rppi - reason foundation · samuel r. staley, ph.d., is director of the urban futures program () at...

RPPIThe Reason Foundation is a national re-search and educational organization thatexplores and promotes the twin values ofrationality and freedom as the basic un-derpinnings of a good society. Since 1978,the Los Angeles-based foundation has pro-vided practical public-policy research,analysis, and commentary based upon theprinciples of individual liberty and respon-sibility and limited government.

about RPPI

contents

about the authors

Reason Public Policy Institute (RPPI), a di-vision of the Reason Foundation, fusestheory and practice to influence public poli-cies. The Institute's research draws uponeconomics, science, and institutional analy-sis to critique public policies and advancenew policy ideas in a variety of policy ar-eas, including education, infrastructure andtransportation, the environment, urbanland use and local economic development,social services, and privatization and gov-ernment reform. To that analysis, the Insti-tute brings a political philosophy that sup-ports rule of law, marketplace competition,economic and civil liberty, personal re-sponsibility in social and economic inter-actions, and institutional arrangementsthat foster dynamism and innovation.

The Reason Foundation is a tax-exempt edu-cational organization as defined under IRScode 501(c)(3). The Reason Foundation nei-ther seeks nor accepts government fund-ing, and is supported by individual, founda-tion, and corporate contributions. Nothingappearing in this document is to be con-strued as necessarily representing theviews of the Reason Foundation or its trust-ees, or as an attempt to aid or hinder thepassage of any bill before any legislativebody.

Photos used in this publicationare Copyright © 1996. Photodisc,Inc. Copyright © 1999. ReasonFoundation. All Rights Reserved.

Samuel R. Staley, Ph.D., is Director of the Urban Futures Program (www.urbanfutures.org) at Reason Public Policy Institute (RPPI).As an economic-development consultant, researcher, and policy analyst, he has authored more than 50 articles, studies, andreports on urban-development issues and policy, including two books. His work has appeared in professional journals such as theJournal of the American Planning Association, Planning and Markets (www-pam.usc.edu), and the Capital University Law Review.He has authored or co-authored the RPPI policy studies The Sprawling of America: In Defense of the Dynamic City; A Line in theLand: Urban-growth Boundaries, Smart Growth, and Housing Affordability (with Jefferson G. Edgens and Gerard C.S. Mildner);Urban Sprawl and the Michigan Landscape (co-published with the Mackinac Center for Public Policy in Midland, Michigan); andMarket-Oriented Planning: Principles and Tools (with Lynn Scarlett).

Gerard C.S. Mildner, Ph.D., is an assistant professor in the School of Urban Studies and Planning at Portland State University inOregon. Mildner’s research on urban-development and policy issues has appeared in conference proceedings as well as journalssuch as Land Economics and Transportation Quarterly and includes Scarcity by Design: The Legacy of New York City’s HousingPolicy (Harvard University Press, 1992, co-authored with Peter Salins). Through the Center for Urban Studies at Portland State,Mildner’s research has covered urban-growth boundaries, affordable housing, city-parking policy, impact fees, solid-waste dis-posal, and sports stadiums. He is also an academic advisor to the Cascade Policy Institute in Portland, Oregon.

Lessons fromPortland

11The Politics ofGrowth-boundaryExpansion

8The PortlandCase

3California andthe BurgeoningGrowth-boundaryMovement

21Urban-growthBoundary

ExecutiveSummary

1Glossary ofTerms

12Endnotes

9

Executive Summary

Urban-growth boundaries are emergingas one of the most popular growth-man-

agement tools in the fight againstsuburbanization. More than 100 cities andcounties have adopted them, and statewidemandates for growth boundaries exist in Or-egon, Washington, and Tennessee.

Urban-growth boundaries, however, havepotentially large, if unintended, negative im-pacts on housing. The burden of these impactswill most likely be felt by low-, moderate-, andmiddle-income households since their hous-ing choices will be the most severely con-strained.

An examination of housing-production trendsand home prices in Napa County, Californiafound that:

n Housing production fell by 74.2 percentwhen strict growth controls in NapaCounty, California were implemented,creating an effective countywide urban-growth boundary; and

n Housing prices soared in rural parts of thecounty as demand outstripped supply, in-creasing the price “premium” for ruralhousing from 16.3 percent in 1985 to 84.8percent over the county average in 1997.

Similar impacts were found in Portland, Or-egon where a regional-growth boundary hemsin 24 cities and three counties. A review ofresearch and housing data found:

n Portland now ranks among the 10 percentleast affordable housing markets in thenation;

n The average housing density has in-creased from five homes per acre to eighthomes per acre while multifamily hous-ing units makeup about half of all newbuilding permits;

n Even with these increases in density, thePortland area is expected to have a hous-ing deficit of almost 9,000 housing unitsby 2040;

n High rates of infill and redevelopment wereassociated with low overall levels of hous-ing production; and

n More than 80,000 single-family homes be-came “unaffordable” to Portland residentsas a result of housing-price inflation.

Several lessons were gleaned from NapaCounty and Portland’s experience with growthboundaries:

n Growth boundaries contribute to higherhousing costs, although the magnitude isuncertain. Metro, Portland’s regional-plan-ning agency, could alleviate housing costsby releasing more low-cost vacant landfor development but chooses not to;

n Growth boundaries encourage consum-ers to buy larger homes with fewer open-space amenities such as private yards;

n Growth boundaries create new special-interest groups that will oppose growth-boundary expansion, including high-in-come hobby farmers who want to protecttheir rural lifestyle;

n Recalls of local officials and the defeat ofnew funding for the regional-rail systemsuggests that public support for urban-growth boundaries in Portland may beweakening; and

n Higher housing prices are contributingto concerns by low- and moderate-in-come households that the growth bound-ary works against their interests, weak-ening overall support for regional-growthmanagement.

