Suburban Decline: The Next Urban Crisis
With their small, outmoded, and deteriorating housing, many post-World War II bedroom suburbs are in urgent need of help.
The old urban crisis, characterized by the decline of central cities, still has not been addressed adequately by federal, state, and local policymakers. The problems facing central cities have proved extremely difficult and frustrating. And as cities continued to muddle along, the “crisis” became familiar, even boring, and hence lost its political cachet. Of course, some cities have seen signs of revitalization, not just downtown but in many neighborhoods. But even this limited progress is likely to be overshadowed by the next urban crisis already rushing toward us. This is the crisis of suburban decline, and it promises to be even more intractable than the crisis in central cities.
The crisis in central cities was dramatic: Business districts shrank; economic, political, and cultural centers diminished in size and function; once-fashionable residential neighborhoods fell into decline; and deterioration, crime, riots, and despair emerged in poverty ghettos. The crisis of suburban decline will be less visibly theatrical and more a backstage phenomenon. It will be concentrated in ordinary single-use residential-only subdivisions of the type constructed in every metropolitan area from the end of World War II through 1970, and it will feature changes that will be nearly invisible to all but their residents.
These bedroom suburbs once were heralded as “the future,” providing opportunities for individual households to improve their housing quality. But most of these suburbs depended excessively on the whims of private housing markets. Weaknesses in these bedroom suburbs have emerged as they have reached middle age. As with human beings, maturity brings some strength. But maturity is also the threshold for deterioration, sometimes rapid, sometimes slow and persistent. Hence, the vitality of these bedroom suburbs is vulnerable to changing fashion and to the next, more geographically dispersed, round of housing opportunities.
Indeed, settlements in the United States continue to evolve in cycles of rapid development and decline. However, public programs can be constructed to reduce this propensity to create successive rounds of neighborhood and community growth and decline. These programs, which are based on combinations of policies that have been successful in other countries but have not been combined in the United States, also can help mitigate the deterioration of middle-aged suburban neighborhoods. But the job will not be easy, as the character of these suburbs typically makes them unlikely natural candidates for reinvestment and reinvigoration.
Common perceptions of metropolitan trends have been dominated by concern about the decline of most large central cities in population, income levels, and in many cases, property values. For example, consider the median income of families in central cities as compared with the median income of all families in their metropolitan area (or “relative income,” in the parlance of economics). In a study we conducted among 147 U.S. metropolitan areas with a single central city, relative income had declined in 89 percent of those cities from 1960 to 1990.
But generally overlooked is the fact that many suburbs also had suffered during this period. Among 554 well-established suburbs in the 24 most populous metropolitan areas, 405 suburbs had declined in median family income, as compared with the entire metropolitan area, from 1960 to 1990. Even more striking, given the frequency and extent of city decline, 112 suburbs, or 20 percent, had declined in relative income at a rate faster than their central cities. And this rate of decline had accelerated during later years. Between 1980 and 1990, 32.5 percent of the suburbs had declined faster in relative income than their central cities. In addition, suburbs had become more polarized. Between 1980 and 1990, the number of suburbs whose family incomes were below 80 percent of the metropolitan median family income had increased fourfold, from 22 to 90, and the average income level in the suburbs with the lowest incomes fell from 82 percent of metropolitan income to 62 percent.
Suburban decline varied considerably among the metropolitan areas in our sample. For example, relative income decline had hit all of Denver’s suburbs but only 43 percent of San Diego’s suburbs. Suburbs declining faster than central cities ranged from 52 percent in the Kansas City area to none in the Baltimore and Milwaukee areas. The suburbs declining fastest were in the Atlanta area: 67 percent of its suburbs declined in relative income by 20 percent or more from 1960 to 1990, and 42 percent declined by 30 percent or more. Yet only 33 percent declined faster than the central city, because Atlanta itself declined so rapidly.
