Growth Management in the Portland Region and the Housing Boom of the 1990's

By Gerard C.S. Mildner, Ph.D.

Portland State University

Gerard Mildner, Ph.D. is an associate professor in the Department of Urban Studies and Planning at Portland State University and a research advisor to the Urban Futures Program at the Reason Public Policy Institute in Los Angeles. This paper was presented at the meetings of the Association of Collegiate Schools of Planning in Ft. Lauderdale, Florida, November 9, 1997. Comments and suggestions should be directed toward Dr. Mildner at the Department of Urban Studies and Planning, P.O. Box 751, Portland State University, Portland, Oregon, 97207.

Working papers are intended to make scholarly research on urban development and planning issues more widely available to urban policy analysts and researchers on urban policy. Working papers represent research in progress and are published to invite comment and discussion as preparation for their submission to academic journals and other professional publications. The authors are solely responsible for the content of their research and analysis.

I. Introduction

In the 1970's, the Oregon state legislature established a system of statewide land use planning which mandates local governments to develop land use plans and regulations in conformity with a number of statewide goals. This overall system of land use planning has withstood several attempts at repeal by referendum and legislative revision, suggesting widely held support for the system.

In the last six years, however, rapid changes in population, housing prices, and land prices, particularly in the Portland area, have put the land use system under considerable economic and political stress. Population growth has tripled since the early 1980's, land prices have doubled since 1990, and housing prices have risen by more than 60%.

This paper will discuss how Oregon's current land use policies have adjusted to the new housing conditions and discuss the degree of the applicability of this model to other settings. In short, this paper argues that many unique conditions in Portland make land use planning more likely to be successful than in the average US city. Then, the paper describes the political coalitions that have formed surrounding those policy changes and assess whether this political coalition in favor of growth management can withstand this kind of economic change.

Finally, in a more normative vein, the paper will argue that despite good intentions, the current decision-making process will do little to remedy the ongoing housing cost problem in the region and will lead to considerable conflict in coming years. And the paper will suggest a more efficient, market-oriented, alternative vision for the region to achieve the environmental benefits, open space protection, housing affordability, metropolitan equity, and regional mobility the public wants.

II. Housing Prices and the Region 2040 Debate

This October, Portland's Metro Council decided upon an expansion of 4,500 acres in the Portland metropolitan area's Urban Growth Boundary based upon state law that mandates that they establish a growth boundary which contains a "20-year land supply". This vote was seen as a compromise between pro-environmental, zero-expansion advocates and pro-development advocates who had been favoring an expansion of at least 10,000 acres. The Metro Executive Officer's recommendation of 7,000 acres was rejected by a coalition of five of the seven Council members.

This debate has taken place within the context of a rapid population growth, with the 1990-95 growth rate more than 3 times the 1980-85 growth rate. Since 1990, housing prices in the Portland region have risen 10 per cent annually, changing the metropolitan area from one of the lowest cost places for housing on the West Coast, to the one of highest priced, small metropolitan areas in the country. The rise in housing prices has created a burden for young and low-income households, particularly those that are seeking new homes or renting.

Of course, the recent price increases are neither a permanent nor an irreversible process; Portland will not become New York City or Los Angeles. Eventually, the rise in housing prices will have a negative effect on the attractiveness of the region to investors and, at some point, will likely prevent continued employment growth. And of course, the relationship between housing markets, regional labor markets, and regional export demand is a complex, simultaneous system. Nevertheless, the lag times between when housing and land prices negatively affect the cost structure of local industries and deter employment growth can be quite long. In the meantime the region experiences higher housing prices.

The table below lists the average existing housing appreciation rate and population growth rate for 51 metropolitan areas in the United States between 1990-1995, using data from the National Association of Realtors. Portland has been among the fastest growing metropolitan areas in the country during this period, along with other cities in the west and southeast. Unlike the other cities, Portland's rate of housing price appreciation was much higher than others.

