Issues in Science and Technology, Online A Publication of National Academy of Sciences National Academy of Engineering Institute of Medicine University of Texas at Dallas
  • Home
  • Back Issues
  • Search


Caught in Traffic

DENA BELZER

GERALD AUTLER

Countering Sprawl with Transit-Oriented Development

Incorporating public transit into planning can help foster efficient land use patterns and create a more balanced set of transportation choices.

Since World War II, two mutually reinforcing processes have characterized U.S. cities: decentralization and an increasing reliance on the automobile. Heavy investment in roads and other implicit subsidies of automobile use, combined with comparatively low levels of transit funding, have facilitated decentralized urban development patterns and inefficient use of land. These development patterns, which we refer to as sprawl, have made transit service unviable or inefficient in most suburban areas and many urban areas and have reinforced automobile dependence.

Although decentralization initially helped bring relief from crowded industrial cities, and the automobile spelled newfound mobility for millions, it has become increasingly apparent that these trends have gone too far. Most Americans now have few, if any, effective transportation choices other than the automobile. The machine that promised freedom has become a trap, diminishing our options rather than expanding them. What's more, our current land use and transportation patterns contribute to a range of environmental ills, economic inefficiencies, health and safety issues, and social inequalities.

Despite some encouraging trends, there is still a need for concerted policy efforts to reshape land use patterns at the regional, local, and neighborhood scales; to expand transportation choices; and most of all, to better integrate transportation and development. The need to promote such integration is driving growing interest in transit-oriented development (TOD), which focuses on better connecting transit systems physically and functionally with the surrounding development.

Effective TOD can help foster more efficient land use patterns and create a more balanced set of transportation choices in which automobiles coexist alongside other options. However, a look at some existing projects that purport to be transit-oriented shows that although many have been quick to embrace the term, most have fallen short of the concept's full potential. A clearer vision of exactly what TOD is meant to accomplish, coupled with a new approach to development, are necessary if TOD is to maximize its contribution to our social, economic, and environmental health.

Costs of the status quo

Sprawl continues to have an enormous impact on our built and natural environments. Overall, U.S. metropolitan areas are expanding their urbanized area at about twice the rate of population growth. Even areas experiencing little or no population increase are continuing to urbanize new land. Driving has also increased at a far faster rate than population. Our inefficient land use and transportation patterns produce economic, environmental, social, and other costs.

First, sprawl costs us money. Water, sewer, and road infrastructure costs more to build and maintain at low densities, and much of the additional cost is borne by the public at large rather than by the residents of low-density areas. An auto-dominated transportation system is economically inefficient, causing households, regions, and the nation as a whole to pay more for transportation than they would if a more balanced system were in place.

U.S. households spend more on transportation than on any other category of expense except housing and more than households in most other wealthy countries. Other developed countries may have higher government spending on transportation, but even so, as a nation the United States spends more of its gross domestic product on transportation than other wealthy countries. This is true because much of the total cost of driving is paid for indirectly. The full cost of automobile transportation includes not just public spending on roads and bridges but also private spending that is included in the cost of development. For example, the provision of parking is a major financial consideration in any form of development, and it raises the cost significantly. Indirectly, we all pay for "free" parking spaces in shopping centers, office parks, and housing.

More difficult to measure is the cost of abandoning older developed areas as growth moves farther and farther out. The abandonment of existing physical infrastructure and building stock is wasteful from an economic and environmental standpoint, but it also contributes to economic decline and social isolation in many of the areas left behind, imposing costs on individuals and society in the form of poverty, poor education, crime, and so on.

Second, sprawl has significant environmental consequences. Farmland and open space are lost to subdivisions, office parks, and shopping centers; development leads to the destruction of sensitive habitats; the increase of impervious surfaces and destruction of wetlands reduces the rate of aquifer replenishment and contributes to degradation of water quality; and the replacement of tree cover with asphalt and buildings creates urban heat islands that raise temperatures (and therefore cooling bills) and accelerate the formation of smog and ozone.