Urban-growthBoundaries

Growth management has risen to the forefront of public debate in the United

States. Both Vice President Al Gore and Presi-dent Bill Clinton are promoting a “LivabilityAgenda” for cities and encouraging so-called

11urban-growth boundaries and housing affordability

Reason Public Policy Institute82

“Smart Growth” initiatives to limitsuburbanization and revitalize central cities.

On the local level, one of the most popularSmart Growth tools is the urban-growth bound-ary, a politically designated line around citiesbeyond which development is either prohib-ited or highly discouraged. Growth boundaries,also called urban-limit lines or rural-limit lines,exist in more than 100 cities, counties, and re-gions across the nation.1 California, in particu-lar, has emerged as one of the most prolificcenters for this approach to growth control.

While popular, the full consequences of adopt-ing these growth controls have not been fullyexplored. While many are geared toward curb-ing suburban development in outlying ruralareas, growth boundaries can have unintendedeconomic and social consequences. By re-stricting land availability for new housing,growth boundaries could increase the price ofland and, ultimately, housing. As affordablehousing disappears, economically vulnerablelow- to moderate-income households sufferthe most. Even middle-income families maybe at risk based on emerging evidence fromPortland, Oregon and California.

Thus, growth boundaries may be a poorgrowth-management strategy, especially inareas concerned about housing affordability.Impacts such as the density versus affordabilitytrade off are rarely discussed adequately inthe heat of local growth-management debates.

California and theBurgeoning Growth-boundary Movement

Ground zero in the growing support forgrowth boundaries may be California. In

some states (e.g., Oregon, Washington, andTennessee), growth boundaries were man-dated by a distant state legislature. Urban-growth boundaries in California are the prod-uct of local efforts. As a result, California’s

growth boundaries are not uniform and thedata on their impacts are mixed. Public sup-port for growth boundaries, however, contin-ues to grow across the state, where 80 per-cent of the local initiatives with growth bound-aries on the ballot passed in November 1998.Frustration with lost open space, traffic con-gestion, and overcrowded schools led morethan 20 counties and 50 cities to adopt urban-limit lines or green belts in the 1970s and1980s.2 In the 1990s, growth-boundary activ-ity increased and proliferated.

Some California cities and counties are evenshrinking their boundaries.3 A movement isafoot in Contra Costa County, just across thebay from San Francisco, to reduce its growthboundary and thwart development plans de-spite a projected deficit of 45,000 homes.4 TheCity of Cotati in Sonoma County (north of SanFrancisco) reduced the amount of developedand developable land inside its boundary byalmost one-third.5

More telling may be the attempts by slow-growth advocates to convert urban-serviceareas—boundaries beyond which public infra-structure services such as roads, sewers, andwater will not be extended—into growthboundaries. Sacramento County is a case inpoint. Antigrowth interest groups are trans-forming the county’s urban-service area intoan urban-growth boundary by lobbyingagainst approval for development projectsoutside the service area.6

The most troubling characteristics of urban-growth boundaries, however, may be the in-equities they create between existinghomeowners and low- and moderate-incomefamilies.

The City of Napa in northern California is anexample. Just north of San Francisco, Napaadopted a limit line in 1975 in an attempt tocap the city’s population at 75,000. Residen-tial development was still allowed in rural ar-eas, providing a relief valve for the residen-tial-housing market. Then, in the early 1990s,

the county clamped down on new develop-ment. Residential-building permits plummetedby 74.2 percent from 1989 to 1996 as a mora-torium on residential permits took effect in thecounty and a regional-housing recession tookhold in the early 1990s.7 As the recession gaveway to the current economic boom, buildingpermits increased, but not as quickly as inother parts of the Bay Area without growthboundaries.8 Well after the end of the reces-sion, building permits remain at half the lev-els that existed before the county growth con-trols were instituted.

An affordability wedge between rural and ur-ban residents is also emerging in Napa County.As the growth controls took hold, the averagevalue of a single-family home in unincorpo-rated areas of Napa County climbed 158.1percent to over $373,000 from 1985 to 1997.9

The average value of a new single-familyhome in the City of Napa increased by lessthan half the rate in rural areas (66.8 percent),but the rate of increase doubled after thecounty growth controls created an effectiveregional-growth boundary.10 Nevertheless,the housing-price “premium” for homes out-side the more densely urbanized areas grewfrom 16.3 percent to 84.8 percent over theNapa County average.

Thus, growth boundaries potentially serve upa double whammy for homebuyers. First, re-ducing land supplies through growth controlsdrives up housing prices. Then slow-growthpolicies outside cities create an open-spacepreserve that can be tapped only by high-in-come households.

Although the data for California and NapaCounty are not definitive, they clearly suggesturban-growth boundaries can have the unin-tended impact of reducing the supply of hous-ing. Moreover, the use of urban-growth bound-aries also suggests local growth controls passthe responsibility for accommodating futurehousing demand and growth onto other com-munities. In the case of Napa County, the bound-ary by the county’s largest city (Napa), coupled

with an apparent unwillingness of other citiesin the county to annex territory and provide forfuture growth, has had the effect of increasinghousing costs and limiting the availability ofhousing for working families.

The case of Portland, Oregon, on the other hand,provides an example of a pure urban-growthboundary that has been in place on a regionallevel over a longer period of time. Portland’sexperience should provide more direct lessonsabout the relationship between urban-growthboundaries and housing affordability.

