Divided We Sprawl

Urban Affairs: Divided We Sprawl

A call for the reinvention of the American city and
suburb that would exploit the infrastructure of the one
 and mitigate the "frantic privacy" of the other

by Bruce Katz and Jennifer Bradley

BY many accounts Baltimore is a comeback city. It has a beautiful piece of calculated nostalgia in the Camden Yards baseball stadium, which draws tens of thousands of visitors throughout the spring and summer. It has a lively waterfront district, the Inner Harbor, with charming shops and hot snacks for sale every hundred yards or so. But although it may function well as a kind of urban theme park (and there are plenty of cities that would love to achieve that distinction), as a city it is struggling. For twenty years Baltimore has hemorrhaged residents: more than 140,000 have left since 1980. Meanwhile, the surrounding suburbs have steadily grown. The population of Howard County, a thirty-minute drive from the city, has doubled since 1980, from 118,600 to 236,000. The people who have stayed in Baltimore are some of the neediest in the area. The city has 13 percent of Maryland's population but 56 percent of its welfare caseload. Only about a quarter of the students who enroll in a public high school in the city graduate in four years.

And Baltimore is not unique. The image of America's cities has improved greatly over the past few years, thanks to shiny new downtowns dotted with vast convention centers, luxury hotels, and impressive office towers, but these acres of concrete and faux marble hide a reality that is in many cases grim. St. Louis, Cleveland, Philadelphia, and Washington, D.C., lost population throughout the 1990s. These cities are also losing their status as the most powerful economies in their regions. Washington started the 1990s with a respectable 33 percent of the area's jobs. Seven years later it had only 24 percent. The rate of population growth in the nation's suburbs was more than twice that in central cities -- 9.6 percent versus 4.2 percent -- from 1990 to 1997. In just one year -- 1996 -- 2.7 million people left a central city for a suburb. A paltry 800,000 made the opposite move. In the major urbanized areas of Ohio 90 percent of the new jobs created from 1994 to 1997 were in the suburbs. Ohio's seven largest cities had a net gain of only 19,510 jobs from 1994 to 1997; their suburbs gained 186,000. The 1990s have been the decade of decentralization for people and jobs in the United States.

Not even cities that are growing -- southern and western boom cities -- are keeping pace with their suburbs. Denver has gained about 31,000 people in the 1990s (after having lost residents during the 1980s), but the counties that make up the Denver metropolitan area have gained 284,000 people -- about nine times as many. In Atlanta and Houston central-city growth is far outmatched by growth in outlying counties. And these cities, too, are losing their share of the jobs in their respective regions. In 1980, 40 percent of the jobs in the Atlanta region were in the city itself; by 1996 only 24 percent were.

Meanwhile, the poor have been left behind in the cities. Urban poverty rates are twice as high as suburban poverty rates, and the implementation of welfare reform appears to be a special problem for cities. Although welfare caseloads are shrinking in most cities, in general they are not shrinking as quickly as they are in the states and in the nation as a whole. Often cities have a disproportionate share of their states' welfare recipients. Philadelphia County, for example, is home to 12 percent of all Pennsylvanians but 47 percent of all Pennsylvanians on welfare. Orleans Parish, in which the city of New Orleans is located, has 11 percent of Louisiana's population but 29 percent of its welfare recipients. This hardly adds up to an urban renaissance.

Cities -- both the lucky, booming ones and the disfavored, depleted ones -- are losing ground for two reasons. First, they push out people who have choices. Urban crime rates have fallen, but they are still generally higher than suburban rates. Some urban school systems are improving, but in most of the nation's twenty biggest urban school districts fewer than half of high school freshmen graduate after four years. City mayors have cut taxes, but urban tax rates (and insurance rates, too) are often higher than suburban ones. Second, suburbs pull people in. This is not a secret. What is less well known -- in fact, is just beginning to be understood -- is how federal, state, and local policies on spending, taxes, and regulation boost the allure of the suburbs and put the cities at a systematic, relentless disadvantage. People are not exactly duped into living in detached houses amid lush lawns, peaceful streets, and good schools. Still, it is undeniable that government policies make suburbs somewhat more attractive and affordable than they might otherwise be, and make cities less so.