However, this is only circumstantial evidence of whether Portland's urban growth boundary is the cause of the region's rapid price appreciation. Although the Oregon system of land use management is unique, all metropolitan areas have some level of local zoning and building code regulation that, if stringent enough, could have a significant effect on the inelasticity of housing supply, which in turn causes rises in housing demand to lead to appreciation rather than new development.

However, data on the harshness of land use regulation on a metropolitan-wide basis is generally not available. In addition, some metropolitan areas may have significant water, mountain, terrain, or public ownership of land that may restrict the availability of local land supply for housing development. Finally, there may be significant variations in local policies that affect the underlying attractiveness or amenities of a metropolitan area. In Portland's case, the implementation of a series of property tax cut measures, beginning in 1991 may have had such an effect.

Population Growth and Housing Price Appreciation PMSA, 1990-95 (cont.)

Another aspect of the housing price appreciation in Portland is the way it compares to housing prices over the past business cycle and across neighborhoods.

As shown in the table below, the Portland area housing market in the early 1980s were quite depressed, as the regional economy was hit by the national recession and low timber prices. During the late 1980's, the economy picked up and housing prices began to rise primarily in the higher income communities, such as Lake Oswego, West and Northwest Portland, Tigard, and Wilsonville. Inner-city Portland saw very little investment and housing price appreciation during this period.

The 1990-95 period, however, has seen a resurgence of inner-city neighborhoods, exemplified by fact that the North, Southeast, and Northeast Portland areas saw the greatest rate of housing price appreciation in the region. For advocates of the growth management system, this turn of events is a symbol of success, believing that investments in transit projects and downtown revitalization have caused an increase in the amenities of close-in neighborhoods, thereby leading to new investment and a a new vitality. The tightness of the urban growth boundary has forced investment inwards and led to inner city revitalization.

A less sanguine view of these developments would describe this reinvestment and

revitalization process as gentrification. As land supplies became tighter on the fringe of the metropolitan area, housing demand is pushed inwards, and the increased land prices have led to higher density development in suburbs and city neighborhoods alike. And as these suburban neighborhoods achieve higher density and become less distinguishable from inner city neighborhoods in physical form, higher income households compete with lower income households for inner city locations.

III. Bubble in the Land Market

The boom in housing prices in the region (as well as rents) has many causes, both local and national, economic and political. Two important factors on the demand side have been the decline in interest rates since 1989 and the precipitous decline in property taxes begun by Measure 5 in 1990. In both cases, those reductions have reduced the cost of home ownership and increased the demand for owner-occupied homes. Thus, the increase in house prices from these causes do not represent a real decline in affordability since they are offset by other benefits.

A third factor has been the intertwined forces of rapid employment growth, new high tech investment, increases in household incomes, and rapid migration into Oregon. Since 1988, population growth in the region has averaged 35,000 people per year, triple the rate the rate for the previous 7 years. This extended boom has brought employment and wage gains for local residents and migrants, both high-skilled and low-skilled.

A final and important factor in Portland's housing price appreciation has been the degree to which the region's Urban Growth Boundary (UGB) has forced the increase in demand for housing to be translated into higher land and housing prices rather than into more houses being built. Originally established in 1979, the region's UGB was set far beyond existing development to provide for 20 years' worth of land supply for future development (as require by law). Since the regional economy soon fell into an extended recession that saw both nominal housing price declines and population stagnation, the original UGB was never a significant constraint on housing supply.

In the current boom, however, the UGB has become a binding constraint. According to a recent Metro study, the amount of vacant land available for new development has been reduced from 75,000 acres to slightly more than 50,000 acres since 1985. The table below attempts to integrate land consumption rates and total inventory. A new land inventory was subsequently conducted by Metro and found 55,000 vacant acres.