Although the correlation is not always exact, declines in relative income also serve as an indirect proxy for a multitude of other conditions that typically mark settlements in transition. For example, threats to public safety, such as increases in crime and the number of fires that occur, often rise as local incomes fall. In a study of the Chicago metropolitan area conducted by Myron Orfield, a state legislator in Minnesota and author of Metropolitics, 13 suburbs had higher rates of serious crimes in 1994 than did the central city. In addition, poorer urban communities, including suburbs, often have schools that are overburdened and in need of repairs, as well as student populations that enter the educational system with serious disadvantages linked to poverty and, in many cases, the lingering effects of racism. Many such communities also face increased public health problems, such as illicit drug use, that challenge the financial ability of local jurisdictions to provide adequate treatment and prevention and hence drain resources that might otherwise be used for community redevelopment.
In our studies, the suburbs suffering decline occurred in varied spatial patterns, which are not always explained simply by proximity to the central city. In some areas, inner suburbs remained strong. For example, in the area surrounding Washington, D.C., only Prince George’s County in Maryland had rapidly declining inner suburbs. In nearby Montgomery County as well as in the counties in northern Virginia, many of the older inner suburbs had rising property values and population as well as stable income. In Virginia, the “third-ring” counties on the metropolitan fringe were the areas with lower incomes and high local property tax rates. Other spatial patterns also are notable. In some areas, a pie-shaped “favored quarter” radiating from the central city often had higher-value housing and higher-income residents. And occasionally, suburbs with high incomes and housing values were located adjacent to low-income suburbs. Rings, favored quarters, and idiosyncratic historic relics occur in combinations that defy desires for clear images of metropolitan spatial patterns. But some causes and consequences of social transitions in metropolitan areas can be explained.
Threat of high mobility
Severe income decline contributes to three related community development problems. First, too little reinvestment in structures occurs, shortening the useful lives of buildings and wasting resources. Second, disparities in local governments’ tax resources increase, leading to wide differences in the quality of local services. These differences are especially serious for public education, diminishing the educational effectiveness of the nation and harming the development prospects of individuals. Third, geographic areas with severe problems expand, augmenting the avoidance aspect of residential location decisions. As people avoid these areas, metropolitan sprawl grows apace, creating spatial patterns that waste resources and increase environmental deterioration.
If strategic planning for local government jurisdictions and for regions is to be meaningful, then the direction and pace of neighborhood and jurisdiction change must be considered. The leading engine of neighborhood change is residential mobility. According to census data, roughly 50 percent of all metropolitan residents (homeowners and renters combined) move within five years, and 50 percent of metropolitan homeowners move in eight years. Thus, each neighborhood and jurisdiction is in danger if it cannot attract new residents or “replacement inmovers.”
Nearly every conventional theory of neighborhood change leads to the prediction that aging housing and neighborhoods, especially the oldest neighborhoods, will be accompanied by relative income decline among residents, compared with income of residents in newer housing and neighborhoods. But in our study of suburbs nationally, as well as in a study of 700 census tracts in Virginia, we found that aging housing was not consistently related to income decreases or increases. In the national study, the relative income decline in suburbs occurred as often in areas dominated by middle-aged housing (built after 1945) as in neighborhoods with older housing (built before 1940). In the Virginia study, relative income decline occurred in more than two-thirds of middle-aged neighborhoods in generally prosperous suburban counties, such as Fairfax near Washington, D.C., and Henrico and Chesterfield near Richmond. In contrast, most neighborhoods with even older housing–in both cities and suburbs–made comebacks between 1980 and 1990. In Richmond, Norfolk-Virginia Beach, and the suburbs of Washington, D.C., more than 60 percent of neighborhoods built substantially before 1940 rose in relative income, often dramatically.
Instead, we believe that housing quality, not housing age, is perhaps the most powerful driver, and the role of quality usually has been missed in most social scientists’ diagnoses of urban and suburban decline. We think that housing size and type, as well as settings for housing, help explain decline. For example, the median size of new single-family houses built in the late 1940s in Levittown, New York, the quintessential picture of post-World War II suburbia, was 800 square feet. By 1970, the median size of new single-family houses nationwide had reached 1,375 square feet. By 1995, the median new house was 1,920 square feet, with the average being nearly 2,100 square feet, and this size has continued to increase. Middle-aged neighborhoods that were constructed from 1945 to 1970, with housing that now is considered small as well as outmoded and in need of repair, are poor prospects for substantial reinvestment. Severe income decline in many such neighborhoods will be difficult to avoid.
