tcrp report 39 - transportation research...

T R A N S I T C O O P E R A T I V E R E S E A R C H P R O G R A M

SPONSORED BY

The Federal Transit Administration

TCRP Report 39

The Costs of Sprawl—Revisited

Transportation Research BoardNational Research Council

TCRP OVERSIGHT AND PROJECTSELECTION COMMITTEE

CHAIRROBERT G. LINGWOODBC Transit

MEMBERSGORDON AOYAGIMontgomery County GovernmentJ. BARRY BARKERTransit Authority of River CityLEE BARNESBarwood, Inc.RONALD L. BARNESCentral Ohio Transit AuthorityGERALD L. BLAIRIndiana County Transit AuthorityROD J. DIRIDONIISTPSSANDRA DRAGGOOCATACONSTANCE GARBERYork County Community Action Corp.ALAN J. GIBBSRutgers, The State Univ. of New JerseyDELON HAMPTONDelon Hampton & AssociatesKATHARINE HUNTER-ZAWORSKIOregon State UniversityJOYCE H. JOHNSONNorth Carolina A&T State UniversityALAN F. KIEPPERParsons Brinckerhoff, Inc.PAUL LARROUSSEMadison Metro Transit SystemEVA LERNER-LAMThe Palisades Consulting Group, Inc.GORDON J. LINTONFederal Transit AdministrationDON S. MONROEPierce TransitPATRICIA S. NETTLESHIPThe Nettleship Group, Inc.JAMES P. REICHERTReichert Management ServicesRICHARD J. SIMONETTAMARTAPAUL P. SKOUTELASPort Authority of Allegheny CountyPAUL TOLIVERKing County DOT/MetroMICHAEL S. TOWNESPeninsula Transportation Dist. Comm.LINDA S. WATSONCorpus Christi RTA

EX OFFICIO MEMBERSWILLIAM W. MILLARAPTAKENNETH R. WYKLEFHWAFRANCIS B. FRANCOISAASHTOROBERT E. SKINNER, JR.TRB

TDC EXECUTIVE DIRECTORLOUIS F. SANDERSAPTA

SECRETARYROBERT J. REILLYTRB

TRANSPORTATION RESEARCH BOARD EXECUTIVE COMMITTEE 1998

OFFICERS

Chairwoman: Sharon D. Banks, General Manager, AC TransitVice Chairman: Wayne Shackelford, Commissioner, Georgia Department of TransportationExecutive Director: Robert E. Skinner, Jr., Transportation Research Board

MEMBERS

THOMAS F. BARRY, JR., Secretary of Transportation, Florida Department of TransportationBRIAN J. L. BERRY, Lloyd Viel Berkner Regental Professor, Bruton Center for Development Studies,

University of Texas at DallasSARAH C. CAMPBELL, President, TransManagement, Inc., Washington, DCE. DEAN CARLSON, Secretary, Kansas Department of TransportationJOANNE F. CASEY, President, Intermodal Association of North America, Greenbelt, MDJOHN W. FISHER, Director, ATLSS Engineering Research Center, Lehigh UniversityGORMAN GILBERT, Director, Institute for Transportation Research and Education, North Carolina

State UniversityDELON HAMPTON, Chair and CEO, Delon Hampton & Associates, Washington, DCLESTER A. HOEL, Hamilton Professor, Civil Engineering, University of VirginiaJAMES L. LAMMIE, Director, Parsons Brinckerhoff, Inc., New York, NYTHOMAS F. LARWIN, General Manager, San Diego Metropolitan Transit Development BoardBRADLEY L. MALLORY, Secretary of Transportation, Pennsylvania Department of TransportationJEFFREY J. MCCAIG, President and CEO, Trimac Corporation, Calgary, Alberta, CanadaJOSEPH A. MICKES, Chief Engineer, Missouri Department of TransportationMARSHALL W. MOORE, Director, North Dakota Department of TransportationANDREA RINIKER, Executive Director, Port of TacomaJOHN M. SAMUELS, VP—Operations Planning & Budget, Norfolk Southern Corporation, Norfolk, VALES STERMAN, Executive Director, East-West Gateway Coordinating Council, St. Louis, MOJAMES W. VAN LOBEN SELS, Director, CALTRANSMARTIN WACHS, Director, University of California Transportation Center, University of California

at BerkeleyDAVID L. WINSTEAD, Secretary, Maryland Department of TransportationDAVID N. WORMLEY, Dean of Engineering, Pennsylvania State University (Past Chair, 1997)