The Portland Case

The State of Oregon implemented one ofthe nation’s most comprehensive state-

planning laws in the early 1970s. In Portland, aregional boundary has hemmed in 24 citiesand three counties for 20 years and has be-come the linchpin to regional urban planning.Portland uses its growth boundary explicitlyto increase housing density and redirect in-vestment into inner cities. Thus, the Portlandeffort provides an excellent opportunity toexamine the intended and unintended conse-quences of adopting growth boundaries.

While originally intended as a dynamic tool foraddressing the needs of a city’s population,the growth boundary in Portland has madeplanning for the needs of a growing popula-tion far more complex. To more fully appreci-ate the relationship between growth bound-aries, housing prices, and housing choice, thefollowing sections review the effects of growthboundaries in Portland, Oregon on:

n Housing cost and prices;

n Housing density;11

n Developable land and infill; and

n Consumer choice in the housing market.

The effects of growth boundaries on housingprices and consumer choice may create a po-



litical environment that may be difficult to sus-tain and further compound the negative, ifunintended, consequences of imposing agrowth boundary.

A. Housing Costs in Portland andOregon

Since the passage of Oregon’s growth-man-agement laws, the cost of housing in urbanareas has increased significantly. Oregon’shousing markets now rank among the nation’sleast affordable in the nation and on the WestCoast (see Table 1).12 Eugene holds the dubi-ous distinction of being the least affordableOregon housing market, ranking among thebottom 3 percent in housing affordability na-tionwide, according to the National Associa-tion of Home Builders. Not far behind are Port-land and Medford, both ranking among thebottom 10 percent.

Another policy goal of planning in Portland andthe rest of the state is to meet the housingneeds of low- and moderate-income families.

Housing-price trends have important implica-tions for affordability. One local real-estateconsultant estimates that, from the secondquarter 1995 through second quarter 1997,housing-price appreciation alone pushed80,000 single-family homes over thresholdsof affordability.13 In other words, 80,000 fewerunits were considered “affordable” in 1997compared to 1995. Meanwhile, from 1990 to1997, just 6,450 single-family homes and 3,530multifamily units were approved.14

Because the relationship between housingprices and land supply is complex, simple ex-planations for Portland’s extraordinary hous-ing-price increases are elusive. Metro hasclaimed that land supply is largely irrelevantfor understanding Portland’s recent housing-price surge, arguing that increased popula-tion growth is the primary factor influencinghome prices.15

Housing prices in Portland, however, are notincreasing in a neutral real-estate market: theamount and quality of vacant land has been

HOUSING AFFORDABILITY FOR OREGON METROPOLITANAREAS: FOURTH QUARTER 1998

table 1

Affordability National Percenile Regional PercentileCity/MSA Index Rank* Rank Rank** Rank

Eugene-Springfield 30.4 185 99.5% 47 97.9%Medford-Ashland 38.5 182 97.8% 45 93.8%Portland-Vancouver 38.7 180 96.8% 43 89.6%Salem 44.8 173 93.0% 37 77.1%

*Nation (N=186) **Western States (N=48)

Source: National Association of Home Builders, http://www.nahb.com. The affordability index is theNAHB’s “Housing Opportunity Index” (HOI) score and represents the share of new and existinghomes that could be purchased by a family earning the metropolitan area’s median-family income.Thus, lower HOI scores represent less affordable areas since the median family could afford topurchase a lower percentage of homes sold in the regional market. Therefore, lower rankings repre-sent less affordable areas.

falling for 20 years as a direct result of theurban-growth boundary. As land becomesmore scarce inside the boundary, the in-creased competition for developable land in-side the growth boundary appears to be con-tributing to higher land prices.16

These effects are evident in the Portland area’slargest suburban county. Washington Countyis the second-largest county in the metropoli-tan area, but has the largest amount of new-home construction and the best data in theregion on housing-lot prices. Lot prices forsingle-family houses in Washington Countylagged inflation, as measured by the Con-sumer Price Index (CPI) from 1985 to 1990,the years the Portland area experienced ahousing recession (Figure 1).17 After 1990,housing prices increased significantly. By 1994,home prices were one and one-half timesgreater than in 1985 (a 140 percent increase).Lot prices more than doubled in five yearswhile the CPI increased by 52.5 percent.18

These trends significantly surpassed Metro’shousing-price forecast, which predicted landprices would rise by 20 percent in real termsfrom 1995 to 2000.19

A glimpse of the growth boundary’s poten-tial impacts on land prices became evidentearly.20 In 1980, just one year after Portland’sregional-growth boundary was established,an analysis of 455 purchases of vacant lotsfor single-family homes found that landprices inside the boundary were significantlyhigher than those outside the boundary.21 Thechanges in market price were tied to expec-tations by builders and developers about thelikelihood the land could be developed forresidential purposes.

Land prices varied by how much local gov-ernments restricted development and devel-opers believed those restrictions were bind-ing.22 Rural-land values outside the boundaryfell as developers recognized its availabilityfor urban development was limited. Land val-ues inside the boundary increased as devel-opers recognized its potential for development

and more people were competing for the sameparcel of land. Thus, strict regulatory adher-ence to the growth boundary resulted in thelargest differences in price. As Metro has be-come even stricter in its adherence to mini-mum-density targets, land prices are likely toincrease even further.