We infer from these trends in housing size and income levels, as well as from field observations, that suburban decline usually occurs where there are large numbers of small houses with little aesthetic charm, where the houses are located in inconvenient settings, where there are few public amenities, and where there often are no alternatives to automobile transportation. These conditions are typical of suburbs developed between 1945 and 1970. Single-use districts were the norm, with residential areas often designed to omit all nonresidential uses. The suburbs usually were remote from employment and from public transportation. They were designed to limit through traffic and featured curvilinear roads terminating in cul de sacs.
In sum, such neighborhoods were designed to be inconvenient. Inconvenience was intended to keep out strangers, especially those deemed undesirable. Everyone living within new residential subdivisions was assumed to engage in constructive or benign behavior. But 30 to 50 years later, these middle-aged, inconvenient suburbs may be populated with a significant number of people with propensities for criminal and other undesirable behaviors. Most middle-income and upper-income people avoid these neighborhoods in choosing residential locations. Lower-income people who often have used such areas as stepping stones to better housing conditions are also faced with living under conditions that make further economic and social advancement more difficult.
And even if current occupants and replacement inmovers, including those with potentially high investment capacity, desire to reinvest in maintenance, upgrades, or expansions of their houses, four institutional and behavioral obstacles must be surmounted. First, homeowners in areas of similar housing often are counseled not to increase the value of their property to more than 20 percent above their neighbors’ housing values. Retrieving such investments at resale is considered unlikely. Therefore, the first potential reinvester confronts risks that later reinvesters may avoid. But later reinvesters will not exist without early ones. This problem is particularly severe when dwellings age in unison from the same starting time, a condition typical of post-World War II suburban subdivisions. Second, sources of reinvestment funds are scarce. The most likely source, a home equity line of credit, will not be available to new purchasers who have not accumulated sufficient equity against which to borrow. Third, builders may be likely to give high estimates for proposed remodeling, anticipating unforeseen structural problems or insisting on time and materials with a substantial profit margin. The problem is magnified in strong economies, when many builders refuse small construction jobs unless the profit potential is high. Ultimately, the cost per square foot charged to expand an existing structure may be more than building a new dwelling. Fourth, most potential reinvesters will be amateurs, engaging the reinvestment process for the first time, which means that many of them will be somewhat baffled by zoning, construction, design, and timing requirements that upgrade and expansion processes involve.
Tyranny of easy development
The growing problems facing middle-aged suburbs often are attributed to the growth of metropolitan sprawl, because demand for housing in older neighborhoods is thought to be too low. But the claim that sprawl is merely a response to consumer preferences in a free market is exaggerated and may be a myth. Instead, we argue that much sprawl occurs because developers and lenders find sprawl development more familiar, convenient, and less risky.
Consumers make purchases where housing of adequate size with contemporary facilities is available. Larger houses of the type increasingly favored by consumers now are easier to build on the metropolitan fringe. Developers also seek easy development opportunities in which risk is low. Developers fear delays, seeing time as money. Fringe development is where opposition from neighbors usually is lowest and where support from public officials often is greatest. Conversely, current residents typically resist proposals for infill projects in older neighborhoods, especially where market demand is strong. Residents fear new projects will alter the housing scale, reduce trees, increase traffic, or increase renters. Such objections occur even when the proposed housing is larger and more costly and where more people will own their homes, as compared to neighborhood norms. Sprawl is the result. Thus, fringe development constitutes what we call a tyranny of easy development decisions. And this tyranny ultimately subjects residents and businesses in closer-in jurisdictions to the consequences of decisions made in more remote government jurisdictions.
When metropolitan sprawl is rapid, and income and resource disparities among government jurisdictions are large, reinvestment in established neighborhoods also faces special obstacles. But some attributes help cities and suburbs overcome these obstacles. In the Washington, D.C., region, Arlington and Alexandria in Virginia, along with Greenbelt in Maryland, have demonstrated that substantial private reinvestment is possible by using various combinations of investing in fixed-rail mass transit, stimulating high-density mixed-use development near transit stations, emphasizing the preservation of historic neighborhoods, and stressing walkable street and pathway patterns with significant public amenities. Arlington and Alexandria had each of these characteristics, whereas Greenbelt had two (walkable patterns and public amenities). Most suburbs have none. A major question facing each metropolitan region is whether suburbs lacking these assets can be stabilized.
