September 13, 2000 | Backgrounder on Smart Growth
The debate over urban sprawl is a healthy sign of Americans' enduring desire for self-renewal. As such, it is ultimately a national discussion about how to live "the good life" in an affluent democracy. At its core, the debate over "sprawl"--which often masquerades as a dispute over technical concerns about traffic congestion, mass transit use, population density, the availability of public works, and so forth--represents a clash of values and visions over federal and state government involvement in largely local issues. Even seemingly narrow policy discussions cannot avoid tacitly endorsing one side or the other in this controversy.
The debate promises to get even livelier with the ascension of Maryland Governor Parris Glendening to the chairmanship of the National Governors' Association (NGA), which recently released a report on urban sprawl titled "Growing Pains: Quality of Life in the New Economy."1 Governor Glendening, who has been credited with coining the term "smart growth," advocates growth control policies that limit individual and community choices in favor of restrictive government directives, a fact that promises to provide intense debate on sprawl issues throughout his term as chairman.
The nation's governors have been at the forefront of innovative policy ideas and reforms on such sensitive issues as welfare, education, crime, and the environment, and it is natural to expect them to provide equally creative thinking on the issue of urban sprawl. To do so, however, they will have to overcome the rigid biases of both the NGA's new chairman and its professional staff that issued the "Growing Pains" report. The remarkable bias in this report in favor of restrictive growth control policies is highly inconsistent with the orientation of a nonpartisan organization that seeks cooperation rather than confrontation among its members.
The NGA acknowledges that governors are "choosing from a wealth of ideas and experiments underway in the United States" to address sprawl,2 yet "Growing Pains" fails to offer fresh thinking or innovative solutions on this issue. Instead, it relies on many of the familiar clichés, half truths, misperceptions, and falsehoods promulgated by the so-called smart growth movement. Though its criticism of current social and community preferences is muted, the report nonetheless calls for "developing public consensus for social and cultural changes"--in other words, for rejecting current suburban preferences.3
Governors and policymakers who seek critical assessments of the problems facing America's communities and the best solutions available should note that, as a piece of policy rhetoric, "Growing Pains" regrettably gives the mistaken impression that the problems of growth can be addressed adequately without making hard choices or difficult tradeoffs.
Despite "key findings" that "local governments lack the resources...and the federal government is too far removed from the diverse needs of regions and states to effectively influence local growth policies,"4 the recommendations in the "Growing Pains" report fall short of offering policy guidance that will effectively address concerns over sprawl. The reason: The report suffers from five serious flaws.
It includes facts, case studies, and anecdotes that are inconsistently sourced or not sourced at all;5 many of its key points have been derived from the rhetoric of the smart growth advocacy groups and are included without question;6 and little original research or critical analysis is offered to bolster these claims.
Many of its directives are contradictory. One section, for example, calls for more emphasis on regional planning while another suggests greater emphasis on local "community" planning; and while the report calls for increased housing choices in one part, elsewhere it condemns large lot development and clearly implies that this one housing choice should be constricted.
It calls for more "intelligent" and "coordinated" land use planning, yet it offers no discussion of the cognitive difficulties facing public-sector attempts at coordination and resource allocation on the scale it contemplates for addressing regional or metropolitan land use.
For several years, the debate over urban sprawl has been dominated by half truths, poorly understood facts, and outright falsehoods. Thus, the conventional wisdom about the causes of and solutions to sprawl that has emerged is based on stubborn misconceptions and superficial appreciation of urban phenomena. Regrettably, the "Growing Pains" report offers little to break the debate out of this rut. To the contrary, it embraces nearly every misguided tenet of the conventional wisdom.
While "Growing Pains" eschews the frothier language that often appears in "smart growth" literature--such as the "paving of America" or the metaphor of urban sprawl as a "cancer" on the land7 --it nonetheless conveys the impression that urban sprawl is an American crisis. It puts forth a blizzard of facts, anecdotes, and case studies to support three broad conclusions, which it calls the "Laws of Growth":8
NGA Law of Growth #2: "As distance from urban cores increases and population density decreases, the rate of growth increases for population, land consumption, residential dwellings, and private vehicles."
As descriptions of the phenomenon of suburban growth, the first two statements are accurate and in fact similar; yet they raise a fundamental question: What is the scale at which this growth in population is occurring, both absolutely and relative to other factors such as total land supply, environmental impacts, and tradeoffs between different forms of development? The third law, to the extent that deteriorating conditions in central cities (high crime rates, poorly performing schools, failing public services, and so forth) have propelled the middle class to flee to the suburbs, is an accurate description of trends in the 1960s, 1970s, and 1980s. But the question remains: To what extent can urban flight be said to cause urban decline today?