B. Housing Density

Another important goal of growth boundariesis increasing housing density within existingurban areas. Despite a national reputation asa growth-management success story,Oregon’s planning system fell sufficiently shortof this goal that many planners andpolicymakers argued for “mid-course correc-tions” as early as the mid-1980s.23 Actual den-sities inside the growth boundaries, for ex-ample, ranged from two-thirds to one-fourthbelow the levels permitted by local plans.24

Although Portland had the highest-allowabledensities, actual densities were one-third lowerthan those allowed by local land-use plans.25

Portland nevertheless increased housing den-sities by putting more people on less land eventhough they were below planners’ targets.Densities for new development increased onaverage from five homes per acre (one-fifthof an acre per home) to eight homes per acre(one-eighth of an acre per home) from 1994 to1997.26 The amount of land used for new hous-ing development has declined as multifamilyhousing units have increased from 25 percentof all building permits in 1992 to 49 percent in1997.27

Even these density increases fall below thelevels needed to meet Metro’s (the regionalplanning agency) projected population in-crease. At current trends, without an expan-sion of the boundary, the Portland metropoli-tan area will experience a 42,060 housing-unitdeficit by the year 2017.28 If densities increaseto achieve those recommended in the Metro2040 Plan, the housing deficit would still be8,590 units.


This creates a problem for local planners: ifdensities are too low (and more land is usedper household), the urban-growth boundarywill have to be expanded sooner than theypredicted. In order to increase densities fur-ther and avoid expanding the growth bound-ary, Metro has recently implemented a man-date for a minimum density of more than sixunits per net acre (which is creating new eco-nomic and political problems and trade offs).29

C. Developable Land and Infill

In order to achieve higher densities within theurban-growth boundary, Metro and Portlandplanners are relying on “refill.” Refill consists oftwo elements: developing vacant land (infill)inside the growth boundary and redevelopingexisting property more intensely at higher den-

sities. An examination of the Portland modelreveals another unintended effect of growthboundaries: the reliance on infill developmentcontributes to higher housing costs and impactsthe availability of affordable housing.

As Metro pushes for more infill development,vacant land within the growth boundary isdisappearing. Vacant land inside the bound-ary has fallen from 75,000 acres in 1985 toless than 55,000 today. Once environmen-tally sensitive and otherwise undevelopableland is considered, the amount falls to lessthan 38,000 acres.30 While almost 40 percentof the land in the boundary was vacant in1980, the share of total vacant land repre-sented just 19.8 percent of the land by 1997,less than 14 percent when undevelopableland is considered.31


CHANGE IN RESIDENTIAL LOT PRICES VS. INFLATION INWASHINGTON COUNTY

figure 1Source: Washington County Tax Assessor.

Metro estimates Portland’s current refill rateat 25.4 percent.32 In other words, about one-fourth of all housing units built inside thegrowth boundary are either infill or redevel-opment of existing property. Metro, however,also found that the rate of residential rede-velopment and infill was negatively relatedto the total housing built when the data werebroken down by neighborhood.23 High refillrates, it turns out, require higher home pricesto justify the development of smaller and rela-tively more-expensive parcels of land.34 Inaddition, if homebuyers cannot trade offhome size for larger lots, they may invest inbuilding bigger and more expansive homes.Thus, Portland was achieving a higher rate ofinfill and redevelopment because land andhousing prices were increasing.

Inner-city Portland appeared to benefit fromthese higher rates of redevelopment. From1990 to 1995, inner-city neighborhoods in Port-land experienced a substantial increase inhome-price inflation: the North, Southeast,and Northeast areas of the City of Portlandsaw their housing prices increase the fastest(Table 2). Home prices in North Portlanddoubled, rising from $41,300 in 1990 to$83,800 in 1995 (in noninflation adjusted dol-lars). The average home price among thesecities increased from $97,684 to $152,700.

For those favoring growth boundaries, thesetrends suggest success. Higher refill rates andrebounding home prices in inner-city neigh-borhoods should mean the goals of revital-ization and increasing density are being met.

These observers, however, may be ignoringthe trade offs implicit in these trends. Higherhousing prices are a double-edged sword. Onthe one hand, housing demand is pushed in-ward as land becomes more scarce on thefringe. Higher land prices create incentivesfor higher-density development in existingsuburbs and inner-city neighborhoods.35 Assuburban neighborhoods achieve higher den-sities and become less distinguishable frominner-city neighborhoods, higher-income

households look at inner city and suburban lo-cations more competitively.

Higher-housing prices, on the other hand, mayalso reflect a supply constraint. With the impo-sition of an effective growth boundary, con-sumers have fewer housing choices than un-der a freely functioning land market since de-velopers are more limited in the kinds of homesthey can provide. Rural and semi-rural settingsfor homes and communities will be reducedfor all except the wealthiest families and long-time residents. In this case, higher-housingprices reduce the overall quality of life for resi-dents, since they must pay more for a homethat provides potentially fewer benefits (e.g.,smaller lots and denser living). Many house-holds are buying what is available in a restric-tive housing environment as new homes onlots larger than one-fifth of an acre are prohib-ited. Higher refill rates come at a cost: higherhousing prices, less affordable housing, andless private open space in the form of yards.

D. Consumer Behavior

Land, as the previous section discussed, is animportant beneficial characteristic of a home.Sometimes this land is privately owned (e.g.,back and front yards), and sometimes it is pub-lic open space (e.g., a neighborhood park). Notsurprisingly, when land is abundant, peopleprefer homes with larger lots. As housing pricesincrease, homebuyers recalculate the benefitsof different housing characteristics. Sincehouses usually provide important, high-prior-ity benefits—bedrooms for personal privacy,shelter from the elements, kitchens, etc.—homebuyers will often choose smaller lotswhen home prices are high. Thus, growthboundaries influence the housing decisions byfamilies and other households as well.

