Portland's Urban Growth Boundary:
Growth Management
or Growth Limitation?
Anthony M. Rufolo,
Portland State University
Summary
Introduction
Portland, OR is noted for its Urban Growth Boundary (UGB). The UGB is basically a line drawn around the urban area, and all urban areas in Oregon are required to have such boundaries. Development outside of the boundary is subject to substantial limitations. Portland also is unique in having a regional government, Metro; and one of Metro's duties is to manage the UGB for the Portland metropolitan area. The UGB is stated to be a tool to help manage growth in the Portland area; however, it is easy to see how a very tight boundary can also be used to try to limit growth. Metro is in the process of determining how much the boundary should be expanded to accommodate the growth expected in the Portland area, but there are many people who favor keeping the boundary where it is. Metro has tried to justify this approach by showing that it would be possible to contain future population within the existing boundaries. Unfortunately, the Metro analysis tends to ignore economics and the likely responses from markets and individuals to the constraints being developed. This is likely to lead to unintended consequences, not the least of which is diversion or limitation of some of the anticipated growth.
Metro has generated an analysis that purports to show that growth could be accommodated within the existing boundary, but many of the assumptions in the analysis are highly questionable. The paper will proceed by outlining the goals for the UGB, the process used to determine how much expansion is needed, and some data that casts doubt on the validity of some of the key assumptions used in the Metro analysis.
Goals
Oregon's land use planning is guided by nineteen goals that have been adopted as administrative rules (DLCD 1995). The UGB is specified as part of the State goal regarding Urbanization (Goal 14), to "provide for an orderly and efficient transition from rural to urban land use." This goal is further elaborated with seven specific (and sometimes conflicting) criteria for establishing and changing the boundaries, including accommodating population growth, maintaining livability, orderly and efficient provision of public services, maximum efficiency of land use, and retention of agricultural land. Other State goals include Housing and Economic Development.
The varied goals and objectives increase the potential for disagreement about how the boundary should be set and changed, but it is important to recognize that the UGB is not intended to be a single-purpose tool. Further, state law requires that the boundary contain enough land to meet housing needs for a twenty-year period.
The requirement for a twenty-year supply of land is perhaps the area most in need of some economic analysis. Since price can adjust to clear the market under almost any feasible circumstances, the question becomes what does this requirement mean? An economic interpretation would be that it would mean enough land that there would not be substantial distortions in the land market.
When government plans reduce the relative supply of land for particular purposes, they raise the price of such land. This increase in price is the clearest signal that the policy is being used to alter relative development patterns rather than to rationalize them in some way. For example, traditional zoning is justified by planners as a method to separate incompatible land uses, but if land-use separation is the only goal, there should not be any impact on the relative price of land for different purposes. Yet in most communities the price of land zoned for higher-density residential development is much higher than the price of land zoned for low-density single-family development. This serves as a sign that local governments are trying to affect the mix of development as well as to separate land uses.
Similar signs are available with the UGB. During much of its history, the price differential for land inside and outside the boundary was minimal. Within the last five years or so, the differential has become large. This price differential itself is significant evidence that the UGB is being used to limit growth rather than manage it, but a review of the analysis used by Metro to determine the amount of land needed offers other evidence that the process is using assumptions that ignore economic analysis.
The Expansion Process
The process to determine whether the UGB should be expanded and by how much is based on a series of forecasts. To somewhat over-simplify, Metro starts by forecasting population and then land use under existing policies. Metro then proposes policy changes that might affect land use and then adjusts land use for the policy changes. Land requirements are then determined. In a related analysis, housing supply and prices are also forecast.
While it is not possible to go into the process in great detail here, several of the key assumptions can be highlighted and evaluated. In effect, Metro has assumed that zoning laws could be changed to allow for higher density and that substantial amounts of land could be redeveloped at higher density to accommodate the expected growth (Metro 1996a). Further, Metro made various assumptions about the availability and price of housing that would be developed over the same time period to accommodate expected growth (Metro 1996b).
It is clear that many urban areas have higher density than the Portland Metropolitan Area. Hence, there is no doubt that it is physically possible to substantially increase the density of the region. However, analysis of the economics suggests the biases that appear in the Metro analysis. Areas with high density typically have very high prices for housing and have large numbers of people living at great distances from the city center to get lower-density housing at lower prices.