Population Growth and Suburbanization. "Growing Pains" cites a litany of alarming statistics about land consumption relative to population growth. For the nation as a whole, it cites the U.S. Department of Agriculture's 1997 National Resources Inventory (NRI), which estimates that the rate of urbanization in the United States doubled from 1992 to 1997. The fact that the report uses the NRI as a valid source of these statistics is curious, since the U.S. Department of Agriculture itself withdrew its NRI report because of admitted errors that may have led to substantial miscalculations. The author of "Growing Pains" acknowledges this withdrawal, but only in a footnote.
More significant are the report's cited facts about sprawl in specific cities. For example, it claims that "[b]etween 1970 and 1990, the population of Chicago increased by 4 percent, but its land area expanded by 46 percent." For Philadelphia, "the land area increased by 32 percent from 1970 to 1990, for a population increase of less than 3 percent, consuming 125,000 acres of open space."9 However, sources for these specific statistics are not included.10 Without such citation of sources, it is impossible to review the methodology used to derive the statistics in order to judge their accuracy.
The statistics given for Chicago and Philadelphia do not match the U.S. Census Bureau's statistics for land area and population growth. For Chicago, the Census Bureau finds that between 1970 and 1990, "urbanized land area" grew by slightly more than 24 percent (not 46 percent, as "Growing Pains" reports) while population grew by slightly over 1 percent (not 4 percent, as claimed in the report).11 Moreover, according to the Census Bureau data, Chicago's urbanized area grew less from 1970 to 1990 than it did from 1950 to 1970 (24.1 percent from 1970 to 1990 compared with 80.4 percent between 1950 and 1970), suggesting that the rapid expansion of land area is a receding problem.
For Philadelphia, the Census Bureau statistics are considerably more dramatic than those reported in "Growing Pains." Between 1970 and 1990, Philadelphia's urbanized area grew 54.8 percent while population grew by 5 percent, according to the Census Bureau, compared with the 32 percent and 3 percent, respectively, reported in "Growing Pains."
Considering changes in population and land area in percentage terms alone is a misleading and superficial measure of suburbanization, since percentages always shrink as the base number against which they are calculated expands. For example, a 10 percent increase in land area in a city that covers 100 square miles is an order of magnitude larger than a 10 percent increase in land area in a city that covers only 10 square miles.12
Some way of putting land area changes in proportion is needed. For example, the 24.1 percent expansion of Chicago's land area from 1970 to 1990 represents 0.6 percent of the total land area of Illinois. The comparable figure for Philadelphia is 0.9 percent of Pennsylvania's land area. The growth in Chicago's land area from 1950 to 1990 (877 square miles, according to the Census Bureau) represents 1.6 percent of the total land area in Illinois.13 Accommodating 40 years of population and economic growth for one of the largest and most dynamic metropolitan areas in the nation on 1.6 percent of the state's land area seems neither profligate nor a crisis.
Moreover, while the overall population of the Chicago area grew by only 1.1 percent from 1970 to 1990, its suburban population grew by 19.7 percent. It is misleading to compare the population growth for an entire metro region with the growth in land area in the suburbs because central city populations have been shrinking worldwide. Considering such facts makes the 24.1 percent increase in Chicago's land area over this period look less disproportionate. In fact, the population density of its suburban areas is higher today than in 1950, according to Census Bureau figures (2,600 people per square mile in 1950 compared with 2,954 people per square mile in 1990).
All of these cities experienced a decline in population density of between 20 percent and 30 percent, compared with the Chicago area's decline of 18.5 percent.14 In other words, European cities, which the smart growth movement frequently praises for their public transit capacities and "compact" development patterns, are sprawling faster than are America's cities.
The common experience of declining population in both American and European cities prompts the question of why such trends are occurring. The first two NGA Laws of Growth imply that the low-density, decentralized suburban form of development that characterized the last generation occurred willy-nilly, with little or no rational relation to the needs of the commercial marketplace or the demands of consumers and homebuyers.15 As an example, "Growing Pains" cites Rochester, New York, where 2 million square feet of big-box retail stores were built in the suburbs even though there were 2 million square feet of empty retail space available close to the urban center. This implies that all retail square feet are equal in functionality and convenience to consumers, or that the outcome is something that the public sector is better suited to decide for retailers and consumers alike.16
The preference for suburban lifestyle and automobility is a function of widening affluence, the result of the rapid expansion of the upper middle class over the past 25 years. Transportation analyst Alan Pisarski concludes that the majority of the increases in automobile ownership and vehicle miles traveled over the past 20 years occurred among working women and minorities, and that this has little to do with the spatial design of American cities and suburbs.17 It is ironic that the smart growth movement deplores an increase in the democratization of auto ownership and autonomy for two groups that liberal opinion typically considers victims.