As this inventory has been reduced, the quality of the vacant land that is available for development has become less attractive, whether in terms of slope, soil, view amenities and transportation access, or would require significant losses in local amenities. For example, of the remaining 55,000 acres, 16,000 were deemed to be either too steep, too wet or environmentally constrained to be usable, and 13,000 acres were seen as necessary for schools, streets, and parks. And although there is some overlap between this category and those listed above, approximately 12,000 vacant acres are zoned for exclusive farm use or forest use and would need to be rezoned to be usable. Therefore, the reduction in vacant land from 75,000 to 50,000 acres is more significant than first appears because of environmental constraints and prior claims on their use.

This increased competition for land has translated into land prices, although the extent is difficult to determine. Good data on land prices is hard to come by. For comparative purposes, the Metro staff has been using data from the Urban Land Institute's ULI Market Profiles, which I present below. However, much of this information is drawn from very small or narrow samples. For example, the Los Angeles data comes from developments in a single place, the Antelope Valley, and the Las Vegas data is a per lot price estimate derived from builders' cost information.

As an alternative, for a report prepared previously by a team headed by this author, we studied information from the Washington County Tax Assessor's reports. Washington County is a most populous suburban county in the 4-county Portland region and the county with the greatest new home production.

The Washington County Tax Assessor reviews sales prices and determines an average ratio of sales price to assessed value for each class of property. These ratios are used to adjust assessments and can be interpreted as an average percentage price increase for property within these property classes. With low transaction volumes, the average percentage increase might be unrepresentative of the class as a whole. However, errors due to the low number of observations do not create either an upward or downward bias and should be minimized over a number of years. The table below reports the annual percentage increases for residential lots.

As the data indicates, land prices were fairly stable in the 1980's, with single digit annual rates of increase. However since 1990, land prices have grown at a rapid pace. Taking the County's assessment ratios as presented, lot prices have doubled in 5 years.

This rate of price increase could have many causes, one of which is the general rate of inflation in the economy. We have adjusted the price index by the Consumer Price Index and report inflation-adjusted price indices in the table below. This adjustment shows that about 20% of the increase in lot prices over the 1990-95 period was due to inflation. However, the conclusion remains that lot prices in Washington County grew by 79% in inflation-adjusted terms.

This work was considerably at odds with Metro's 1996 "Housing Needs Analysis". In their model, real land prices are expected to rise by 20% in real terms over 1995-2000 in their "2040 Growth Scenario" which was most like the scenario ultimately chosen by the Metro Council. Data on lot price inflation from 1990-1995 from Washington County suggests that land prices have already been rising at a much faster rate than this scenarios would project for the next five years.

IV. Density and Zoning

A large part of the growth boundary debate reflects the size of the single family development lot sizes. Since achieving higher density is a formal goal of the planning system, planners need to know the degree that zoning changes or land price changes can induce housing developers (and ultimately buyers) to build at high density.

To answer these questions, we conducted an analysis of subdivisions in each of the four counties in the region: Multnomah, Clackamas, and Washington Counties in Oregon, and Clark County in Washington state. Washington has implemented a milder form of growth management in recent years, but that seems to have done little to slow down housing development. Clark County, Washington rivals Washington County, Oregon in housing production rates and is well know for its larg lot subdivisions.

Two types of analyses were conducted. In Clackamas and Multnomah County, a random sample of all subdivision plat maps approved between 1991-95 were analyzed to measure land consumption and landuse constraints. In Clark County, Washington and Washington County, Oregon, an GIS analysis of all single-family, subdivision housing production between 1990 and June, 1995 was performed to compare both trends in land consumption and its relation to zoning characteristics.

Between 1990 to 1995, there were trends towards smaller lot sizes in three out of the four counties over the time period analyzed. The analysis of Multnomah and Clackamas County indicated that much of the trend towards smaller lots was mitigated by increases in open space and street area uses. Some of this trend may also be explained by increased row house development, particularly in Multnomah County. For the sample in Clackamas and Multnomah County as a whole, street area required approximately 16% of the subdivision land area; dedicated open space required 12%, and a further 6% was set aside as easements for slope, environmental, access, or utility purposes.

