Managing Space to Manage Growth
Daniel R. Mandelker*
Oregonians don’t like sprawl, but they don’t like high density either.
—John A. Kitzhaber, Governor of Oregon
As growth management programs come of age, experience can tell us how they work and what can make them more effective. Strategies that manage space to manage growth are important elements in these programs. Space management directs development to one part of an urban area, but limits it elsewhere, to attain the policies that growth management adopts.
This article examines two longstanding growth management programs that rely on space management: the tiered growth program in San Diego, and the urban growth boundary program in Oregon, as carried out in Portland. The article first reviews the goals that growth management seeks to achieve. It then discusses the San Diego and Portland programs, focusing on the strategies adopted in these programs and the extent to which they were successful. Finally, this article concludes with recommendations for improving space management strategies.
I. Urban Sprawl and the Growth Management Movement
Growth management began in the 1960s to provide new techniques for managing rapid and uncontrolled growth through urban sprawl. Though there is no consensus on a definition of sprawl, commentators usually characterize it as low-density development that expands as leapfrog noncontiguous development from the core of metropolitan areas. Regulations for the Florida state land use planning program define urban sprawl as premature and poorly planned conversion of rural land, and development that does not relate to adjacent land uses and does not make maximum use of existing public facilities.
Critics of sprawl would point to its many problems. These include higher capital and operating costs for private and public facilities, higher transportation and travel costs, and the excessive consumption of agricultural and sensitive lands. They also include the deterioration in the quality of life, and social impacts, such as suburban exclusion and a mismatch of jobs and housing.
The criticism that urban sprawl increases capital facility and service costs gained major support in an early influential study. It showed the cost of servicing scattered and low-density development is much higher than the cost of servicing compact development at higher densities. Critics contested these findings, but most studies conclude that lower densities and urban sprawl do result in higher capital facility costs. Studies have also found modest, but cumulatively significant, reductions in operating costs for compact rather than sprawl development. These findings are important to the legal basis for growth management programs that remedy this problem. Courts have held, and are likely to continue to hold, that land use programs requiring the orderly provision of services and facilities at optimal cost is a legitimate governmental objective in growth management.
A related timing problem is that rapid development may overwhelm a community so that it cannot provide facilities and services when new development needs them. Local governments can handle this problem by providing necessary facilities in advance before growth occurs, but few have the resources to do so. Growth management can time development so that local governments can budget and plan for needed services and facilities.
Policy makers who became concerned with urban sprawl soon realized that zoning cannot handle the sprawl problem. Originally a static system that designated where development could occur, zoning gradually became a more flexible process in which local governments could review development proposals as they were presented for review. Though this process could have controlled the rate, timing, and character of growth, it did not do so because comprehensive plans, and thus land development regulations, did not consider these issues.
Because the problems that drive growth management programs vary, it is difficult to define what growth management does. The conventional understanding is that growth management influences the rate, amount, type, location and quality of growth. One topology lists four types of controls: adequate public facilities programs that prohibit development unless adequate public facilities are available, phased growth programs that determine when to allow development, urban growth boundary programs that set limits on urban growth, and rate-of-growth programs that establish a defined growth rate.
These strategies reflect the various origins of the growth management movement. Some focus on the provision of public facilities and try to time the provision of these facilities with new development. Other strategies manage space, and attempt to regulate the rate of growth or determine where development should locate. Programs with spatial dimensions, such as urban growth boundary programs, control the shape and form of development. Space management is new to American land use planning, though it has long been a key element of land use planning in other countries. An example is the British Green Belt program, which limits the growth of cities to preserve agricultural land and prevent urban sprawl.
Space management programs are especially critical because they dramatically affect the spatial form of development in ways not typical in American tradition. They are good faith efforts to modify development patterns to provide a more desirable pattern for urban growth. Urban growth boundaries, for example, establish a boundary line beyond which new development cannot occur. These programs have major effects on the land market because they prevent development where it otherwise might occur and because they direct development to areas where it might not otherwise occur. Space management programs also have an implicit preference for higher-density, compact urban development in areas where development can occur. This high-density preference is a corollary to the criticism of low-density sprawl, which is considered wasteful and difficult to service.
Two major growth management programs in western cities have made use of space management for over a quarter of a century. They deserve study as examples of how these programs work. One is the tiered system of growth management in San Diego, California. The other is the urban growth boundary in place in Portland, Oregon, which the state planning program requires.
II. Tiered Growth in San Diego
San Diego, though it has cycled through boom and bust periods, is one of the fastest-growing cities in the country. It is also one of the most desirable. Growth management became a major issue in the 1970s, when growth accelerated. The city was large enough, and had enough undeveloped area, so that a growth management strategy made sense within the city limits. When growth became a major problem in the 1970s, the city called in a national consultant who had prepared and successfully defended a phased growth program in New York State. This consultant proposed a tiered growth management program for San Diego. The introductory chapter of his report details the purpose and strategy of the program:
The growth strategy supports neither extreme of unrestrained expansion nor the complete cessation of growth. Rather, it conceives that urban growth will occur in logically defined increments phased with and/or adjusted to the City’s capacity for accommodating such increments.
This statement shows that the principal objective of the program was the provision of facilities needed to serve new development. Its principal concerns were the staging and timing of growth, the timely provision of public facilities within areas where growth could occur, and a requirement that new development should pay the capital costs it requires.
The city also faced several space management problems. Downtown and inner city areas were not attracting enough new development, while excessive development threatened the northern tier. This type of growth pattern would ultimately have produced low-density sprawl in outlying areas, while the inner city declined. The city also has important wetlands, canyons and other natural resource areas that residents value, which new development threatened.
As adopted, the program has three growth tiers: an urbanized tier, a planned urbanizing tier, and an urban reserve tier. The consultant’s proposal encouraged growth in the urbanized tier, staged growth in the planned urbanizing tier, and deferred growth for fifteen to twenty years in the urban reserve. It also included an environmental tier intended to protect the area’s canyons, steep slopes and other natural resources, but the city did not adopt it. The growth management program only applies to residential development, because it assumed nonresidential development will carry its fair share of needed improvement costs and does not affect the need for schools, parks and libraries.
The consultant’s proposal included different policies and objectives for each tier, most of them regulatory, though it proposed other measures, such as redevelopment, where it was necessary in the urbanized tier. There was no strategy for allocating growth to designated areas within the tiers where the program allowed growth. Neither was enough attention paid to the need for capital improvements in the urbanized tier, though there was a brief discussion of a capital improvements program. In the planned urbanizing tier the city adopted a special benefits assessment, which the courts eventually upheld, that carried out the program’s proposal to shift the cost of new facilities to developers. In the urban reserve the principal control was large lot zoning at a minimum of ten acres for each dwelling unit. This type of zoning protects land from urban growth because the density it allows is too low to allow development at an intensive scale.
The San Diego plan creatively used several standard land use measures to manage the rate and direction of growth, though the options then available limited its choice. Today, for example, there is greater support for programs that protect threatened environmental areas. An innovative development exaction shifted the cost of new public facilities to developers in the planned urbanizing area. Problems that arose later reflect, to some extent, the political climate in which the program began. Its growth restrictions were partly a response to an initiative proposal that would have limited growth in the city. Yet the decision to make large areas unavailable for development was both novel and dangerous, as the development industry had never faced the obstacle that large areas of a municipality were off limits. To reduce opposition to this policy, the city made concessions. It removed a substantial area from the future urbanizing to the planned urbanizing tier and dropped the open space tier from the program. Neither did the city adopt legislation protecting natural resource and sensitive areas until 1990.
At first, the program succeeded. Development increased dramatically in the urbanized areas, and growth in the planned urbanizing area declined. A major factor in this shift in development preferences was the absence of a development exaction in the urbanized areas.