Urban Flight. The author of "Growing Pains" points out that central cities must be made more "livable" if they are to become competitive with their suburbs, and he rightly acknowledges that without central city revitalization, "Actions that only attempt to limit suburban growth are not likely to be effective and may have negative impacts."18
Yet there is a troubling ambiguity in this third so-called Law of Growth--namely, the implication that suburbanization causes the deterioration of urban cores.19 This is not necessarily a normal occurrence.
First, there are numerous examples of rapidly suburbanizing metro areas where the urban core is also thriving (consider Sacramento, Seattle, San Francisco, Denver, and Houston, among others).20
Second, where the phenomenon of suburban growth and urban deterioration has occurred, it is not a chicken-and-egg dilemma that allows the observer to discern causation. It is manifestly the case that bad government in the urban core accelerated the exodus of the rising urban middle class.21
"Growing Pains" contains no acknowledgment of the legacy of poor government policies that plagued major American cities in the postwar decades. There has been a hopeful turnaround in the quality of urban government over the past decade, visible most dramatically in falling crime rates,22 but even this improvement says more about public policy and urban decline than population growth or suburbanization.
Disputes over the facts of urban phenomena and how they should be understood will go on for a long time to come.23 Yet, the most significant defect of "Growing Pains" lies not in its purported facts or their interpretation, but in its general prescription for more intensive planning.
"Growing Pains" calls for "more intelligently and sensitively coordinating, steering, and shaping growth to better serve immediate and longer-term needs of states."24 Aside from recommending "channeling more growth into areas already developed, principally urban centers and older suburbs"--in other words, achieving growth chiefly through infill--it fails to explain how "more intelligent" and "coordinated" planning should work in practice.
Yet the report provides scant evidence that population growth and consumer demands can be accommodated through infill. As James Frank, professor of urban and regional planning at Florida State University, notes:
In most cities, not much surplus capacity is lying idle, so the duration of the payoff from infill development may be quite short. It would be desirable to have some sense of how much available unused capacity actually exists in typical communities....
[An] area of desirable research should be aimed at the quantitative measurement of the amount of available capacity that exists in typical communities to evaluate the magnitude of opportunities for infill.25
"Growing Pains" offers only one statistic: "One hundred eighteen cities said they could add more than 5.8 million people without adding appreciably to their existing infrastructure."26 But no source is given for this figure, so it is impossible to know whether the population densities assumed for this projection are realistic or whether "infrastructure" in this case comprises land area and roads. The population of the United States is growing by 2.5 million people a year; thus, these 118 cities could accommodate less than three years of growth. Infill does not appear to be a serious long-term policy solution for accommodating population growth.
The few case studies of rapidly growing cities are not encouraging. There is some information available for Phoenix, Arizona, generated by its Infill Housing Program in 1997.27 This program offers a sliding scale of fiscal incentives for developers to build within a 162-square-mile zone inside the city limits. The program has identified 1,365 parcels of vacant land larger than 0.7 acre within this zone, which sounds like a substantial supply of land; but if it were all built out at the higher range of residential densities, it would accommodate only about one year's growth in the city's population.
A similar estimate exists for Albuquerque, New Mexico. The Albuquerque Transportation Evaluation Study identified 34,000 acres of empty land in the city limits but failed to look at how this land was classified or how much of it had existing infrastructure. A more precise picture is offered by the Albuquerque Geographic Information System (AGIS). The most recent AGIS map identifies 20,624 acres of open land within Albuquerque's urban service area that is zoned for single-family housing and 2,609 acres of land zoned for multifamily housing. However, the map designates 11,822 acres of vacant land zoned for single-family housing within the existing water service area and only 1,576 acres of land zoned for multifamily housing in this area. Though this amount of land sounds substantial, if built out at the higher range of residential densities, it would accommodate less than 10 years' growth in the city's population. Infill is not an adequate long-term growth management policy; it promises more than it can deliver.
The plethora of adjectives in the third Law of Growth, such as more "intelligently and sensitively coordinating, steering, and shaping growth," slides over a more significant difficulty than just the limitations of infill. The intensive regional land use planning called for in the report sets out some large objectives that are highly detailed in their outcome, including jobs-housing balance, higher density residential development, more mass transit usage, and more open space preservation. This kind of intensive land use planning represents a quantum jump from traditional zoning. As such, it represents a major increase in public-sector resource allocation, chiefly in the amount of land and strict regulation of its uses.