EX OFFICIO MEMBERS

MIKE ACOTT, President, National Asphalt Pavement AssociationJOE N. BALLARD, Chief of Engineers and Commander, U.S. Army Corps of EngineersANDREW H. CARD, JR., President and CEO, American Automobile Manufacturers AssociationKELLEY S. COYNER, Administrator, Research and Special Programs, U.S. Department ofTransportationMORTIMER L. DOWNEY, Deputy Secretary, Office of the Secretary, U.S. Department ofTransportationFRANCIS B. FRANCOIS, Executive Director, American Association of State Highway andTransportation OfficialsDAVID GARDINER, Assistant Administrator, U.S. Environmental Protection AgencyJANE F. GARVEY, Administrator, Federal Aviation Administration, U.S. Department ofTransportationCLYDE J. HART, JR., Maritime Administrator, U.S. Department of TransportationROBERT A. KNISELY, Deputy Director, Bureau of Transportation Statistics, U.S. Department ofTransportationGORDON J. LINTON, Federal Transit Administrator, U.S. Department of TransportationRICARDO MARTINEZ, National Highway Traffic Safety Administrator, U.S. Department ofTransportationWALTER B. MCCORMICK, President and CEO, American Trucking Associations, Inc.WILLIAM W. MILLAR, President, American Public Transit AssociationJOLENE M. MOLITORIS, Federal Railroad Administrator, U.S. Department of TransportationKAREN BORLAUG PHILLIPS, Senior Vice President, Association of American RailroadsVALENTIN J. RIVA, President, American Concrete Pavement AssociationGEORGE D. WARRINGTON, Acting President and CEO, National Railroad Passenger CorporationKENNETH R. WYKLE, Federal Highway Administrator, U.S. Department of Transportation

TRANSIT COOPERATIVE RESEARCH PROGRAM

Transportation Research Board Executive Committee Subcommittee for TCRPSHARON D. BANKS, AC Transit (Chairwoman)LESTER A. HOEL, University of VirginiaTHOMAS F. LARWIN, San Diego Metropolitan Transit Development BoardGORDON J. LINTON, U.S. Department of TransportationWILLIAM W. MILLAR, American Public Transit AdministrationWAYNE SHACKELFORD, Georgia Department of TransportationROBERT E. SKINNER, JR., Transportation Research BoardDAVID N. WORMLEY, Pennsylvania State University

T R A N S I T C O O P E R A T I V E R E S E A R C H P R O G R A M

Report 39

The Costs of Sprawl—Revisited

ROBERT W. BURCHELLNAVEED A. SHADDAVID LISTOKINHILARY PHILLIPS

Center for Urban Policy ResearchRutgers University

ANTHONY DOWNSThe Brookings Institution

SAMUEL SESKINJUDY S. DAVIS

Parsons Brinckerhoff Quade and Douglas, Inc.

TERRY MOOREDAVID HELTON

MICHELLE GALLECONorthwest

Subject Areas

Public Transit

Research Sponsored by the Federal Transit Administration inCooperation with the Transit Development Corporation

TRANSPORTATI ON RESEARCH BO ARDNATIONAL RESEARCH COUNCIL

NATIONAL ACADEMY PRESSWashington, D.C. 1998


The nation's growth and the need to meet mobility,environmental, and energy objectives place demands on publictransit systems. Current systems, some of which are old and inneed of upgrading, must expand service area, increase servicefrequency, and improve efficiency to serve these demands.Research is necessary to solve operating problems, to adaptappropriate new technologies from other industries, and tointroduce innovations into the transit industry. The TransitCooperative Research Program (TCRP) serves as one of theprincipal means by which the transit industry can developinnovative near-term solutions to meet demands placed on it.