Federal mortgage-interest and property-tax deductions give people a subtle incentive to buy bigger houses on bigger lots, which almost by definition are found in the suburbs. States also spend more money building new roads -- which make new housing developments and strip malls not only accessible but financially feasible -- than they do repairing existing roads. Environmental regulations make building offices and factories on abandoned urban industrial sites complicated and time-consuming, and thus render untouched suburban land particularly appealing.

Together these policies have set the rules of the development game. They send a clear signal to employers, householders, builders, and political leaders: build out on open, un-urbanized, in some cases untouched land, and bypass older areas. These policies were never imagined as a coherent whole. No individual or committee or agency wrote the rules of development as such. No one stopped to consider how these rules, taken together, would affect the places where people live and work. The rules are simply the implacable results of seemingly disparate policies, each with unintended consequences.

When the policies that made it easier for people to flee the cities and move to the suburbs hurt only urban neighborhoods, the people who chose or had to stay behind suffered. Now, however, these policies, together with the problems of decay and decline in the cities and rapid suburban development, are causing problems for suburbanites, too -- most notoriously the problem known as sprawl.

Thus much of the unhappiness of the cities is also the unhappiness of the suburbs. The familiar image of a beleaguered urban core surrounded by suburban prosperity is giving way to something more realistic and powerful: metropolitan areas in which urban and suburban communities lose out as a result of voracious growth in undeveloped areas and slower growth or absolute decline in older places. The idea that cities and suburbs are related, rather than antithetical, and make up a single social and economic reality, is called metropolitanism.

METROPOLITANISM describes not only where but also in some sense how Americans live -- and it does this in a way that the city-suburb dichotomy does not. People work in one municipality, live in another, go to church or the doctor's office or the movies in yet another, and all these different places are somehow interdependent. Newspaper city desks have been replaced by the staffs of metro sections. Labor and housing markets are area-wide. Morning traffic reports describe pileups and traffic jams that stretch across a metropolitan area. Opera companies and baseball teams pull people from throughout a region. Air or water pollution affects an entire region, because pollutants, carbon monoxide, and runoff recognize no city or suburban or county boundaries. The way people talk about where they live reflects a subconscious recognition of metropolitan realities. Strangers on airplanes say to each other, "I'm from the Washington [or Houston or Los Angeles or Chicago or Detroit] area." They know that where they live makes sense only in relation to other places nearby, and to the big city in the middle. Metropolitanism is a way of talking and thinking about all these connections.

The old city-versus-suburb view is outdated and untenable. We can no longer talk about "the suburbs" as an undifferentiated band of prosperous, safe, and white communities. There are two kinds of suburbs: those that are declining and those that are growing. Declining suburbs, which are usually older and frequently either adjacent to the city or clustered in one unfortunate corner of the metropolitan area, are starting to look more and more like central cities: they have crumbling tax bases, increasing numbers of poor children in their schools, deserted commercial districts, and fewer and fewer jobs. For such suburbs to distance themselves from cities makes about as much sense as two drowning people trying to strangle each other.

Growing suburbs are gaining, sort of. They are choking on development, and in many cases local governments cannot keep providing the services that residents need or demand. Loudoun County, a boom suburb in northern Virginia, epitomizes this kind of place. The county school board predicts that it will have to build twenty-three new schools by 2005 to accommodate new students. In February of this year the board proposed that the next six new schools be basic boxes for learning, with low ceilings, small classrooms, and few windows. "We cannot ask the voters to keep voting for these enormous bonds," a county official told The Washington Post earlier this year, referring to a $47.7 million bond issue in 1998 for the construction of three new schools. "Nor can we continue to raise taxes every single year to pay for school construction." Predictably, parents complained about the cutbacks in amenities -- after all, they had moved there for the schools. "I just think they have to maintain their standards," a disgruntled parent told The Washington Post. But these suburbs cannot maintain their standards. There are simply too many new people who need too much new, expensive infrastructure yesterday -- not just schools but also sewer and water lines, libraries, fire stations, and roads.