High land prices appear to be a necessary impetus to higher density. To some extent land, labor, and materials are substitutable with each other in producing housing. When any input becomes expensive relative to the others, builders will try to conserve on the expensive input and use more of the less expensive inputs. When land is expensive, less is used; and housing is typically built at higher density. Further, lot size itself is an amenity, and people will typically pay more for houses on slightly larger lots, all else being equal. With higher land prices, less land would be purchased. Hence, both from the production and consumption sides, high land prices lead to higher density. However, land is also a component that adds to housing cost. Higher land prices directly raise housing cost by the higher cost of the input and indirectly raise housing cost by forcing substitution of labor and materials for land in producing housing. In other words, it is typically more expensive per square foot to build higher-density buildings. To the extent that high land prices are offset by using little land, the final cost of the housing may not rise by as much, but the final unit is also not worth as much to the consumer.
One response to higher-priced housing is for people to search for lower-cost housing in distant locations and then commute. Hence, the high-density scenario is likely to imply that fewer people actually live within the UGB due to the higher housing cost. Some of the growth may simply be diverted to other metropolitan areas, but at least some of it will show up as people living in distant communities and commuting long distances. It is reasonable to question whether this outcome is intended as the result of "growth management." Yet we have ample evidence from other areas that people will indeed commute long distances to achieve low-cost, low-density housing.
Another outcome of high land prices is to reduce the amount of housing available to lower- income families. Typically, lower cost housing is the older part of the housing stock; but in areas with high housing prices, higher-income families also compete for the existing stock. METRO's (1996b) assumptions that the private market will provide large numbers of new single-family and apartment units at relatively low prices do not appear to be based on any analysis of the Portland housing market. The outcome will be that there will be fewer lower-cost houses among the existing stock and few if any additions in the lower price range from new construction. At the very lowest end of the income distribution, the higher housing prices will translate into increased rates of homelessness.
The Evidence
Mildner et al (1996) looked at several of these issues and found that the evidence relating to the proposed changes did not support the Metro conclusions. For example, Metro assumes that by zoning land for smaller lots, less land will be used per unit in the future. However, a comparison between the minimum lot size specified by zoning and the actual lot size for recent construction found several problems with this assumption. First, many units are built on lots well above the minimum required in zoning. Second, the ratio of actual to minimum is higher for lots with low minimum lot sizes. Third, in at least one county there was virtually no difference in the actual lot size between large and small minimum zoning requirements. In Washington County, OR lots zoned for a 5000 sq. ft. minimum had an actual average of close to 10,000 sq. ft. while those zoned for a 10,000 sq. ft. minimum average only a little over 10,000 sq ft.
There is evidence that higher land prices lead to lower lot sizes and that lot sizes in the Portland area have been decreasing over time due to rising land values. However, Metro used a prediction of little if any increase in land prices in generating its forecasts of housing cost and availability. Hence, there appears to be a basic inconsistency between the goal of affordable housing and the goal of higher density, which is simply ignored in much of Metro's analysis.
Implications
METRO's analysis may best be described as showing one set of outcomes that would make it possible to accommodate expected future growth without expanding the UGB; however, it fails to adequately address the market conditions that would make these outcomes occur and the responses of people to those market conditions. In particular, it appears that high land prices are necessary to induce high density for new construction, but such high land prices contribute to high housing prices. These high housing prices in turn serve to limit the actual growth and, perhaps more importantly, to divert some of the population that would have settled within the UGB to locate outside of it. Further, these high housing prices are likely to impose substantially more hardship on lower-income families than is acknowledged in the analysis.
Hence it appears that METRO's analysis entails a fundamental contradiction. It is possible to achieve the density that they project only with high land and housing prices, but high land and housing prices will induce more people to locate outside of the UGB and price more out of housing entirely. Thus, either the density will occur because of very high growth rates, with much of that growth diverted to rural or other urban areas, or the high prices will prevent the growth from occurring within the UGB. It does not seem possible, however, to both attain the density projected and keep growth from spilling over into lower housing-cost areas.
References
Oregon Department of Land Conservation and Development (DLCD) (1995), Oregon's Statewide Planning Goals and Guidelines, 1995 edition (Salem, OR 1995).
Metro (1996a), "Urban Growth Report: Discussion Draft," (March) (Portland, OR).
Metro (1996b), "Housing Needs Analysis: Discussion Draft," (March) (Portland, OR).
Mildner, Gerard C.S. et al (1996), "Impact of the Urban Growth Boundary on Metropolitan Housing Markets," (Center For Urban Studies, Portland State University: Portland, OR).
Return to List of Papers
Return to List of Abstracts
