Downsizing Cities

To make cities work better, make them smaller

URBAN America is changing, and as has so often happened, the change is largely unplanned. Fields turn into housing developments; housing developments turn into suburban towns; suburban towns grow into cities; commercial strips and malls grow into full-fledged urban centers. Moreno Valley and Santa Clarita, both in California, not even incorporated in 1980, are today among the fastest-growing cities in the country. Forty years ago Mesa, Arizona, was a town outside Phoenix; today, with 325,000 inhabitants, Mesa is bigger than Louisville or Tampa. As for Phoenix, it is now the eighth largest city in the United States, with more than a million people, and makes up only part of a metropolitan area that covers more than 400 square miles.

A metro area—or a metropolitan statistical area (MSA), as the Census Bureau calls it—is defined as a central city of at least 50,000 inhabitants together with adjacent communities with which it has a high degree of social and economic integration. A metro area without a central city must contain at least 100,000 people in all (or, in New England, 75,000, as the term is defined there); major metro areas of more than a million, which in practice always contain central cities, may be designated consolidated metropolitan statistical areas. Although the concept of a metro area was introduced in 1949 as a demographic convenience, the city proper remains a distinct legal entity.

In Cities Without Suburbs (1993), David Rusk, a former mayor of Albuquerque, convincingly demonstrates that those central cities that have expanded their limits to annex suburbs, or that have enough vacant land to accommodate suburban growth, do better than static cities in a number of significant ways. The two Ohio metropolitan areas of Columbus and Cleveland serve as an example. Since 1950 the city of Columbus has been aggressively expanding, and now covers about 200 square miles; Cleveland's area—seventy-seven square miles—is almost unchanged. Although metro Cleveland is comparable in income level and racial composition to metro Columbus, and both metro areas have grown significantly in the past four decades, the present situation of the two cities is very different. Cleveland has fallen behind Columbus in economic growth and job creation; it is more racially segregated; and it has a significantly lower per capita income, more poverty, and a lower municipal-bond rating. The city of Cleveland, despite being the center of a consolidated metropolitan statistical area, is anything but integrated with its surrounding communities. By every measure (income, education, employment, poverty) the city is less well off than its suburbs. The same pattern is repeated across the country in cities like Detroit, St. Louis, and Chicago, which are increasingly isolated—economically and racially—within their metro areas.

CAN old manufacturing cities and their suburbs be unified? Rusk concedes that true metropolitan government is unlikely (the only American metro area with a directly elected regional government is Portland, Oregon) and suggests revenue sharing between rich and poor jurisdictions, metro-wide housing-assistance programs, and economic-development plans. But helping ailing center cities by transferring funds from the suburbs is unlikely to garner much political support.

What is more likely is a federal program, such as the recently created Empowerment Zones, aimed at revitalizing inner-city slums. This is another version of what used to be called urban renewal, or enterprise zones, and it is as unlikely to succeed as its predecessors. Nicholas Lemann made this point forcefully last year in a New York Times Magazine article titled "The Myth of Community Development." "Nearly every attempt to revitalize the ghettos has been billed as a dramatic departure from the wrongheaded Government programs of the past," he wrote, "even though many of the wrongheaded programs of the past tried to do exactly the same thing."

Even if it were possible to expand the tax base of isolated central cities to capture at least some of the wealth of surrounding suburbs, doing so would not solve another problem. A city like Cleveland is not just less dynamic, poorer, and less racially heterogeneous than it used to be; it is also considerably smaller. Since 1970 Cleveland's population has decreased by 33 percent—one of the most precipitous declines of any large city. (During the same period Columbus grew by 17 percent.)

Cleveland is an example of an urban phenomenon that is increasingly common: the shrinking city. From 1970 to 1990 the total population of the 200 U S. cities with more than 100,000 inhabitants apiece increased from 59 million to 64 million—a nine percent growth rate. (The nation's population grew by 22 percent in the same period.) But growth was not experienced equally by all cities. In fact, cities fell into two distinct categories: about two thirds grew vigorously, and the other third actually lost population. Over the two decades the average population decline was 12 percent and the average growth was an astonishing 81 percent. The cities that lost population tend to be the older manufacturing cities; the gainers are newer suburban-style cities, chiefly in the West and the Southwest. The trend has continued; today nine of the twenty largest cities in the country have smaller populations than they did even in 1950, and some cities, such as Boston and Buffalo, are smaller today than they were in 1900.

These and many smaller cities now have what is referred to in the real-estate business as a low occupancy rate. In a shopping mall with a low occupancy rate the owner can decide to refurbish the mall or to offer special leases. If these strategies don't work, in the short run the owner will absorb the loss; in the long run either rents must be raised or the owner will go bankrupt. But if rents are raised too high, more tenants will leave and bankruptcy will only be accelerated.