Problems arose with the facilities benefits assessment in the planned urbanizing area. Judicial approval of the assessment took seven years, and the collection of fees then lagged infrastructure needs. Problems also occurred in the urbanized area. An obsolete zoning code allowed eight-plex apartment buildings in single-family neighborhoods, front-yard parking, and other undesirable design practices that provoked neighborhood objections. Alarmed residents put pressure on the city to adopt legislation to protect inner area neighborhoods from multifamily development.
Problems also arose with the adequacy of public facilities in the urbanized area. This area attracted development, as the city did not require impact fees there, so only general budget revenues were available for improving new facilities to adequate standards. These revenue sources became inadequate soon after the city adopted the growth management program, when state constitutional initiatives limited real property tax rates and spending growth. The constitutional limitations made it impossible for the city to finance needed capital improvements in the urbanized area, so services and facilities deteriorated or became obsolete. Another problem was that built-up neighborhoods began to demand higher public facility standards. Planners had assumed that existing infrastructure in these neighborhoods would be sufficient.
Demands for more development put pressure on the urban reserve, an area where the program planned for development later. Citizens became concerned when the city council began to shift too much urban reserve land to the planned urbanizing area, where development could occur. In 1985, voters adopted an initiative that requires voter approval for any shift from the urban reserve to the planned urbanizing area, but this victory was short-lived. Voters have approved two projects under this initiative, and later initiatives intended to limit growth failed.
A new form of low-density development that escapes the 1985 initiative has also become popular. Developers took advantage of a city policy that allows clustered developments on four-acre lots in the urban reserve. This type of development does not need voter approval under the 1985 initiative because it does not require reclassification from the urban reserve to the planned urbanizing area. It also has a ready market among affluent homebuyers who seek an exclusive residential environment.
The city’s response to these problems has been slow and inadequate. It has delayed the implementation of ordinances that protect sensitive lands and limit the introduction of multifamily development in residential neighborhoods. The city hired its original consultant late in 1989 to work with a growth management team on improvements in the program, but the council rejected their proposals.
The San Diego history illustrates some common problems faced by spatial growth management programs. First, events outside the program had a major effect, especially on the fiscal side. Judicial delay in the approval of the assessments for capital facilities is one example. Fiscal measures must receive legal approval before a city can use them safely, which means that innovation, though necessary, is risky. Innovative regulatory controls may also face a legal challenge that delays implementation.
The San Diego experience also shows that space management can arouse damaging resistance if it modifies market expectations in land development. San Diego’s tier program conflicts with the American preference for minimum development controls. The density curve is normally less pronounced than what the San Diego program requires, as development is usually less intensive in the core and more intensive in outlying areas. Cutting against this preference meant, over time, that unexpected coalitions would unite against the program. Developers tried to undermine the urban reserve, while inner city residents protested the development and infrastructure problems the program brought. Political support weakened.
In the urban reserve, large lot zoning selected to carry out the program may have made it vulnerable to new development. Because the program preserved this area by limiting growth, its open character attracted low-density development, and the voter initiative did not prohibit it. Although the 1985 initiative did slow development in the urban reserve by requiring voter approval to shift land to the planned urbanizing area, voters ultimately approved two projects.
The San Diego example also shows that attention to implementation detail is essential. One problem was that the program made development policy choices in each tier, but did not have a strategy for allocating and phasing development inside the tiers. There was no strategy, for example, for allocating development within the inner urbanized area. This omission created difficulties when the time came to make development decisions in the tiers, and the city delayed the adoption of a development strategy that could deal with these problems. It finally adopted a plan for the urban reserve in the early 1990s that made strategic choices in that area, and that called for the preparation of subarea plans. The city has adopted some of these plans.
III. The Oregon Urban Growth Boundary Program
Oregon’s state land use and urban growth boundary (UGB) programs are well‑known growth management systems. A set of state planning goals adopted by the state Land Conservation and Development Commission (LCDC) are its critical elements. LCDC reviews local plans and land use regulations and approves them if they comply with the state goals. Local land use regulations and decisions must be consistent with the approved plan. A special tribunal, the Land Use Board of Appeals (LUBA), hears appeals on land use decisions after appellants exhaust all local appeals.
The principal state planning goal that mandates growth management is an urbanization goal that requires incorporated municipalities to adopt urban growth boundaries. Local governments must draw a clear line between areas that can urbanize and areas that must remain non-urban. Local governments must apply seven factors contained in the urbanization goal to decide on the size of the urban growth boundary. Incorporated municipalities apply these factors to designate enough growth within their UGB to provide an adequate land supply for twenty years. A UGB can, and usually does, extend beyond municipal boundaries. The Portland regional planning agency administers this program in the Portland metropolitan area and is responsible for making decisions about the boundary. The state housing goal, supplemented by legislation, requires local governments to provide needed affordable housing within UGB boundaries.
A key purpose of the state program is the preservation of the Willamette Valley in western Oregon, which has most of the state’s valuable agricultural land and most of its population. A complementary agricultural goal requires the preservation of agricultural areas, and the statutes authorize adoption of exclusive farm use zones to reinforce this goal. The statutes also require a minimum eighty-acre lot size in exclusive agricultural zones. Enforcement is the primary problem. Growth can occur outside UGBs in agricultural areas known as “exception lands.” These are lands either committed to urbanization or needed for other uses.
Observers agree that the preservation of agricultural and other natural resource areas were the primary motivation behind the urbanization goal and the UGB policy. These priorities mean that the UGB, unlike San Diego tiers, is not primarily a measure to shape urban growth. The state planning goals also do not include a strategy for allocating development within a UGB.
An important measure of the program’s success is the extent to which growth has occurred inside, rather than outside, UGBs. Unlike San Diego, Portland provides public facilities and subsidies inside the urban growth boundary to encourage development, although highway congestion is a problem. Studies of the UGBs, some limited to Portland, do find that a substantial portion of new development has occurred within UGBs. A study of development inside the UGBs also showed a substantial amount of development occurring in or next to the urban core, as intended. Density increases inside the Portland UGB are impressive,  but densities are lower than the program intended. Lower densities have occurred even though zoning that discourages housing or makes it more costly is prohibited by statute, and though LCDC requires six to ten units per acre for the Portland area on undeveloped, residentially-designated lands.
One of the reasons why higher-density development has not occurred inside the UGBs is that opposition to this type of development has become increasingly common. Developers became disillusioned when they could not build at the expected densities promised by the program at its adoption.
Development has continued to occur at low densities in so-called exception areas outside UGBs, often as spurious farms. This development is substantial and undercuts the urban growth boundary program, though it has slowed in recent years. The conversion of land contiguous to UGBs to low-density development is especially troublesome because it makes the extension of UGBs difficult. If low-density development occurs on land next to the UGB, it will not be available for high-density development when the boundary expands. The UGB must then expand further than it should have been, and higher-density development must leapfrog over the low-density development that is in the expansion area. This is the very type of urban sprawl the urban growth boundary program tries to prevent.
Oregon legislation now allows local governments to designate “urban reserve areas” that are next to UGBs. These areas provide for the long-term urban expansion and cost-effective provision of public facilities and services when the UGB expands. Local governments are to give priority to urban reserve areas when expanding urban growth boundaries.
Battle lines over development are more clearly drawn in the Oregon UGBs than in San Diego because the UGB marks the boundary between areas where urban development can and cannot occur. In San Diego, large lot zoning in the urban reserve area permits low density development. In Oregon the program does not allow urban development outside the growth boundary, though low-density residential development in exception areas undermines this objective. Even so, there are significant price differences between land inside and outside the urban growth boundary. The UGB is not responsible for all of the price increase that has occurred inside the boundary, but price increases aggravate affordable housing problems.
A significant problem in the Oregon UGB program is deciding where development should occur and at what densities. Development at low densities inside the UGBs accelerates demands for boundary expansion, which can damage the goal of preserving agricultural and forest lands. Higher densities within the UGB reduce demand for boundary expansion but create opposition from existing neighborhoods. Housing at higher densities inside the UGB can be expensive and push lower-income housing outward. Balancing these competing claims requires a carefully orchestrated strategy, which is more difficult to secure. The statutes now authorize density increases within a UGB to meet housing needs as an alternative to a boundary expansion.
