much on proto-commuters as they did on intercity travelers. The London and Greenwich railway opened in 1836, and new lines quickly began radiating out of central London. Cities across Europe and the United States soon followed.
Steam trains couldn’t stop very frequently, because of their slow acceleration, and they were comparatively expensive. But once in motion, they could move passengers at an unprecedented rate: 10 miles or more in a half-hour. Instead of gradually extending the city, they created little villages around their stations a couple miles apart—railroad suburbs. Because access to the city was expensive, they were limited to all but the very wealthy and those who served them. The upper middle class lived within walking distance of the station; the truly rich lived on estates a short coach ride away.
Today, many of these villages remain enclaves of affluence: The Main Line suburbs of Philadelphia, Scarsdale near New York, Brookline near Boston, the “Stockbroker Belt” west of London, and Saint-Germain-en-Laye near Paris are excellent examples. America’s railroad city, Chicago, spawned several of the most legendary railroad suburbs, such as Evanston, Oak Park, and Lake Forest.
But for the urban masses, a different mobility technology would be needed. The horse-drawn omnibuses (or horsecars, their successors on rails) would not cut it. Thousands of workhorses pulling hundreds of wagons around the city created a serious environmental hazard, as the piles of manure on the streets could be literally as high as the people walking past. The smell of 19th-century city was beyond description, and the public desire for a faster and less malodorous form of mass transportation was fervent.
The city of bicycles and streetcars: 1880s-1950s
The invention of bicycles and electric streetcars provided such an alternative. The “safety bicycle,” which began supplanting towering “high-wheelers” in 1885, allowed any reasonably physically fit person to travel considerably faster than on foot. The first electric streetcar system entered service in 1888 in Richmond, Virginia, and the technology spread at the speed of electricity.
In practice, streetcars and bicycles could cover about 4 miles in a half-hour—much slower than the steam railroads carrying executives to their semi-rural estates, but much quicker than the horsecar and the human pedestrian. Suddenly the city itself was no longer limited to a few square miles, and developable areas grew exponentially. A walking city 2 miles in diameter covered only a little over 3 square miles; a streetcar city 8 miles in diameter could cover 50 square miles.* It was the first great mass-market suburban boom.
Cities raced to grant streetcar franchises to entrepreneurs, and the industry experienced a boom much like that of the internet in the 1990s. Within a few years, every American city had at least a modest trolley network. Bigger cities had a web of rails sprawling outward in all directions, producing the urban typology that defined the turn-of-the-century era: the streetcar suburb. The middle class was no longer confined to crowded tenements; new neighborhoods sprouted on cheap farmland. They often took the familiar form of a main street lined with shops, with small streets full of single-family homes and townhouses running off that central spine.
The slower spread of the streetcar in Europe, let alone the rest of the world, contributed to the comparatively compact prewar design of their cities. Even in European cities like Berlin that grew quickly in the late 19th century, most new buildings were walk-up apartments.
The city underground: 1900s-1950s
London was a major exception. The world’s largest city needed a faster means of transportation. The poor still clustered in extreme density and squalor near their places of employment in the city center and docks, but railroads, by then quite high-capacity and frequent, enabled the middle class to spread into an array of suburbs, making London a more low-rise and less-dense city than its European neighbors.
The problem was getting from the various London train terminals to jobs in London’s traditional business district and elsewhere in the central city. Unlike in the poorer areas to the east and south, building elevated viaducts through some of the world’s most valuable real estate was infeasible. Charles Pearson promoted an ingenious solution: go underneath.
The first London Underground line opened in 1863, using steam trains. The route was a sensation, but clouds of smoke made the system a bit impractical, as riders came out covered in black soot. The solution—electricity—helped create the Tube we know today.
In America, the comparatively new cities tended to develop on a grid of straight roads, which meant that trains could sometimes be elevated over top of the streets without unacceptable curves rather than buried underground, as in central London and most old European cities. El trains were an ugly solution, but they worked. Cities like Chicago, Boston, Philadelphia, and New York City covered their streets with steel skeletons, and trains were soon rumbling above. They carried the growth of New York, for example, up the island of Manhattan. If trolleys and streetcars had brought midtown within a reasonable commute, elevated trains enabled nearly the entire island to be developed. A web of els spread throughout Brooklyn, connected by the Brooklyn Bridge to Manhattan. The system predated the city’s subway by several decades.
Elevated trains and (later) subways meant that even the working class could live as far as eight miles from their places of work and still get there in a half-hour. They could afford it, too, because many of the systems had a cheap flat fare regardless of distance. The teeming blocks of tenements began to thin out: New homes could be built on land cheap enough for the working class to afford a decent apartment, though racism built into the market limited the avenues of escape for many immigrants and, especially, African Americans.
Those who did flee left more room for businesses, like the skyscrapers that emerged in Chicago around the turn of the century, and the department stores like Macy’s and Marshall Field’s that occupied the most prominent corners in every city. Along with attractions like palatial vaudeville and movie theaters, these urban features draw people from miles around, since people were no longer restricted to a close radius around their homes.
The combination of technologies created the form of cities before World War II. Dense downtowns had high-capacity rail transit radiating outward—subways and elevated trains in big cities, which produced apartment suburbs that were relatively high-density, but still much lower than the old tenement slums; and streetcar suburbs, with single-family homes and townhouses. Some of the richest people lived beyond, in small towns and on estates accessible to the central city by railroad.
But this order was not to last.
The city of the expressway: 1950s-present
Like many technologies, the car began as a toy for the rich, used to reach their country estates or to race around on tracks. As a means of basic transportation, it was severely limited by high cost and lack of paved roads. That started to change in the U.S. with the introduction of Ford’s Model T in 1908, which was mass produced at a price that was affordable to this country’s comparatively affluent middle class.
