2002 Utah L. Rev. 1
Utah Law Review
2002
Articles
*1 SPRAWL, GROWTH BOUNDARIES AND THE REHNQUIST COURT
Michael Lewyn [FNa1]
Copyright ? 2002 Utah Law Review Society; Michael Lewyn
I. Introduction
Over the past several decades, metropolitan America has been revolutionized by suburban sprawl--"low-density, single-use development, married with strip and auto-oriented commercial land uses, at the very edges or beyond the fringe of existing urbanization." [FN1] Sprawl typically involves: (1) low-density development requiring dependence on automobiles; (2) segregated land uses (that is, commercial uses are far from residential uses); (3) long distances and poor non-automotive access between housing, jobs, and schools; (4) consumption of land for suburban development occurring at a faster rate than population growth; and (5) "consumption of agricultural and/or environmentally sensitive land for suburban development." [FN2] At the end of World War II, roughly 70% of metropolitan Americans lived in central cities. [FN3] By contrast, in 1990, only about 40% of metropolitan Americans, and only 31.3% of all Americans, lived in central cities. [FN4] *2 In recent years environmentalists, [FN5] many planners, [FN6] and a few politicians [FN7] have sought to limit suburban sprawl. A recent survey by the General Accounting Office (GAO) revealed that 84% of county officials and 64% of city officials rate sprawl as a "high," "very high," or "moderate" concern. [FN8]
Critics of the status quo argue, inter alia, [FN9] that suburban sprawl:
1. Destabilizes urban neighborhoods, causing "a middle class exodus from the central city, causing in turn crime and the segregation of poorer and minority populations in the declining central city." [FN10]
2. Destroys farmland and open space; for example, "from 1982 to 1992 [America] lost to urban and suburban development an average of 400,000 acres per year of 'prime' farmland, the land with the best soils and climate for growing crops." [FN11]; and
*3 3. Forces Americans to drive in order to meet their basic needs, increasing vehicle traffic, thus increasing air pollution [FN12] and traffic congestion. [FN13]
*4 The most stringent [FN14] anti-sprawl measure adopted by any American state is Oregon's "urban growth boundary" (UGB) [FN15] program. Urban growth boundaries are "lines on maps, surrounding areas already marked by 'urban-type' development, within which that type of development is to be channeled and encouraged, *5 and beyond which such development is to be discouraged or forbidden." [FN16] Outside the boundary, rural industries (such as logging) and open space are promoted. [FN17]
This Article briefly describes Oregon law, and then focuses on three issues: (1) whether UGBs are constitutional under the Supreme Court's 2001 decision in Palazzolo v. Rhode Island [FN18] and other relevant precedent, (2) whether the UGB has in fact saved Portland (Oregon's largest city) [FN19] from the social problems caused by sprawl, and (3) whether the side effects of UGBs make them a cure worse than the disease of sprawl.
II. Background: What Oregon Law Requires
In 1973, Oregon enacted Senate Bill 100, which established a comprehensive statewide growth management program. [FN20] Senate Bill 100 was enacted based on legislative findings that "[u]ncoordinated use of lands within this state threaten the orderly development, the environment of this state and the health, safety, order, convenience, prosperity and welfare of the people of this state," [FN21] and that "[t]he promotion of coordinated statewide land conservation and development requires the creation of a statewide planning agency to prescribe planning goals and objectives to be applied by state agencies, cities, counties and special districts *6 throughout the state." [FN22] The Oregon Legislature accordingly created a new citizen commission to oversee the planning program, the Land Conservation and Development Commission (LCDC). [FN23] The Legislature also created the Department of Land Conservation and Development as LCDC's staff for implementing the program, [FN24] and directed the LCDC to appoint the Department's director and supervise its staff. [FN25] The Legislature further required all Oregon cities and counties to adopt comprehensive land use plans and regulations implementing such plans. [FN26] The Legislature defined a comprehensive plan as a "generalized, coordinated land use map and policy statement . . . that interrelates all functional and natural systems and activities relating to the use of lands, including but not limited to sewer and water systems, transportation systems, educational facilities, recreational facilities, and natural resources and air and water quality management programs." [FN27] Such plans must be consistent with statewide planning goals [FN28] to be drafted by the LCDC. [FN29] Between 1974 and 1976, the LCDC adopted nineteen planning goals. [FN30] Local governments then adopted land use plans, which were reviewed by LCDC. [FN31]
Although the Oregon Legislature did not directly state that local governments must establish UGBs, it implicitly did so by using the term "urban growth boundary" in numerous statutes. [FN32] So the LCDC satisfied legislative intent when *7 it enacted Planning Goal 14, which specifically requires every incorporated community to establish a UGB [FN33] that excludes rural land. [FN34] The Oregon courts have interpreted Goal 14 "to prohibit urbanization outside existing UGBs" [FN35] and therefore to "preclud[e] the conversion of rural land to urban use" [FN36] outside UGBs. [FN37] Neither Goal 14 nor any other LCDC rule defines the term "urban use." [FN38] However, the Oregon courts have held that an outdoor performing arts center seating 15,000 people is an "urban" [FN39] land use, as are "areas of half-acre residential lots to be served by community water and sewer." [FN40] On the other hand, "residential density of one house per ten acres is generally not an urban intensity." [FN41]
Land outside UGBs may be converted to "urban use" only if such land is physically developed to the extent that it is no longer available for rural land uses [FN42] or is "irrevocably committed to urban levels of development" [FN43]--that is, if "it is impracticable to allow any rural uses in the . . . area." [FN44] To take advantage of these exceptions to Goal 14 (or to "take an exception," in the terminology used by Oregon courts), [FN45] a local government must show not only that commercial farm or forest use is impractical, but that all other rural uses (including sparse settlement, small farms, etc.) are impractical. [FN46]
*8 A UGB must contain the urban core and sufficient undeveloped land to accommodate population growth. [FN47] UGBs are drawn and amended based on seven factors enunciated in Goal 14: (1) the need to accommodate long-term population growth 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) maximum efficiency of land uses, (5) the environmental, energy, economic, and social consequences of a plan, (6) the retention of agricultural land uses, and (7) the compatibility of urban land uses with nearby agricultural activities. [FN48]
Portland's UGB is administered by a regional agency, the Portland Metropolitan Service District (commonly known as "Metro"). [FN49] Metro has the power to force cities and counties to amend their land use plans to conform with the boundary set by Metro. [FN50] The Portland metropolitan UGB was adopted by Metro in 1979, approved by LCDC in 1980, [FN51] and encompasses twenty-four municipalities, portions of three counties, [FN52] and 369 square miles (or 236,000 acres) of land. [FN53]
III. Constitutionality: Are UGBs an Unconstitutional "Taking" of Land?
The United States Constitution provides that private property may not "be taken for public use, without just compensation." [FN54] It is not clear that the Framers of the Takings Clause intended it to restrict regulation of property (as opposed to *9 actual physical appropriation). [FN55] Nevertheless, the Supreme Court has held that "there will be instances when government actions do not encroach upon or occupy the property yet still affect and limit its use to such an extent that a taking occurs." [FN56] Some commentators have suggested that UGBs are intrusive enough to constitute a "taking" of land, and thus cannot be implemented unless the government compensates all property owners adversely affected thereby. [FN57] Although no Supreme Court case is directly on point, lower court decisions have indirectly addressed the constitutionality of growth boundaries.