To help resolve these conflicting pressures, the Portland regional planning agency has adopted an urban growth management plan as part of its Metro 2040 Growth Concept, although opposition has slowed implementation. The plan requires local governments to increase housing densities and meet housing capacity standards set by the plan. The Growth Concept, and the statute giving priority to urban reserve areas in boundary expansions, are the bases for agency regulations for the review of growth boundary expansions. These regulations supplement the state planning goals.
The regulations create a category of “first tier urban reserves” that have a priority in boundary expansions because they are areas where urban services are most effectively provided. A proposed boundary expansion amendment also requires an urban reserve plan that provides for an average minimum residential density of ten dwelling units to the acre and a diversity of housing stock. Plans must ensure the orderly, economic and efficient provision of urban services through annexation to a city, a city and county agreement on planning and zoning, or an urban services agreement. These regulations reinforce the UGB program by giving priority to in-boundary expansions to adjacent lands and by requiring reasonable densities with assurances that adequate services are available.
The Oregon experience shows how expectations about development opportunities and coalition shifts can affect program performance. The development industry welcomed the UGB because it seemed to provide a commitment to higher-density development inside UGBs. When opposition arose within the UGB, the industry felt betrayed. Outside the UGB the agricultural sector, though it usually supports the conversion of farmland, joined environmentalists in a campaign to preserve even marginal agricultural lands from development. The UGB may have become a symbol that polarizes opposing interests and prevents meaningful consensus on growth management strategies.
IV. Defending and Improving Strategies for Growth Management
Problems in the San Diego and Portland space management programs do not diminish their importance as major innovations in land use controls. Many problems are political. Opposition arose when program costs not apparent up front became obvious. Opposition hardened in Portland, for example, to expansion of the growth boundary and more intensive development within it. Framing a program around environmental protection, as in Oregon, can also encourage a rigid defense of environmental areas that prevents compromise. In San Diego, concessions were made at the beginning, political support diminished, and the development community applied pressure to weaken the program. Voters responded with initiatives that limited the city’s freedom to make program changes, but later initiatives lost and the use of the ballot box underscores the loss of political support. The local basis for the program may also make it more vulnerable. In Oregon, growth management through growth boundaries is state-mandated, popular support has continued, and statewide initiatives to weaken the program have lost.
Improvements in growth control techniques cannot eliminate political opposition, but they can improve political acceptance by providing clearer and more effective strategies for growth management. The San Diego and Oregon experience shows that growth management programs need improvement in their strategies for subordinate, second-level policies. The San Diego tiers and the Oregon UGBs provide a framework for growth, but subordinate strategies are not as well developed.
A. The Legal Defense of Growth Restrictions
1. Some Conceptual Problems
A perennial problem for growth management programs is to develop strategies for limiting development where the program does not want development to take place. Traditional land use controls cannot accomplish this objective because they are usually lenient. Local governments do not use zoning and other controls to place limits on where development can occur, impose boundaries that identify growth opportunities, or forcefully direct development to core areas. The San Diego and Oregon programs made major changes in this system. They adopted boundaries that decide where development can and cannot occur, and deliberately directed development to the inner core. Strategies of this kind can create windfalls for wipeouts problems because they dramatically affect development opportunities and land prices on each side of the boundary line. These problems, in turn, can create significant new legal difficulties for the development control system.
Legal problems are most difficult on the side of the boundary where development cannot occur. Development restrictions raise the familiar cry that one group of landowners must accept deep losses in property value to benefit the rest of the community. This is a well-known takings problem.
It is significant that there has not been a successful legal attack against growth restrictions in either San Diego or Portland. One reason may be the tradeoffs in development opportunities these programs provide. They restrict development in some areas but provide development opportunities elsewhere. Another explanation may be that the programs provided enough opportunities for development so that legal attacks on growth restrictions are not necessary. Shifts of urban reserve land in San Diego to allow their development, and development opportunities on exception lands in Oregon, are examples.
The boundary may also affect expectations in ways that diminish taking of property objections. The urban growth boundary in Portland, for example, must allow enough land for twenty years of growth. Since the boundary is likely to expand when this land supply is no longer available, adjacent land near the boundary has a good expectation of development in a reasonable time after twenty years. The question is whether a restriction on development for this period is a taking of property. If courts will accept that a delay in the development of property of this length is not a taking, then restrictions on development in growth management programs are safe from a takings attack.
The Ramapo case upheld delays in development to carry out a growth management program. There, a growth management program deferred growth for as long as eighteen years and permitted development only when designated public facilities and services were available. The New York Court of Appeals rejected a takings claim because it held the delay was part of a reasonable program for controlling growth. A court could apply this kind of reasoning to delays in development that occur in areas near growth boundaries because those delays are also limited in time. How much delay a court will accept is another question.
Supreme Court cases decided since Ramapo raise other questions. It is now clear the courts require compensation for a temporary taking for the period a regulation was in effect before a court holds it violates the takings clause. In addition, the Court’s Lucas decision held a land use regulation that does not allow an economically viable use is a taking per se. That case found a per se taking when a beach setback in a coastal management program deprived a landowner of all economically viable use of his land. The Court would not consider the purpose statement in the act as a basis for upholding the restriction. The Lucas per se takings rule, and the requirement that compensation is payable for a temporary taking when a land use regulation is unconstitutional, could invalidate temporary restrictions on development in growth boundary programs. If the program does not allow any development on land outside the growth boundary, a court could find a temporary taking for the period during which this restriction was in effect.
In areas further from the boundary line, the delay in development may be substantial, so an argument that the program requires only a temporary delay in development may not apply. In these areas, however, the economics of a taking claim may discourage a takings attack. If the landowner believes the discounted development value of her land at a future date is worth more than what she might recover as compensation in a takings suit then she will not sue. Discounted value may be higher because land values will rise as development occurs in adjoining and surrounding areas.
If a landowner brings a takings claim, the critical question is whether the land use restriction is a per se taking because it denies all economically viable use of the property. What is a denial of economically viable use is not clear. In Lucas, the Court did not decide whether there must be a developmental use of the property to avoid a claim that a regulation does not allow an economically viable use. Some cases have held the key question is whether there is a competitive and realistic market for the land that is subject to restriction. This means there must be a market of buyers who are willing to buy the land for development, not for speculation. It does not mean the land use regulation allows a developmental use of the property. If this market exists, there is no taking, but this view may not be a correct interpretation of the Supreme Court’s Lucas decision.
2. Control Techniques for Limiting Development
This discussion of takings problems in growth restriction areas can provide guidance on the type of controls that can limit development yet not create takings problems. Exclusive agricultural zoning, as in Oregon, is one option. Agricultural protection is clearly a legitimate governmental objective, and the courts uphold agricultural zoning against takings claims when land is presently used or suitable for agriculture. A Lucas per se taking does not arise.
Controls that restrict development in sensitive environmental areas are other possibilities. These controls can present takings problems under the Lucas decision because they often prevent any developmental uses, but this problem has not yet been serious. In the wetlands cases, for example, courts have dismissed takings claims where the property was not all wetlands, and the landowner could carry out an economically viable use on that part of the land. Takings problems have arisen only when a permit denial or regulation prohibits development on small single lots. This problem should not occur in undeveloped areas outside growth boundaries where land holdings are likely to be extensive.
When agricultural zoning or environmental land use regulations are not an option, growth management programs may have to rely on large lot, single-family residential zoning, as in San Diego. Large lot zoning can be an effective restriction on development when it zones densities so low that they discourage development, but that kind of zoning can raise legitimacy and takings problems.