The need for TCRP was originally identified in TRB SpecialReport 213—Research for Public Transit: New Directions,published in 1987 and based on a study sponsored by the UrbanMass Transportation Administration—now the Federal TransitAdministration (FTA). A report by the American Public TransitAssociation (APTA), Transportation 2000, also recognized theneed for local, problem-solving research. TCRP, modeled after thelongstanding and successful National Cooperative HighwayResearch Program, undertakes research and other technicalactivities in response to the needs of transit service providers. Thescope of TCRP includes a variety of transit research fieldsincluding planning, service configuration, equipment, facilities,operations, human resources, maintenance, policy, andadministrative practices.

TCRP was established under FTA sponsorship in July 1992.Proposed by the U.S. Department of Transportation, TCRP wasauthorized as part of the Intermodal Surface TransportationEfficiency Act of 1991 (ISTEA). On May 13, 1992, amemorandum agreement outlining TCRP operating procedureswas executed by the three cooperating organizations: FTA; theNational Academy of Sciences, acting through theTransportation Research Board (TRB); and the TransitDevelopment Corporation, Inc. (TDC), a nonprofit educationaland research organization established by APTA. TDC isresponsible for forming the independent governing board,designated as the TCRP Oversight and Project Selection (TOPS)Committee.

Research problem statements for TCRP are solicitedperiodically but may be submitted to TRB by anyone at any timeIt is the responsibility of the TOPS Committee to formulate theresearch program by identifying the highest priority projects. Aspart of the evaluation, the TOPS Committee defines funding levelsand expected products.

Once selected, each project is assigned to an expert panel,appointed by the Transportation Research Board. The panelsprepare project statements (requests for proposals), selectcontractors, and provide technical guidance and counselthroughout the life of the project. The process for developingresearch problem statements and selecting research agencies hasbeen used by TRB in managing cooperative research programssince 1962. As in other TRB activities, TCRP project panels servevoluntarily without compensation.

Because research cannot have the desired impact if products failto reach the intended audience, special emphasis is placed ondisseminating TCRP results to the intended end users of theresearch: transit agencies, service providers, and suppliers. TRBprovides a series of research reports, syntheses of transit practice,and other supporting material developed by TCRP research.APTA will arrange for workshops, training aids, field visits, andother activities to ensure that results are implemented by urbanand rural transit industry practitioners.

The TCRP provides a forum where transit agencies cancooperatively address common operational problems. The TCRPresults support and complement other ongoing transit research andtraining programs.

TCRP REPORT 39

Project H-10 FY'95ISSN 1073-4872ISBN 0-309-06306-XLibrary of Congress Catalog Card No. 98-61200

1998 Transportation Research Board

Price $62.00

NOTICE

The project that is the subject of this report was a part of the TransitCooperative Research Program conducted by the Transportation ResearchBoard with the approval of the Governing Board of the National ResearchCouncil. Such approval reflects the Governing Board's judgment that theproject concerned is appropriate with respect to both the purposes andresources of the National Research Council.

The members of the technical advisory panel selected to monitor thisproject and to review this report were chosen for recognized scholarlycompetence and with due consideration for the balance of disciplinesappropriate to the project. The opinions and conclusions expressed orimplied are those of the research agency that performed the research, andwhile they have been accepted as appropriate by the technical panel, theyare not necessarily those of the Transportation Research Board, theNational Research Council, the Transit Development Corporation, or theFederal Transit Administration of the U.S. Department of Transportation.

Each report is reviewed and accepted for publication by the technical panelaccording to procedures established and monitored by the TransportationResearch Board Executive Committee and the Governing Board of theNational Research Council.

To save time and money in disseminating the research fundings, the reportis essentially the original text as submitted by the research agency. Thisreport has not been edited by TRB.

Special Notice

The Transportation Research Board, the National Research Council, theTransit Development Corporation, and the Federal Transit Administration(sponsor of the Transit Cooperative Research Program) do not endorseproducts or manufacturers. Trade or manufacturers' names appear hereinsolely because they are considered essential to the clarity andcompleteness of the project reporting.