Types of successful policies
Private markets will remain the overwhelming means by which land and structures are developed, purchased, and sold. If spatial patterns (too much sprawl), fragile structures (too little reinvestment), and income and resource characteristics (wide disparities in resources among local government jurisdictions) are to be changed significantly, it will only be by altering the incentives, risks, and responsibilities that builders, developers, and lenders face in these markets, as well as by changes in consumer preferences.
Public policies that can construct or reconstruct healthy suburbs are the same as policies that can nurture cities. Useful policies limit extremes in income and taxable-resource disparities and focus on creating places where people value local experiences and culture in addition to engaging in market transactions. Substantively, useful policies concern housing, transportation, land use, education, local government structure, and finance. Geographically, they aim at limiting sprawl, augmenting reinvestment, and restraining or compensating for income and fiscal disparities.
Six types of policies need to be developed, expanded, and preferably integrated:
Reinvestment. Reinvestment capacity must be enhanced through a variety of private and public institutions, incentives, and regulations. As one step, states can remove impediments to reinvestment that are embedded in building codes. Experiences with New Jersey’s housing rehabilitation code since its implementation in 1998, for example, indicate that numerous reinvestments have occurred because unreasonable reconstruction requirements and uncertainty for redevelopers have been reduced. An example is that wider stairways like those required in new construction no longer are required in rehabilitated buildings; this is a particularly important change for rehabbing multistory structures.
On a larger scale, communities can devise development plans that specify changes across a number of fronts. For example, Albemarle County, a growing county that surrounds the city of Charlottesville, Virginia, recently completed a three-year planning process, spearheaded by local citizens. The planners, drawing on substantial community input, recommended that the county try to emulate the city’s pedestrian-friendly environment, especially that of its downtown. As a first step, according to the planners, the county would have to revise its zoning and subdivision ordinances concerning new developments, which currently require single-use residential districts, wide streets, and large setbacks for residences, and which favor streets ending in cul de sacs. The planners have called for mixed-use development, narrower streets that connect with one another, and smaller residence setbacks.
Although other changes in government policies are needed–such as tax abatements and tax credits; subsidized loans; infrastructure and facility investments; technical assistance, and, occasionally, acquisitions for redevelopment–the main thrust of government policies should be to guide and energize private-sector decisions that will increase reinvestment in structures and neighborhoods relative to construction in new settlements that add to metropolitan sprawl. Most of all, this policy requires widespread belief that this subject is important and that political courage and government resources should be mustered to cope with it. A transformation in public awareness and political salience is needed before widespread policy adaptations will occur.
In some districts, even these public steps to promote reinvestment may not be enough. Here, government support for neighborhood planning, technical assistance, low-interest loans, public school investments, tax abatements, and secondary mortgage purchases must be complemented by foundation- and church-supported nonprofit organizations, as well as by profit-sector redevelopers, who will collaborate in tackling the much tougher problems of reviving areas that outsiders ignore and insiders leave. Nonprofits can undertake such efforts because, by definition, they can make investments where payoff is not the motivating impulse, and they can focus on improving community institutions that serve the public good. For example, nonprofit housing corporations might reinvest in adaptive uses, such as converting abandoned or underutilized churches and other public buildings to community centers that provide day care, meeting rooms for local groups, and activities such as community theater performances.
Transportation and settlement patterns. Transportation should support the development of and reinvestment in viable settlements. Such settlements require transportation alternatives for diverse purposes, from automobiles to public transport to walking and bicycling. Because many people are located (perhaps trapped is a better term) in disconnected settlements from which they can function moderately well only by using automobiles, too many people assume that more highways and continued auto dependence constitute the best practices. But viable settlements cannot be achieved by responding to the current preferences of a dispersed majority of affluent and semiaffluent households. Hence this dilemma: Policies favoring highway construction are self-defeating and perpetuate the recurring processes of settlement decline in outward rolling waves, but they are the policies with the greatest short-range political support.