Courts may accept large lot zoning at low densities when it implements a growth management program contained in a comprehensive plan. An important pre-Lucas Maryland case upheld restrictive large lot zoning that implemented a growth management program in Montgomery County, next to Washington, D.C. The county downzoning required two acre zoning to protect watersheds and a green belt in a fifty-square mile area around a satellite community designated in a master plan. The master plan promoted the physical isolation of the community from urban sprawl, and carried out a corridor plan adopted for the metropolitan region. The court relied on the purpose of the master plan to hold that downzoning was not a taking of property.
Zoning of this type is more problematic after Lucas because a court can hold that it deprives a landowner of all economically viable use of her land. This issue remains open, though a post-Lucas Maryland case upheld comprehensive five acre zoning adopted to implement a comprehensive plan. This zoning protected the Baltimore watershed from unsuitable development and prevented urban sprawl. The court held the Lucas per se taking rule did not apply because the zoning restriction did not leave the property economically idle.
Densities at two or five acres, which the Maryland cases approved, may prevent development effectively in areas permanently restricted from growth in growth management programs. Low residential densities may still be vulnerable to a takings attack, though judicial tolerance for very low density zoning when used to implement a growth management program may be higher than expected. Low-density zoning may also provide an escape hatch for affluent housing in areas where growth is not supposed to occur.
Transfer of development rights (TDR) programs are another strategy that can help avoid Takings Clause claims in areas where a growth management program prohibits development. TDR programs help avoid these problems because they provide for the transfer of development rights from restricted areas to areas where development can occur. Compensation paid by buyers of rights to sellers can fully or partially mitigate a takings claim. The best example of a TDR program that supports growth management is the Montgomery County, Maryland program that protects the county’s agricultural area from development. Nevertheless, TDR programs in extensive agricultural or natural resource areas are difficult to implement if the market will not generate the trades necessary to provide adequate compensation to sellers of rights. The volume of rights for sale in restricted areas must be in balance with the growth allowed in growth areas to make a TDR program work. If this balance does not occur, additional public intervention through development rights banks that can buy and hold development rights may be necessary, but banks may be expensive to create and difficult to manage. Montgomery County may be unique because its location next to the national capital creates a strong demand for office space. This demand supports the market for development rights.
B. Strategies to Intensify Development Where it is Needed
Space management programs usually encourage higher density development in areas inside the boundaries to offset restrictions on development outside the boundaries. Higher density development often occurs as infill in existing neighborhoods and usually requires upzonings. Residents of these neighborhoods may object if they believe higher densities will have a negative effect on their neighborhoods and may attack upzoning for higher density development as spot zoning. Courts strike down upzonings as spot zoning if they are incompatible with the surrounding area and do not provide a public benefit.
This problem has two dimensions. One problem is at the design and scale level. A zoning ordinance may allow intensive development in existing neighborhoods with poor design or out of scale with its environment. Neighbors then resist and oppose proposals for new development. Attention to design and scale in land development regulations can allow more intensive development that does not destroy existing neighborhood character.
Even good design is not enough if higher density development is the real objection, which is often the case. A growth management strategy for urbanized areas must provide a framework for new development that existing residents can accept. Adequate design review, density floors and ceilings in residentially designated areas, periodic review of how the municipality has dealt with new residential uses, and the funding of infrastructure “banks” the municipality can draw on for capital facilities illustrate measures that local governments can consider.
Local governments may also have to face the impact fee issue in built-up areas, though levying impact fees in these areas may make it more difficult to attract new development. The basic issue is fairness, and deciding when a municipality can shift the cost of new capital facilities to the private sector. If new development in established neighborhoods creates a demand for new or improved facilities, the case for shifting costs to the private side is compelling.
Municipalities also need to adopt plans for the development of urbanized areas that decide where new development will go and at what densities. There are attempts to do just this in plans, such as urban village and urban center plans, which allocate new development within cities. Courts will uphold upzonings that implement a comprehensive plan against objections that they improperly favor an individual landowner at the expense of his neighbors. Specifying the intensity of growth that must occur within growth boundaries can also help. Model legislation proposed by the American Planning Association requires urban growth areas to contain land at “minimum densities and intensities.” They must accommodate a designated percentage of the growth expected to occur within the region or county in which the urban growth area is located.
Programs to allow new development in urbanizing areas are less difficult to manage. Raw land converts to development in this process, and usually there are no neighbors who can object that development is too intensive. Techniques such as floating zones and planned unit development regulations can authorize new development as it occurs, and require compliance with the growth management program.
Assuring adequate public facilities in urbanizing areas is a more difficult problem. In San Diego and Portland, new development has overwhelmed public facilities, especially highways, despite attempts to ensure adequate public facilities as growth occurs. The courts finally upheld the facilities benefit assessment in San Diego, but exactions on development are more difficult to defend following Supreme Court decisions that place the burden on municipalities to justify exactions. Some transportation facilities, such as highway interchanges and mass transit, arguably are a public responsibility and not subject to exaction.
Attempts to resolve this problem by requiring adequate public facilities before a local government approves new development do not always succeed. Defining adequate service levels is difficult, matching incremental development to public facility planning is not easy, and service deficiencies have encouraged sprawl by forcing development to outlying areas. A similar “concurrency” requirement that likewise attempts to require adequate facilities when new development receives approval has also proved difficult to manage.
There is no magic fix that can ensure the provision of necessary capital facilities and services in growth management programs. There must be adequate public facilities budgeting, firm fiscal support and developer exactions that have an adequate legal basis.
Space management strategies that dramatically shape the pattern of development are powerful control measures in growth management programs. Their primary purpose is to designate areas where growth cannot occur and where it is encouraged. Like other land use programs with single-issue or limited objectives, they are overlays on existing land development regulations. They may pay some attention to development densities in areas where growth can occur and to development restrictions in areas where growth cannot occur. They do not pay enough attention to more detailed subordinate strategies that can manage growth in growth-designated areas and prohibit growth where the program restricts it.
This strategic failure makes it difficult to strike a program balance that can link the decision on how much area to commit to growth with the decision on how much area to restrict. Finding the correct balance between growth and growth restriction is the key that will determine the success or failure of a space management strategy.
The imperative of managing change is another important lesson from San Diego and Oregon. Planning for growth management before the fact, in a political environment that may be uncertain at best, is clearly not the easiest task. Governments must adopt the most effective strategies available when they create these programs, but they must also monitor and respond to change. The alternative, as the Governor of Oregon warns, is the impossible.
* Stamper Professor of Law, Washington University in St. Louis. The author would like to thank Nico Calavita, Bob Einsweiler, Frances Foster, Bob Freilich, Wendie Kellington, Stuart Meck, Doug Porter, and Ed Sullivan for their valuable comments on an earlier draft of this article. Of course, the author’s analysis and opinions are his own.
 The text adopts the definition of sprawl used throughout Transit Cooperative Research Program, Rep. 39, The Costs of Sprawl—Revisited (1998). See also Robert W. Burchell & Naveed A. Shad, The Evolution of the Sprawl Debate in the United States, 5 Hastings West-Northwest J. Envtl. L. & Pol’y 137, 140-42 (1999) (defining sprawl as “low-density residential and nonresidential intrusions into rural and undeveloped areas, and with less certainty as leapfrog, segregated, and land-consuming in its typical form”).
 “Urban sprawl” means urban development or uses which are located in predominantly rural areas, or rural areas interspersed with generally low-intensity or low-density urban uses, and which are characterized by one or more of the following conditions: (a) The premature or poorly planned conversion of rural land to other uses; (b) The creation of areas of urban development or uses which are not functionally related to land uses which predominate the adjacent area; or (c) The creation of areas of urban development or uses which fail to maximize the use of existing public facilities or the use of areas within which public services are currently provided. Urban sprawl is typically manifested in one or more of the following land use or development patterns: leapfrog or scattered development; ribbon or strip commercial or other development; or large expanses of predominantly low-intensity, low-density, or single-use development. See Fla. Admin. Code Ann. § 9J-5.003(134) (1999).