In Clark County, the number of lots developed grew by 80% during 1992-94, while the size of the average lot fell dramatically. Only in Washington County did the average lot size seem to rise over the time period being analyzed. This reflects an increase in housing development in the unincorporated sections of the county where lot sizes tend to be much greater.

The Washington County and Clark County data also allowed us to analyze the role that zoning played in reducing the average lot sizes in each jurisdiction. By matching parcel level information on lot size to the minimum lot size zoning categories, we can determine whether zoning designations are a key constraint keeping densities as low as they are. The relationship between zoning and actual density is important since Metro is seeking to use local zoning codes as a mechanism to increase densities and meet the expressed desire of the public for a more compact metropolitan area.

In our data, we distinguished between 3 zoning categories depending upon the minimum lot size for permitted. Small lot zoning had a minimum standard between 5,000 to 7,000 square feet. Medium lot zoning had a minimum lot size between 7,001 to 10,000 square feet. And large lot zoning required a minimum of between 10,001 to 40,000 square feet. As can be seen below, while lots in all three zoning categories are seen in Clark County, large lot zoning in Washington County either doesn't exist.

However, a comparison of actual lot sizes with the zoning designation indicates that development is occurring at much lot sizes compared to the minimum that is permitted. In Washington County, there is hardly any different between average lot sizes in the small and large lot zoning areas. For Clark County, small lot zoned areas are being developed at significantly higher densities, although given the general lack of density in Clark County, the true anomaly is the very large lots in the other two categories.

However, this evidence indicates that in fact, zoning designation is not an important determinant of average lot size. In neither of the two counties where we could make this comparison were builders adopting the highest level of density that the zoning code would allow. Rather average lot sizes were falling, we suspect, more because of the increase in land prices as buyers were substituting structure for land.

This degree of "underbuild" has had some important effects. For the environmental, anti-growth community, represented by 1,000 Friends of Oregon and Livable Oregon, this degree of underbuild represents excessive land consumption and the need to restrict the size of new development. For the regional government officials, the failure of easing traditional zoning restrictions to be the cause of higher density has led to the proposal for maximum lot size regulation rather than minimum lot size regulation. This kind of form of regulation has very little history and its effect is uncertain.

One possible market response might be greater movement of population to exurban and neighboring communities such as Clark County or Salem, Oregon where Metro regulations do not have effect. Another might be for large lot housing consumers to purchase double lots to build a single housing unit, thereby evading the intent of regulation. The first evasion might lead to an unexpected rise in long distance commuting and decentralization, what might be termed the "sprawling effects of growth containment". The latter evasion might lead to thinly settled subdivisions which are designed for infill housing to be built once demand for more density had caught up with the subdivision's planned density.

V. Metro's 2040 Plan

In response to legislative requirements, Metro and local governments in the region have planned for adjustments in the UGB to account for the region's future growth in the so-called "2040 Process". Metro planners and officials have been hesitant, however, to view the UGB decision in terms of its impact on housing markets and housing prices, but rather in the narrow legal issue of establishing a 20 year land supply and preserving farmland and forest land. Public surveys taken early in the process showed strong support for zero expansion.

Due to these political pressures, Metro planners have sought new and innovative ways to accommodate housing development within the available land supply, thereby minimizing the need for expansion. These methods include assuming a rising proportion of homes be built as redevelopment properties and as infill housing and requiring local governments to increase zoning on vacant land by an average of 60%. Local city planners, who view planning at Metro as the ultimate career objective, have readily acceded to this ideas and have "volunteered" their cities for increased housing capacity.

This shift in regulation has led to a number of conflicts. An attempt to rezone

single-family neighborhoods in Southwest Portland and suburban Milwaukie for row houses and apartments brought a fierce rejection by local residents. Thus, the attempt by regional planners to force densification plans on local homeowners may not be politically feasible.

Yet Metro and local government has been amazingly quiet in discussing the rise of housing prices in inner city neighborhoods and their gentrification by higher income groups. As new housing has become more difficult to build on the suburban fringe, middle and upper income home buyers have shifted their demand more towards the traditional neighborhoods of inner city Portland.