Portland, Oregon provides an interesting ap-plication of this housing “substitution” effect.When Metro began implementing its land-useplan in the early 1990s, developers were al-lowed to build more houses on each acre ofland as maximum densities allowed by local


zoning codes increased. By the mid-1990s, theeffects on the local real-estate market werebecoming apparent. Researchers at Metroexamined almost 8,000 home sales from Janu-ary 1996 through June 1997 to determine howhouseholds responded to higher housingprices and whether a lot size/building sizetradeoff existed. 36 Portland-area homebuyers,it turned out, did not buy larger lots as homeprices increased for houses of the same size.37

As home prices increased, homebuyersbought larger homes on smaller lots. Forhomes in different sizes categories (large vs.small), researchers also found that consum-ers traded off lot size for home size.38 Ofcourse, these trade offs were made in an arti-ficially constrained and manipulated market.The growth boundary is a politically desig-nated barrier, not one chosen by consumerstrading in an open market.

The Politics of Growth-boundary Expansion

Metro claims that the urban-growthboundary is not intended to prevent

housing development. Rather, the goal is tomanage growth. The boundary was originallyexpected to be a dynamic, growth-manage-ment tool: as land became more scarce, localofficials and planners would expand theboundary to accommodate more growth.

Portland politics is making expansion of thegrowth boundary difficult. In 1997, Metro’sexecutive officer recommended a 7,000-acreexpansion of the growth boundary to accom-modate new growth.39 The Metro Council voteddown the recommendation five votes to two.Metro eventually expanded the boundary by5,300 acres in 1998,40 but the expansion rep-


HOUSING PRICES BY PORTLAND SUB-MARKET, 1982-95

table 2

Prices Appreciation RateType 1995 1982-85 1985-90 1990-95

Lake Oswego/West Linn suburb $244,400 -0.25% 54.55% 33.12%West Portland inner city $210,200 -2.27% 51.11% 46.58%NW Portland inner city $195,900 0.50% 44.78% 35.85%Tigard/Wilsonville suburb $174,900 -9.86% 61.65% 39.25%Milwaukie/Gladstone suburb $144,800 -4.79% 43.29% 54.04%Oregon City/Mollala suburb $144,500 10.60% 37.87% 62.00%Beaverton/Aloha suburb $141,700 -3.05% 44.52% 34.31%Hillsboro/Forest Grove suburb $134,500 -6.62% 34.05% 54.60%Gresham/Troutdale suburb $132,900 -5.97% 34.49% 49.49%NE Portland inner city $114,500 -9.15% 24.42% 78.35%SE Portland inner city $109,700 -8.83% 22.06% 85.30%N. Portland inner city $83,800 -17.48% 16.67% 102.91%Unweighted Averages $152,700 -4.76% 39.12% 56.32%

Source: Gerard C.S. Mildner, “Growth Management in the Portland Region and the Housing Boomof the 1990s,” Urban Futures Working Paper No. 98-1, Reason Public Policy Institute, May 1998, http://www.urbanfutures.org.


resents about two years of the average landtake-up through development.41 More signifi-cantly, the expansion was a compromise be-tween environmental activists, zero-expan-sion advocates, and prodevelopment groups.Even while local planners may favor a moredynamic approach to land supply within thegrowth boundary, regional politics may pre-vent its timely expansion. A close examina-tion of the Portland case reveals that, onceimplemented, growth boundaries make plan-ning decisions more difficult and political. As aresult, responding to increased housing de-mand and population growth becomes a func-tion of local and regional political forces, ratherthan land-use planning.

Local politics is making growth-boundary ex-pansion difficult in part because the planningprocess itself is creating new winners andlosers.

A. Low-income Households

Low-income households, for instance, wereonce considered big winners from statewideplanning. Oregon’s planning law included aplank supporting the housing needs of all resi-dents in the state, and many believed a prop-erly planned region would yield affordablehousing.42 Strong demand for inner-city hous-ing, however, has pushed up prices in inner-city Portland neighborhoods faster than insuburban communities, displacing some low-and moderate-income families.

Low-income housing strategies such asinclusionary zoning are unlikely to significantlyimprove affordable housing. Portland has con-sidered an inclusionary-zoning program thatwould require developers to provide afford-able housing. If the plan had been enacted dur-ing the 1990s, “affordable” homes and apart-ments would have represented just two per-cent of the units that “disappeared” as a re-sult of escalating housing prices.43 Programssuch as the one considered by Metro wouldmake little headway in improving affordabilityin the Portland metropolitan area.

B. �Hobby� Farmers

The growth boundary has had another unin-tended side effect. A new interest group con-sisting of noncommercial farmers, sometimescalled “hobby farmers,” has emerged that ishighly resistant to expansions of the growthboundary. Many of these households appearto be circumventing Metro’s restrictive land-use policies by buying properties outside thegrowth boundary and calling them farms.Ninety percent of the farms under 160 acreswhere new homes were authorized in the1980s reported no farm receipts.44 Half of thefarm operations with new homes were in theWillamette Valley, which contains the City ofPortland and about 60 percent of Oregon’s resi-dents.45 By planting a field of Christmas treesor a large patch of strawberries, these land-owners have been able to get ruralhomebuilding permits under the exemption forfarmers. Ironically, by building their homes onlarge rural parcels, these hobby farmers arecreating the exurban sprawl that many ofgrowth-boundary advocates wish to avoid.Hobby farmers are hostile to boundary expan-sions because higher-residential densities anddevelopment of nearby open space would di-minish their quality of life.

Lessons from Portland

Several events suggest that grassroots op-position to regional planning is emerging

as an important political force. Metro is expe-riencing significant resistance to higher-den-sity residential development and the proposedtransportation plan by grassroots groups. Forexample:

n A regionwide referendum to fund theregion’s light-rail extension was rejectedin November 1998; and

n Residents of the suburban community ofMilwaukie, Oregon recalled all city coun-cil members that voted to accept Metro’shigh-density zoning mandate to accommo-date future population growth.