A. Are UGBs a "Taking" Under Supreme Court Precedent?
1. The Key Cases: A Summary
In the 1978 case of Penn Central Transportation Co. v. New York City, [FN58] the Supreme Court announced a three-part balancing test governing "regulatory takings" cases: in determining whether a land use regulation is intrusive enough to constitute a "taking," the court weighs the regulation's economic effect on the landowner, the extent to which the regulation interferes with a landowner's reasonable investment-backed expectations, and the character of the government action. [FN59] In the Palazzolo case, the Supreme Court reiterated its commitment to the Penn Central balancing test. [FN60]
*10 Of all the Court's post- Penn Central cases, the case of Agins v. Tiburon [FN61] is most closely on point. The Agins plaintiffs challenged municipal zoning ordinances that prohibited the construction of more than five single-family homes on the plaintiffs' five-acre tract of land. [FN62] Specifically, the plaintiffs argued that the zoning ordinances were a "taking" because they "forever prevented [their land's] development for residential use" [FN63] and thus "completely destroyed the [land's] value." [FN64]
The Supreme Court upheld the zoning ordinances, for two reasons. First, the ordinances "substantially advance legitimate governmental goals" [FN65]-- specifically, "protect[ing] the residents of [the city] from the ill effects of urbanization." [FN66] Second, the ordinances did not "prevent the best use of [plaintiffs'] land . . . nor extinguish a fundamental attribute of ownership" [FN67] because the plaintiffs "may be permitted to build as many as five houses on their five acres of prime residential property." [FN68]
The zoning ordinances upheld in Agins are quite similar to Oregon's urban growth boundaries. Both UGBs and the zoning laws at issue in Agins limit development in rural areas in order to prevent urbanization of those areas. [FN69] And like the ordinances at issue in Agins, Oregon law allows some residential development on rural land: the Oregon Supreme Court has stated that residential development outside UGBs is not an "urban use" (and is therefore allowed) if the developer builds no more than one unit every ten acres, [FN70] and has reserved judgment on whether the one-acre lots contemplated by Agins [FN71] are "urban" (and thus presumably forbidden outside UGBs). [FN72] In other words, the major difference *11 between the ordinances upheld in Agins and Oregon law is that the former allowed landowners to build one lot per acre, while Oregon law allows landowners to build one lot (and maybe more) per ten acres--a distinction of no obvious constitutional significance.
It could be argued that Agins is no longer dispositive, because Rehnquist Court rulings have sharply curtailed government regulation of land use. [FN73] Two decisions in particular are arguably relevant: Lucas v. South Carolina Coastal Council [FN74] and Palazzolo v. Rhode Island. [FN75]
The Lucas plaintiff purchased two beachfront lots in South Carolina on which he intended to build houses. [FN76] Two years after he purchased the lots, the state enacted a law which prohibited such beachfront construction. [FN77] The Court assumed for the purposes of its decision that this statute "rendered [plaintiff's] parcels 'valueless."' [FN78] The Court stated that as a general rule, [FN79] a taking occurs when land use regulation (1) "'does not substantially advance legitimate state *12 interests"' [FN80] or (2) "'denies an owner economically viable use of his land."' [FN81] Because the statute at issue prevented the plaintiff from gaining any financial benefit from his land, the Court held that the plaintiff had suffered a constitutional taking unless the state could "identify background principles of nuisance and property law that prohibit the uses he now intends." [FN82] The Court reasoned that
regulations that leave the owner of land without economically beneficial or productive options for its use--typically, as here, by requiring land to be left substantially in its natural state--carry with them a heightened risk that private property is being pressed into some form of public service under the guise of mitigating serious public harm. [FN83]
The Court accordingly remanded the case to the South Carolina courts so the state could have a chance to "identify background principles of nuisance and property law that prohibit the uses [Lucas] now intends in the circumstances in which the property is presently found." [FN84] The Court added in dicta that a landowner "whose deprivation is one step short of complete" [FN85] may be entitled to compensation under certain circumstances, explaining:
Such an owner might not be able to claim the benefit of our categorical formulation, but, as we have acknowledged time and again, '[t]he economic impact of the regulation on the claimant and . . . the extent to which the regulation has interfered with distinct investment-backed expectations' are keenly relevant to takings analysis generally. [FN86]
The Palazzolo court elaborated on Lucas by clarifying the Court's definition of a "total taking" [FN87] and by explaining the relationship among the factors relevant *13 to "partial taking" cases. [FN88] In 1959, the Palazzolo plaintiff formed a corporation to develop waterfront land in Rhode Island. [FN89] The corporation's applications were denied three times by various Rhode Island agencies--first for lack of essential information, and later on environmental grounds. [FN90] In 1971, Rhode Island enacted legislation creating the Rhode Island Coastal Resources Management Council ("the Council"), an administrative agency charged with the duty of protecting the state's wetlands. [FN91] In 1978, the corporation's charter was revoked for failure to pay corporate income taxes, and title to the property passed to the plaintiff as the corporation's sole shareholder. [FN92] The plaintiff then filed two additional development applications with the Council. Both were denied because of their adverse impact on salt marshes within the property; [FN93] under the Council's regulations, such salt marshes were "coastal wetlands" on which development must be limited. [FN94] The Rhode Island Supreme Court affirmed the decision of the Council. [FN95] The plaintiff appealed this decision to the U.S. Supreme Court, which reversed in part and remanded. [FN96]
The Supreme Court began its opinion with a general description of the law of regulatory takings. The Court held that "a regulation which 'denies all economically beneficial or productive use of the land' will require compensation under the Takings Clause." [FN97] The Court added that
where a regulation places limitations on land that fall short of eliminating all economically beneficial use, a taking nonetheless may have occurred, depending on a complex of facts including the regulation's economic effect on the landowner, the extent to which the regulation interferes with reasonable investment-backed expectations, and the character of the government action. [FN98] *14 Thus, the government generally must compensate landowners for regulations that are "total takings," [FN99] that is, those that deprive landowners of all "economically beneficial use" of their land. [FN100] By contrast, "partial taking[s]," [FN101] that is, regulations that do not eliminate all economically beneficial use of the plaintiff's land, are judged by the Penn Central three-part balancing test. [FN102]
As to the "total takings" half of this test, the Court affirmed the Rhode Island court, holding that the Rhode Island regulations were not a "total taking" because even after the Rhode Island authorities' adverse decision, the plaintiff could still build a $200,000 residence on his eighteen-acre parcel. [FN103] The Court explained that "[a] regulation permitting a landowner to build a substantial residence on an 18-acre parcel does not leave the property 'economically idle."' [FN104] Further, "it is undisputed that the parcel retains significant worth for construction of a residence." [FN105]
The Court went on, however, to hold that the Rhode Island Supreme Court had erroneously analyzed the plaintiff's "partial takings" claim, and to remand the case to the Rhode Island courts so that they could address that claim under the Penn Central balancing test. [FN106] The Court reversed for two reasons. First, the Rhode Island Supreme Court erred in holding that the plaintiff's takings claim was not yet ripe for judicial review. [FN107] Second, the state court erred in holding that the plaintiff "had no reasonable investment-backed expectations that were affected by [the state's] regulation . . . [because he] did not become the owner of the land until 1978 . . . [when] there were already regulations in place limiting Palazzolo's ability to fill the wetlands for development." [FN108]
As to the latter issue, the Court explained that an otherwise unreasonable taking
do[es] not become less so through passage of time or title. . . . [Otherwise] the post enactment transfer of title would absolve the State of its obligation to defend any action restricting land use, no matter how extreme or unreasonable. A State would be allowed, in effect, to put an expiration date on the Takings Clause. [FN109] *15 In other words, a regulation may be a compensable "taking" even if it is enacted before the plaintiff buys the land being "taken."