Some changes in public policies and policy analyses indicate that the attention paid to transportation effects and settlement patterns has been increasing. Leverage provided under the Air Quality Act to the U.S. Environmental Protection Agency (EPA) has been applied in the Atlanta region, blocking spending of federal highway funds for new construction until land use plans and alternative transportation plans that can improve air quality are approved. EPA has required a revision of Northern Virginia’s 2020 Transportation Plan because it will contribute to sprawl. Modeling of transportation and land use options for the Los Angeles area revealed that only an option bringing home and work into closer proximity would reduce traffic congestion and improve air quality. Transportation should be viewed as a community development function at least as much as a means of moving individuals quickly from home to work or from one metropolitan area to another.
Admittedly, a transition to patterns in which walking, bicycling, and public transportation will be effective alternatives for large percentages of residents will take 20 to 30 years or more. Each incremental step toward mixed-use redevelopment will provide relief for some households. But if this shift in transportation and land use priorities is not begun now, the prevailing forces favoring roadways will become even harder to change. Also, it may well be that communities need not be completely transformed before they begin to attract attention as more livable spaces. Simply getting the process in motion, backed up by long-term plans and firm public commitments, may encourage potential new residents to become pioneers in “rediscovering” these older areas.
Places. Viable places depend on easy, useful, and pleasurable walking. If there is nothing worth walking to, the life spans of dwellings and settlements are shortened. Public policies, therefore, should nurture mixed land and building uses, respect nature and history, create beauty, and facilitate convenient access through engaging public realms. Some cities, such as Portland, Oregon, and some suburbs, such as Arlington, Virginia, have used mass transportation and mixed-use development near transit stops to create more walkable and vibrant neighborhoods. Other suburbs, such as Alexandria, Virginia, with its 18th- and 19th-century Old Town district, and Greenbelt, Maryland, with its 1930s New Town environment and generous public spaces linking residential and civic areas, have met housing market tests effectively, being reconstituted by replacement inmovers who are as well off as their outmovers. And in an even more radical concept, a California urban designer has proposed converting massive suburban retail parking lots into gridded street networks with mixed-use development along them.
Compact regional development. Compact development requires public guidance and rules that encourage geographic limits. Incentives for compact development, with appropriate transportation links and access to nature, should be combined with limitations on sprawl by withholding infrastructure investments and prohibiting environmentally damaging water supply and sewage disposal practices. Portland has gone furthest in adopting a growth boundary, an elected limited regional government, greater density requirements, and a long-range infill redevelopment plan. In 1999, Tennessee adopted legislation requiring each metropolitan county to adopt a growth boundary and granted cities annexation rights out to the growth boundary at the option of city councils. Clean separations between urbanized and agricultural areas should be the norm, rather than rare.
Education. Funding for schools is inadequate in many middle-aged suburbs, as well as in many working-class suburbs and central cities. Investment in buildings, equipment, and teachers should be as high or higher in middle-aged suburbs and cities as in recently developed outer areas. State aid policies should reward balanced local and regional education funding decisions. Otherwise, desires to avoid public schools are added to desires to find satisfactory housing, creating another severe obstacle to reinvestment in mediocre suburban neighborhoods. If land development on the metropolitan fringe is rewarded with superior schools serving better-prepared students, massive incentives for sprawl by families with school-age children augment disincentives for reinvestment toward the center, middle rings, and intermediate sectors.
Revenue sharing. More annexation by cities of suburban areas is needed, preferably using procedures followed in a few states, such as North Carolina, where annexation has been used extensively. Sharing revenues may be a more practical, although still difficult, alternative to annexation. Where local government boundaries are difficult to alter, which typically is the case, revenue-sharing systems that discourage sprawl and reward reinvestment would be useful. For example, even in the generally prosperous quadrant northwest of Philadelphia, the property tax rate in Upper Darby is six times higher than in nearby Upper Merion, contributing to Upper Darby’s 13 percent population loss in 10 years and its 17 percent relative income decline in 30 years. Revenue-sharing systems can be created explicitly through formulas implicating each unit of general government in a metropolitan region. This approach has been followed in the Minneapolis-St. Paul and the Charlottesville, Virginia, metropolitan areas. Revenue-sharing systems also can be imposed incrementally through state spending decisions, because state governments inherently constitute structures for sharing revenues throughout each state and within each metropolitan area. In Maryland, for example, the centerpiece of the state’s Smart Growth policy involves targeting spending for highways, public transportation, water and sewer systems, and public school buildings toward cities and growth areas of counties.