Published reports of the


are available from:

Transportation Research BoardNational Research Council2101 Constitution Avenue, N.W.Washington, D.C. 20418

and can be ordered through the Internet athttp://www.nas.edu/trb/index.html

Printed in the United States of America

FOREWORDBy Staff

Transportation ResearchBoard

TCRP Report 39 will be of interest to individuals involved in ongoingdiscussions and debates about urban sprawl and its effects. This report is a literaturereview that represents the culmination of the first phase of TCRP Project H-10, "TheCosts of Sprawl—Revisited." The report was prepared by Rutgers University'sCenter for Urban Policy Research, in conjunction with The Brookings Institution,Parsons Brinckerhoff Quade and Douglas, Inc., and ECONorthwest. Urban sprawl isa topic that interests urban planners, economists, environmentalists, sociologists,transportation professionals, policymakers and public officials, academics in manyfields, and the general public.

Regardless of the focus of any particular debate or discussion on urban sprawl,most such discussions attempt to define sprawl and address whether it is "good" or"bad." Consequently, Chapter 1 of Section I of TCRP Report 39 provides a workingdefinition of sprawl and its associated costs. The next chapter provides historicaldiscussion, dating back to the early 1920s when zoning acts were initiallydeveloped, and to the 1950s when the term "sprawl" entered the planning literature.As indicated by the title of this research project, the seminal 1974 report The Costsof Sprawl, prepared by the Real Estate Research Corporation, serves as aspringboard for this research effort.

Section II of the report contains the Literature Synthesis. This sectionsystematically presents the literature on sprawl in chapters that focus on thefollowing major areas of impact:

• Public/private capital and operating costs,• Transportation and travel costs,• Land/natural habitat preservation,• Quality of life, and• Social issues.

Throughout this section, the research team discusses the literature and identifies theextent to which there is agreement and disagreement about the premises andconclusions.

Section III of the report presents annotations of studies, organized in chaptersthat focus on the same five major impact areas as Section II.

While this report will not resolve the debate on the benefits and costs of urbansprawl, it provides an important repository of information for the debaters.

CONTENTS i PREFACE

1 SECTION I Introduction

5 CHAPTER 1 Defining Sprawl

9 CHAPTER 2 Historical Overview

41 SECTION II Literature Synthesis

45 CHAPTER 3 Public/Private Capital and Operating Costs

61 CHAPTER 4 Transportation and Travel Costs

73 CHAPTER 5 Land/Natural Habitat Preservation

83 CHAPTER 6 Quality of Life

103 CHAPTER 7 Social Issues

113 CHAPTER 8 Overall Summary of the Literature

133 SECTION III Annotations of Studies

135 CHAPTER 9 Public/Private Capital and Operating Costs

157 CHAPTER 10 Transportation and Travel Costs

183 CHAPTER 11 Land/Natural Habitat Preservation

195 CHAPTER 12 Quality of Life

205 CHAPTER 13 Social Issues

231 REFERENCES

257 LIST OF ANNOTATED STUDIES BY AUTHOR

265 INDEX

http://nationalacademies.org/trb/publications/tcrp/tcrp_rpt_39-f.pdf

http://nationalacademies.org/trb/publications/tcrp/tcrp_rpt_39-e.pdf

http://nationalacademies.org/trb/publications/tcrp/tcrp_rpt_39-d.pdf

http://nationalacademies.org/trb/publications/tcrp/tcrp_rpt_39-c.pdf

http://nationalacademies.org/trb/publications/tcrp/tcrp_rpt_39-b.pdf

COOPERATIVE RESEARCH PROGRAMS STAFF

ROBERT J. REILLY, Director, Cooperative Research ProgramsSTEPHEN J. ANDRLE, Manager, Transit Cooperative Research ProgramDIANNE S. SCHWAGER, Senior Program OfficerEILEEN P. DELANEY, Managing EditorJAMIE M. FEAR, Associate Editor