Because the Court so held by a five to four margin, [FN110] Justice O'Connor's separate concurrence is especially significant as to the "investment-backed expectations" issue. Justice O'Connor wrote that "the Rhode Island Supreme Court erred in effectively adopting the sweeping rule that the preacquisition enactment of the use restriction ipso facto defeats any takings claim based on that use restriction," but that nonetheless, the Court's holding "[did] not mean that the timing of the regulation's enactment relative to the acquisition of title is immaterial to the Penn Central analysis." [FN111] Instead, "interference with investment-backed expectations is one of a number of factors that a court must examine. . . . [T]he regulatory regime in place at the time the claimant acquires the property at issue helps shape the reasonableness of those expectations." [FN112] In support of this view, Justice O'Connor reasoned that if
existing regulations dictate the reasonableness of [investment-backed] expectations in every instance, then the State wields far too much power to redefine property rights upon passage of title. . . . [But] if existing regulations do nothing to inform the analysis, then some property owners may reap windfalls and an important indicium of fairness is lost. [FN113] Justice O'Connor accordingly wrote that rather than adopting "per se rules," [FN114] courts "must attend to those circumstances which are probative of what fairness requires in a given case." [FN115] In other words, "investment-backed expectations" are *16 one of the three factors examined under the Penn Central balancing test, and a landowner's decision to acquire property already governed by the challenged regulations is in turn highly relevant to the presence or absence of such expectations.
Justice O'Connor also discussed the third of the Penn Central factors: the "character of the governmental action" challenged by a Takings Clause plaintiff. [FN116] She explained that under this prong of the Penn Central test, a "use restriction on real property may constitute a 'taking' if not reasonably necessary to the effectuation of a substantial public purpose." [FN117] This language echoes the Agins Court's requirement that a regulation "substantially advance legitimate governmental goals" [FN118] in order to not constitute a taking. Because Justice O'Connor supplied the crucial fifth vote for the Court's decision, her concurrence is likely to be followed by future courts. [FN119]
In sum, Palazzolo, interpreted in light of Justice O'Connor's concurrence holds:
1. That "a regulation which 'denies all economically beneficial or productive use of land' will require compensation under the Takings Clause." [FN120]
2. That a regulation allowing a landowner to build one house on eighteen acres of land is not such a "total" taking. [FN121]
3. That regulations other than "total takings" are governed by the Penn Central balancing test, under which the courts must balance: (a) "the regulation's economic effect on the landowner," [FN122] (b) "the extent to which the regulation interferes with reasonable investment-backed expectations," [FN123] and (c) "the character of the government action" [FN124] (that is, the extent to which a "substantial public purpose" [FN125] supports the regulation at issue).
*17 2. Applying the Case Law to UGBs
(a) Are UGBs an Unconstitutional "Total Taking"?
As noted above, a regulation which "denies all economically beneficial or productive use of land" [FN126] will generally require compensation under the Takings Clause. UGB statutes modeled on Oregon law are clearly not such a "total taking" under Palazzolo. [FN127] The Palazzolo plaintiff sought to persuade the Supreme Court that he suffered a "total taking" [FN128] because he could only build one $200,000 residence on his eighteen-acre parcel. [FN129] The Supreme Court flatly rejected this theory, holding that a regulation "permitting a landowner to build a substantial residence on an eighteen-acre parcel does not leave the property 'economically idle."' [FN130]
Landowners in Oregon can do far more with their land than the Palazzolo plaintiff could do with his, for two reasons. First, the Oregon courts have held that houses on ten-acre lots are not "urban uses" [FN131] and thus may be built outside UGBs. [FN132] By contrast, the Palazzolo plaintiff could only build one house on an eighteen-acre parcel. [FN133] Second, Oregon landowners are not even limited to building one house on a large parcel of land: instead, they may also use land outside UGBs for agricultural purposes, [FN134] and may even convert such land to urban uses if "it is impracticable to allow any rural uses" [FN135] on such land. Thus, Oregon law is apparently less restrictive than the regulations challenged in Palazzolo. It follows that if the Rhode Island regulations at issue in Palazzolo did not "leave the [plaintiff's] property 'economically idle,"' [FN136] Oregon's less *18 restrictive laws do not do so either, and thus are not a "total taking" under Lucas and Palazzolo.
Some commentators endorse a more pro-plaintiff interpretation of Supreme Court precedent. For example, one commentator writes that because UGBs "prevent development consistent with the otherwise predictable use of the property," [FN137] they automatically deprive landowners of "all or mostly all of the value of their property." [FN138] But this need not be the case. For example, suppose that a zoning ordinance, by preventing Blackacre from being turned into a subdivision, reduces its fair market value [FN139] from $20 per square foot to $12 per square foot. Blackacre will have lost 40% of its value--more than the owners would like, but certainly less than "all or mostly all of the value of their property." [FN140] By contrast, in Lucas, the value of plaintiff's property was allegedly reduced to zero. [FN141]
It has also been suggested that a taking occurs whenever property suffers "a significant dimunition in value." [FN142] If this is the case, UGBs, although not per se unconstitutional, [FN143] may lead to a large number of successful takings claims, because
the creation of the boundary itself will cause dramatic differences between the value of undeveloped property inside the boundary and the value of similar property outside. . . . [T]hose on the inside can rest assured that the supply of developable land is now artificially limited, and hence more valuable. Those on the outside will be prohibited from entering the market. [FN144] *19 The "significant dimunition in value" standard, however, was rejected in both Lucas and Palazzolo. The Lucas Court held that "in at least some cases the landowner with 95% loss will get nothing." [FN145] Moreover, a year after Lucas, the Supreme Court cited cases rejecting challenges to regulations that diminished land value by 75% and 92.5%, in a decision unanimously reaffirming that "our cases have long established that mere diminution in the value of property, however serious, is insufficient to demonstrate a taking." [FN146] As a result of such statements, lower courts applying Lucas have "not usually found a taking because the land use regulation attacked did not deny the landowner all economically viable use." [FN147]
The Palazzolo Court similarly held that a regulation that "places limitations on land that fall short of eliminating all economically beneficial use" [FN148] is not automatically invalid, but rather is subject to the Penn Central balancing test. [FN149] Thus, a regulation that significantly reduces the value of a landowner's property is not per se unconstitutional.
(b) Applying the Penn Central Balancing Test to UGBs
As explained above, [FN150] even a "partial taking" (that is, one that reduces the value of a landowner's property by less than 100%) may be a compensable regulatory taking if it "interferes with reasonable investment-backed expectations" [FN151] or is not "reasonably necessary to the effectuation of a substantial public purpose." [FN152] Neither of these factors turns Oregon's UGB statutes into a compensable "taking."
i. Investment-Backed Expectations
Justice O'Connor, writing as the crucial fifth vote in support of a five to four decision, [FN153] wrote that "the state of regulatory affairs at the time of acquisition" [FN154] *20 is relevant to the extent of landowners' investment-backed expectations, and that "the nature and extent of the permitted development under the regulatory regime vis-a-vis the development sought by the landowner may also shape legitimate expectations." [FN155] It follows that a UGB is unlikely to interfere with landowners' investment-backed expectations where, as in Oregon, UGBs have existed for over two decades, [FN156] because a property owner will rarely have an "investment-backed expectation" in resisting enforcement of statewide statutes and regulations enacted decades ago. [FN157]
Even in states where UGBs have not yet been enacted, proposed UGBs are likely to interfere with a landowner's reasonable investment-backed expectations only if a landowner reasonably relied on their absence--for example, where property outside the UGB has previously been zoned for (or was about to be zoned for) intensive development, and a landowner bought the property in reliance upon such zoning. [FN158] Otherwise, the landowner's reasonable expectations will not have been affected by the newly enacted UGB. State and local governments can easily minimize such problems by including land already zoned for development within UGBs. [FN159] It therefore appears that even if UGBs do diminish rural property values, rural landowners will rarely have constitutional claims under Supreme Court precedent.
ii. Substantial Public Purpose
Finally, a "partial taking" may also be a compensable taking if it is not supported by a "substantial public purpose." [FN160] As noted above, the Agins court held that limiting the spread of urbanization is a legitimate state interest justifying restrictions upon development of rural and suburban land. [FN161] Thus, UGBs (which, in Oregon, were enacted at least partially in order to protect rural land from the *21 spread of suburbia) [FN162] are clearly supported by a "substantial public purpose" under Agins.