Whether they moved to these places for rural tranquility, lovely views, and open space, or for good schools, or for the chance to buy a nice house, or just because they wanted to get away from urban hassles, residents of growing suburbs sense that frantic, unchecked growth is undermining what they value and want to keep. The old paradigm of cities and suburbs as opposites, or partisans in a pitched battle, doesn't explain the relationship between these gaining suburbs and their declining older cousins a few exits back on the highway.

Suburbs are not the enemies of cities, and cities are not the enemies of suburbs. That is the first principle of metropolitanism. Cities and suburbs have a common enemy -- namely, sprawl. The cycle described above, of draining the center while flooding the edges, is familiar to almost anyone who has driven from one edge of a metropolitan area to another. It is endlessly repeatable, at least potentially: the center just gets bigger, and the edges move out. Metropolitanism is a way of thinking that might break this cycle.

Alas, the city-suburb dichotomy is alive and well in law and in policy. The result is a tangle of regulations and programs that are excellent at throwing growth out to the edges of metropolitan areas and ineffectual at bringing it back to or sustaining it in the metropolitan core. One reason the problem of growth has not been solved is that the city-versus-suburb analysis doesn't properly describe it. The metropolitan reality requires different kinds of policies -- ones that take connections and the varying impact of growth into account.

THE metropolitanist policy agenda has four basic elements: changing the rules of the development game, pooling resources, giving people access to all parts of a metropolitan area, and reforming governance. These are interlocking aspects of how to create good places to live; they are closely related and can be hard to distinguish. To understand the cascade of consequences that policies can have, consider the policy chain reaction that would begin if the rules of the development game were changed to fit the metropolitanist paradigm. Those rules are mainly the policies that guide transportation investments, land use, and governance decisions, all of which are themselves entangled. Start at one end of the knot: transportation. Major highways, built by federal and state dollars, act as magnets for new development. This has been clear ever since the 1950s, when the interstate-highway system made the suburbs widely accessible and hugely popular. A metropolitanist viewpoint recognizes that these highways will probably pull lots of investment and resources away from the metropolitan core. New development, spawned by highways, will necessitate expensive state-funded infrastructure, such as sewer systems, water pipes, and new side roads. Meanwhile, existing roads, pipes, and sewers, which already cost taxpayers plenty of money, are either not used to the fullest or starved of funds for repair.

A metropolitanist transportation policy might eschew a new beltway and instead direct federal transportation dollars to public transit, which draws development toward rail stations rather than smearing it along a highway, or to repairing existing roads rather than building new ones. That is what Governor Parris Glendening, of Maryland, and Governor Christine Todd Whitman, of New Jersey, have proposed for their states, and what elected officials are working on or have accomplished in the metropolitan areas of Boston, Chattanooga, and Portland, Oregon. New businesses and housing developments will be steered toward where people already live and public investments have already been made.

At this point land use comes into play. "Land-use planning" may sound a little soporific, but it is simply a brake on chaos. It allows communities to prepare for growth in a way that avoids gridlock and preserves public resources. It connects the basic places of life: where people work, where they live, where they play, drop off their dry cleaning, check out a library book, buy a box of cereal. A metropolitanist land-use scheme would preserve open spaces and create parks and other public areas, thereby taking big parcels of suburban land off the development market. Where, then, would all the new development go? An enormous amount of vacant land already exists inside the boundaries of metropolitan areas, which generally have developed in leapfrog fashion, with big gaps between one subdivision or strip mall and the next. Parks and open spaces will not fill all those gaps, which could support development -- as could the abandoned urban properties known as brownfields.

Land-use decisions can affect how as well as where things are built. Zoning policies can call for transit-oriented development -- clusters of shops, apartment buildings, and offices around bus or rail stops, so that people will drive a little less. They can require or at least encourage varied housing near office buildings and supermalls, so that everyone who works there, from the receptionist to the escalator repairer to the middle manager to the chief financial officer, can live near his or her workplace.