Housing price appreciation in inner city areas of Southeast Portland, Northeast Portland, and North Portland have been 85%, 78%, and 103%, respectively, between 1990-95, while the suburban communities have averaged 45%. While this form of gentrification may conform with the planners' desire for creating mixed income communities, the burden is being carried by Portland's poor.

VI. The Politics of Zero Growth

Early on in the 2040 Process, advocates of zero-expansion, headed by the environmental group, 1,000 Friends of Oregon, formed a pressure group, Coalition for a Livable Future, that sought to merge the efforts of environmental advocates and low-income housing advocates. From the perspective of the environmental advocates, the Coalition was a vehicle to prevent advocates of low-income housing from forming a coalition with developers and favoring a expansion of the boundary.

To attract the housing advocates, the Coalition added to their recommended policies the idea of enhancing local trust funds for low-income housing with real estate transfer taxes and land sales taxes and imposing an inclusionary zoning mandate on new housing development in the UGB expansion areas.

Both of these planks had appeal to the environmental community in that they further diminished the possibility that the metropolitan area would be expanded at all. Developers on the urban fringe would be required to build at a inner-city density, pay for local service expansion, build low income housing, or pay funds into a low-income housing trust fund.

The economic effect of these policies would be to raise the threshold price at which any new housing gets built. High density housing is more expensive to build, and the burden of impact fees and low-income housing subsidies will be shared by reduce prices for land owners and higher rents and prices for housing consumers. And by raising the supply price of new housing, this policy would unambiguously raise the market price of housing in the region and create excess burdens on low-income households, for whom housing is a large percentage of their incomes.

VII. The Politics of Low-Income Housing

Why did the housing advocates accept this package of policies? First, the housing trust fund promises the creation of an income stream to support their development plans. As federal funds for low-income housing have been reduced in recent years and long-term federal subsidy commitments have expired, the advocates have faced a choice of a contraction of their activities or an appeal to create a state or local subsidy. Since the state budget has growing demands due to increased prison sentences and local property tax cuts, this local fund source was the only alternative to contraction.

Second, the political ideology of low-income housing advocacy is distinctly anti-capitalist, and hence unfavorable to the idea of a coalition with private housing developers. For years, housing advocates have painted a picture helpless low-income households vulnerable to escalating rents on the private market. Accepting a policy which causes rents and prices to continue to rise perpetuates the housing crisis.

Third, the connection between development constraints and rising housing prices is an indirect effect and easy to ignore. Easing development constraints would largely cause new housing for the middle and upper-income households to be built. Those households would move out of less desirable housing that would, after several iterations, would filtered down to lower income households.

This phenomenon, known as the "filter hypothesis", is either dismissed or ignored by housing advocates, who usually focus on the unaffordability of new housing to low-income households. However, just as an "affordable" automobile is usually a used car, an affordable home is typically a house built years ago that has depreciated. New development for upper-income households hastens the day that the older housing is vacated.

Four, the interests of low-income households and non-profit housing providers are not overlapping. Federal studies have consistently pointed out that housing vouchers and portable certificates are a better way of serving the poor, while non-profit providers are better served by receiving project-based assistance. And while housing subsidies are required for all the providers new developments, over 80% of low-income renters find housing in the private market. Thus, the advocates represent the select 20% of low-income renters in assisted housing, not the 80% suffering from rising rents.

Finally, many low-income housing advocates are also political environmentalists, and the easiest way to resolve a policy that involves a trade-off between those two causes is to deny that any trade-off exists at all.

Thus, in a recent memo describing how housing affordability is going to decline, a team of Metro planners asserted, "The Housing Needs Analysis is intended to service as a technical examination of housing conditions in the region...[but] does not serve as a policy document." That is, we project that housing prices will be rising, but we certainly are not advocating that they do.