In sum, three factors are relevant to whether a land use regulation constitutes a taking: the weight of the state interests favoring the regulation; whether the regulation renders plaintiff's property valueless or nearly so; and whether the regulation interferes with plaintiff's reasonable investment-backed expectations. [FN163] None of these factors supports a finding that UGBs generally constitute a "taking" that requires governments to compensate adversely affected landowners. Thus, UGBs are constitutional under Rehnquist Court case law.
B. Lower Courts Address UGBs (Sort Of)
Although the Oregon courts have upheld Oregon's land use laws, [FN164] they have never specifically addressed the constitutionality of UGBs. However, one federal decision indirectly supports the constitutionality of UGBs. In Buckles v. King County, [FN165] plaintiffs challenged a county's decision to zone their property as "Rural Area" pursuant to the Washington growth management statute (which established "urban growth areas" similar to the Oregon UGBs). [FN166] Plaintiffs *22 argued, inter alia, that this "designation 'fails to advance a legitimate county interest and thereby takes Plaintiffs' property without payment of just compensation."' [FN167]
The U.S. Court of Appeals for the Ninth Circuit began its discussion by noting that, as a general rule, "a land use regulation does not constitute a taking if the regulation does not deny a landowner all economically viable use of the property and if the regulation substantially advances a legitimate government interest." [FN168] The court then held that the county had advanced numerous substantial interests in support of its decision, one of which was "complying with the Growth Management Act's limitation on urban development outside of the urban growth boundary because the [plaintiffs'] property lies outside of King County's urban growth boundary." [FN169] If "limitation on urban development outside of the urban growth boundary" is a legitimate state interest, it arguably follows that a UGB statute specifically designed to limit such development advances a legitimate state interest. Thus, Buckles supports the constitutionality of UGBs.
IV. Have UGBs Cured the Mischiefs of Sprawl?
Although UGBs are probably constitutional, they may nevertheless be unwise if they fail to prevent problems associated with sprawl, such as urban decay, destruction of farmland, and automobile dependency. Oregon's UGBs appear to have prevented the urban decay that other cities have suffered as a result of sprawl. However, the impact of the UGB upon the Portland region's land use and transportation patterns has been less dramatic.
A. Saving the City
As suburbs have grown, older central cities throughout America have emptied out. Between 1900 and 1950, every large American city gained *23 population. [FN170] But most older cities have been devastated by sprawl: of the eighteen American cities with over 500,000 people in 1950, thirteen lost population in the 1950s, fifteen in the 1960s, and sixteen in the 1970s. [FN171] For example, by 2000 St. Louis had lost almost 60% of its 1950 population, while Buffalo and Cleveland had lost nearly half of their 1950 population. [FN172] Those cities that have gained population have done so either by becoming hubs for foreign immigration or by annexing newly developed areas that would be considered suburbs in other cities. [FN173] As cities have become smaller, they have become poorer. In 1990, thirty-one of America's thirty-seven largest cities had poverty rates above the national average, [FN174] and central cities contained half of America's poor (as opposed to one-third in 1960). [FN175] Jobs, as well as people, have fled to suburbia: about 95% of the fifteen million new jobs created in the 1980s were in suburbs. [FN176]
In some regions, city life is not a viable choice for any residents other than the poorest or most adventurous. For example, only 4% of Cleveland-area households earning over $100,000 live in the city of Cleveland. [FN177] Even institutions that would be urban in other communities are suburban in Cleveland: the office of EcoCity Cleveland, one of the area's leading anti-sprawl groups, is located in the suburb of Cleveland Heights. [FN178]
Middle-class flight feeds upon itself: as cities become more and more dominated by the poor, who cannot afford to move to suburbia, city neighborhoods become more dangerous (because poverty-packed areas have higher crime *24 rates), [FN179] city schools become less prestigious (because children from poor backgrounds are typically less academically proficient than students from affluent backgrounds), [FN180] and city tax bases dwindle due to the city's poverty (thus causing higher taxes and inadequate services) [FN181]--thus causing additional middle-class flight. [FN182]
While some older cities have decayed, Portland has grown and prospered. From 1980 (when the Portland UGB was created) [FN183] to 2000, the city of Portland's population grew by over 40% [FN184] after declining for several decades. [FN185] Moreover, Portland's population growth compared favorably to that of the central cities of the most comparable western metropolitan areas. Table 1 compares Portland's growth to that of Denver, Salt Lake City, and Seattle--three regions *25 that, although otherwise comparable, [FN186] have lacked UGBs for most or all of the past two decades. [FN187]
*26 TABLE 1: POPULATION GROWTH
City population % Metropolitan area
increase/decrease population % increase
1950-80 [FN188] 1980-2000 1980-2000 [FN190]
[FN189]
Portland -1.5 43.7 43.8
Denver 18.4 12.3 47.5
Seattle 5.6 14.0 46.2
Salt Lake 10.4 11.0 46.5
City
As Table 1 shows, the UGB allowed Portland to gain a proportionate share of the region's population growth, while Denver, Seattle, and Salt Lake City were left in the dust by their suburbs. By contrast, in the decades before the UGB was adopted, Portland was losing population while Denver and Seattle gained population--even though Portland annexed thirty-nine miles of land between 1950 and 1980, increasing its land mass by nearly 60%. [FN191]
Moreover, Denver, Seattle, and Salt Lake City are far healthier than some other cities in high-growth regions. For example, the Atlanta metropolitan area's population grew by over 80% (far faster than metropolitan Portland) during the 1980s and 1990s [FN192]--yet while the Atlanta region's population surged, Atlanta's central city population actually declined from 425,000 to just over 416,000. [FN193] *27 Similarly, metropolitan Washington, D.C. grew by over 40% from 1980 to 2000 [FN194] while the central city population tumbled from 638,000 to just over 572,000. [FN195]
Because Portland annexed almost twenty-two miles of previously suburban territory in the 1980s, [FN196] it could be argued that annexation rather than UGBs caused Portland's startling growth. This argument is probably incorrect, for two reasons. First, as Table 2 shows, Salt Lake City and Denver annexed even more territory than Portland did, yet they failed to experience comparable urban growth.