 Although the anti-sprawl position has considerable appeal, defenders of sprawl dispute the arguments that sprawl threatens natural resources and creates higher government costs, and deny that it is a serious social problem. See, e.g., Samuel R. Staley, The Sprawling of America: In Defense of the Dynamic City 14-15 (1999) (claiming that the “sprawl index” is declining, that urban development does not threaten agriculture, that the effect of suburbanization on local government costs is exaggerated, and that air quality deteriorates at higher densities). See generally, e.g., Peter Gordon & Harry W. Richardson, Are Compact Cities a Desirable Planning Goal?, 63 J. Am. Plan. Ass’n 95 (1997) (explaining benefits of urban sprawl, including opportunities for infill development). See also Ivonne Audriac et al., Ideal Urban Form and the Dilemma of the Good Life: Florida’s Growth Management Dilemma, 56 J. Am. Plan. Ass’n 470 passim (1990) (noting that sprawl is a response to market preferences, and that attempts to control it will likely drive up land and housing values); Gregg Easterbrook, Suburban Myth, New Republic, Mar. 15, 1999, at 18 (arguing that sprawl is not entirely negative because besides the fact that the alternatives and proposals to remedy sprawl are unrealistic to implement, people actually enjoy some of the effects of sprawl).
 See Reid Ewing, Is Los Angeles-Style Sprawl Desirable?, 63 J. Am. Plan. Ass’n 107, 117-18 (1997). See generally Transit Cooperative Research Program, supra note 1; Symposium, Urban Sprawl, 29 Urb. Law. 157, 158-251 (1997) (citing specific examples of increased costs and pollution caused by sprawl).
 See generally Real Estate Research Corp., The Costs of Sprawl (1974).
 See, e.g., Alan A. Altshuler, Book Review, 43 J. Am. Plan. Ass’n 207, 208 (1977) (asserting that the study underestimated demand for services from higher-density development and mixed density and unit size effects). See generally, e.g., Duane Windsor, A Critique of The Costs of Sprawl, 45 J. Am. Plan. Ass’n 279 (1979) (book review) (commenting on the failure to disentangle density from other factors and a failure to credit sprawl as a response to market preferences).
 See Transit Cooperative Research Program, supra note 1, at 46-49. See also Robert W. Burchell & David Listokin, Land, Infrastructure, Housing Costs and Fiscal Impacts Associated with Growth: The Literature on the Impacts of Sprawl v. Managed Growth 10 (1995) (claiming that planned development and growth can result in reduced costs to communities); Jerry Weitz & Terry Moore, Development Inside Urban Growth Boundaries: Oregon's Empirical Evidence of Contiguous Urban Form, 64 J. Am. Plan. Ass’n 424, 430-34 (1998) (asserting that sprawl and scattered development costs more than contiguous and planned development because it is an inefficient use of land and resources). See, e.g., The Sierra Club, The Dark Side of the American Dream: The Costs and Consequences of Suburban Sprawl (visited Nov. 14, 1998) <http://www.sierraclub.org/sprawl/report98/costs.html#who> (“Providing services to new development has grown so costly in Prince William County, Virginia, near Washington, D.C., that even though the county has the highest property-tax rate in the Commonwealth, every new house brings a $1,688 shortfall.”).
 See Transit Cooperative Research Program, supra note 1, at 50-52.
 See Golden v. Ramapo Planning Bd., 285 N.E.2d 291, 301-03 (N.Y. 1972), appeal dismissed, 409 U.S. 1003 (1972).
 See Eric Damian Kelly, Community Growth: Policies, Techniques, and Impacts 20 (1993).
 See Edward J. Kaiser & David R. Godschalk, Twentieth Century Land Use Planning, 61 J. Am. Plan. Ass’n 365, 365-66 (1995).
 See Eric Damian Kelly, Planning, Growth, and Public Facilities: A Primer for Local Officials 16 (1993).
 For further discussions of urban growth boundary programs see Tom Daniels, When City and Country Collide 187-209 (1999), V. Gail Easley, Staying Inside the Lines: Urban Growth Boundaries, 16-27 (1992), Douglas R. Porter, Managing Growth in America’s Communities 61-69 (1997), and Ned Farquhar, Zoning Fallout: The Implications of Urban Growth Boundary Designations, Zoning News, Mar. 1999, at 1.
 For an early account of this program see generally Daniel R. Mandelker, Green Belts and Urban Growth (1962).
 See generally Dowell Myers & Alicia Kitsuse, The Debate Over Future Density of Development: An Interpretive Review (1999) (discussing the density issue and conflicting reports on the impact of sprawl on urban development in California) (on file with author).
 For more in-depth commentary on the San Diego program see generally Robert H. Freilich, Battle Against Sprawl: Smart Growth Systems Using the Ramapo Approach (1999); Douglas R. Porter et al., Profiles in Growth Management 81-87 (1996); Nico Calavita, Growth Machines and Ballot Box Planning, 14 J. Urb. Aff. 1 (1992) [hereinafter Calavita, Ballot Box]; Douglas R. Porter, San Diego’s Brand of Growth Management: A for Effort, C for Accomplishment, 48 Urb. Land, at 21 (1989) [hereinafter Porter, Effort]. See also Nico Calavita, Vale of Tiers, Planning, Mar. 1997, at 18 (criticizing San Diego’s growth management program as well-intentioned but ineffective against development interests).
 Professor Robert H. Freilich prepared the program as consultant to the city of San Diego. Professor Freilich was renowned for recently having won a major case, Golden v. Ramapo Planning Bd., 285 N.E.2d 291, appeal dismissed, 409 U.S. 1003 (N.Y. 1972), in the New York Court of Appeals, thereby sustaining a growth management program he had developed for the Town of Ramapo (a suburb of New York City). The Ramapo program allowed new development only when adequate public facilities and services were available.
 See Robert H. Freilich, A Growth Management Program for San Diego (1976) (on file with author).
 Id. at 2-5.
 See John W. Witt & Janis Samartino-Gardner, Growth Management v. Vested Rights, One City’s Experience: A Case Study of San Diego, 20 Urb. Law. 647, 650 (1988).
 Professor Freilich stated that another purpose of the program was to organize growth in the planned urbanizing tier through a transportation corridor approach. See generally Freilich, supra note 16.
 See Freilich, supra note 18, at 4-1, 5-1, 6-1.
 See id. at 5-2 (“The objective in identifying . . . [planned urbanizing] areas is to channel new growth into them in an orderly, logical sequence that enables the City to expand facilities and services commensurate with growth.”).
 Professor Freilich did not propose a transfer of development rights program for the environmental tier, probably because development rights transfer was then a new and untried idea. For two discussions of transfer of development rights and other techniques the city could have used in the environmental tier, see generally Rick Pruetz, Saved By Development (1997) and Jerold S. Kayden, Market-Based Regulatory Approaches: A Comparative Discussion of Environmental and Land Use Techniques in the United States, 19 B.C. Envtl. Aff. L. Rev. 565 (1992).
 See Witt & Samartino-Gardner, supra note 20, at 651.
 See Freilich, supra note 18, at 6-19,
 For an analysis of the legal issues suggesting statutory authority and constitutional problems in the program see generally John M. Winters, An Independent Legal Analysis of a Growth Management Program for San Diego (1978) (on file with author).
 See, e.g., Ann Louise Strong, Transfer of Development Rights to Protect Water Resources, Land Use L. & Zoning Dig., Sept. 1998, at 3 (discussing TDRs generally, and specifically considering the use of TDRs in four water resource protection programs in the United States). See generally, e.g., Jeanne S. White, Beating Plowshares into Townhomes: The Loss of Farmland and Strategies for Slowing Its Conversion to Nonagricultural Uses, 28 Envtl. L. 113 (1998) (discussing the importance of farming to communities and the tools being used to preserve farmlands in several areas).
 See Calavita, Ballot Box, supra note 16, at 7-8.