It is an unremarked irony of the smart growth movement that at a time when virtually the entire world is deregulating markets for fundamental goods, such as the telecommunications, transportation, banking, and energy production and distribution sectors, America should be considering stringent new government regulations on the market for the basic commodity of land. Moreover, there is no serious consideration given in "Growing Pains" or elsewhere to how planners can overcome the chronic failings of other kinds of public-sector resource control and allocation--failings rooted in the inability of planning processes to assimilate sufficient information in order to make correct local decisions.
This problem is compounded as the time horizon lengthens, which is why long-term "comprehensive planning"--as called for by the smart growth movement--is inherently destined to produce perverse results. (Consider, for example, the failure of repeated attempts to regulate cable television; the number of factors involved in long-term comprehensive land use regulation dwarfs those of cable TV.)
It is doubtful, to say the least, that the public sector can produce more efficient (however defined) land use without introducing a host of unwanted side effects (chiefly higher housing prices) and unintended consequences (chiefly worse traffic congestion).28 Nobel laureate Friedrich Hayek wrote in 1960 that
We must not overlook the fact that the market has, on the whole, guided the evolution of cities more successfully, though imperfectly, than is commonly realized and that most proposals to improve upon this, not by making it work better, but by superimposing a system of central direction, show little awareness of what such a system would have to accomplish, even to equal the market in effectiveness.29
Professors Harry Richardson and Peter Gordon of the University of Southern California's School of Urban and Regional Planning have commented: "Any claims for the cost of market failure [in land use] have to be balanced by an assessment of the costs of government failure.... [I]mperfect markets work better than imperfect government."30
When this problem is seriously considered, it becomes evident that the principle underlying the third Law of Growth ("more intelligently and sensitively coordinating, steering, and shaping growth") conceals deep confusion. Like much of the smart growth literature, the report states that the object of policy change is to provide more choices for people. But it downplays the implication that aggressive planning will deliberately constrict choice. For example, "Growing Pains" says that one object of smart growth is to "[c]reate a range of housing opportunities and choices."31 Yet another section of the report strongly criticizes the trend of larger houses on larger lots because such development uses too much land,32 implying that smart growth policies would make this choice available to fewer people. (The report, for example, praises a minimum housing density requirement in Maryland,33 which clearly rules out the choice of large lot development.)
In the end, "Growing Pains" does not offer policy guidance commensurate with its estimation of the dimensions of the problem of sprawl. The report singles out numerous state and local initiatives to address particular aspects of the problem (such as neighborhood revitalization programs, open space preservation ordinances, and regional planning initiatives34 ), yet the examples seem both small next to the seriousness of the problem described in the report and incommensurate with its clarion call for "articulating a statewide vision" for growth.
When the word "vision" appears in contemporary discourse, it is usually either a sign that difficult tradeoffs are being avoided or a sign of belief that changes can be made without making hard choices. The difficulty in discussing urban growth is that different people and different communities have different "visions" for their future. It is not possible for governors to "articulate a statewide vision" that can harmonize all the knowledge, values, preferences, and passions these different people and different communities have for their own future.
In practice, therefore, having a statewide vision for growth will mean having to impose one vision over competing visions. Between the unacknowledged tradeoffs, conflicting goals, and uncertainties of the smart growth vision, governors may wish to think twice before embracing the perspective offered in the NGA's "Growing Pains" report.
Steven Hayward is a former Senior Fellow at the Pacific Research Institute for Public Policy in San Francisco. In 1997 and 1998, he was a Bradley Fellow at The Heritage Foundation, where he studied urban issues.
1. Joel S. Hirschhorn, "Growing Pains: Quality of Life in the New Economy," National Governors' Association Report No. 13467, June 2000, at http://www.nga.org/Pubs/IssueBriefs/2000/GrowingPains.pdf.
2. See NGA Online, "Quality of Life Issues Move to the Top of Governors' Agenda," News Report, June 7, 2000, at http://www.nga.org/Releases/PR-05June2000Growth.asp.
3. The most egregious example is the report of the Pennsylvania 21st Century Environment Commission, which concludes: "We must find ways to prompt Pennsylvanians to explore their personal lifestyle choices--where they choose to live and work, and how much they travel each day, how much energy they consume or save, and consider changes in those patterns that will not only improve their long-term quality of their lives, but also contribute to a better quality of life for all the citizens of the Commonwealth." See Report of the Pennsylvania 21st Century Environment Commission, September 1998, p. 57, available at http://www.21stcentury.state.pa.us/2001/final.htm.