Growth management in the Portland region

is becoming less stable politically as new in-terest groups and coalitions emerge to sup-

port specific aspects of the regional plan. Asland inside the growth boundary becomes in-

creasingly scarce, housing prices have in-creased significantly. Portland still retains

many of the characteristics of suburban liv-ing that the growth boundary was intended

to discourage. Ironically, achieving densitytargets may only be possible through high

housing prices that make expensive and inef-ficient parcels of land profitable to develop

through the private sector.

While some believe a substantial amount ofland still exists within the urban-growth bound-

ary for land development, Metro is predictinga housing deficit even if significantly higher

densities are achieved on current land. In fact,the remaining land inside the boundary is gen-

erally less productive and more expensive todevelop. Combined with a political climate sup-

portive of zero growth, Metro will not likelyexpand the growth boundary significantly to

moderate upward pressure on housing prices.

Several lessons can be gleaned from Portland’sexperience with growth boundaries.

n First, urban-growth boundaries canachieve goals such as encouraging higherresidential densities and infill, but these

outcomes come at a cost. If the growthboundary is successful, it will constrain

vacant land and require housing develop-ment on more-expensive land and on lots

much smaller than consumers would oth-erwise prefer.

n Second, growth boundaries encourageconsumers to trade off land for largerhomes with fewer open-space amenitiessuch as private yards and reduce their over-

all quality of life by constraining land sup-ply and contributing to higher land costs.

n Third, growth boundaries contribute tohigher housing costs, although the mag-nitude is uncertain. Metro could help alle-

viate housing costs by releasing morelow-cost vacant land for development (al-

though it chooses not to).

n Fourth, growth boundaries will encour-age the creation of new interest groupsopposed to growth-boundary expan-sion. Local policymakers will be encour-

aged to maintain the boundary as a bind-ing constraint on land development, opt-

ing for increasing densities in existingareas rather than expanding the bound-

ary significantly.

n Fifth, the political support for growthboundaries and growth management ingeneral will change as the full conse-quences of the policies become evident.Higher housing prices, for example, are

contributing to concerns by low- andmoderate-income households that the

growth boundary may work against theirinterests.

To avoid the unintended side effects of urban-

growth boundaries, state and localpolicymakers should consider alternative,

market-based approaches to growth manage-ment. Market-based approaches substitute a

regulatory approach to development that re-stricts consumer choice for one where the

real-estate market and incentives are createdto achieve the same goals. Examples of mar-

ket-oriented approaches include reformingzoning ordinances to allow for market-deter-

mined densities, allowing for administrativereview of development projects without sig-

nificant negative impacts on neighboring prop-erty owners and the community more gener-

ally, privatization or full-cost pricing for publicinfrastructure, and voluntary conservation

easements and privately funded purchase-of-development rights to protect open space. ❑


Glossary of Terms

Density: number of people or households peracre of land.

Green belt: a strip of dedicated open spacearound cities where land development is pro-hibited except for agricultural, park land, andopen-space uses.

Growth management: the direction, control,channeling, or guidance of commercial andresidential development through public policy.

Housing amenity: quality or characteristicof a home.

Housing-substitution effect: the trade offsconsumers make among housing character-istics such as lot size, bedrooms, garage ar-eas, bathrooms, etc.

Infill: the development of vacant land in al-ready urbanized areas with existing homesand buildings.

Infrastructure: public services such as roads,sewers, water, schools, etc.


Low density: low number of people or house-holds per acre of land. Single-family, detachedhomes with large lots are often defined as lowdensity, although the actual size of the lot israrely specified.

Metro: the regional planning agency over-seeing Portland’s growth boundary and theimplementation of the region’s 2040 long-range plan. Metro’s board is the only electedregional government in the nation and is re-sponsible for regional transportation andland-use planning.

Refill: the combination of in-fill developmentand redevelopment of existing land in urban-ized areas.

Urban-growth boundaries (or urban-limitlines): politically designated line around citiesbeyond which development is either prohib-ited or highly discouraged.

Urban-service area: a boundary beyondwhich public infrastructure services will not beextended.

endnote

s 1 Estimate based on known growth boundaries providedby Samuel R. Staley, Director of the Urban FuturesProgram, Reason Public Policy Institute, Los Angeles,California. Most of these growth boundaries are in Or-egon, Washington, California, and Lancaster County,Pennsylvania. These numbers do not include urban-ser-vice boundaries. For a discussion of the distinction, seeSamuel R. Staley, Jefferson G. Edgens, and Gerard C.S.Mildner, A Line in the Land: Urban-growth Boundaries,Smart Growth, and Housing Affordability, Policy StudyNo. 263 (Los Angeles, Reason Public Policy Institute,October 1999).

2 William Fulton, “Sliced on the Cutting Edge: GrowthManagement and Growth Control in California,” in GrowthManagement: The Planning Challenge of the 1990’s, ed.Jay M. Stein (Newbury Park, California: Sage Publica-tions, 1993), pp. 113-126.

3 A summary of this trend can be found in Staley, Edgens,and Mildner, A Line in the Land. Growth controls have along history in California, beginning in earnest withland-use regulation in coastal areas. See H.E. Frech III,“The California Coastal Commissions: Economic Im-pacts,” in Resolving the Housing Crisis: GovernmentPolicy, Decontrol, and the Public Interest, ed. M. BruceJohnson (San Francisco: Pacific Research Institute forPublic Policy, 1982), pp. 259-274.

4 For housing deficit projections, see Phil Serna, Tri-Val-ley Housing Needs Assessment, Home Builders Asso-ciation of Northern California, June 1998, Table C.