 See id. at 16. See also Interview with Kenneth E. Sulzer, Executive Director, San Diego Association of Governments (Jan. 25, 1999) (noting that habitat protection areas adopted under the Endangered Species Act, 16 U.S.C. §§ 1531-1544 (1994 & Supp. IV 1998), also restrict growth). See generally Craig Manson, Natural Communities Conservation Planning: California's New Ecosystem Approach to Biodiversity, 24 Envtl. L. 603 (1994) (discussing California’s implementation of the Natural Community Conservation Planning program and the possibility of a balance between endangered species and economic concerns in communities).
 See Letter from Nico Calivita, Professor, San Diego State University (Mar. 31, 1999) [hereinafter Calavita Letter] (on file with author).
 See J.W. Jones Co. v. City of San Diego, 203 Cal. Rptr. 580 (Cal. Ct. App. 1984).
 See Calavita, Ballot Box, supra note 16, at 11 (noting that the considerable time lag before facilities were actually built was a result of the lead-time needed to plan, design, engineer, and actually construct public improvements).
 See Porter, Effort, supra note 16, at 25.
 See Calavita, Ballot Box, supra note 16, at 16.
 See Roger W. Caves, Land Use Planning: The Ballot Box Revolution 140-53 (1992).
 The hardening of public opposition to development in areas reserved for development occurred in both San Diego and Portland. Without careful examination of growth trends and land availability, however, it is difficult to determine whether public opposition was justified. In Oregon, moreover, popular initiative could not change the program because it was mandated by state law. See id. at 152-53.
 See Telephone Interview with Nico Calavita, Professor, San Diego State University (Jan. 20, 1999) [hereinafter Calavita Interview]. In November 1998 voters turned down an initiative that would have established an urban growth boundary for the county. See id.; Caves, supra note 36, at 153-62.
 See Calavita, Ballot Box, supra note 16, at 17. The proposals included citywide impact fees, level of service (LOS) standards and a capital facilities plan to meet LOS standards, and phasing of new development if it exceeded the demand for transportation facilities beyond what could by accommodated on the basis of the capital facilities plan. In 1987, the city had also adopted an interim development ordinance that set limits on residential construction for eighteen months. See id. at 12.
 See Dan Eggen, Local Controls Fail to Restrict Growth, Wash. Post, Aug. 9, 1998, at B1.
 See Porter, Effort, supra note 16, at 22-24. The need to seek voter approval of initiatives to limit the city council’s control over the program indicates the extent to which interest groups perceived a lack of political support. See generally Christopher Leo et al., Is Urban Sprawl Back on the Political Agenda? Local Growth Control, Regional Growth Management, and Politics, 34 Urb. Aff. Rev. 179 (1998) (arguing that attempts to control sprawl have failed, in large part, because regional growth management initiatives have not successfully been distinguished from unpopular growth controls).
 Large lot zoning is a problematic zoning technique, and courts have found it unconstitutional when used for exclusionary purposes. See National Land & Inv. Co. v. Kohn, 215 A.2d 597, 612-13 (Pa. 1965). This problem does not seem to have arisen in San Diego.
 Developing this kind of strategy was not part of the work program for the growth management plan. See Freilich, supra note 18, at 1-5.
 See Calavita Letter, supra note 31.
 For surveys of the Oregon program, see generally American Planning Ass’n, Legislative Guidebook Phases I & II, Interim Edition 6-43 (1998); Gerrit Knaap & Arthur C. Nelson, The Regulated Landscape: Lessons on State Land Use Planning from Oregon (1992); Carl Abbott, The Portland Region: Where City and Suburbs Talk to Each Other—and Often Agree, 8 Hous. Pol’y Debate 11 (1997); Robert L. Liberty, Oregon’s Comprehensive Growth Management Program: An Implementation Review and Lessons for Other States, 22 Envtl. L. Rep. (Envtl. L. Inst.) 10,367 (1992); and Edward J. Sullivan, Marking the Twenty-Fifth Anniversary of SB 100, Or. L. Rev. (forthcoming). For a critique of state land use systems, see generally Jerry Weitz, Evolution of State Sponsored Land Use Planning (forthcoming 1999).
 These goals were legislatively mandated upon creation of the UGB. See Or. Rev. Stat. §§ 197.225-.245 (1991).
 The term “land use decision” is defined in id. § 197.015(10), and a rich and varied case law that tends to lead the Land Use Board of Appeals to review most local actions affecting land use in case of doubt.
 The seven factors are:
(1) the demonstrated need to accommodate long‑range urban population growth requirements consistent with LCDC goals;
(2) the need for housing, employment opportunities, and livability;
(3) the orderly and economic provision for public facilities and services;
(4) the maximum efficiency of land uses within and on the fringe of the existing urban area;
(5) the environmental, energy, economic, and social consequences;
(6) the retention of agricultural land as defined, with Class I the highest priority for retention and Class VI the lowest priority; and
(7) the compatibility of the proposed urban uses with nearby agricultural activities.
Department of Land Conservation & Dev., Oregon’s Statewide Planning Goals and Guidelines (1995). The first two factors are called the “need” factors.
 See Or. Rev. Stat. §§ 197.301-.302.
 The housing requirements are at id. §§ 197.303-.314. Section 197.307(3)(a) provides an example of one such requirement:
When a need has been shown for housing within an urban growth boundary at particular price ranges and rent levels, needed housing, including housing for seasonal and year-round farmworkers, shall be permitted in one or more zoning districts or in zones described by some comprehensive plans as overlay zones with sufficient buildable land to satisfy that need.
 See id. § 215.203.
 See Or. Rev. Stat. § 215.780 (Supp. 1998). Smaller lot sizes are allowed as exceptions under strict conditions. See id. §§ 215.780(2)(a)-(b). The statute also requires an 80-acre lot size minimum for forest zones, and a 160-acre minimum for the rangeland agricultural area. See id. §§ 215.780(1)(b)-(c).
 See §§ 197.732(1)(a)-(c). See also 1000 Friends of Oregon v. Land Conservation and Development Commission, 724 P.2d 268, 279 (Or. 1986) (explaining the three types of exceptions local governments can use under sections 197.732(1)(a)-(c)).
 See Or. Rev. Stat. §§ 197.732(1)(a)-(c) (1991 & Supp. 1998).
 See Wendie L. Kellington, Oregon’s Land Use Program Comes of Age: The Next 25 Years, Land Use L. & Zoning Dig., Oct. 1998, at 3-4; Weitz & Moore, supra note 7, at 431. But see Easley, supra note 13, at 5 (quoting purposes of the UGB for Salem, which also include the efficient and economic provision of services, and the matching of services with population growth).
 See Calavita Interview, supra note 38.
 See Kellington, supra note 55, at 4.
 See, e.g., id. (indicating that the Portland UGB is essentially full); Weitz & Moore, supra note 7, at 424 (“[M]ore than 90 percent of Oregon’s new residents between 1980 and 1989 located inside UGBs.”).
 See Weitz & Moore, supra note 7, at 429 tbl.3.
 See Rachel L. Schowalter, Reuse, Restore, Recycle: Historic Preservation as an Alternative to Sprawl, 29 Envtl. L. Rep. (Envtl. L. Inst.) 10,418, 10,421 (1999) (“Since Portland, Oregon, adopted its urban growth boundary in 1975, Portland’s population has grown by almost 50 percent, but it has used only 2 percent more land.”).
 See Or. Rev. Stat. § 197.307(6) (1998). The statute provides that local governments must also have “approval standards” and “special conditions” which are “clear and objective and shall not have the effect, either in themselves or cumulatively, of discouraging needed housing through unreasonable cost or delay.” Id. See also Rogue Valley Ass’n of Realtors v. City of Ashland, LUBA No. 97-260 (Or. Land Use Bd. App. Sept. 24, 1998) (invalidating approval standards in a steep slope ordinance).
 See Or. Admin. R. 660-007-035 (1998).