Pooling resources is the second element of a metropolitanist agenda. In most metropolitan areas a new office complex or amusement park or shopping mall tends to confer benefits on a single jurisdiction by adding to its property-tax coffers. Meanwhile, neighboring communities are stuck with some of the burdens of development, such as additional traffic and pollution and the loss of open space. Pooling resources -- specifically, a portion of the extra tax revenue from development -- means that development's benefits, like its burdens, are spread around. The Twin Cities area has a tax-base-sharing scheme whereby 40 percent of the increase in commercial and industrial property-tax revenues since 1971 is pooled and then distributed so that communities without substantial business development are not overwhelmed by needs and starved of resources. In other parts of the country regional jurisdictions have agreed to tax themselves to support cultural and sports facilities; this makes sense, because the entire region benefits from those facilities.

The third element of a metropolitanist agenda is giving everyone in the metropolitan area access to all its opportunities. Access is easy for people with decent incomes and decent cars. They can live where they wish, and they can get from their houses to their jobs without enduring extraordinary hassles. Poor people do not have this kind of mobility.

There are three ways to solve the access problem: make it easier for urban workers to get to suburban jobs; provide affordable housing (through new construction or vouchers) throughout a metropolitan region; or generate jobs in the metropolitan core or at least near public-transportation routes. State and federal governments are now implementing programs that help people to overcome core-to-edge transportation problems, and through housing vouchers are giving low-income people more choices in the metropolitan housing market. Across the country churches and nonprofit organizations are running jitney services and private bus lines to get people to work. A group of Chicago business leaders has called on major employers to weigh affordable-housing options and access to public transit in their business location and expansion decisions. Businesses and nonprofit groups are also trying to bring jobs and people closer together. Housing vouchers administered by nonprofit organizations with a metropolitan scope allow low-income families to move into job-rich municipalities. The nonprofits counsel families about their options and develop relationships with landlords. In the Atlanta region BellSouth will soon consolidate seventy-five dispersed offices, where 13,000 people have worked, into three centers within the Atlanta beltway, all of which are easily accessible by mass transit. After studying where employees lived, the company picked locations that would be of roughly equal convenience for commuters from the fast- growing northern suburbs and from the less-affluent southern suburbs.

The final element of the metropolitanist agenda has to do with governance. Whereas markets and -- more important -- lives operate in a metropolitan context, our governmental structures clearly do not. They hew to boundaries more suited to an eighteenth-century township than to a twenty-first-century metropolis. Chicago's metropolitan area, for example, encompasses 113 townships and 270 municipalities. This fragmentation works against sustainable metropolitan areas and facilitates segregation by race, class, and ethnicity. Welfare-to-work programs are hindered when public transportation stops at the city-suburban border, for example. Issues that cross jurisdictional borders -- transportation, air quality, affordable housing -- need cross-jurisdictional solutions and entities that bring together representatives from all the places, small and large, within a metropolitan area to design and implement these solutions. Some such entities already exist: In every urban region in the country a metropolitan planning organization coordinates the local distribution of a chunk of federal transportation funds. Oregon and Minnesota have established metropolitan governments for their largest urban areas, Portland and the Twin Cities. But informal metropolitan governance, in which local governments coordinate their policies and actions, is possible and efficacious. Also, it's necessary.

METROPOLITANISM is a genuinely different view of the American landscape, and politicians from both parties are beginning to think that a majority of voters might find something to like in it. Like Governor Glendening and Governor Whitman, Governor Thomas Ridge, of Pennsylvania, has laid out land-use objectives for his state that include linking new development to existing infrastructure and encouraging metropolitan cooperation. Governor Roy Barnes, of Georgia, has proposed a strong metropolitanist transportation authority for Atlanta, and in March the state legislature approved it. Governors and state legislators are central to the metropolitanist agenda, because states control an important array of tax, land-use, governance, transportation, work-force, and welfare issues.

Vice President Al Gore clearly recognizes the political potential of this issue and is trying to establish it as one of his signature issues. "We're starting to see that the lives of suburbs and cities are not at odds with one another but closely intertwined," he said in a speech last year. "No one in a suburb wants to live on the margins of a dying city. No one in the city wants to be trapped by surrounding rings of parking lots instead of thriving, livable suburban communities. And no one wants to do away with the open spaces and farmland that give food, beauty, and balance to our post-industrial, speeded-up lives." For more than a year Gore has been talking about America's growing "according to its values," and has even implied that development is not always welcome.