Like a mall owner, the administration of a city with a low occupancy rate can try to increase its tax base by refurbishing the downtown area to make it more attractive to business. It can organize riverboat gambling and build aquariums and world trade centers. It can stimulate employment by enlarging the public sector. (It cannot, however, create the sort of manufacturing jobs that were the basis for the earlier prosperity and growth of great cities like New York, Chicago, and Philadelphia.) It can also try to balance its budget by raising revenues through higher property taxes, business taxes, and income taxes, and by curtailing services—although these tactics, like raising rent, will eventually only hasten population decline.

The mall owner who has tried everything and finds that there is simply no demand for space has a final option: make the mall smaller. Consolidate the successful stores, close up an empty wing, pull down some of the vacant space, and run a smaller but still lucrative operation. Many cities, such as New York, Detroit, and Philadelphia, don't stand a chance of annexing surrounding counties. Downsizing has affected private institutions, public agencies, and the military, as well as businesses. Why not cities?

TWO things happen when a city loses population. The reason for the first is that although a city is often said to be shrinking, its physical area remains the same. The same number of streets must be policed and repaired, sewers and water lines maintained, and transit systems operated. With fewer taxpayers, revenues are lower, often leading to higher taxes per capita, an overall deterioration of services, or both. More people depart, and the downward spiral continues.

The reason for the second is that urban vitality has always depended on an adequate concentration of people. In 1950 the average density in cities like Detroit, Cleveland, and Pittsburgh was more than 12,000 people per square mile; by 1990 it was around 6,000 or 7,000—a dramatic decline. The reality is even worse than it sounds, because the decline was not distributed equally across the city, and certain areas experienced much more dramatic reductions.

Without sufficient concentrations of people, not only is the provision of normal municipal services extremely expensive but urban life itself begins to break down. There are not enough customers to support neighborhood stores and services, or even to provide a sense of community. Empty streets become unsafe, and abandoned buildings become haunts for drug dealers and other criminals. A national study of housing abandonment found that the "tipping point" in a neighborhood occurred when just three to six percent of the structures were abandoned. Vacant lots and empty buildings are more than just symptoms of blight—they are also causes of it. Central cities of metro areas that have aggressively expanded their borders face these problems too, even if the cities have a broader and richer tax base.

The first need of a city whose population has declined radically is to consolidate those neighborhoods that are viable. Rather than mounting an ineffectual rear-guard action and trying to preserve all neighborhoods, as is done now, the de facto abandonment that is already in progress should be encouraged. Housing alternatives should be offered in other parts of the city, partly occupied public housing vacated and demolished, and private landowners offered land swaps. Finally, zoning for depopulated neighborhoods should be changed to a new category—zero occupancy—and all municipal services cut off. Efforts should be made to concentrate in selected areas resources such as housing assistance and social programs.

Inevitably consolidation would involve the movement of individuals and families from one part of the city to another. It is true that private freedoms would be sacrificed for the common good in the process, just as they are when land is expropriated to build a highway or a transit system. Does this sound heartless? Surely it is less so than the current Pollyannaish pretense of providing services to many poor and depopulated neighborhoods, which are occasionally half revived with community-development projects and then left on their own to decay even further.

The comprehensive downsizing of cities faces formidable obstacles—not only the lack of legal and bureaucratic tools (town planning has traditionally dealt with growth, not decline) but also political opposition. Historic preservationists will likely oppose the demolition of many old buildings. Local politicians whose electoral base would be eroded can also be expected to resist attempts at consolidation.

In 1976 Roger Starr, a past administrator of New York City's Housing and Development Administration, published an article suggesting "planned shrinkage," a reduction in New York's municipal services. "I profoundly wish I'd never coined that phrase," he says today. "It was a most unfortunate term that was misinterpreted by many people as a plan to drive poor blacks out of New York, and I still receive calls that accuse me of being a racist." It is important to underline that the act of consolidating neighborhoods would not move people out of the city, and no loss of benefits and services would ensue. Nevertheless, since the most depopulated parts of the city tend to be the poorest, and since the poorest city inhabitants are predominantly black and Hispanic, relocation would undoubtedly affect members of these groups more than others. Much would depend on the ability of minority leaders to see that given the lack of real alternatives, abandoning half-empty neighborhoods is not necessarily a political defeat.