5 Staley, Edgens, and Mildner, A Line in the Land.

6 Janice Fillip, “Stretching the Limits,” Comstock’s (Feb-ruary 1998), pp. 47-48. A more detailed discussion ofthis project can be found in Staley, Edgens, and Mildner,A Line in the Land.

7 Building permit data from Construction Industry Re-search Board, Burbank, California.

8 This is more fully discussed in Staley, Edgens, andMildner, A Line in the Land.

9 Construction Industry Research Board, Burbank, Cali-fornia.

10 Ibid.

11 Increasing density is implied in adopting the growthboundary. By preventing development outside a geo-graphic boundary, investment is expected to be chan-neled into existing areas through redevelopment of infill.This increases the residential density of neighborhoods.This goal has become more explicit in areas such asPortland that are explicitly using growth management toachieve more compact forms of development by reduc-ing lot sizes and increasing the multifamily share of thelocal housing market.

12 The homebuilders examined interest rates, median-family income, and home sales in each metropolitanarea to determine what proportion of homes could bepurchased by a family with the median income in themetropolitan area. See also the discussion in RichardH. Carson, Paying for Oregon’s Growth (Beaverton,Oregon: New Meridian Press, October 1998). Rapidhousing-price appreciation is only circumstantial evi-dence of whether urban-growth boundaries reduce hous-ing affordability. Although the Oregon system of land-use management is unique, all metropolitan areas havesome level of local zoning and building code regulationthat, if stringent enough, could have a significant ef-fect on the supply of housing. Lower-housing supplycauses demand for existing homes to increase, lead-ing to increased home appreciation. However, compa-

rable data on the harshness of land-use regulation ona metropolitan-wide basis are generally not avail-able, although economist Gerrit Knaap has examinedthe differences within the Portland metropolitan area.See Gerrit J. Knaap, “The Price Effects of UrbanGrowth Boundaries,” Land Economics, vol. 61, no. 1(February 1985), pp. 26-35. In addition, some metro-politan areas may have significant water, mountain,terrain, or public ownership of land that may restrictthe availability of local land supply for housing de-velopment. Finally, there may be significant varia-tions in local policies that affect the underlying at-tractiveness or amenities of a metropolitan area. InPortland’s case, the implementation of a series ofproperty tax-cut measures, beginning in 1991, mayhave had such an effect.

13 See the discussion in Staley, Edgens, and Mildner, ALine in the Land, pp. 17-19.

14 Ibid.

15 Metro notes that land prices are rising as fast or fasterthan in cities without growth boundaries, including SaltLake City, Phoenix, and San Diego. A number of factorsin addition to the growth boundary are probably alsocontributing to the boom in housing prices, including:the national decline in interest rates since 1989; theprecipitous decline in property taxes in Oregon begunthrough a 1990 ballot-box initiative; Portland’s rapidemployment growth; investment in new high-tech busi-nesses; and rising household incomes.

These effects, however, should be temporary as thebuilding industry adjusts to higher levels of demandand new housing supply comes on line and pricesmoderate. For the Portland metropolitan area, popula-tion is increasing between 2.0 percent and 2.5 percentper year. See Portland Metro, Urban Growth Report,December 1997, p. 42. While larger than the nationalaverage, this is less than other metropolitan areassuch as Las Vegas (about 6.5 percent per year), Phoe-nix (about 3.0 percent), Riverside (about 2.7 percent),and Orlando (about 2.7 percent). These fast-growingcities did not experience significant housing priceappreciation during the same period. While Portland’shousing prices increased by 61.5 percent from 1990 to1995, housing prices in Las Vegas—the fastest grow-ing city among the cities listed—increased by aboutone third as fast (22.0 percent during this period).

16 Good data on land prices are difficult to obtain. TheMetro staff uses data from the Urban Land Institute’sULI Market Profiles to compare Portland to othercities. Much of this information, however, is drawnfrom very small or narrow samples. For example,the Los Angeles data come from development in asingle place, the Antelope Valley, and the Las Vegasdata are a per lot price estimate derived from build-ers’ cost information.

17 Washington County is the most-populous suburbancounty in the four-county Portland area and the countywith the greatest new-home production. The Washing-ton County Tax Assessor reviews sales prices anddetermines an average ratio of sales price to as-sessed value for each class of property. Theseratios are used to adjust assessments and can beinterpreted as an average-percentage price increasefor property within these property classes. With lowtransaction volumes, the average-percentage increasemight be unrepresentative of the class as a whole.However, errors due to the low number of observa-tions do not create either an upward or downwardbias and should be minimized over a number of years.



18 We have adjusted the price index by the ConsumerPrice Index and report inflation-adjusted price indices.This adjustment shows that about 20 percent of theincrease in lot prices over the 1990-95 period was dueto inflation. Thus, lot prices in Washington County grewby 79 percent in inflation-adjusted terms.

19 Gerard C.S. Mildner, “Growth Management in the Port-land Region and the Housing Boom of the 1990s,” UrbanFutures Working Paper No. 98-1, Reason Public PolicyInstitute, May 1998, http://www.urbanfutures.org.

20 Gerrit J. Knaap and Arthur C. Nelson, “The Effects ofRegional Land Use Control in Oregon: A Theoretical andEmpirical Review,” The Review of Regional Studies,vol. 18, no. 2 (1988), pp. 37-46.

21 Knaap, “The Price Effects of Urban-growth Boundariesin Metropolitan Portland, Oregon,” pp. 26-35.

22 The growth boundary was new enough that the 20-yearsupply of vacant land provided little constraint on de-velopment. Land is converted to residential uses aslocal zoning boards permit development. See Knaap,“The Price Effects of Urban-growth Boundaries,” p. 33.