Comprehensive solutions needed
These types of policies are common in Europe and Canada, although their design and implementation vary, as does their success. In the United States, such policies are rare as elements and do not exist as a set. Some states have limited disparities, primarily through generous annexation opportunities for cities and substantial state finance of public education and other services. Success in limiting sprawl has been rare, with Portland perhaps doing best. Suburban reinvestment has been so little studied that most successes are obscure, but Alexandria and Arlington, in Virginia, have achieved a great deal.
The limited number of promising federal, state, and local policies has led us to imagine a framework in which state and local governments could collaborate, perhaps with the federal government or with private foundations, in financing an integrated attack on sprawl, disparities, and disincentives for reinvestment. We call this framework a Sustainable Region Incentive Fund (SRIF). Local governments would be challenged to achieve a variety of specific objectives, reflecting the six goals described above. Trends toward achieving goals measured with indicators over time would be rewarded with payments from a SRIF. Local governments would be responsible for achieving these goals, working separately or together. Small governments would be less likely than large ones to have significant capacity within their boundaries to achieve the goals. Hence, incentives for collaboration among local governments would be inherent in the policy framework. Methods of collaborating, however, would be left to local initiative.
If a SRIF system were set up statewide or nationally, experience would accumulate that could help inform local governments about successful practices elsewhere. Some incentives, such as organizational, staffing, and research funding, could be provided at the launching moment for this process. Some meaningful intermediate funding also could be provided. Most of the funding, however, would come through rewards for successful performance rather than for promises. Selection of the indicators with which to measure progress and success would be crucial. These could vary among metropolitan areas. They should involve a limited number of indicators, perhaps 5 to 10, to make the goals comprehensible and to permit a formula for reward payments to be communicated and understood. For example, rewards could be given to communities for increasing the median age of their housing; increasing the number of residents, jobs, and commercial sales volume within half a mile of fixed-rail mass transit stops; increasing the use of mass transit by bus and car pools; reducing the loss of nearby farm lands; and increasing test scores in public schools while decreasing the disparity of test scores between sections of the metropolitan area as well as between population subgroups within particular neighborhoods.
This approach may be overly ambitious, but it could be scaled down or, it might be hoped, scaled up, to suit any region’s potential. But there is some evidence, based on examples where a results-oriented strategy has been used, that such an approach will help. For example, the focus on results is similar to the growing interest in benchmarking in public budgeting and in measuring the effects of social programs funded by the public sector and by foundations, such as the Pew Partnership’s program called Wanted: Solutions for America. The standards of learning that have been adopted for public education in 20 states may be the most expansive linkage between client results and public policy consequences. In some of these 20 states, failure on standardized tests by individual students may prevent their graduation from high school, and insufficient percentages of school systems’ students passing may lead to loss of accreditation for specific schools or entire school systems.
The flexibility embodied in proposing goals but not mandating programs is similar to the trend toward pollution prevention through private-sector innovations rather than through publicly mandated best practices. Identifying relationships among results is similar to EPA’s blocking of federal spending for highways in areas that show little promise of meeting air quality goals based on current conditions and past trends. Collaboration could focus on a primary goal, such as meeting water supply needs, as in the Grand Rapids, Michigan, area, but with spinoffs into growth management, sprawl, conservation, and reinvestment occurring as a consequence. The implementation of Virginia’s Regional Competitiveness Act of 1996 has demonstrated that small amounts of state funds can be sufficient to convince local governments, public educators, and private business managers to plan together for economic development. By inference, one can hope that similar or larger amounts of funds targeted to improving the quality of life and enhancing environmental conservation and economic prospects simultaneously might be more effective in achieving results.
Overall, then, the United States faces two challenges in coping with suburban decline. The first challenge is to recognize that such decline already is commonplace and growing worse, as well as to understand the real forces that foster it. The second challenge is to begin to develop needed policies and to take coordinated steps at the local, state, and federal levels to reverse the downward spiral in the nation’s middle-aged bedroom suburbs. In a fortunate alignment, many of these same policies and actions can also be used to extend inroads into the original urban crisis by helping to revive the nation’s cities, including the most poverty-stricken ghettos.