PROJECT PANEL H-10GORDON AOYAGI, Montgomery County Government, Rockville, MD (Chair)SCOTT BAKER, KPMG Peat Marwick, McLean, VARICHARD G. BICKEL, JR., SEPTA, Philadelphia, PATHOMAS BLACK, Lincoln Institute of Land Policy, Falls Church, VALEWIS BOLAN, Bolan Smart, Vienna, VAGARY L. BROSCH, Center for Urban Transportation Research, Tampa, FLGREG BYRNE, City of Fort Collins, ColoradoDON CHEN, Surface Transportation Policy Project, Washington, DCELIZABETH DEAKIN, University of California, BerkeleyPATRICK T. DECORLA-SOUZA, FHWAPATRICIA S. NETTLESHIP, The Nettleship Group, Inc., Santa Monica, CAJANET OAKLEY, NARC and AMPO, Washington, DCKENNETH J. WARREN, Milwaukee County Transit SystemRICHARD P. STEINMANN, FTA Liaison RepresentativeJAMES A. SCOTT, TRB Liaison Representative

The Costs of Sprawl–Revisited Preface

Rutgers • Brookings • Parsons Brinckerhoff • ECONorthwest i TRANSIT COOPERATIVE RESEARCH PROGRAM(TCRP) H-10

PREFACE

In 1974, the Real Estate ResearchCorporation published a three-volumestudy entitled The Costs of Sprawl. Thestudy consisted of an Executive Summary,Detailed Costs Analysis (Volume I), andLiterature Review/Bibliography (VolumeII). It encompassed more than onethousand pages. From the time of itspublication until today, it has beenregarded by the social science communityas one of the most significant critiques ofsprawl and among the most influentialstudies ever undertaken. The Costs ofSprawl has been cited in countlessenvironmental and planning reports andjournals; it has been reviewed—bothpositively and negatively—by more thanone hundred journals and magazines; andit has been presented as the seminal studyon growth impacts to numerousCongressional committees and bodies.The Costs of Sprawl was funded jointly bythe U.S. Council on EnvironmentalQuality, the Department of Housing andUrban Development, and theEnvironmental Protection Agency.

The Costs of Sprawl, like no other studybefore, sought to isolate both density andlocation of development as significantcontributors to the costs of development.

The study analyzed six hypothetical newcommunities of 10,000 dwelling unitseach, from high density (19-20 units peracre) to intermediate density (3-4 units peracre); from communities with high levelsof planning and design to those lackingsignificant planning. The study analyzedimpacts on infrastructure, housing,transportation, energy, environmental, andquality of life costs of sprawl (Real EstateResearch Corporation [RERC] 1974).

Although The Costs of Sprawl wasinfluential, it was also flawed. Theanalyses of community types allowed unitsize and number of occupants to vary, andthe savings attributed to differentcommunity types were actually a functionof the differing size (and types) of unitsand numbers of people found there. Theabsence of sprawl was not the reason forthe savings; smaller units and fewerpeople to service were the cause of thesavings. Yet, even though theseshortcomings were uncovered, thedirection of the findings so paralleled pastand current intuitive feelings that thestudy continues to be used twenty-fiveyears later as one of the most cogentarguments against sprawled developmentpatterns.

The Costs of Sprawl–Revisited Preface

Rutgers • Brookings • Parsons Brinckerhoff • ECONorthwest ii TRANSIT COOPERATIVE RESEARCH PROGRAM(TCRP) H-10

Why such interest in sprawl? AlthoughAmericans like their single-familyresidences, automobiles, and suburbanlifestyles, there is a nagging feeling thatboth the aesthetics of how communitiesdevelop and the efficiency of movementwithin and between them could beimproved. In addition, buried down deepis a recognition that Americans arewasteful in their consumption ofmanmade (infrastructure) and natural(land) resources, and that theirdevelopment choices are selfish in termsof impacts on central cities and thepopulations within them. But first it mustbe shown to the citizenry at large thatthere is a problem, because life is goodand "If it ain't broke, don't fix it." Issuburban sprawl different from analternative form of development? Is it lessefficient? Does it cause resources to beneedlessly consumed? Is there analternative? What do those who havestudied this issue say? How substantivelystrong is the evidence they bring to bear?