5. Hirschhorn, the author of "Growing Pains" and policy studies director for the natural resources division of the NGA Center for Best Practices, repeatedly refused to supply missing citations and sources for specific facts included in his report, despite this author's requests.
6. Advocacy groups whose claims are cited in "Growing Pains" include the Surface Transportation Policy Project, American Rivers, 10,000 Friends of Pennsylvania, Natural Resources Defense Council, Sierra Club, and American Farmland Trust. Several other unsourced claims also appear to come from advocacy groups.
11. A tabulation and comparison of U.S. Census Bureau data for the 34 metropolitan areas with a population of 1 million or more in any decennial census from 1950 to 1990 is available at http://www.demographia.com/dm-uaix.htm#intro.
12. To understand the shortcomings of measuring growth in percentage terms alone, consider this example. If we start with an area of 10 square miles that has a population of 100 and add another 10 square miles, we have increased the land area by 100 percent; moreover, adding 50 people has increased the population by 50 percent. Adding another 10 square miles to this 20-square-mile area represents a 50 percent increase, even though the land area added is identical to the first increment; and adding another 50 people to the 150 of the previous increment represents a 33 percent increase in population, though the population growth increment is identical. By the seventh iteration, the land area would expand by 14.3 percent, while the population would grow by 12.5 percent, even though absolute land area and absolute population growth in each iteration are identical to the first, when the land area expanded by 100 percent and the population by 50 percent.
14. Data derived from Jeffrey Kenworthy, Felix B. Laube, et al., An International Sourcebook on Automobile Dependence in Cities, 1960-1990 (Boulder, Colo.: University Press of Colorado, 1999); tabulation available at http://www.demographia.com/db-intlua-data.htm.
17. Alan E. Pisarski, Commuting in America II: The Second National Report on Commuting Patterns and Trends (Lansdowne, Va.: Eno Transportation Foundation, 1996); see also Alan E. Pisarski, Cars, Women, and Minorities: The Democratization of Mobility in America (Washington, D.C.: Competitive Enterprise Institute, 1999).
19. Other authors make this argument explicitly, especially David Rusk in Cities Without Suburbs (Washington, D.C.: Woodrow Wilson Center, 1993), and Inside Game, Outside Game (Washington, D.C.: Brookings Institution, 1999). For a thorough discussion of this issue, see Ronald D. Utt, "The Relationship of Cities and Suburbs," in Jane S. Shaw and Ronald D. Utt, eds., A Guide to Smart Growth: Shattering Myths, Providing Solutions (Washington, D.C.: The Heritage Foundation and Political Economy Research Center, 2000), pp. 77-92.
20. It should also be observed that many of the "inner-ring" suburbs supposedly threatened by sprawl are thriving. In the Los Angeles area, for example, consider Pasadena, Glendale, Burbank, and Long Beach.
21. See, e.g., Fred Siegel, The Future Once Happened Here: New York, D.C., L.A., and the Fate of America's Big Cities (New York: Free Press, 1997), and Steven Hayward, "Liberalism's Urban Legacy," Policy Review, January-February 1998.
22. See especially John O. Norquist, The Wealth of Cities: Revitalizing the Centers of American Life (New York: Addison-Wesley, 1998); Stephen Goldsmith, The Twenty-First Century City: Resurrecting Urban America (Washington, D.C.: Regnery, 1997); Steven Hayward, "Fixing the Dysfunctional Central City," in Randal Holcombe and Samuel Staley, eds., Land Use Planning for the 21st Century (Westport, Conn.: Greenwood Press, forthcoming).
23. A short list of the other inaccurate or incorrect "facts" and conclusions adduced in "Growing Pains" includes the nexus between low-density development and air quality, the relationship between density and traffic congestion, the theory of "induced congestion" through road-building, the annual decline in wetlands, farmland trends, the cost of public works in low-density suburbs, and vehicle-miles-traveled trends in Portland, Oregon.
28. The planners in Portland, Oregon, who are praised in "Growing Pains," admitted these tradeoffs. In a 1994 report entitled "Metro Measured," Portland's regional government observed that, "By the same token, the data suggest a public welfare tradeoff for increased density, reduced VMT [vehicle miles traveled] and higher nonauto travel. The downside of pursuing such objectives appears to be higher housing prices and reduced housing output." See Metro Measured (Portland, Ore.: Metro, 1994), p. 45.