Dense, Denser, Densest

Last fall, Foreign Policy published what it called a global cities index, a list of 65 world cities ranked according to a variety of economic, cultural, and social indicators. Compiled by the consulting firm A. T. Kearney and the Chicago Council on Global Affairs, the index measures business activity, the size of capital markets, and the flow of goods through airports and ports. It also takes into account cultural and information resources such as the number of performance venues, the extent of broadband access, international coverage in the local press, the degree of political engagement as measured by the number of think tanks and conferences, and university enrollment and education levels. The 2010 list predictably included global powerhouses and national capitals such as London, Paris, and Tokyo, but the United States had no less than six cities in the top 20—New York, Chicago, and Los Angeles (in the top 10), as well as San Francisco, Washington, and Boston.

Lists such as these have become commonplace, and American cities are often among the top ranked. It is hardly surprising that the United States contains so many leading global cities; after all, it is an economic superpower and a very large country. What is striking is that these cities are physically so different—large as well as small, old as well as new, horizontal as well as vertical, and sprawling as well as concentrated. Clearly there is no one-size-fits-all American urban template.

Consider New York, Los Angeles, and Washington. The popular image of New York, the oldest of the three, is of a small island crammed with skyscrapers. The reality is different. Manhattan is at the heart of a metropolitan region that stretches over parts of three states, more than 3,000 square miles of cities, suburbs, and small towns. Even within the five boroughs there is considerable variety between, say, Queens, where homeownership is the norm, and Manhattan, where a majority of residents are tenants.

Compared to New York, Los Angeles is very new; 100 years ago the city had barely 100,000 inhabitants. Metropolitan Los Angeles has a reputation as a sprawling, spread-out place, yet its urbanized area is half the size of New York’s. The Angeleno city fathers have worked hard to create a distinct downtown—with limited success so far—and Los Angeles continues to be a city of many subcenters (in that sense, at least, it resembles London). Unlike London, Los Angeles is not a walkable city, yet it is dense, with mile upon mile of cheek-by-jowl dingbats, boxy two- and three-story apartment buildings.

Washington resembles neither Los Angeles nor New York. Although there are tall buildings in Rosslyn, Virginia, just across the Potomac, downtown Washington has no high buildings at all, thanks to the District of Columbia’s roughly 10-story height limit. Historically, many American cities had height limits—Los Angeles as late as 1957. The difference is that in Washington, thanks to congressional inertia, the height restriction has persisted, making Washington look more Parisian than American. The skyline consists of civic landmarks rather than skyscrapers—the Washington Monument standing in for the Eiffel Tower—and downtown is dominated by bulky office and apartment buildings. This mid-rise pattern extends quite far toward the periphery, instead of dropping off quickly to single-family houses, as it does in most American cities.

What about the other top-rated cities? Chicago’s downtown Loop is as clearly defined as Manhattan—and similarly vertical—but the flat midwestern topography has allowed the urbanized area to extend unchecked in three directions—north, south, and especially west, far past O’Hare Airport. As a result, Chicago covers a larger area than any of the other six leading American cities except New York. The historic center of Boston is compact and walkable, and although there are some skyscrapers, there is no memorable skyline. On the other hand, metropolitan Boston spreads out more than either Los Angeles or Houston. Of the six global American cities, San Francisco is the outlier; not just the hilliest city, it is also the smallest in area. Hemmed in by water on three sides (as Oakland is contained by mountains), metropolitan San Francisco is less than a third the area of Boston, and packs in more inhabitants per square mile than any of the six global cities.

For that great analyst of urban life, Jane Jacobs, density was a critical measure of a city’s vitality. Indeed, density affects the energy of streets and other public spaces, as well as the variety of amenities a city can support. In addition, density dictates the type of mass transit that is viable—buses, streetcars, light rail, or subways. Yet density is not always what it appears to be. Los Angeles, somewhat counterintuitively, is extremely dense. So are San Francisco and New York. Chicago is somewhere in between; Washington, given its height restriction—and the low density of its suburban fringe—is further down the list, although denser than either sprawling Chicago or Boston, a small city surrounded by extremely low-density suburbs.

Of course, the gross density of an entire urbanized region is a crude measure. Boston (low gross density) and San Francisco (high gross density) both have walkable, high-energy centers and a relatively large number of downtown residents. The low and compact historic cities of Georgetown and Alexandria in the D.C. metropolitan area also have high residential densities, and the highest employment density is in the capital’s mid-rise center, not Virginia’s thriving Rosslyn-Ballston corridor or the outlying edge city of Tysons Corner, as one might expect. Although Los Angeles has a high gross density, its small downtown has only half as many residents as Chicago’s. And of the six major cities, Los Angeles has the lowest share of workers using mass transit, since, unlike New York, Washington, and San Francisco, it lacks sufficiently high concentrations of people living within walking distance of transit stops.