 See Editorial, Growth Land-Use Plans Must Be Enforced, Atlanta J., Nov. 12, 1998, at A26.
 See Interview with Duane Desiderio and Grant Madsen, Representatives of the National Association of Homebuilders, Dallas, Tex. (Jan. 13, 1999).
 Or. Rev. Stat. §§ 197.73(1)(a)-(c) (1991 & Supp. 1998). These are areas that are already developed for rural residential homesites or for commercial or industrial uses, or are areas “committed” to development because of parcelization or installation of services or because surrounding development makes farming and forestry impracticable. See Liberty, supra note 45, at 10,387. See also 1000 Friends of Oregon v. Land Conservation & Dev. Comm’n, 724 P.2d 268, 277-79 (Or. 1986) (explaining genesis, application, and mechanics of exception areas).
 See Nyran Rasche, Protecting Agricultural Lands in Oregon: An Assessment of the Exclusive Farm Use Zone System, 77 Or. L. Rev. 993, 997 (1998).
 See Interview with Edward J. Sullivan, Attorney, Portland, Or. (Jan. 31, 1999).
 See Or. Rev. Stat. § 195.145 (Supp. 1998). The statute does not specify the techniques that local governments must use to carry out this purpose, but states instead that the Land Conservation and Development Commission “shall provide to local governments a list of options, rather than prescribing a single planning technique, to assure the efficient transition from rural to urban use in urban reserve areas.” Id. Designation of urban reserve areas is voluntary, but the LCDC may require a designation in major metropolitan areas. Prohibiting the subdivision of land is one technique that can keep urban reserve land clear for future development.
 See id. § 195.145(4).
 See id. § 197.298(1)(a).
 See generally Freilich, supra note 18.
 See Gerrit J. Knaap, The Price Effects of Urban Growth Boundaries in Metropolitan Portland, Oregon, 61 Land Econ. 26, 32-33 (1985); Arthur C. Nelson, Using Land Markets to Evaluate Urban Containment Programs, 52 J. Am. Plan. Ass’n 156, 160 (1986).
 See Knaap, supra note 72, at 31; Nelson, supra note 72, at 162.
 Housing affordability is problematic because of the prevailing, fairly low, average income in the area. The rise in housing prices is also due, in part, to the emergence of a fairly high-income segment of the labor force that wants large homes, even on small lots. See Letter from Douglas R. Porter, President, Growth Management Institute (Feb. 22, 1999) (on file with author).
 See id.
 See Or. Rev. Stat. § 197.296(4)(b) (Supp. 1998). Model legislation proposed by the American Planning Association requires monitoring of an urban growth boundary to ensure that land supply remains adequate. See American Planning Ass’n, supra note 45, § 6.201.1(9).
 See Robin Franzen & Brent Hunsberger, Have We Outgrown Our Approach to Growth?, Oregonian, Dec. 13, 1998, at A1.
 See Metropolitan Regional Service District, Urban Growth Management Functional Plan (last modified Feb. 2, 1999) <http://www.multnomah.lib.or.us/metro/growth/tfplan/ funcplan.html>.
 See Portland, Or. Metropolitan Code § 3.01.010(e) (1993) available at <http://www.multnomah.lib.or.us/metro/glance/metcode/metcode1.html>.
 See id. §§ 3.01.012(e)(4)-(5).
 See id. § 3.01.012(e)(1).
 See id. § 3.01.012(e)(2). In addition, the expansion must assist in compliance with the Growth Concept or other statutory or regulatory requirements for land within an urban growth boundary. See id.
 Opposition to high-density development near transit stations is an example. See Douglas R. Porter, Transit-Focused Development: A Progress Report, 64 J. Am. Plan. Ass’n 475, 485 (1998) (noting that transit-focused development has encountered neighborhood resistance and little governmental leadership in four regions). Porter notes that some transit-oriented development has occurred in the Portland area. See id.
In San Diego, transit-oriented development has been limited by the location of transit lines, limited land availability, market problems, and concerns about residential development that was perceived as not paying its own way. See Marlon G. Boarnet & Nicholas S. Compin, Transit-Oriented Development in San Diego County: The Incremental Implementation of a Planning Idea, 65 J. Am. Plan. Ass’n 80, 90-92 (1999).
 See Daniel R. Mandelker, Environmental Policy: The Next Generation, 64 Town Plan. Rev. 107, 108-09 (1993).
 See id.
 See id.
 It is not clear, of course, that urban growth boundaries and development tiers are entirely responsible for the price effects that occur where these controls are used. An answer to the cause-and-effect question is not important, however, to landowners inside and outside the boundary. Inside the growth boundaries, higher land prices create pressure for intense development, and can create equity problems if higher prices mean that lower income households cannot find adequate housing. It should also be pointed out that dramatic price differences on either side of the boundary are only different in degree from the price differentials that always occur in developing areas. In the absence of a growth boundary, however, the price curve will slope gradually. There will not be the dramatic fall-off that occurs at the boundary line. See Tom Daniels, When City and Country Collide: Managing Growth in the Metropolitan Fringe 190 (1999); Keith W. Dearborn & Ann M. Gygi, Planner’s Panacea or Pandora’s Box: A Realistic Assessment of the Role of Urban Growth Areas in Achieving Growth Management Goals, 16 U. Puget Sound L. Rev. 975, 977-78 (1993).
 In Portland, the metropolitan district has defined the “urban reserve” as “land likely to be needed . . . for a 30 to 50 year period.” Portland, Or. Metropolitan Code § 3.01.010 (1998), available at <http://www.multnomah.lib.or.us/metro/glance/metcode/ metcode1.html>. However, under the rules that apply to the program, land closest to the urban growth boundary are likely to be considered first in any urban growth boundary amendment.
 See Golden v. Ramapo Planning Bd., 285 N.E.2d 291 (N.Y. 1972), appeal dismissed, 409 U.S. 1003 (1972).
 See id. at 304-05.
 See First English Evangelical Lutheran Church v. Los Angeles County, 482 U.S. 304, 318-19 (1987). Language in First English suggests that a temporary moratorium on development might not be a taking. See id. at 321 (noting that the Court did not have before it a “case of normal delays” in obtaining building permits, zoning changes, and the like).
 See Lucas v. South Carolina Coastal Council, 505 U.S. 1003, 1014-19 (1992).
 See id.
 See id. at 1031 (“We emphasize that to win its case South Carolina must do more than proffer the legislature’s declaration that the uses Lucas desires are inconsistent with the public interest . . . .”).
 Similar issues arise when communities impose moratoria, either to provide an opportunity to revise comprehensive plans and land use regulations, or to prohibit development until public facilities are adequate to serve the new development. If the moratorium is a taking because it prohibits all development during the moratorium period, a court could award compensation for the time the moratorium was in effect. See Daniel R. Mandelker, Land Use Law § 6.11 (4th ed. 1997).
 If a landowner delays in bringing a takings claim, the delay may work against her because a court may hold that a self-imposed delay means a landowner does not have investment-backed expectations that were frustrated by the development restriction. See Dodd v. Hood River County, 136 F.3d 1219, 1230 (9th Cir.); Tahoe-Sierra Preservation Council, Inc. v. Tahoe Reg’l Planning Agency, 34 F. Supp.2d 1226, 1240-41 (D. Nev. 1999) (appeal pending). Courts consider investment-backed expectations when they apply the Penn Central balancing test to takings claims. See Penn Cent. Transp. Co. v. New York City, 438 U.S. 104, 123-35 (1978) (providing three relevant factors to consider when examining a takings claim: (1) the economic impact of the regulation on the landowner, (2) the extent to which the regulation interferes with investment-backed expectations, and (3) the character of the governmental action).