In addition, private vehicles use about twice as much fuel per passenger mile as public transportation and emit significantly more carbon dioxide. Transportation now accounts for roughly 43 percent of total energy consumption in the United States and 33 percent of total greenhouse gas emissions. Although we have made great strides in reducing vehicle emissions, private automobiles are still a significant contributor to smog and air pollution.

Third, sprawl has significant equity implications. The abandonment of the metropolitan core leaves inner cities and first-ring suburbs struggling to provide adequate services with an eroded tax base even as growth continues on the periphery. Inner areas sometimes end up effectively subsidizing areas being built farther out. Further, because of the lack of adequate alternatives to the automobile, many low-income households either spend an exorbitant portion of their total income on transportation or suffer the effects of decreased mobility. The problem is exacerbated by the fact that so much employment is located in suburban areas that are difficult to reach without a car. And it is not just low-income people who are hurt by limited transportation options; a rapidly increasing number of senior citizens have outlived their ability to drive.

Many projects billed as transit-oriented fall far short of yielding the full range of benefits.

Fourth, health and safety also suffer from sprawl. Automobile use is a significant generator of air pollution, which is a factor in asthma and other respiratory diseases and which has a detrimental impact on life expectancy. Traffic accidents result in more than 40,000 deaths and hundreds of thousands of injuries every year in the United States, a fatality rate per passenger mile 17 times higher than that of transit buses.

Although it may be overstating the case to claim that our land use and transportation patterns cause the health problems associated with lack of exercise, they certainly exacerbate them. Exercise, including walking, can help reduce the risk of a wide range of diseases. It can also lead to increased longevity as well as higher quality of life in old age. Yet residents of areas that do not facilitate walking and bicycling as part of a daily routine are deprived of a form of exercise that is easy, convenient, and free. In addition to their human cost, these health and safety problems also take an economic toll in the form of health care costs and shortened productive life spans.

Finally, there is increasing dissatisfaction with sprawl because of its negative effect on quality of life. The lack of places with a distinct identity in the sea of identical subdivisions and strip malls is one aspect, but there are more concrete ones. The average U.S. driver spends 443 hours driving every year, the equivalent of more than one eight-hour workday every week. This is not limited to commuters: The average mother spends more than an hour per day in the car, a significant increase from the mid-1980s. This is a function of the increasing distance between destinations and the lack of mobility choices for children. The number of foot and bicycle trips taken by children has fallen steadily as traffic has increased and more suburbs have been developed without adequate pedestrian infrastructure.

Most people would probably agree that having choices is a key component of quality of life, and in fact many residents of walkable and transit-oriented neighborhoods cite the existence of mobility choices as a quality of life feature and an important factor in their selection of neighborhood. Yet most Americans don't enjoy such choices.

TOD as part of the solution

The exact nature and magnitude of any of the individual costs described above can be debated. Nevertheless, it is clear that changing our land use and transportation patterns can yield substantial benefits that can be captured by individuals, regions, and the nation as a whole. For example, a new financial instrument called a location-efficient mortgage allows people who live in neighborhoods in which they can spend less on transportation and who take advantage of that fact to obtain a larger mortgage than they would normally be eligible for. And if Americans used transit for just 10 percent of their travel needs, we could reduce greenhouse gas emissions by more than one-fourth of what would have been our target under the Kyoto Accord, we could reduce smog and carbon monoxide significantly, and we could stop importing the amount of oil that we buy annually from Saudi Arabia.

Although TOD is only one part of a larger tool kit of policies (often referred to collectively as "smart growth") that are necessary for reshaping our regions, it is an important strategy for maximizing the benefits of transit investments and providing a real alternative to traditional development at the local level. Moreover, as a model of development in which efficient land use and transportation work together, TOD can help lay the groundwork for larger regional efforts.