TABLE 2: CENTRAL CITY GROWTH THROUGH ANNEXATION, 1980-1990
% increase in city land area % increase in population
[FN197] [FN198]
Portland 21.0 43.7
Seattle 0 12.3
Denver 37.7 14.0
Salt Lake City 47.7 11.0
Second, Portland annexed land during the 1950s, 1960s, and 1970s just as it did in the 1980s, yet failed to gain population because the city's land area failed to keep up with the migration of city residents to suburbia: Portland encompassed 64 square miles in 1950 and 103 square miles in 1980--yet the city *28 actually lost population. [FN199] Without UGBs to limit sprawl, Oregonians moved to suburbia faster than the city of Portland could expand its boundaries. [FN200]
As Portland's population has grown, its economy has grown as well. The number of private sector jobs increased by 21.4% in the city of Portland between 1992 and 1997 [FN201] while jobs in Salt Lake City, Denver, and Seattle increased by 6.6%, 8.8%, and 8.4% respectively. [FN202]
Because of Portland's growth and prosperity, Portland is one of the few central cities that has not become a dumping ground for the region's poor. Table 3 compares poverty trends in Portland and other western cities to poverty trends in their suburbs.
*29 TABLE 3: POVERTY TRENDS, 1979-97 [FN203]
Poverty rate [FN204] 1979 Poverty rate 1997
Denver City 13.7 16.4
Denver suburbs 5.4 6.5
Seattle City 11.2 12.9
Seattle suburbs 6.1 5.9
Salt Lake City 14.2 16.2
Salt Lake City suburbs 6.3 6.7
Portland City 13.0 12.6
Portland suburbs 7.2 7.8
In Denver, Seattle, and Salt Lake City, the central city poverty rate is more than twice the suburban poverty rate, and the gap between city and suburb has stayed roughly the same or grown. [FN205] By contrast, Portland has actually narrowed the city-suburb economic gap: its poverty rate was 1.8 times the suburban poverty rate in 1979, and was only 1.61 times the suburban poverty rate in 1997. [FN206]
In sum, Portland's UGB has arguably enhanced rather than reduced consumer choice, by making the city of Portland a viable option for middle-class households. Before the UGB was created in the 1980s, Portland was losing population while other western cities were gaining population. But over the past two decades, Portland has (unlike other western cities) grown as fast as its suburbs, and has narrowed the city-suburb economic gap. So for city- dwellers and would-be city-dwellers, the UGB has been a success on this ground alone.
B. Protection of Natural Resources
The preservation of agricultural and other natural resources areas was a *30 primary motivation behind Oregon's planning statutes. [FN207] For example, Oregon law justifies the creation of LCDC by stating that the "promotion of coordinated statewide land conservation and development requires the creation of a statewide planning agency." [FN208] Presumably, the UGB's success in this respect can be measured by Oregon's success in preserving farmland.
Table 4 shows the amount of farmland in the counties that are part of the Portland, Denver, Seattle, and Salt Lake City metro areas.
TABLE 4: ACRES OF FARMLAND IN METRO-AREA COUNTIES [FN209]
1987 1992 1997 [FN210] % loss 1987-97
Portland 605,201 571,588 596,903 1.4
Seattle 154,628 135,969 118,141 23.5
Salt Lake City 418,138 414,542 263,120 37.0
Denver 1,311,463 1,312,770 1,308,710 0.2
*31 Table 4 suggests that Oregon's policies have been only moderately successful at preserving rural land. The Portland region has slowed farmland loss, but one of the three comparable metro areas (Denver) has been slightly more successful in doing so. On the other hand, the Portland region actually gained agricultural land between 1992 and 1997, while Denver's progress in farmland preservation was reversed. Thus, it is too early to tell whether the UGB is increasing the supply of farmland and/or rural land.
The inconclusive results of the UGB were hardly unforeseeable, because at the time of its creation, the Portland UGB included some vacant buildable land, thereby allowing developers to urbanize rural land without running afoul of the UGB. [FN211] This "land cushion" constituted 37% of the buildable land within the UGB, [FN212] giving developers ample opportunity to urbanize inside the UGB land. Presumably, a stricter UGB would have reduced development of rural land, but might have created negative side effects (such as land shortages leading to higher housing prices). [FN213] Moreover, thousands of "exurban" houses have been built outside the UGB, either on lands zoned for commercial farm and forest production or on "exception lands" (that is, land outside the UGB but identified as "committed" to urban land uses). [FN214] Thus, the UGB was lenient enough to allow development of a significant amount of rural land.
In sum, the Oregon UGB has been at best moderately successful in slowing the urbanization of rural areas: metropolitan Portland lost farmland in the 1980s and early 1990s, but may have slowed or halted those losses in recent years.
C. Sprawl, Automobile Dependency, and Portland
Thanks in part to suburban sprawl, [FN215] Americans drive more than ever. Between 1983 and 1995, the length of the average American commute increased by 37%. [FN216] As a result, vehicle miles traveled (VMT) grew four times faster than the driving-age population, as Americans spread farther and farther apart. [FN217] Also, because more Americans than ever live far from bus stops and other transit *32 facilities, [FN218] public transit usage in the U.S. is far lower than in other affluent countries, [FN219] and many jobs are simply inaccessible to the carless urban poor. [FN220] By making automobile ownership virtually mandatory for a normal life, [FN221] sprawl limits consumers' transportation choices [FN222] while arguably increasing traffic congestion [FN223] and air pollution. [FN224]
One measure of automobile dependency is the growth or decline of public transit use. Table 5 addresses this issue.
*33 TABLE 5: ANNUAL TRANSIT BOARDINGS, 1980-96 (IN MILLIONS) [FN225]
1980 boardings 1996 boardings % increase
Denver 46.9 69.8 48.7
Portland 50.3 79.8 58.7
Salt Lake City 19.1 23.8 24.9
Seattle 98.2 115.5 17.6
All four metropolitan areas experienced increased transit ridership at a time when the majority of metropolitan transit systems actually lost riders [FN226]--but transit ridership grew faster in Portland than in Denver, Seattle, or Salt Lake City. Because cities tend to be more transit-accessible than suburbs, [FN227] UGB-related growth in city population [FN228] may have contributed to Portland's surge in transit ridership.
It does not appear, however, that automobile use has decreased in Portland. Table 6 compares vehicle miles traveled ("VMT") in Portland to VMT in other urbanized areas.
*34 TABLE 6: DAILY VMT PER CAPITA IN URBANIZED AREAS, 1989-99 [FN229]
1989 1999 % increase
Denver 17.3 23.2 34.1
Portland 18.9 20.8 10.0
Salt Lake City 18.5 22.6 22.1
Seattle 24.2 25.9 7.0
Table 6 shows that Portlanders drive less than other westerners, and that their per capita VMT has grown more slowly than that of Salt Lake City and Denver commuters. Table 6 also shows, however, that Portland's VMT has grown more quickly than that of Seattle commuters. Although city residents drive less than suburbanites, [FN230] the number of city residents arguably added by the UGB (161,000 if every single city resident added between 1980 and 2000 would have lived in the suburbs otherwise) [FN231] comprised only 8.3% of the region's population [FN232]--not enough to massively reduce automobile use, especially since most residents of the city of Portland drive to work. [FN233]
In sum, Portland's transportation patterns differ only modestly from those of its western neighbors: transit use has risen, but automobile use has risen as well (albeit more slowly than in some other metropolitan areas).
*35 V. Are UGBs a Cure Worse Than the Disease?
The case against the UGB has been stated succinctly by the Almanac of American Politics, an authoritative guide [FN234] to American politics:
[Portland's] 'livable community' comes at a price. The Portland area was recently rated the fourth-least affordable place in the nation to purchase a new home. Portland's policies guarantee greater traffic congestion; even optimistic planners acknowledge that its population growth will be much larger than the number of people who can be persuaded to use mass transit. Greater population densities produce more intense pollution . . . . [FN235] In other words, the Almanac suggests that the UGB: (1) reduces housing affordability by limiting development; and (2) creates traffic congestion and pollution by increasing density. Other commentators have suggested that the UGB reduces housing quality by increasing density, and that regardless of their beneficial effects, such land use controls are an inappropriate government intrusion into property rights. [FN236] Each of these arguments will be addressed in turn.