VIII. Longer Run Politics of Growth Management.

Although many have painted the upcoming UGB expansion decision as momentous and

unprecedented, whether the decision will be long-lasting is questionable. Already a coalition of pro-expansion advocates are gathering a petition for a state-wide vote on whether to abolish Metro, the regional planning body. A recent state-wide referendum on state funding for Portland area light rail suggests that a base of reflexively anti-Portland votes await to be mobilized.

Part of this anti-UGB coalition will surely be the landowners with properties surrounding the Portland UGB who have considerable wealth gains to be made from converting their land to urban uses. A recent Metro report suggests that raw land just inside the UGB sells for six times the value of land just outside the UGB where urban housing and commercial development is permitted. Thus, where large gains exist for the expansion of the UGB, large political contributions can be raised to further that cause.

This analysis suggests that growth management in Oregon's Willamette Valley is likely to be politically unstable, unlike, for example, the experience of Britain with greenbelt planning. In its 50-year policy of greenbelts, the UK government compensated landowners for the loss of the development rights. Thus, the pig farmer in Essex would see no benefit from being designated part of the London area since he did not hold the right to build a subdivision of housing.

In Oregon, however, land use planning was sold as a way to enhance the profitability of farming, by guaranteeing permanence and "protecting" farmers from urban encroachment. However, to keep this regulation less costly and to insure support from the agricultural community, farmers retained the right to develop their properties should their land be included in the UGB at a later point. In the future, the agricultural community is likely to divide between the landed farmers who want to see their land reach its highest value and farming implements and supplier lobby who want to preserve their customer base.

IX. Bursting the Bubble

How and when will the rapid escalation in regional housing prices come to a halt? This analysis suggests several alternative scenarios, some economic and some political.

The political scenarios would begin with a more vocal articulation of the interests of low-income households in the Portland region who are suffering disproportionately from the rise in housing prices. This scenario would require a mobilization that would be distinct from their current handlers in the community development community.

Another variant of this scenario involves some enlightenment of the non-profit housing advocates. The current environmental-housing advocate coalition is something of an imbalance trade between no growth of the UGB now, in exchange for a promise of housing assistance later. Thus, while Metro will be making is its decision to expand or hold the UGB this October, the decision to impose a real estate transfer tax or development linkage fees will come years in future. When and if this deal fails to materialize, the position of the low-income housing advocates may change.

The other political scenarios involve changes in the state political climate or changes in public attitudes. These kind of changes are inherently more difficult to predict, but some of the fault lines are clear. Landowners, property rights groups, and communities outside the Willamette Valley have been agitated by the Oregon planning system for some time. As the UGB in the Portland region becomes ever tighter, they may find additional support from first time home buyers, renters, and developers.

The economic scenarios revolve around when metropolitan growth will come to a halt. Already, some analysts have noted the rapid growth in suburban and exurban communities not constrained by the Portland UGB. That is, as housing prices are pushed up inside the Portland UGB, an increasing number of home buyers have turned to Clark County, Washington, the Salem, Oregon metropolitan area, or exurban communities like McMinnville, Newberg, and Mollala. Ironically, in this scenario, growth management is seen as an inducement to creating greater regional sprawl.

This kind of diffusion of excess housing demand in the wider regional market in reaction to tight zoning restrictions has long been noted in analysis of growth management in individual towns. That is, if Pasadena has exclusionary zoning, housing prices don't rise disproportionately in Pasadena since consumers can choose a close substitute in Los Angeles, Glendale or another nearby community. Growth management in Boulder, Colorado, has led potential residents to commute from Denver or nearby towns.

The difference in Oregon, is that unlike Pasadena and Boulder, the Portland UGB dominates the wider housing market. Sharply rising house prices in Portland mean sharply rising house prices in Salem and McMinnville. And this housing market escape valve is limited by the cost of long-distance commuting and state's unwillingness to accommodate it with new highway construction.