The study that follows is a detailedexamination of most of the informationthat can be assembled on both sprawl andits costs in an effort to answer the abovequestions. The monograph views the costsof sprawl (with lower-case letters) asinvestigated in a variety of types andforms of about 500 studies. These studiesvary between those that: (1) focusspecifically on sprawl, and those that dealwith suburban or exurban development;(2) are highly quantitative, involvingmodeling or econometric analyses, andthose that are qualitative and purelydescriptive; (3) concern the "harder" orphysical/engineering aspects of sprawl,and those that substantively involve"softer" or quality of life/social issues; (4)are primary analyses and break newground, and those that are secondaryanalyses of the works of others that addvery little; and (5) vilify sprawl and see nopositive effects, and those that champion

the development form as purely andunequivocally "American" with few, ifany, negative impacts.

With regard to the latter, this assemblageof material identifies and providesevidence for both negative and positiveimpacts of sprawl in each of five impactcategories. These are: (1) public andprivate capital and operating costs; (2)transportation and travel costs; (3)land/natural habitat preservation; (4)quality of life; and (5) social issues.

The work contained in this monograph isdivided into three sections and thirteenchapters. Section I contains two chaptersthat provide an introduction to, and anhistorical overview of, sprawl's "growth."Chapter One contains an introduction tothe concept of sprawl, including itsdefining traits; Chapter Two highlightssignificant events in the evolution of thesprawl literature. Section II is a synthesisof the literature of sprawl's impacts: Towhat degree can the impact be recognized,and what is its relation to sprawl? Itdivides sprawl's impacts—more than 40 intotal, two-thirds negative and one-thirdpositive—into the above five impactcategories. The first five chapters of thissection discuss each of the abovecategories of impacts. The sixth chapter inthis section presents in summary forminformation from the previous chaptersboth quantitatively and qualitatively.Section III individually annotatesapproximately one-quarter of the sprawlliterature. Again, this section is brokendown into five chapters according to thefive sprawl impact categories.

The review of the sprawl literature isdesigned to be historical, substantive,comprehensive, and integrative. Presentedin this way, the reader will be drawn intothe argument about sprawl from its originsto the present.

The Costs of Sprawl–Revisited Introduction

Rutgers • Brookings • Parsons Brinckerhoff • ECONorthwest 1 TRANSIT COOPERATIVE RESEARCH PROGRAM(TCRP) H-10

SECTION

INTRODUCTION

The literature review that follows is ananalysis of the writings and studiesconcerning a pattern of land developmentin the United States termed "sprawl."Sprawl is the spread-out, skipped-overdevelopment that characterizes the non-central city metropolitan areas and non-metropolitan areas of the United States.Sprawl is one- or two-story, single-familyresidential development on lots ranging insize from one-third to one acre (lessacreage on the West Coast), accompaniedby strip commercial centers and industrialparks, also two stories or less in heightand with a similar amount of land takings(Ewing 1997).

Sprawl occurs on a micro basis in almostevery county of the United States(although it occurs in significant amountsin only about one-fifth of the nation's3,200 counties). Sprawl also occurs inWestern and Eastern European, LatinAmerican, and Asian countries inresponse to increased affluence and

growing dependence on the automobile asthe preferred method of intra- and inter-metropolitan travel. Most United Statescounties that contain sprawl have it in itsresidential form—i.e., low-densityresidential development in rural andundeveloped areas. Some counties arecharacterized by nonresidential sprawl,commercial and industrial developmentwith floor-area ratios less than 0.2 locatedin the same types of areas (Burchell andShad 1998).

Sprawl is the spread-out,skipped-over developmentthat characterizes the non-centralcity metropolitan areas andnon-metropolitan areasof the United States. — Ewing 1997

Sprawl occurs, in part, because localgovernments in the United Statesencourage this form of development viazoning and subdivision ordinances which,in turn, reflect the desires of a large shareof their citizenry. This type of

I



development is favored by the generalpublic because it (among other factors):

1) dilutes congestion whileaccommodating unlimited use of theautomobile;

2) distances new development from thefiscal and social problems of oldercore areas;

3) provides a heterogeneous economicmix;

4) fosters neighborhoods in whichhousing will appreciate;

5) fosters neighborhoods in whichschools provide both education andappropriate socialization for youth;and

6) requires lower property taxes to payfor local and school district operatingexpenses than locations closer in.(Burchell 1997a)

Sprawl is so well-accepted by the publicthat the AAA-rated locations for bothresidential and nonresidentialdevelopment are increasingly farther outrather than closer in, and more rather thanless segregated by type of land use(Gordon and Richardson 1997a). Gatedcommunities, farmettes, research parks,