An exception to the Lucas per se taking rule may also bar landowners who bought land after the UGB was adopted from making takings claims. Lucas held that property was subject to “background principles” of state law. See Lucas at 1029. Although the Court did not explain what this term means, some courts have held that a landowner who buys land takes title subject to legislation adopted before she acquired title. See, e.g., Gazza v. New York Dep’t of Envtl. Conservation, 679 N.E.2d 1035, 1040-42 (N.Y. 1997) (refusing to find a taking where a homeowner was denied a setback variance for wetland regulations known by him to be in existence prior to his purchase of the property); Hunziker v. State, 519 N.W.2d 367, 371 (Iowa 1994) (“[T]he ‘bundle of rights’ the plaintiffs acquired by their fee simple title did not include the right to use the land contrary to the provision of those three Iowa Code sections. . . . These sections and their resulting prohibitions concerning the use of land ran—so to speak—with the land.”).
 See Lucas at 1019 n.8.
 See, e.g., Del Monte Dunes v. City of Monterey, 95 F.3d 1422, 1433 (9th Cir. 1996) (finding that since a sufficient number of people would buy the property for the restricted use it is “commercially marketable”), aff’d on other grounds, 119 S. Ct. 1624 (1999); Florida Rock Indus. v. United States, 18 F.3d 1560 (Fed. Cir. 1994) (finding that when determining fair market value a detailed inquiry into motivation and sophistication of buyers is not necessary).
 Controls explicitly adopted to defer development on a temporary basis are another possibility. The well-known Ramapo case upheld a growth management program that delayed development for up to eighteen years, and authorized the approval of new development only when served by adequate public facilities and services. See Golden v. Ramapo Planning Bd., 285 N.E.2d 291, 302 (N.Y. 1972), appeal dismissed, 409 U.S. 1003 (1972). The town based its growth management plan on a comprehensive plan, and the New York court relied on the plan in upholding these restrictions on development. See id. Moratoria and other controls that delay development temporarily are now vulnerable under the Supreme Court’s recent takings cases as invalid temporary takings. Some courts since First English have held that interim planning moratoria are not takings. See, e.g., First English Evangelical Church v. County of Los Angeles, 258 Cal. Rptr. 893 (Cal. Ct. App. 1989) (holding that an interim ordinance was not a taking); Williams v. City of Central, 907 P.2d 701 (Colo. Ct. App. 1995) (holding that a gaming moratorium was not a taking absent extraordinary delay); Woodbury Place Partners v. City of Woodbury, 492 N.W.2d 258 (Minn. Ct. App. 1992) (holding that there was no taking in an adoption of a two-year moratorium).
However, although a delay in developing land outside a UGB may be temporary, the restriction on land that prevents development is not adopted as a temporary restriction. This fact may lead a court to hold, as it did in Tahoe-Sierra Preservation Council, 34 F. Supp. 2d at 1248-51, that the restriction is not the equivalent of a moratorium, and that a taking has occurred.
 See Boundary Drive Assocs. v. Shrewsbury Township Bd. of Supervisors, 491 A.2d 86, 90 (Pa. 1985). See also Agins v. City of Tiburon, 447 U.S. 255, 261 (1980) (refusing to find a compensable taking in California’s open-space plans).
 See Mandelker, supra note 95, § 12.10. See, e.g., Still v. Board of County Comm’rs, 600 P.2d 433 (Or. Ct. App. 1979) (upholding rejection of non-farm development that violates agricultural preservation policy in exclusive agricultural zone even if economically unfeasible to farm land and no interference with farming in surrounding area).
 For examples of excellent treatises on the law of takings, see generally R. Meltz et al., The Takings Issue (1999) and Stephen J. Eagle, Regulatory Takings (1996).
 The Court in Lucas clearly indicated that regulations denying a property owner all “‘economically viable use of his land’” are per se takings. See Lucas v. South Carolina Coastal Council, 505 U.S. 1003, 1016-19 (1992) (quoting Agins, 447 U.S. at 260).
 See, e.g., K & K Constr., Inc. v. Department of Natural Resources, 575 N.W.2d 531, 535-38 (Mich. 1998) (holding that for a permit application where there was a single comprehensive development comprised of several tracts in common ownership, the parcel under consideration included those tracts).
 See, e.g., Bowles v. United States, 31 Fed. Cl. 37 (1994) (upholding a takings claim for denial of a permit for development of wetlands under the Clean Water Act, 33 U.S.C. §§ 1251-1387 (1994)).
 Support can be found for this position in Justice Brennan’s opinion in Penn Central, although it was not fully developed. See generally Penn Cent. Transp. Co. v. New York City, 438 U.S. 104 (1978). See also Daniel R. Mandelker, Waiving the Taking Clause: Conflicting Signals from the Supreme Court, in 1994 Proceedings of the Institute on Planning, Zoning, and Eminent Domain 57 (Southwestern Legal Foundation ed., 1995) (noting the convoluted nature of takings jurisprudence).
 See Norbeck Village Joint Venture v. Montgomery County Council, 254 A.2d 700 (Md. 1969). It should be noted that in this author’s opinion the Maryland courts are more receptive than most to local government land use regulations.
 See id. at 703.
 See id. at 703-04.
 The court held that the landowner “did not, by an Olney plan country mile, meet their heavy burden of showing that the rezoning they dispute confiscates their property.” Id. at 706.
 See Security Management Corp. v. Baltimore County, 655 A.2d 1326, 1328 (Md. Ct. App.). The court also rejected equal protection and due process claims. See id. at 1330-33.
 See id. at 1329.
 See id. at 1330. The county council had rejected a rezoning to 16 dwelling units to the acre in what the developer called an “environmentally sensitive” community.
 One commentator has suggested clustering techniques for developments in large lots that can leave room for future infill development when a UGB expands. See Easley, supra note 13, at 14. How effective this technique can be is problematic.
 The courts have upheld zoning for up to five-acre lots when necessary because of environmental conditions. See Mandelker, supra note 95, § 5.26.
 See Julian Conrad Juergensmeyer et al., Transfer of Development Rights After Suitum, 30 Urb. Law. 44l, 444-48 (1998).
 See id. at 444-46.
 See Pruetz, supra note 24, at 210-14.
 See Barry C. Field & Jon M. Conrad, Economic Issues in Programs of Transferable Development Right, 51 Land Econ. 331, 338 (1975).
 See Sarah J. Stevenson, Note, Banking on TDRs: The Government’s Role as Banker of Transferable Development Rights, 73 N.Y.U. L. Rev. 1329, 1371 (1998).
 See Mandelker, supra note 95, §§ 6.27-6.30.
 I am indebted to Ed Sullivan for these suggestions.
 There may be legal problems in levying impact fees in established neighborhoods if the municipality does not have the authority to levy these fees in connection with the issuance of a building permit or other development approval, such as a conditional use. See, e.g., Bringle v. Board of Supervisors, 351 P.2d 765 (Cal. 1960) (holding that a zoning board had implied power to attach condition requiring dedication of land). The opportunity to levy fees as part of subdivision approval does not usually exist because inner city areas are built-up.
 The plan for King County in Seattle, Washington is an example. See Douglas R. Porter, Profiles in Growth Management 230-55 (1996).
 See Mandelker, supra note 95, §§ 6.32-6.34.
 See id. § 201.1(6)(a).
 See, e.g., Dolan v. City of Tigard, 512 U.S. 374 (1994) (requiring that government’s permit condition does not impose more than a proportionate degree of burden on the landowner).
 See Porter, supra note 13, at 122-133.
 See S. Meck, Model Statutes on Uniform Development Standards, Concurrency and Smart Growth Technical Assistance (Apr. 19, 1999) (draft on file with author); Porter, supra note 13, at 131. See e.g., Fla. Dept. of Transp., Report of Transportation and Land Use Study Committee 19-32 (1999) (on file with author). The concurrency requirement, as stated in Florida legislation, is intended to provide “that public facilities and services needed to support development shall be available concurrent with the impacts of such development.” Fla. Stat. Ann. § 163.3177(10)(h) (West 1999). In Florida, the concurrency requirement has tended to force new development to the urban fringe, where service levels are higher. See Fla. Dept. of Transp., supra.