A. Portland's So-Called Affordability Crisis
By most measures, Portland is not one of America's most expensive metropolitan areas. According to the National Association of Home Builders (NAHB), Portland's median home sale price in 2000 was $168,000, far less than *36 San Francisco ($505,000), New York ($237,000), Seattle ($220,000), Denver ($183,000) and literally dozens of other metropolitan areas. [FN237]
It could be argued that even though Portland is not incredibly expensive, the UGB has nevertheless triggered an affordability crisis by creating a rapid increase in home prices. [FN238] Table 7 tests this theory by comparing Portland's housing price increases to those of comparable metropolitan areas.
TABLE 7: HOUSING PRICE INCREASES IN THE URBAN WEST, 1991-2000 [FN239]
1991 median price (in 2000 median price (in % increase
thousands of dollars) thousands of dollars)
(1st quarter) (3rd quarter)
Denver 84 183 117
Portland 80 168 110
Salt Lake 76 151 98
City
Seattle 130 220 69
Table 7 shows that home prices in Portland have increased more slowly than home prices in Denver, and that Portland-area housing price appreciation has been only slightly more rapid than price appreciation in other western regions.
It could be argued that despite Portland's perfectly ordinary rates of housing appreciation, Portland's houses are nevertheless relatively more expensive than houses in other regions because wages are lower in Portland. Table 8 sets forth the ratio between median income and median home price for Portland and comparable metropolitan areas.
*37 TABLE 8: MEDIAN INCOMES AND MEDIAN HOME PRICES, 2000 (in thousands of
dollars) [FN240]
Median income Median home price Income as % of home price
Denver 62.1 183 33.9
Portland 53.7 168 31.9
Salt Lake City 53.4 151 35.3
Seattle 65.8 220 29.9
Table 8 shows that even after Portland's lower wages are accounted for, Portland is more affordable than Seattle and only slightly less affordable than Denver or Salt Lake City. In all four areas, the median home price exceeds median income by about a three to one margin.
If Portland's housing appreciation rate is lower than Denver's and its overall prices are lower than Seattle's, then why is it so often argued that Portland is unaffordable? UGB critics rely on the NAHB Housing Opportunity Index (HOI) [FN241] which, in recent years, has consistently labeled Portland as one of the nation's most expensive metropolitan areas. [FN242] The HOI, however, yields preposterous results. For example, metropolitan New York City's median income is only $2500 higher than Portland's ($56,200 as opposed to Portland's $53,700), and New York City's median home price is $69,000 higher ($237,000 as opposed to Portland's $168,000). Yet, the NAHB considers New York City more affordable than Portland! [FN243]
Why does the NAHB claim that New York is more affordable than Portland? The NAHB's affordability index is the percentage of homes allegedly "affordable" to a hypothetical household earning the median regional income. [FN244] Thus, NAHB reasons, New York is more affordable than Portland because New *38 York households earning the median income can allegedly afford 29.3% of homes, while Portland households earning Portland's median income can only afford 27.6% of homes. [FN245] This means that the prices of homes above the NAHB-designated "affordability level" (as well as the price and desirability of homes below this "affordability level") are irrelevant to the NAHB index.
Portland's apparent unaffordability may be related to the fact that the gap between Portland's most expensive and least expensive housing is smaller than similar gaps in New York or other cities. Thus, a homeowner who cannot afford the average house also cannot afford many of the area's cheaper-than-average houses. [FN246] For example, only about one-third of Portland-area homeowners have unusually high (50% over the regional median) or unusually low (50% below the regional median) housing costs. [FN247] In contrast, a majority of Los Angeles-area homeowners [FN248] and over 40% of New York-area homeowners have unusual housing costs. [FN249]
It follows that under the NAHB's criteria, the metropolitan areas with the largest gap between the most expensive and least expensive homes may seem *39 quite affordable, because such areas will have a large number of low-cost homes affordable to median-income homeowners. It further follows that the most dangerous cities are by definition relatively affordable under the NAHB's methodology, because the very cheap houses in expensive cities may be located in run-down, dangerous areas. [FN250]
The following hypothetical illustrates how the NAHB test may lead to absurd results. Imagine two cities, City A and City B. In both, the median housing price is $150,000. A household earning the median income can afford a $50,000 to $100,000 house, and any house costing under $50,000 is likely to be in a crime-ridden neighborhood and thus undesirable to all but the most adventurous (or desperate) homebuyers. Suppose that City A is a socially homogeneous city with very low crime rates, and that as a result the price gap between the best and the worst neighborhood is lower than in most big cities (as appears to be the case in Portland). [FN251] City A's housing prices are thus distributed as follows:
Over $200,000: 5%
$150-200,000: 40%
$100-150,000: 25%
$50-100,000: 20%
Under $50,000: 5%
Although City A has very few "high-end" houses, the remaining houses are so clustered in the $100,000 to $200,000 range that a median income household can afford only 25% of them. Thus, City A has an affordability index of 25 under the NAHB methodology. However, most of the low-end houses are not in crime-infested neighborhoods, so 20% of houses are both affordable and relatively desirable (i.e., not in a dangerous neighborhood). Thus, the real affordability index--that is, the percent of houses that are both affordable to middle-class homebuyers and located in neighborhoods that middle-class homebuyers would actually be willing to consider--is 20, almost as high as the NAHB affordability rating.
*40 Portland closely resembles City A. Portland's murder rate (5.3 murders per 100,000 residents in 1998) is lower than that of most big cities. [FN252] The Portland metropolitan area is significantly safer than most cities. [FN253] Portland's urban poverty rate is lower than the national poverty rate [FN254] and (as in City A) the price gap between Portland's most and least expensive houses is lower than in other metropolitan areas. [FN255]
By comparison, City B is a two-class city. It has a number of relatively safe elite neighborhoods and/or suburbs, an equally large number of cheap, crime-infested slums, and a tiny middle class. As a result, its housing prices are distributed as follows:
Over $200,000: 40%
$150-200,000: 10%
$100-150,000: 5%
$50-100,000: 5%
Under $50,000: 40%
Because 45% of City B houses are theoretically affordable to a median-income household (i.e., one who can only afford a $100,000 house), the NAHB would give City B an affordability index of 45--nearly twice that of City A. But because of City B's crime problem, many of its housing units are simply unthinkable to the average homebuyer concerned about his or her safety. In fact, when the dangerous "under $50,000" neighborhoods are excluded, City B's affordability rating drops to 5, one-fourth that of City A.
Thus, the NAHB index yields absurd results because it systematically disfavors relatively safe, homogenous cities and metropolitan areas (like City A and Portland), and favors cities and metropolitan areas sharply divided into rich and poor, safe and unsafe (such as city B). The relative affordability of America's most dangerous regions supports this interpretation of NAHB's tabulations. The five metropolitan areas with the highest violent crime rates in 1998 all had higher *41 "affordability scores" than Portland. [FN256] Table 9 compares Portland with two of these "City B" -type areas.
