The question is not whether we want to live in cities. Obviously, a growing number of us do—otherwise we would not build so many of them. The real question is: In what kind of cities do we want to live? Compact or spread out? Old or new? Big or small?
Judging from the direction that American urbanism has taken during the second half of the 20th century, one answer is unequivocal—Americans want to live in cities that are spread out. Decentralization and dispersal, the results of a demand for private property, privacy, and detached family homes, have been facilitated by a succession of transportation and communication technologies: first, the railroad and the streetcar; later, the automobile and the airplane; lastly, the telephone, television, and the Internet. In addition, regional shopping malls, FedEx, UPS, the Home Shopping Network, and Amazon.com have helped people to spread out. Even environmental technologies—small sewage treatment facilities and micro power plants—have allowed people to live in more dispersed communities than in the past.
This is not simply suburbanization. All the cities that have experienced vigorous population growth during the second half of the 20th century—Houston; Phoenix, Ariz.; Dallas; San Jose, Calif.; Atlanta, Ga.—have grown by spreading out. These are horizontal cities, with generally low population densities, typically fewer than 10 people per acre compared with 15 to 20 people per acre in the older, vertical cities. Horizontal cities depend on automobiles for mass transportation and on trucks for the movement of goods. In a horizontal city, the difference between city and suburb is indistinct. People in both live chiefly in individual houses rather than in flats or apartment buildings, and the houses are organized in dispersed, semi-autonomous planned communities that are different from the urban neighborhoods of the past. Versions of the dispersed city can be found in large cities such as Los Angeles, small cities such as Las Vegas, and in the metropolitan areas surrounding all cities, old and new.
The Technological Reshaping of Metropolitan America, a 1995 report of the federal Office of Technology Assessment, concluded, "Given the technological and economic trends toward decentralization, America's central and inner cities are unlikely to regain their earlier dominance." Decentralization suits an economy that depends on flexibility, adaptability, and rapid change. Dispersal also suits an increasingly heterogeneous society, which is the exact opposite of what is implied by that misleading term global village. Despite living in an urban civilization, Americans are not more alike today, they are more dissimilar, and dispersal handily accommodates different lifestyles, values, and, of course, incomes.
Horizontal cities have another characteristic—they tend to be new. During the industrial era, advanced infrastructure, good ports, and large workforces gave established cities a head start. In a postindustrial era, however, being old is no longer an advantage; a new community needs but to lay down fiber-optic cables, build a walkable downtown, and entice a Whole Foods and a Target to be competitive. I am always struck, returning from Seattle or Denver, Colo., to my home in Philadelphia, by the urban contrasts. The difference is not just a century's worth of accumulated industrial grime compared with unsullied newness, or that so much of Philadelphia's infrastructure—the stone viaducts, the narrow turnpikes, the commuter railroad lines—are relics of the past. In Philadelphia, the new always seems slightly makeshift, shoehorned into an old mold that doesn't quite fit. The past is part of the city's charm—and what keeps me here—but it exacts a price in terms of efficiency, convenience, and endless maintenance. An old city is like an old car: It still runs, it will get you there, but it doesn't have the safety features, conveniences, and efficiencies of a newer model.
Newness has other advantages. New cities generally have new, and often streamlined, building regulations, new social compacts between management and labor, and new ways of doing things. Houston's absence of zoning, its less restrictive building regulations, and its lower construction costs, for example, mean that according to the 2006 census, the average owner-occupied dwelling cost $126,000, compared with $496,000 in New York City. Since incomes in greater Houston are only slightly lower than in New York, that makes the city much more affordable for the middle class and, according to Harvard economist Edward L. Glaeser, accounts for Houston's greater appeal. Between 2000 and 2007, the city grew by 19.4 percent, compared with just 2.7 percent for New York. The presence of a large middle-class workforce also explains why Houston has more blue-collar manufacturing jobs than New York.
Another answer to what kind of cities we want seems to be: "warmer." A 2008 national survey of the major metropolitan areas that appealed most to people found that the favorites—Denver, San Diego, Seattle, Orlando, and Tampa—all shared "warm weather, a casual lifestyle and rapid growth." In fact, all of the Top 10 cities on the list were in temperate climates, seven in the West and three in the South. It also helps if a city is close to attractive natural amenities such as lakes, mountains, beaches, or the desert. Given the popular interest in the outdoors, and outdoor activities, nearby wilderness areas have become important urban amenities. As David Brooks put it, "These [favorite cities] are places where you can imagine yourself with a stuffed garage—filled with skis, kayaks, soccer equipment, hiking boots and boating equipment. These are places you can imagine yourself leading an active outdoor lifestyle." Industrial cities didn't need beautiful settings; postindustrial cities do.