Critics contend that TOD will not work or that it is not the way people want to live. However, there is abundant evidence that when good alternatives to the automobile and traditional suburban development are provided, many people choose them. New York, Boston, Chicago, and San Francisco have high levels of transit ridership (and lower levels of household spending on transportation) compared to many other metropolitan areas and are also among the country's most desirable real estate markets. Within those regions, there are many neighborhoods that embody TOD features: They provide an array of housing, shopping, and mobility choices; households own fewer automobiles, use them less, and spend less on transportation; and residents ride transit, walk, and bicycle more. This is true even when income levels are taken into account. In fact, many of those neighborhoods are extremely desirable places to live, commanding high rents and housing prices. These places show that development and transit really can work together and that neighborhoods that provide mobility choices and convenient services really can lead to different travel practices and concrete benefits for individuals and society alike.

The challenge now is to replicate existing transit-oriented neighborhoods and bring the benefits of TOD to more people. There are a number of encouraging trends that suggest that it may be possible to do so. Sprawl and its consequences have increasingly been the focus of attention during the past decade. Movements to change the status quo of development, such as the Congress for the New Urbanism, have achieved notable successes in theory and in practice, and many urban areas have experienced revitalization and population growth after decades of loss, indicating a renewed interest in urban living.

On the transportation side of the equation, there are also new developments. Investments in transit systems were up in the 1990s, with bus and rail systems upgraded and expanded, new rail systems built even in unlikely places such as Dallas, and existing systems gaining riders. Transit ridership increased steadily during the late 1990s through 2001, not only reversing decades of decline but also growing more quickly than automobile use during that period.

Arlington County, Virginia, just outside Washington, D.C., is one of the most successful attempts to merge these trends into working TOD. For more than 30 years, the county has pursued land use and transportation policies intended to use land efficiently and reduce automobile dependency for those who live and work along the corridor of the Metro subway system. Policies have concentrated much of the county's residential and job growth along the corridor, ensuring access to transit, a rich mix of land uses, and incentives for avoiding automobile use. Surveys and ridership statistics show high levels of transit use by residents of the corridor and in-commuting employees alike, a much lower rate of vehicle-trip generation by residential development than in other suburban areas, and significant vehicle-trip reduction by employers that have actively pursued transportation demand management strategies supported by the county. In some residential projects, 40 to 60 percent of the tenants do not use cars on a daily basis. Office parking standards are significantly lower than in nearby suburban locations, with a maximum requirement of 1.7 spaces per 1,000 square feet (and even less in some developments). Metro has gradually been able to replace station parking with housing and commercial space even as ridership has increased, because a majority of riders walk to the train.

However, Arlington County is the exception rather than the rule. TOD is neither as prevalent as it might be nor as effective as it could be. Many transit stations, both new and old, are still surrounded by parking, cut off from the adjacent neighborhoods, or lacking a mix of land uses. Moreover, many projects that are billed as TOD fall short of yielding the full range of potential benefits. They constitute transit-adjacent development or transit-related development rather than a true integration of land use and transit.

A Miami project clearly shows this. The project is called TOD by virtue of its location next to rail and bus transit, and although it is certainly better than standard suburban development and likely does encourage transit use, it falls short of the Arlington success story. The parking standard is 3.3 spaces per 1,000 square feet of development, a fairly typical level for suburban development. Moreover, there is a lower priority on housing than on office or retail, and the supermarket provides 200 surface parking spaces. All this suggests that the project does not constitute a balanced neighborhood in which most people can accomplish most daily tasks without a car. To be fair, a new project cannot be expected to yield the same benefits as an area that has evolved over 30 years, but the point remains.

If the potential benefits of TOD are so compelling, why has it not yet become part of standard development practice? Equally important, why do so many projects labeled TOD fall short of their potential?