Still, car commuting was not much more convenient or fast than existing mass transit. Roads built for carriages, wagons, and pedestrians soon became severely overcrowded. Cities in the 1920s and 1930s spread a little bit more than they had when everyone relied on the streetcar, especially in places like Detroit and Los Angeles where roads were wide and auto ownership was highest, but the age of the automobile would await another technology—the expressway—and the huge infrastructure transformation that adapted the city to the car.
After World War II, two dominant urban planning ideas contended to shape the city of the future. Both sought to solve what was seen as the great urban problem: overcrowding. The first was the “towers in a park” model famously espoused by Swiss-French architect Le Corbusier, who envisioned replacing crowded old cities with huge skyscrapers surrounded by acres of greenery. This vision was implemented in various adaptations in the post-war new towns in Europe, like La Défense and Clichy-sous-Bois outside Paris, Thamesmead outside London, and Vällingby near Stockholm. In the United States, his designs found favor mostly in government projects, like the monumental Empire State Plaza in Albany, and especially in public housing.
For the most part, Americans took another path. Frank Lloyd Wright developed the idea of the Broadacre City in 1932. He believed that the individual mobility provided by the car had made the city obsolete and proposed dispersing the population onto one-acre plots of land. It was a Jeffersonian vision that appealed to the deep-rooted American suspicion of the city. In 1939, a more commercial version was presented at General Motors’ Futurama exhibit at the New York World’s Fair. Broad expressways carried people into the countryside, where they could live in houses on lots that allowed plenty of room for a yard.
It is rare that one finds a past vision of the future that came true so precisely—abetted by lavish government assistance to build highways and widen urban streets into automobile-oriented thoroughfares. After all, Wright had a point: Why crowd into downtowns when we could all drive anywhere we wanted?
The car on the expressway enabled large numbers of people to travel long distances on a day-to-day basis. Instead of small railroad suburbs, where housing was restricted to a short radius around stations, drivers spread out across suburbs could now commute 20 miles in 30 minutes. If the streetcar city covered 50 square miles, the 40-mile-diameter expressway city could cover over 1,250 square miles.*
Even with growing population, the supply of cheap farmland available for housing development appeared endless, and the cost of a detached house collapsed to be within the reach of the working class. In 1946, a house in Levittown, New York, the prototypical postwar automobile suburb, cost $78,469 in today’s dollars, with little or no money down required. The age of urban sprawl began.
Many older, denser cities began to lose population as residents chased the bonanza of cheap land. The great housing crisis of the 1970s was not a shortage—it was a surplus. When Jimmy Carter visited Charlotte Street in the Bronx in 1977, he saw a whole street of apartment houses lying almost completely vacant. And African Americans—kept out of the suburbs for decades by legal and extra-legal measures—were abandoned within declining central cities.
The problem with dispersion is that people can’t simply live next to where they work, especially in dual-income families. Good luck finding a home next to both a bank headquarters and a steel mill. And, of course, many industries like finance and media thrive on concentration. In the U.S., dispersed cities like Los Angeles and Atlanta ended up having even worse traffic than the relatively concentrated ones like New York City, as hordes of far-flung commuters pressing in daily to get to their jobs.
Today, like its predecessors, the expressway has struck a technological threshold. Its downfall is its limited capacity. Even the biggest highway can move at most a couple hundred thousand cars per day—sufficient for small cities and for long-distance trips, but inadequate in a city of millions. And while people were once content—or at least powerless to resist—the demolition of thousands of homes and entire neighborhoods for new urban highways, those days are also over.
The city of the future
The constant in transportation technology from the mid-19th to the mid-20th century was change. New modes repeatedly extended the boundaries of cities and changed the way we lived. But that was an aberration over the course of human history. Since then? Not too much has changed. A person can navigate New York City almost perfectly with a fifty-year-old map.
This has had real consequences. In spread-out metros that are growing in population, highways quickly become overcrowded; expanding them is costly and ultimately ineffective. Commute speeds are slowing inexorably as congestion increases. In North America, at least, rail transit is too expensive to build in meaningful amounts and it faces formidable ideological opposition. And the vaunted self-driving car, as imminent and yet illusory as nuclear fusion, will not transform the basic geometry of road capacity. Could they squeeze out a few percent more from the legacy of the 1950s and 60s? Maybe. But that will buy us at most a few years.
The greatest promise for matching technology to the modern worker has always been the idea of divorcing work from transportation entirely: telecommuting. The tools that would enable white-collar workers to clock in remotely have been available for decades and have improved dramatically in the digital era. That could theoretically finally enable Wright’s vision of the complete dispersal of population. But would it? Despite evidence to the contrary, employers remain skeptical of the productivity of remote workers. And any number of human drives keep people stubbornly collocating to be closer to family, friends, and cultural amenities as well as their workplaces. These drives are unlikely to change with technology—and thus, our transportation dilemma is likely to endure.
The best option is to densify our cities. This is hard, too: Adding housing in established neighborhoods will always be more complex and expensive than building on empty farmland. Real estate also remains comparatively cheap in declining or decentralized cities. When many people don’t really care how close they are to the historic urban center, like in Atlanta or Houston, cities can sprawl basically ad infinitum. But the environmental cost is huge, and it’s simply not an option in parts of the country where cities have already grown into each other, like the Northeast Corridor and Southern California.
For a century, we lived off the legacy of rapid innovation. It allowed our cities to grow exponentially and, therefore, the cost of our housing to decrease dramatically. But we’ve now pretty much burned through the benefits of these gains and there aren’t obvious technological saviors on the horizon. We must make do mostly with building up and densifying the urban areas we already have. As transportation goes, so go our cities.
*CORRECTION: A previous version of this article incorrectly calculated the size of the urban footprints associated with streetcars and expressways.