Virtually every technological innovation of the last 50 years has facilitated, if not actually encouraged, urban dispersal. But the long-term effects of new technologies are often unpredictable. The telephone is, on the face of it, a decentralizing device. Yet telephone communication made working in high-rise office buildings practicable, which in turn produced the concentrated central business district. When people started commuting in cars, an unexpected fallout was the death of the evening newspaper, which used to be read on the streetcar trip home. People couldn't read in their cars, but they could listen to the radio, and that moribund medium found new life. Who could have foreseen that the combination of car radios and cellular phones would give rise to a new format, talk radio, and a new and potent political force, the talk-radio host? Laptops, personal digital assistants, and cell phones are held to be the tips of a great dispersal iceberg, but the migration of work to the motel room and the home office has been accompanied by a countervailing trend: the need for face-to-face contact. That is why there are more conferences, retreats, and conventions than ever before. That is also why so many industries—finance, publishing, entertainment, high tech, and communications—cluster together geographically.
Thus even as dispersal appears to be the order of the day, concentration is making a comeback. Sometimes, concentration takes new forms: power centers, office parks, theme parks, and villagelike planned communities in the middle of nowhere. Sometimes the results of concentration are more familiar: downtown entertainment districts, historic neighborhoods, waterfront esplanades, and urban parks. All such gathering places are evidence of the age-old desire for human contact, crowds, variety, and expanded individual choices. This desire has breathed new life into many small cities, especially college towns, which, with their attendant research facilities, office parks, university hospitals, and cultural amenities, have blossomed and are among the fastest-growing and most attractive places to live and work. Part of this blossoming is the result of technology. Cable television, regional airlines, catalog shopping, and the Internet have brought big-city conveniences to small cities. But when college towns succeed as attractive and vital places to live—and by no means all do—the result is a potent synergy between higher education, information-age industries, and people's preferences for smaller, more intimate communities.
The appeal of the college town as a place to live and work is part of a larger trend that Irving Kristol identified as early as 1970. "If we are a 'nation of cities,' " he wrote, "we are also becoming to an ever greater degree a nation of relatively small and middle-sized cities." That was true then, and it is even truer today. In 1970, slightly more Americans lived in small (between 25,000 and 250,000 inhabitants) cities than in big (larger than 250,000) cities, reversing the situation of a decade earlier, when big cities dominated, as they had done for years. By 2006, although the total urban population had increased, the gap had considerably widened, and more than half again as many people lived in small cities as in large. Indeed, since 1970 the proportion of the urban population living in large cities has steadily declined, while the percentage living in small cities has grown, suggesting that what Americans don't want is to live in large metropolitan areas. This was confirmed by a 2008 Pew poll, which found "not a single one of the 30 [largest] metropolitan areas was judged by a majority of respondents as a place where they'd like to live."
Thus, for many Americans, the answer to "What kind of cities do we want?" seems to be not just dispersed, new, and warmer, but also smaller. In some cases, much smaller. In 2007, the fastest-appreciating residential real estate values in the nation were not in San Francisco, Boston, and New York City but in Corvallis, Ore. (population 53,000); Grand Junction, Colo. (population 46,000); and Wenatchee, Wash. (population 28,000). These small cities are examples of what Joel Garreau has christened the Santa Fe effect, referring to the New Mexico city that, while small (population 62,000), has a variety of big-city amenities including many restaurants and shops, an opera company, a chamber-music festival, and a renowned film festival. Places such as Wenatchee, a hundred miles from Seattle at the eastern foot of the Cascade Mountains, are characterized by attractive natural surroundings (deserts, lakes, mountains, rivers), moderate climates, and a degree of urbanity that belies their small size and their often remote location. (Wenatchee boasts a symphony orchestra, a community theater, and the Apollo Club, a men's choral group.) "The core premise of the Santa-Feing hypothesis is that the same kind of choice is now available to millions of Americans," writes Joel Garreau. "Today, people of means are attracted by this scenario of dramatically different settlement patterns that feature new aggregation—dispersed."
There isn't a single answer to the question "What kind of cities do we want?" because different people want so many different things. While the majority of us appear to prefer dispersed small cities, a significant minority want to live in concentrated big cities, and a tiny fraction is prepared to pay the price of living in the very center of things. Most of us want lively downtowns, at least to visit if not to live in. Nor is it simply a question of individual preferences; we want different things at different times: an exciting big city when we are young, beginning a career, and looking for a mate; a dispersed small city close to nature when we are raising a family; a culture-rich downtown when we are empty nesters; and a walkable small city in a warm climate when we retire. Since American cities are shaped by popular demand, one can expect them to exhibit a variety that is no less rich and diverse than the variety of Americans themselves.