Much of the housing in older industrial cities is as dense as it is because of high land costs, a history of property speculation, and the greed of the original tenement builders. Consolidating neighborhoods would provide an opportunity to relieve congestion and encourage variety in kinds of housing. Two-story attached and semidetached houses can be built at net densities of sixteen to twenty-four houses per acre, instead of the higher densities that characterize nineteenth-century neighborhoods of three-to five-story buildings. There may even be an opportunity, in some cases, for cities to compete directly with the suburbs by providing detached single-family housing. In cities such as New York and Philadelphia detached single-family housing, built with a mixture of public and private support, is appearing in previously high-density areas.

Oscar Newman wrote in Community of Interest (1980) that "architects and architectural historians have been damning the suburban tract development since the 1930s, but social scientists and realtors will tell you that tract houses continue to be the most sought after and the most successful form of moderate-and middle-income housing ever built." The success of the tract house is one reason that cities have been losing people to suburbs. If cities are to attract—or at least keep—moderate- and middle-income citizens, they will have to provide the sort of housing those citizens want. If cities cannot annex suburbs, they can do the next best thing: rezone areas to permit the construction of suburban-style housing.

THE consolidation of residential neighborhoods would produce stretches of empty blocks where buildings would have to be knocked down, services stopped, streets closed, and bus lines rerouted. It would be comforting to think that such vacant land would find new use as recreational parks and wilderness areas. On a small scale, empty lots could be converted into allotment gardens or playing fields. But there is little likelihood of creating large urban parks on the scale of Frederick Law Olmsted's nineteenth-century creations. The costs associated with the removal of basements and old foundations, soil amelioration, drainage, and planting would probably be out of the reach of cities already in financial difficulties. Vacated urban land would more likely be left empty, streets and buried infrastructure in place, available for use at some future date.

The objection may be raised that zero-occupancy zoning would create large tracts of empty land whose presence would disrupt the proper functioning of a city. In fact many cities have grown up around cemeteries, or have enveloped earlier industrial areas such as quarries and railroad yards, and most cities already have large areas of land such as tank farms and container depots that are cut off from everyday use. The only difference between these areas and zero-occupancy zones would be that the latter would be unused, and would not create noise or air pollution.

Cities have another option: divestiture. Contiguous parcels of, say, at least fifty acres could be put up for sale, with one of the conditions of sale being that the land would cease to be part of the city. According to Peter Linneman, the director of the Wharton School's Real Estate Center, large tracts in proximity to cities but without the burden of city taxes and bureaucracy would very likely attract developers who would otherwise shun them. The new developments, whether residential or commercial, would be responsible for their own municipal services, as new developments already are in suburban locations. Assuming that questions of ownership of rights-of-way and underground infrastructure could be worked out, the prospect is attractive. Cities would increase their income (although they would not gain taxpayers), and they would divest themselves of unproductive land. At the same time, people and economic activities would be attracted back into the urban vicinity.

The idea of downsizing goes against the progressive, optimistic American grain. Surely, conventional wisdom says, one should attack the root of the urban problem, not merely its symptoms. The solution to urban decline has always been assumed to be a combination of economic development (leading to renewed growth, more jobs, a larger population, and a larger tax base) and financial aid from federal and state governments. The problem is that population loss has gone on for too long: generally cities that have stopped growing and are losing population have been doing so for more than forty years. If metro-wide tax sharing were to be implemented, and if a more equitable allocation of financial resources finally gave cities a chance to deal more effectively with problems of poverty and immigration, the rate of population decline might be reduced or, at best, the population might be stabilized. Great manufacturing cities like St. Louis, Detroit, Philadelphia, and Cleveland will not return to their earlier predominance or their earlier size. Downsizing alone would not solve the problems of unemployment and urban poverty, but it would permit shrinking cities to marshal their resources more effectively as they make their way to a stable—albeit smaller—future.

Presented by

What Happened to the Milky Way?

Light pollution has taken away our ability to see the stars. Can we still save the night sky?

Join the Discussion

After you comment, click Post. If you’re not already logged in you will be asked to log in or register with Disqus.

Please note that The Atlantic's account system is separate from our commenting system. To log in or register with The Atlantic, use the Sign In button at the top of every page.

blog comments powered by Disqus


What Happened to the Milky Way?

Light pollution has taken away our ability to see the stars. Can we still save the night sky?


The Faces of #BlackLivesMatter

Scenes from a recent protest in New York City


Desegregated, Yet Unequal

A short documentary about the legacy of Boston busing


Ruth Bader Ginsburg on Life

The Supreme Court justice talks gender equality and marriage.


Social Media: The Video Game

What if the validation of your peers could "level up" your life?


The Pentagon's $1.5 Trillion Mistake

The F-35 fighter jet was supposed to do everything. Instead, it can barely do anything.

More in National

More back issues, Sept 1995 to present.

Just In