TABLE 9: HOUSING COSTS IN NEW YORK, PORTLAND, AND LOS ANGELES (4th quarter
1998) [FN257]
Median income (in Median home price (in NAHB
thousands) thousands) score affordability
Portland 49.6 155 38.7
Los Angeles 49.8 179 50
New York 49.8 160 56
Los Angeles and New York had median incomes virtually identical to those of Portland, and higher median home prices. Yet according to the NAHB, Los Angeles and New York (whose violent crime rates were nearly double those of Portland) [FN258] are far more affordable, presumably because they have more extremely inexpensive neighborhoods. For example, New York and Los Angeles have higher median home prices than Portland, [FN259] but also have more homes that cost under $200 per month than Portland. [FN260] Thus, the NAHB index appears to *42 reward cities with a profusion of dangerous, low-cost slums. The NAHB index of affordability may be flawed in other respects. The index fails to consider Oregon's lack of a sales tax (thus underestimating buyers' income). It also fails to account for the fact that second- and third-time buyers benefit from the rising resale value of their first homes, because they can make larger down payments on later homes. [FN261]
Even though Portland's housing prices have exploded in recent years, its overall cost of living has not. Between 1995 and 1999, Portland's consumer price index for all items increased by 12.7%--no more than in Denver (12.7%), and less than in Seattle (13.5%). [FN262] Thus, there is no reason to believe that Portland consumers are being impoverished by high housing prices.
In sum, Portland's property values have indeed gone up over the past decade, but its price increases and home values are in line with those of comparable metropolitan areas. The UGB may have had a marginal effect on Portland's housing costs, but it has hardly been the disaster that some have claimed. [FN263]
B. Portland's So-Called Congestion Crisis
Environmentalists argue that suburban sprawl increases traffic congestion, by increasing the number of cars on the road and the distances that drivers travel. [FN264] UGB critics argue, on the other hand, that anti-sprawl policies such as UGBs actually increase congestion by forcing a constant or growing number of *43 cars onto the same amount of land. [FN265] It follows, according to UGB critics, that Portland's UGB, by increasing density, has increased traffic congestion. [FN266]
The Texas Transportation Institute (TTI), a state research agency affiliated with Texas A & M University, [FN267] regularly conducts "urban mobility studies" that evaluate the extent of congestion in America's largest metropolitan areas. Its 2001 study [FN268] contains numerous measures of congestion trends, including increases in annual delay per person. [FN269] TTI ranked Portland number eleven (out of sixty-eight metropolitan areas ranked) in the 1982-99 increase in delay caused by traffic congestion--that is, TTI found that congestion increased at a more rapid rate in Portland than in all but ten metropolitan areas. But two of those ten were comparable areas which lacked UGBs for some or all of the relevant period--Seattle (no. 5) and Denver (no. 8). [FN270] In other words, congestion increased dramatically in Portland, but no more dramatically than in other areas without UGBs.
Moreover, there is no reason to believe that Portland's congestion increase was the result of increased density. According to TTI, Portland's population density actually decreased slightly from 1982 to 1999, from 3230 people per square mile to 3040. [FN271]
*44 Seattle's brief experience with UGBs supports the view that UGBs do not cause congestion. Seattle instituted UGBs in 1994 [FN272]--so if UGBs cause congestion, congestion in Seattle should have risen more dramatically in recent years than in earlier years. Instead, Seattle's delay per person exploded from nineteen hours per driver in 1982, to fifty-five hours in 1992, and then decreased to fifty-three hours in 1999. [FN273] Thus, the Seattle UGB has not increased traffic congestion.
It could be argued that the UGB has failed to dramatically increase congestion only because Portland has not yet become particularly dense, and if Portland does not dilute or eliminate the UGB in the future, density may increase and thus cause congestion. [FN274] The link between density and congestion, however, is weak. Table 10 lists the urbanized areas with more congestion than Portland, and compares their density to Portland.
TABLE 10: DENSITY AND DELAY PER PERSON, 1999 [FN275]
Delay per person (hours) Population per square mile
Los Angeles 56 5575
Atlanta 53 1585
Seattle 53 2295
Houston 50 1830
Dallas 46 1455
Washington 46 3420
Austin 45 1585
Denver 45 2240
St. Louis 44 1775
Orlando 42 1780
Miami 42 3785
Boston 42 2605
San Jose 42 4340
Nashville 42 1085
San Francisco 42 3205
San Berardino 38 2600
Minneapolis 38 1900
San Diego 37 3575
Indianapolis 37 2050
Louisville 37 2060
Tampa 35 1530
Portland 34 3040
*45 Table 10 shows no link between density and congestion: of twenty-one areas with more congestion than Portland, fifteen (including Seattle and Denver) are less densely populated. Thus, density either (a) does not cause traffic congestion or (b) increases traffic congestion so minutely that the congestion-causing effect of density is overwhelmed by other relevant factors.
In sum, traffic congestion in metropolitan Portland has grown about as fast as in other western metropolitan areas without UGBs--a fact that suggests UGBs neither cause nor cure traffic congestion to a significant extent. And if the UGB increases regionwide density in the future, traffic congestion will not necessarily increase as a result, because there is no clear correlation between regionwide population density and regionwide traffic congestion.
C. Density, UGBs, and Air Quality
Air quality is widely acknowledged to be linked in some way to density, sprawl, and UGBs, but there is no consensus as to how these problems are linked to UGBS. Environmentalists assert that sprawl creates pollution by increasing automobile use, [FN276] and that Oregon's land use policies might therefore limit pollution by limiting sprawl. [FN277] UGB opponents argue, however, that UGB-induced increases in density will lead to increases in pollution. One commentator reasons that "[a]s density rises, so does congestion." [FN278] "[G]iven the fact that air pollution rises as urban automobile speeds decline and as 'stop and start' operation increases, the result is greater air pollution." [FN279]
As noted above, [FN280] Portland's traffic congestion is no worse than that of less densely populated metropolitan areas, including otherwise comparable regions such as Seattle and Denver. [FN281] If Portland's policies are not certain to *46 cause increased traffic congestion, those policies obviously will not cause congestion-related pollution.
Moreover, air pollution in Portland is comparable to that of nearby metropolitan areas. Table 11 lists trends in carbon monoxide emissions for Portland and comparable western metropolitan areas.
TABLE 11: CARBON MONOXIDE EMISSIONS IN PARTS PER MILLION, 1989-98 [FN282]
1989 1998 % REDUCTION
Denver 7.8 3.9 50.0
Portland 8.2 5.1 37.8
Salt Lake City 7.7 4.9 36.3
Seattle 8.5 4.5 47.0
Table 11 reveals that all four metropolitan areas reduced carbon monoxide pollution by roughly similar amounts, and that Portland was slightly less successful in reducing carbon monoxide than Seattle and Denver but slightly more successful than Salt Lake City. Thus, it appears that Oregon's land use and transportation policies, to the extent those policies differ from those of other western states, have not significantly affected air quality. Portland's rather ordinary air quality performance should not be surprising, in view of Portland's apparent failure to reduce automobile use during the 1990s. [FN283]
It could be argued that even if the UGB has not yet increased pollution, it will do so in the future by encouraging higher density (which will in turn allegedly increase pollution by increasing traffic congestion). [FN284] To be sure, a scintilla of evidence supports this theory: Los Angeles has a higher population density than any other urbanized area [FN285] and has more carbon monoxide pollution *47 than most other large urbanized areas. [FN286] But Los Angeles is unusual in that it combines high density with high automobile dependency, [FN287] perhaps because its central core is far less dense than that of less automobile-oriented cities: the central city of Los Angeles had only 7426 people per square mile in 1990. [FN288] By contrast, each of the eight large metropolitan areas included in the study where at least 10% of commuters used public transit in 1990 [FN289] included or was near a central city with 9500 persons or more per square mile. [FN290]
Metropolitan areas with high levels of public transit use tend to have relatively clean air. Table 12 lists the five metropolitan areas with the highest public transit ridership, and their pollutant levels, and compares them to Portland and other comparable metropolitan areas.