Redefining TOD

The existing TOD literature focuses on barriers to TOD. Such barriers include the conservative approach of lenders and developers and their preference for known formulas, zoning codes that prohibit the type of development in question, neighbors' objections to the density or some other aspect of the project, and so on. These are all legitimate problems, but most of them appear as obstacles to a wide range of mixed-use projects and, indeed, to many development projects regardless of their type. They may explain why projects don't get built, but they do not adequately explain the disappointing results of many projects that do get built.

In the case of the Miami project, when the developer was asked about barriers, he was hard-pressed to name any; his overall sense was that the project had gone smoothly. Financing was easy, zoning codes were supportive, and there was no significant public opposition. Removing the kinds of barriers mentioned above clearly is not enough to guarantee an optimal outcome.

Part of the problem is that most existing definitions of TOD are physical; they focus on characteristics such as density next to transit. This leads to a binary notion of success or failure: Projects that are built with those characteristics succeed, whereas those that are not built fail. Missing is a clear notion of the desired functional attributes of the project, and therefore a straightforward way to measure success--and different degrees of success--by way of performance standards and outcomes. Some projects, such as the one in Miami, meet a purely physical definition but fall short when judged on a functional basis, because they do not permit significantly lower levels of auto use and ownership.

In other cases, there may be significant benefits, but information necessary to judge a project on a functional basis is simply not available. The city of Mountain View, California, in Silicon Valley, has actively encouraged development around transit stations on brownfield sites and in its downtown. It has successfully combined New Urbanist principles of development with proximity to transit, increasing housing and transportation choices, and significantly raising the quality of development over prevailing standards in the area. For its efforts the city has won a number of awards. However, the city is unable to provide any data on vehicle ownership, transit ridership, trip generation, vehicle miles traveled, and so on. No targets were used in the planning process, and no comprehensive functional evaluation of the projects has been carried out. Despite this, an award from the American Planning Association cites one of the benefits of the city's efforts as "improved air quality and reduced road congestion." Although plausible, there is no evidence to back up this claim.

Planners have few guidelines for translating TOD rhetoric into reality.

If our goal is to address the costs of sprawl through TOD, we first need to clearly define what the outcomes should be. Then we need to be able to assess whether our efforts are really working and learn from successes and failures alike. A new definition of TOD that focuses on desired functional outcomes, not just physical characteristics, is an important next step. Such a definition can help us better judge the success of TOD projects on the basis of their ability to provide measurable benefits appropriate to the particular context. Equally important, it can guide planning efforts by making goals and tradeoffs explicit from the beginning. Our proposed framework focuses on three main outcomes or goals of TOD: location efficiency, choice, and value recapture/financial return.

Location efficiency is simply the ability to minimize automobile dependency (and more generally the need to travel long distances by any mode) by maximizing the potential synergies between, on the one hand, different land uses and, on the other hand, development and transit. Location efficiency can be quantified through measures such as parking demand, automobile ownership, mode split (the percentage of people using a given mode of transportation), and vehicle miles traveled. If a project does not yield tangible benefits in these areas, it would be difficult to make the case that it's meaningful TOD.

Choice in housing, mobility, and shopping means effective alternatives to the automobile, a land use program that generates synergies so that more daily needs can be fulfilled close to home, and a range of housing types, from single-family houses to apartments, to accommodate diverse incomes and family structures.

Finally, value recapture/financial return means that TOD can actually create value for developers, communities, and households. For developers, TOD can be not just financially viable but quite profitable, especially because such projects generally require less costly parking. For individuals, households, and communities, the benefits can include lower transportation spending and housing costs.

This discussion may still leave readers wondering what TOD will actually look like, particularly if they hear the descriptions offered by some of the more extreme opponents of smart growth, who sometimes paint a frightening picture of a world in which the government will force everyone to live in crowded apartments, abolish automobiles, and restrict personal choice. In reality, there is no single prescription for how to build TOD. The form will necessarily vary depending on the particular environment and location; an urban neighborhood in Chicago will develop differently than will a suburban rail station outside Dallas. It is certainly true that in general TOD will have to be built at higher densities than are typically found in the suburbs, with less private space and less space for automobiles. But TOD can function at a range of scales and in a variety of contexts.