Of course, the idea of cities and suburbs coming together to solve common problems has been around for decades. No one ever before thought of using it to propel a presidential campaign, because the idea of metropolitanism had yet to prove its appeal, in referenda or in elections or in state legislatures. This is no longer the case.

These ideas are only just beginning to penetrate a recalcitrant real-estate-development industry, however. Christopher Leinberger is a managing director of one of the nation's largest real-estate advisory and valuation firms and a partner in a new urbanist consulting company. He has thought a lot about how the industry works, and he has concluded that sprawl is extremely attractive to the industry, because the kind of development it involves is simple and standardized -- so standardized that it is sometimes hard to tell from the highway whether one is in Minneapolis or Dallas or Charlotte. These cookie-cutter projects are easy to finance, easy to build, and easy to manage. Builders like the predictability of sprawl. They know how much a big parking lot is worth, but they aren't sure how to value amenities in older communities, such as density, walkability, and an interesting streetscape. More or less the same can be said of big retail chains. For example, they often overlook the fact that although people in core neighborhoods may have low incomes, they are densely concentrated, which works out to a significant amount of purchasing power. Developers and retailers will have to be willing to think differently if development is to come back to the core. There are encouraging signs. Magic Johnson Theaters, Rite Aid pharmacies, and Pathmark supermarkets are all recognizing that the people left in core communities need places to earn money and to spend it; each of these companies has opened outlets in central cities in recent years. The National Association of Homebuilders has joined with Gore, the U.S. Conference of Mayors, and the U.S. Department of Housing and Urban Development to encourage the development of a million new owner-occupied homes.

Of course, for the politicians' plans to work and the developers' projects to take off, urban core communities will have to win people over. Unless these places have good schools, safe streets, and efficient governments, people will not move from the edge back toward the center. Some mayors have realized this and are trying to make their cities better places to live. Richard Daley, of Chicago, and Stephen Goldsmith, of Indianapolis, are finding innovative ways to address issues that have bedeviled cities for decades: schools, crime, public services, and taxes. It is hard and often unpleasant work; it means privatizing some services, eliminating others, and ending wasteful patronage. But cities must be ready to take advantage of the opportunities that metropolitanist policies offer them.

ACADEMICS, architects, and bohemians may decry the soullessness of sprawl, but people seem to like it. Why put up such a fight to save dying places, whether they are called cities or older suburbs or metropolitan cores? After all, as people who see no harm in sprawl like to point out, Americans are living on a scant five percent of the land in this country. Why not just keep sprawling?

There are several reasons to defend not cities against suburbs but centeredness against decentralization, metropolitanism against sprawl. One reason to encourage development in metropolitan cores is a familiar one: the people who live there are among the poorest in their regions -- indeed, in the country -- and they need these opportunities and this investment. It is not fashionable to talk about having a moral obligation to poor people, but that doesn't mean that the obligation has disappeared. John Norquist, the mayor of Milwaukee, is fond of saying "You can't build a city on pity"; but disinvestment and the resulting lack of good schools, good jobs, and good transportation options is also impractical. The U.S. economy needs workers, and there are people in the metropolitan cores who are not getting into the work force. The need for workers will only increase as the average age of the population rises. By 2021 almost 20 percent of the American people will be over sixty-five, as compared with about 12.7 percent today. Whatever Social Security and Medicare reforms are enacted, these elderly people will need an abundance of payroll-tax-paying workers to support them.

The aging of the U.S. population will soon make it clear that sprawl is of no benefit to people who cannot drive. For a seventy-five-year-old without a car, sprawl can be uncomfortably close to house arrest. But metropolitan core communities where public transportation is available and distances are shorter between homes, pharmacies, doctors' offices, and libraries are navigable for older people in a way that settlements on the metropolitan fringe are not. Apartment buildings for the elderly are being built in the suburbs, with a variety of services under one roof, and vans to get people from here to there. But there should be choices for elderly men and women who do not want to be segregated from neighborhoods where babies and teenagers and middle-aged people also live.