TABLE 12: PUBLIC TRANSIT RIDERSHIP AND CARBON MONOXIDE, 1998
% of commuters 1998 using Carbon monoxide emissions
public transit [FN291] (parts/million) [FN292]
New York City 47.3 3.7
San Francisco 19.5 3.5
Chicago 17.1 3.4
Boston 14.2 2.9
Washington 13.7 3.3
Denver 4.4 3.9
Portland 6.0 5.1
Salt Lake City 3.0 4.9
Seattle 7.4 4.5
*48 The most public transit-friendly metropolitan areas are all less polluted than Portland, Denver, Seattle, or Salt Lake City. Therefore, if Portland can ever densify its central core and reduce automobile dependency to the extent that those metropolitan areas have done, air pollution is unlikely to increase.
D. Do UGBs Reduce Housing Quality?
UGB opponents argue that even if the Portland UGB does not increase housing costs, traffic congestion, or air pollution, it has reduced the quality of the housing stock by reducing the amount of buildable land, thus reducing home and lot sizes. For example, one anti-UGB article is titled: Squeezed Out: No Swingset, No Sandbox, No Space Left for the American Dream. [FN293] The article asserts that because few new homes are built on half-acre lots, "[t]he American Dream is dead in Portland." [FN294] Similarly, one anti-UGB activist asserts that if the UGB is not expanded or eliminated, "[p]eople will be crowded together, living on small lots." [FN295]
But in fact, the specter of a hyper-dense Portland is imaginary; Portlanders' homes and lots are as large as ever. Between 1986 and 1995, the median-sized owner-occupied home in Portland grew from 1674 square feet [FN296] to 1764 square feet. [FN297] The median lot size grew slightly, from .22 acre [FN298] to .23 acre. [FN299] The number of houses sitting on over half an acre of land grew from 74,000 [FN300] to 98,700. [FN301] And during the 1990s, "nearly 800,000 of the 1.4 million people living in the Portland area . . . saw no change in the density of their neighborhoods." [FN302] As a result, "only 5% of the region's residents live in areas with population density greater than 10,000 people per square mile." [FN303] In *49 Portland, as in other areas, more people are building bigger houses on more land. [FN304]
Critics argue that even if Portland has no density crisis today, a refusal to expand the UGB could eventually force Portlanders to live in an overcrowded, congested environment. [FN305] But in fact, Portland has plenty of room to grow. The city of Portland has just over 529,000 people living within its 124.7 square miles. [FN306] If the city's population tripled, it would have about 1.587 million people living within those 124.7 square miles, thus creating a density of 12,726 people per square mile, which is still a lower density than that of San Francisco. [FN307] In other words, if a million people moved to Portland in the next few decades, every single one of them could, in theory, be placed in the city of Portland without either making Portland inordinately dense or increasing suburban densities by one iota.
E. The Libertarian Argument Against UGBs
Libertarians argue that regardless of its utilitarian benefits, Portland's UGB should not be imitated because restricting landowners' use of their land violates their property rights. [FN308] This argument makes sense in principle: in a society that truly valued minimal government, any form of land use regulation would obviously be inappropriate. [FN309]
However, the United States is not such a society: even in colonial times, states and cities enacted planning statutes similar to UGBs. For example, the *50 Massachusetts Bay Colony "prohibited dwellings more than one-half mile from town meeting houses without court permission." [FN310] Today, as UGB critic Clint Bolick admits, "minimum lot sizes and restrictions on multiple uses, often popular tactics among suburban governments to keep their communities pristine and exclusive, definitely contribute to 'sprawling' suburbs." [FN311] A typical zoning ordinance has separate zones for: single-family large lot, single-family medium, single-family standard, multi-family low density, multi-family medium density, multi-family high density, general office, neighborhood commercial, community commercial, service commercial, central business district, limited industrial, and heavy industrial. [FN312] These regulations contribute to sprawl by separating residential and commercial uses, thereby making it difficult for people to go from their residences to shops and workplaces without driving. [FN313] Municipalities also reduce housing supplies and force Americans to drive by mandating minimum lot sizes and house sizes within zones. [FN314] These regulations were frequently enacted for the purpose of increasing housing prices. [FN315] They also inadvertently reduce transit use, because as residences are spread farther apart, fewer people can conveniently walk to bus and train stops. [FN316] If pro-sprawl land use restrictions such as separation of land uses, minimum house sizes, and minimum lot sizes do not unduly infringe landowners' rights, neither do UGBs.
Similarly, UGB critics often support pro-sprawl government transportation policies. For decades, government has accelerated sprawl by building new roads into suburbia, thus encouraging development to shift from cities to suburbs. [FN317] But many critics of UGBs and other anti-sprawl measures are less than *51 critical of road spending. For example: (1) The Independence Institute, which purports to address "public policy issues from a free-market, pro-freedom perspective" [FN318] published an anti-UGB paper asserting that "[f]or traffic congestion to be mitigated . . . [r]oadway expansions will be necessary." [FN319] (2) Steven Hayward of the Heritage Foundation complained in a 1998 article that "Portland's planners . . . are substituting political decisions for marketplace decisions," [FN320] yet claimed in the same article that Portland will suffer increased traffic congestion because "the region deliberately avoids road-building." [FN321] (3) Grant Gulibon, a policy analyst for Pennsylvania's conservative Commonwealth Foundation, complains that the UGB "created an artificial scarcity of land . . . . Whenever a commodity becomes scarce the price goes up." [FN322] Yet Gulibon asserts that "[t]he reason [new roads] fill up right away is because you didn't build enough in the first place." [FN323]
UGBs undeniably increase government's voice in land use decisions--but so do a variety of well-established government policies, some of which are supported even by prominent UGB critics. Thus, UGBs are not uniquely intrusive.
VI. Conclusion
As explained above, UGBs are probably constitutional under Supreme Court Takings Clause precedent for two reasons. First, a UGB will rarely, if ever, cause a complete deprivation of economically beneficial use, and is thus unlikely to be a compensable "total taking" under Lucas and Palazzolo. Second, UGBs *52 will generally not be compensable "partial takings" under the Penn Central balancing test, because they are typically justified by a legitimate state purpose (that of preventing the urbanization of rural areas) and will, if prudently drafted, rarely interfere with landowners' investment-backed expectations.
The question of whether UGBs are wise public policy is a more difficult one. The Oregon UGB has apparently had one major benefit: enhancing consumer choice, by making the city of Portland a viable lifestyle choice for middle-class Portlanders, instead of merely a holding pen for people too poor to move to suburbia. And by making it easier for Portlanders to live in the city, the UGB has arguably made it somewhat easier for them to use public transit instead of being enslaved by their cars.
In other ways, Portland has continued to evolve in the same directions as nearby metropolitan areas without UGBs. On the negative side, this means that the UGB has not had an enormous effect upon automobile use, traffic congestion, or urban consumption of farmland. Although Portland's air is cleaner than it was a decade ago, the same is true for other western cities without UGBs. On the positive side, the UGB has not led to the horrors imagined by UGB critics. Portland's real estate is not significantly more expensive than that of comparable western cities, its traffic congestion is no more annoying, and its air is no more polluted. In sum, UGB supporters and opponents alike may have overestimated both the positive and negative effects of UGBs.
(footnotes deleted due to length of article)
Posted by lewyn
at 3:05 PM EST
Updated: Monday, 17 January 2005 5:19 PM EST