The other economic scenario involves an end to the regional economic boom. For 10 years and longer, Portlanders have echoed the mantra that growth management, dense settlement, and rural preservation will enhance the quality of life in the region and perpetuate the economic investment and employment growth. However, recent decisions by Nike and high tech firms in the region to consider expansion in places like Austin, Salt Lake City, and Phoenix where housing and development costs were important factors suggest the investment in the region may be slowing down. Oregon's land use planning system is still a novel system that has yet to experience a full business cycle.

X. Overselling Paradise

Part of this amenity package, dense settlement, runs contrary to national trends towards lower density living and avoidance of the problems of congestion. Already the loss of inner city natural areas through infill has led Metro to develop a parks acquisition plan called the Metro Greenspaces program. And rising traffic congestion in the region has caused the regional government to develop a task force looking into congestion pricing to manage local highways. Will migrants continue to come to high density Portland if that means a 4,000 square foot lot or a rented row house is the only affordable housing?

To be sure, the other aspects of the region's quality of life are undeniable. Portland has beautiful mountains and forests. Portland's downtown is indeed a vital place to work and shop. . Portland's low-income neighborhoods appear reasonably safe and inviting. The Portland region has one of the highest rates of public school attendance of any metropolitan area. And Portland's planning regime has an unusual amount of suburban and central city coordination.

However, the causes of many of these benefits are either God-given as in the case of the forests and mountains, or an accident of history. Portland is still a small metropolitan area where suburban, edge-city developments are not well developed. One can commute from the fringe of Portland to its center in less than 40 minutes.

Portland enjoys an unusual topography whereby large hills surround the downtown area to the west. The West Hills created a haven for the Portland elite to move to and retain their connection to downtown, when the elite in other metropolitan areas were fleeing to the suburbs.

Portland has a relatively low degree of economic and ethnic diversity compared to most metropolitan areas. Caused in part by geographic isolation and a 19th century prohibition on African-American settlement, Portland has a low percentage of blacks and hence did not face the issues of urban riots and white flight suffered by many other cities in the United States.

Another aspect of Portland's lack of diversity is the relatively low percentage of Catholics compared to other cities. Thus, unlike other communities which have long established networks of private, Catholic schools, the white, middle class of Portland had fewer opportunities to the inner city schools.

Hence, despite the problems outlined in much of this essay, the planning system in the Portland area has a number of things working in its favor, a good economy, a vital center, and a relative homogeneity that allows issues to discussed at the metropolitan level without the jealousies of income difference and racial fear intruding to the degree that might occur in other areas.

Whether the system can also sustain the loss of these conditions is another matter. Portland is becoming a larger and more diverse place, and is facing the social problems created by a rapid escalation of housing prices.

And this distinctiveness certainly raises questions about whether the Oregon planning system is replicable in other places. The test of its general applicability would require its introduction in a region with the small town atmosphere of Los Angeles, the hills and valleys of Houston, the ethnic homogeneity of Cleveland, and the economic vitality of St. Louis.

XI. Conclusion

This essay has discussed some of the ways in which the Portland region has responded to the pressures of the economic boom of the last 7 years. The region has witnessed both rapid population growth and a significant rise in housing demand, engendered in part by favorable national conditions. Due to constraints on the supply side, that demand has led to significant rises in regional housing prices.

The political response to these economic and social changes has been both ameliorative and self-denying. The political and economic requirement for greater land supply has been partially recognized and will apparently be ameliorated by an expansion of the urban growth boundary. However, the push to expand the region's UGB has been headed largely by the development community rather than by those most affected by housing price escalation. And that expansion decision has been hamstrung by the requirements being placed on the housing to be built.

At least for now, the liberal coalition of environmental and housing advocates has held together by the denial of any tradeoffs between open-space protection and housing affordability. However, as I have argued, while the environmental side of this bargain appear to have been secured, the housing benefits seem only to have been promised. Thus, while the Portland region seems to have avoided the specter most feared, Los Angeles-style suburban sprawl, the region appears headed instead for the San Francisco model of inner city gentrification and suburban exclusivity.