The role of mass transit in cities is a reminder that urban density affects sustainability. It has been estimated that a Manhattanite’s carbon footprint is a third smaller than that of the average American. Dense urbanization conserves resources in many ways: Urban buildings, whether apartments or row houses, are more compact and energy efficient; amenities are concentrated, which encourages walking; and public transit becomes an option. But even the densest American urban regions are not very dense compared to those of Europe. Greater Paris, for example, covers only about 1,000 square miles, and has a gross density of 10,000 inhabitants per square mile. Greater London, confined to 600 square miles, is even denser, with about 12,850 inhabitants per square mile. And the density of Asian cities such as Singapore is even higher than that. By comparison, the gross density of a typical American urbanized area is about 2,500 inhabitants per square mile.

There are a number of ways in which American cities could become denser. In vertical downtowns, tall buildings could simply get taller, or older office towers could be converted to residential use, as has happened in some business districts. The most common form of urban densification is the conversion of disused waterfronts, decommissioned Navy yards, and obsolete industrial areas into housing and office developments, an attractive strategy since it does not displace existing residents. In older city neighborhoods, taller structures could progressively replace three- or four-story row houses and low apartment buildings, though community resistance makes this a slow process.

Suburban densification is more challenging. In Philadelphia, where I live, it was common practice in the mid-20th century to subdivide large suburban estates into communities of single-family houses, but such open spaces in the suburbs are increasingly rare. Neighborhoods of single-family housing can be made denser by building clusters of smaller houses on what were previously large single-house lots, or by introducing row houses or low-rise apartment buildings. Both strategies involve radical changes to neighborhood identity, however. Perhaps the greatest challenge will be to increase density in the large planned-unit communities that have proliferated in the past few decades. In these, any change is constrained by homeowner associations in which even a small minority of members can effectively block alterations they find objectionable.

But after a century of spreading out, will Americans change their minds and draw together? Some observers maintain (hope) that the current economic recession will encourage (force) home buyers to demand smaller homes and more densely planned communities. This result would be unusual, since previous recessions have not had similar effects. Consumers generally have short memories. Following the energy crisis of 1973, for example, Americans switched to smaller cars, but by 1984, when prices at the pump had dropped, gas-guzzling minivans appeared, soon to be followed by SUVs. In any case, choosing where one lives has never been a strictly economic proposition. It is always a trade-off among the affordability of housing, the length of commutes, the quality of neighborhood amenities—especially schools—and preferred lifestyles.

During the last decade, proponents of downtown living pointed to an increase in downtown residential construction as a harbinger of an urban renaissance, but empty condominiums in cities such as Miami and Chicago suggest that this boom was a product of the housing bubble rather than a signal of a significant change in home buyers’ preferences. Similarly, the fact that the average size of new suburban houses—and lots—has recently shrunk for the first time in decades may be less meaningful than it is made out to be. In a recession, the only customers are first-time buyers who can afford only modest homes (which qualify for Federal Housing Administration mortgages, the chief form of housing finance during economic downturns). Meanwhile, larger houses are not being built because move-up home buyers are unable to sell their homes in today’s weak housing market.

New York, Los Angeles, San Francisco, and Boston grew modestly during the last two decades. (The size of Chicago and Washington hardly changed.) While these cities did better than the rustbelt cities, whose populations continue to shrink, younger cities such as Colorado Springs, Fort Worth, Atlanta, and Charlotte almost doubled in population during the same period, and Phoenix, San Antonio, and Albuquerque grew by more than a third. What is striking about these new growing cities is that they are much less rather than more dense. New York (the city, not the metropolitan region) contains 26,000 inhabitants per square mile, San Francisco 16,000, Chicago and Boston 12,000, and Washington and Los Angeles slightly fewer, but the new cities of the South and West rarely surpass 3,000 inhabitants per square mile, and the fast-growing city of Charlotte, North Carolina, has fewer than that. Jane Jacobs’s teaching notwithstanding, we appear to be spreading out.

The latest U.S. Census figures reveal that during the last decade, suburban areas outstripped cities in population gains. Moreover, The New York Times reports that “more than a third of all 13.3 million new suburbanites were Hispanic, compared with 2.5 million blacks and 2 million Asians.” At the same time, the immigrant populations of small towns and suburbs increased the most, while those of the big cities remained flat, reversing the historical pattern of the past. Jobs and cheaper housing are strong magnets.

Changes in migration trends are a reminder that Americans have always shown a capacity to adapt. For example, when energy prices spiked in the summer of 2008, people quickly tightened their belts, driving less, walking more, turning down their air conditioners, and shutting off the lights. It is certainly possible that the cold recessionary shower will dampen earlier exuberance and accelerate a shift to urban living, at least among young college graduates and higher-income retirees. The question is whether the rest of us will embrace denser and more compact suburbs and cities, or whether we will depend on technological fixes such as electric and hybrid cars, more efficient heating and cooling systems, and alternative energy sources. I suspect the answer will be a bit of both. Some people will embrace urban density, but many will make sacrifices in order to continue the decentralized way of life they prefer. The heterogeneity that has always characterized American cities will continue to produce many different solutions to suit a large and diverse nation.

* * *

Witold Rybczynski is the Martin and Margy Meyerson Professor of Urbanism at the University of Pennsylvania. He is the author of many books, including, most recently, Makeshift Metropolis: Ideas About Cities (2010).

Photo courtesy of Flickr/Jaume Escofet