23 Jerry Weitz and Terry Moore, “Development Inside Ur-ban-growth Boundaries: Oregon’s Empirical Evidence ofcontinuous Urban Form,” Journal of the American Plan-ning Association, vol. 64, no. 4 (Autumn 1998), p. 427.

24 Deborah A. Howe, “Growth Management in Oregon,” inGrowth Management: The Planning Challenge of the1990’s, ed. Jay M. Stein (Newbury Park, California:Sage Publications, 1993), p. 70. See also T. Moore andA.C. Nelson, “Lessons for Effective Urban-containmentand Resource-land Preservation Policy,” Journal of UrbanPlanning and Development, vol. 120, no. 4 (December1994); V. Gail Easely, Inside the Lines , Planning Advi-sory Service Report Number 440 (Chicago, Illinois:American Planning Association, 1992).

25 Howe, “Growth Management in Oregon,” p. 65.

26 Portland Metro, Urban Growth Report Addendum, Au-gust 1998, pp. 14-15.

27 Ibid., Figure 13, p. 15. Land consumption has declinedfrom 2,900 acres in 1995, to 2,300 in 1996, to 1,700 in1997. The share of multifamily housing units amongapproved units tends to vary considerably among andwithin metropolitan areas. See Gerrit Knaap and ArthurC. Nelson, The Regulated Landscape: Lessons on StateLand Use Planning from Oregon (Cambridge, Massa-chusetts: Lincoln Institute of Land Policy, 1992), pp.87-91.

28 Portland Metro, Urban Growth Report Addendum, Fig-ure 2, p. 3. This includes development of new land,infill, and projected redevelopment of existing land.

29 Section 3.07.120 of Metro’s function plan establishesminimum density criteria of between 5,000 and 6,500square foot lots and that “no development application .. . may be approved unless the development will resultin the building of 90 percent or more of the maximumnumber of dwelling units per net acre permitted by thezoning designation for the site.” Section 3.07.810 states“All cities and counties within the Metro boundary arehereby required to amend their comprehensive plansand implementing ordinances to comply with the provi-sion of this function plan within twenty-four months ofthe effective date of this ordinance.”

30 An initial survey by Metro estimated vacant land hadfallen to 50,000 acres. A new land inventory subse-

quently conducted by Metro found 55,000 vacant acres.See the discussion in Mildner, “Growth Management inthe Portland Region and the Housing Boom of the 1990s.”

31 Ibid. Metro estimates that 15,950 acres are unbuildablebecause they are in flood plains, environmentally sen-sitive areas, have steep slopes, or are part of a 200 footriver buffer. Another 15,080 acres will be needed forstreets, schools, parks, churches and other public fa-cilities. Thus, just 38,370 acres are available for resi-dential, commercial, and industrial development withinthe growth boundary. See Portland Metro, Urban GrowthReport, December 1997, pp. 11-17.

32 Sonny Conder, Residential Refill Study, Growth Man-agement Services Department Technical Report, Port-land Metro, February 10, 1999.

33 Conder, Residential Refill Study.

34 Ibid.

35 Ibid.

36 Sonny Conder and Karen Larson, “Residential Lot Val-ues and the Capital-Land Substitution Parameter—SomeRecent Results from the Portland Metro Area,” GrowthManagement Services Division, Portland Metro, un-published paper, May 1998. The sample consisted of5,500 records from Clackamas, Multnomah, and Wash-ington Counties in Oregon and 2,200 records fromClark County, Washington.

37 Ibid., p. 6. Conder and Larson speculate that this may bethe result of smaller household sizes and the increasein multiple-income households.

38 Ibid., pp. 6-7. The estimates ranged from 0.642 to 0.8,where 1 represents a 1:1 willingness to trade-off lotsize for home size and 0 represents no substitution. Inother words, a ratio of 1 would imply that, as homeprices increase, a 10% increase in home size would beequivalent to a 10% reduction in lot size.

39 Mildner, “Growth Management in the Portland Regionand the Housing Boom of the 1990s.”

40 “Metro Council’s Urban Growth Boundary ExpansionDecisions,” http://www.metro.dst.or.us/growth/ugbursa/ugbupdate.html.

41 Local real-estate consultant Jerald Johnson estimatesland take-up of 2,850 acres per year. See Jerald Johnson,Hobson, Johnson, and Associates, memorandum datedOctober 24, 1997. This estimate is higher than Metro’sestimate of about 2,000 gross buildable acres per year.See Portland Metro, Urban Growth Report Addendum,p. 14.

42 This was Goal 10 in the statewide growth-managementlaw. See Gerrit J. Knaap, “Self-Interest and Voter Sup-port for Oregon’s Land Use Controls,” Journal of theAmerican Planning Association, vol. 53, no. 1 (Winter1987), pp. 92-97; Gerrit J. Knaap, “The Political Economyof Growth Management in Oregon.”

43 For homes, the affordable housing threshold was peggedat $125,000. For apartments, the threshold was $500per month. Jerald Johnson, “Issues Associated with theImposition of Inclusionary Zoning in the Portland Metro-politan Area,” Hobson, Johnson, and Associates, Port-land, Oregon, unpublished paper, December 1, 1997.

44 Arthur C. Nelson, “Preserving Prime Farmland in theFace of Urbanization: Lessons from Oregon,” Journal ofthe American Planning Association, vol. 58, no. 4 (Au-tumn 1992), p. 479.

45 Nelson, “Preserving Prime Farmland in the Face ofUrbanization,” p. 479.

Reason Public Policy Institute [RPPI]3415 S. Sepulveda Blvd., Suite 400 ◆ Los Angeles, CA 90034

310-391-2245 ◆ 310-391-4395 [fax]www.rppi.org

rppi - reason foundation · samuel r. staley, ph.d., is director of the urban futures program () at...

Documents