Moreover, in order to work properly and make the tradeoffs (such as higher density in exchange for greater access) attractive, TOD should also contain amenities such as pedestrian-friendly shopping streets, high-quality design, and more and better public space to compensate for the smaller amount of private space. It should offer different housing and transportation choices even within the same development: Single-family homes and townhouses can exist alongside apartments and condominiums, and automobiles will be part of the mix, although a less dominant part than in standard new development.

In short, TOD can resemble many desirable existing urban and suburban neighborhoods. Contrary to the myth reinforced by opponents of TOD and smart growth, it is sprawl, not its antidote, that limits choice by providing only one viable transportation mode and, for the most part, one type of housing.

Bolstering TOD

What can be done to further the cause of TOD? Some policies have begun to move in the right direction. The federal government is funding transit systems more generously, tying funding more closely to land use decisions, and giving transit agencies more flexibility in how they use the land they own around stations. Many state governments, regional entities such as metropolitan planning organizations, and local jurisdictions are also using tools at their disposal to encourage better land use decisions. All of these are important steps that should be built on.

In our view, TOD projects face three significant challenges: factors that result in projects with suboptimal outcomes. But there are practical ways of dealing with those challenges.

First, TOD involves an even broader array of actors than other development, all of which bring very different goals to the table. In addition to local governments, developers, lenders, and community groups, the TOD process involves transit agencies and, indirectly, the federal government, which places some restrictions on the transit agencies that it helps to fund. Very often these actors have disparate goals and consequently pursue strategies that work at cross-purposes to each other. For example, community groups may see limiting local traffic as their primary concern and object to any intensification of land use even if it generates less traffic than would an equivalent amount of standard development. The transit agency may see its primary objective as maximizing the revenue from joint development projects on its property, even though the highest-revenue project may not be the best from the standpoint of functional outcomes such as ridership.

A special fund should be created to help finance TOD projects that do not meet standard underwriting criteria.

The lack of definitional clarity exacerbates this problem, because there is no agreement about what TOD should accomplish from a functional standpoint. For example, a project may yield minimal or ambiguous quantifiable benefits, because such outcomes were not explicitly included in the project's original formulation. In such cases, there is clearly a flaw in the way the goals are defined. Moreover, without a clear functional definition, there is no framework for weighing tradeoffs and making choices about how to balance competing goals.

Second, planners have few guidelines for translating TOD rhetoric into reality. TOD is more difficult than other development because it must achieve a functional integration of transit and the surrounding development as well as synergy among all the uses. Despite a fair amount of piecemeal research on travel behavior, shopping patterns, and so on, planners have fairly little comprehensive information about what is needed in order to maximize the synergies between transit and development. Although we can recognize location efficiency when we see it, we have few concrete prescriptions for creating it in different settings.

What scale of development is necessary to make TOD work in different contexts? How much retail and of what type is needed to serve the resident population and employees so that they are less car-dependent? How can a project strike a balance between providing some parking without allowing it to either place a significant financial burden on the development or detract from the overall pedestrian and transit orientation of the neighborhood? These are the kinds of questions to which planners have few answers.

Third, TOD typically suffers from a very fragmented regulatory, institutional, and policy environment. Despite the need for careful coordination and synergy among the different parts of TOD projects, projects must move forward in an environment of complex and sometimes contradictory regulations. Moreover, there is often no comprehensive plan or vision for station areas and usually no single actor that can set the agenda or coordinate the various actors. In addition, although local governments are usually in the best position to lead the process, many suffer from a significant leadership gap. Many local officials do not understand the potential benefits of TOD, or if they do, they are unsure of how to move forward. As a result, TOD projects often suffer from a lack of vision. Without ambitious goals, a coordinated approach, and a streamlined regulatory framework, many projects fall far short of yielding the full potential benefits of synergy between transit and development.