Unlimited suburban development does not satisfy everyone. Metropolitanism will probably provide a greater range of choices, for the elderly and for everyone else. Policies that strengthen the metropolitan core lead to safer, more viable urban neighborhoods for people who prize the density and diversity of city life. These policies can reinvigorate older suburbs, with their advantages of sidewalks and public transit and a functioning Main Street. And, of course, they allow for brand-new, sizable single-family houses with yards.

It is also possible to argue against sprawl because of a commitment to community. Throughout this essay we have used the word "community" interchangeably with "township," "suburb," "municipality," "jurisdiction," "city," and "place." But "community" also designates a feeling, an ideal -- as in "a sense of community," which many people worry that they have lost and would like to re-create. And they are trying to re-create it. Newspaper dispatches from the suburbs of Detroit and Washington, D.C., report that developers are trying to build what people left behind in older places: town centers, with wide sidewalks and big storefronts, where a person can perhaps run into a friend or an interesting stranger and have a place to hang out in public. In a 1998 essay titled "The City as a Site for Free Association," the political philosopher Alan Ryan writes, "If people are to be self-governing, they must associate with each other in natural and unforced ways from which their political association can spring." By "political association" Ryan means involvement in public life and public decision-making. The underlying assumption of a democracy is that this involvement is a wonderful thing. Yet it is unclear whether the new town centers can generate the unforced interactions that make municipalities feel like communities. For all the good intentions of the developers who build them and the government officials who support them, they are not natural centers. They are places where people are invited to go and be social or civic-minded, but, as Ryan says, "Telling people to go to such and such a café in order to promote political cohesion and political activity is like telling people to be happy; there are many things they can do that will make them happy, but aiming directly at being happy is not one of them."

These town centers are actually some of the few places where suburban people might mingle with crowds and see people who are not like themselves. They are, along with shopping malls, the public spaces of sprawl. Free association, in the sense of unexpected, unplanned encounters that draw us out of ourselves, is hard to come by in decentralized environments. Driving alone or with family members or close friends from one destination to another leaves little opportunity for spontaneity. Sprawl can create a kind of cultural agoraphobia that depletes public life.

Certainly, the outer edges of metropolitan areas are not the only places that are finding public life difficult to sustain. The most depleted neighborhoods of metropolitan cores, with their forlorn "community centers," dingy streets, and empty sidewalks, are not fertile ground for free association either. And yet the architecture and layout of these places are at least supposed to facilitate interaction. Moreover, Ryan writes, there are still "galleries and concert halls, city parks, monuments, and other such places" in our urban cores for communities to come together, group by group and interest by interest.... To the degree that this is irreplaceable by seeing and hearing it all on television or on the stereo system, it encourages people to understand themselves as members of one society, engaged in a multitude of competing but also cooperative projects. A society that does not understand this about the basis of its cultural resources is a society in danger of losing them. At present, we seem to be such a society. Suburbs are not new. They have been in existence in the United States since the nineteenth century. But hypersuburbanization, decentralization, and sprawl are new -- less than two generations old. Americans are now discovering how hard it is to live without a center. In a typical attempt to move simultaneously in opposite directions, they are moving out but also trying to come back. This is not merely nostalgia for some dimly remembered era of civility and good cheer. People are honestly trying to balance the frantic privacy of the suburbs with some kind of spontaneous public life. By now it seems clear that continued sprawl will make this public life very hard to achieve -- at the edges of metropolitan areas, where there are no places to gather, and at the cores of metropolitan areas, where the gathering places are unsafe or abandoned. Is this really a good trade for a big back yard?


Bruce Katz is a senior fellow at the Brookings Institution and
the director of its Center on Urban and Metropolitan Policy.
Jennifer Bradley is a senior analyst at the Brookings
Institution's Center on Urban and Metropolitan Policy.

Illustration by David McLimans.

Copyright © 1999 by The Atlantic Monthly Company. All rights reserved.
The Atlantic Monthly; December 1999; Divided We Sprawl - 99.12 (Part Two); Volume 284,
 No. 6; page 26-42.