These three challenges can be overcome. The following are the most important steps that need to be taken. All actors involved must

  • Forge agreement on a functional definition of TOD.
  • Articulate quantifiable goals and measurable performance standards that drive all aspects of design and implementation of any given project.
  • Explicitly consider the impact of different choices and tradeoffs on the outcome of any given project.

Advocacy and research organizations, universities, and key federal agencies should

  • Provide information and technical assistance and work with planners, local governments, developers, lenders, and transit agencies to build intellectual capital, to develop realistic visions for TOD, and to implement those visions.
  • Conduct more focused and systematic research on how people make transportation choices, what mix of land uses leads to the best functional outcomes, and so on; with the goal of translating that research into planning guidelines and performance standards.
  • Create a typology of TOD projects and scenarios for different contexts, as well as performance criteria for each project type, such as appropriate parking standards and design standards for dealing with whatever parking is created.
  • Create a TOD fund to help finance TOD projects that cannot obtain traditional financing because they do not meet standard underwriting criteria or are not recognized as part of the traditional development portfolio.

Local governments must

  • Take the lead in conducting comprehensive planning and coordinating the myriad components of projects. Create plans that focus on a long-range vision and the incremental steps needed to implement it.
  • Establish TOD area plans around all transit stations, including comprehensive parking strategies and policies for incorporating mixed-income housing into TOD projects.
  • Closely coordinate actions with the transit operator(s), neighborhood groups, and other key actors.

Transit agencies should

  • Plan for long-term value, not short-term return.
  • Recognize the need for every station to be an integral part of a larger area.
  • Plan for TOD at the system-wide scale, assessing how each station fits into the larger network.

Finally, developers and lenders should

  • Become educated about the financial structure and performance of existing TOD and mixed-use projects.
  • Revise underwriting practices that are incompatible with the goals of TOD, such as those that require standard parking ratios for TOD projects.
  • Design flexible projects and use phasing to demonstrate market viability, test assumptions, and allow for the evolution of projects over time.

We can reap enormous benefits by changing our transportation and land use patterns. TOD is not a panacea, but it is an important starting point. Concrete goals, research into the means of achieving those goals, and mechanisms for holding projects to measurable standards are needed to articulate a vision of what is possible. Careful planning and strong leadership are necessary to implement that vision.

Recommended reading

Dena Belzer and Gerald Autler, Transit-Oriented Development: Moving from Rhetoric to Reality (Washington, D.C.: Brookings Institution Center on Urban and Metropolitan Policy, June 2002) (www.brookings.edu/dybdocroot/es/urban/publications/belzertodexsum.htm).

Michael Bernick and Robert Cervero, Transit Villages in the 21st Century (New York: McGraw-Hill, 1996).

Robert J. Shapiro, Kevin A. Hassett, and Frank S. Arnold, Conserving Energy and Preserving the Environment: The Role of Public Transportation (Washington, D.C.: American Public Transportation Association, August 2002).

Surface Transportation Policy Project, Center for Neighborhood Technology, Driven to Spend (Washington, D.C., 2000).

Transportation Research Board, Costs of Sprawl--2000 (TCRP Report 74, Washington, D.C.: National Academy Press, 2000).

Transportation Research Board, The Costs of Sprawl--Revisited (TCRP Report 39, Washington, D.C.: National Academy Press, 1998).

U.S. Environmental Protection Agency, Our Built and Natural Environments: A Technical Review of the Interactions between Land Use, Transportation, and Environmental Quality (Washington, D.C.: EPA 231-R-01-002, 2001).


Dena Belzer is principal and Gerald Autler (gautler@strategiceconomics.com) is associate at Strategic Economics, a consulting and research firm in Berkeley, California.