Large regional malls, initially located alongundeveloped highway interchanges, stimulaterapid additional surrounding development.Source: Constance Beaumont, NTHP.

law offices, medical groups,megahardware and home improvementstores, theatrical and comedy clubs, newand used car lots, and restaurants all nowseek peripheral locations in pursuit of

their markets. The move to the far reachesof the metropolitan area began withsingle-family subdivisions; shoppingcenters and garden apartments sprang upnext; then research and industrial parks;then restaurants and entertainmentfacilities; and finally, discounters of everyform.

The unique aspect of all this developmentis that few entities have ever failedbecause their outward locational decisionswere in the wrong direction. Occasionally,a retailer or a residential development hasgone under because an exit on theinterstate or beltway wasn't developed asplanned, but rarely has an economic entityfailed in the United States because it wasdeveloped too far out.

The move to the far reaches of themetropolitan area began withsingle-family subdivisions;shopping centers and gardenapartments sprang up next; thenresearch and industrial parks; thenrestaurants and entertainmentfacilities; and finally, discountersof every form.

The newest and soon-to-be one of themost successful airports in the UnitedStates is 33 miles from the city of Denver;a taxi ride from the airport baggage claimto the downtown Hyatt costs $40. Is thisan anomaly? No. Cincinnati's new airportis so far from the downtown that it is noteven in the same state! Both airports havealready drawn nonresidential developmentand are now drawing residentialdevelopment to their edges. Both are tensof miles from the nearest existingdevelopment of these types. But neithercan justify its location solely on flightpattern interference with residentialenvironments. Instead, the locations werechosen for exactly the same reason otherland use locations are chosen: anabundance of land was available, and it



was both relatively inexpensive and easyto assemble.

If sprawl is so desirable, why should thecitizens of the United States acceptanything else? The answer is that they nolonger can pay for the infrastructurenecessary to develop farther and fartherout in metropolitan areas. In the state ofSouth Carolina, if sprawl continuesunchecked, statewide infrastructure costsfor the period 1995 to 2015 are projectedto be more than $56 billion, or $750 percitizen per year for the next twenty years.In addition to a massive infrastructureconservation program and the adoption ofnumerous technological cost savers,funding infrastructure in this state couldrequire an increase in the gasoline tax of

c2/ /gallon; an increase in the state salestax of 0.5%; an increase in property taxesof 12.5%; the tolling of all interstates at30-mile intervals; impact fees onresidential and nonresidentialdevelopment of $2,000 per unit and per1,000 square feet, respectively; and amandatory 10 percent set-aside forinfrastructure in all state, county,municipal, and school district generalfunds and intergovernmental transferrevenues (Burchell 1997b).

Despite massive road expenditures, I-395 inArlington, Virginia, slows to a gridlock duringrush-hour traffic.Source: Virginia Department of Transportation.

The big-ticket item in all infrastructureprojections is roads. In South Carolina,roads are expected to cost $25 billion,almost half of the total $56 billioninfrastructure budget. In South Carolina,

roads will cost 2.5 times what will bespent on primary, secondary, and highereducation infrastructure; three times whatwill be spent on health infrastructure,including all hospitals, institutions, and allwater-sewer treatment systems; ten timeswhat will be spent on public safety,administration, and justice infrastructure;fifteen times what will be spent onenvironmental protection infrastructure;and twenty-five times what will be spenton all cultural and recreationalinfrastructure.

Dually supporting and underutilizing twosystems of infrastructure—one that isbeing abandoned in and around centralcities and close-in suburbs, and one that isnot yet fully used in rural areas justbeginning to be developed—is causinggovernments to forgo the maintenance ofmuch infrastructure and the provision ofanything other than growth-relatedinfrastructure. The United States, in otherwords, is funding road infrastructure by:

1) not funding all infrastructure;2) not fully funding developmental

infrastructure;3) not repairing or replacing most types

of infrastructure; and4) not taking advantage of the

technological improvements inrehabilitation, repair, and provision ofinfrastructure that could be passed onto taxpayers as savings.

Still, by no means is an alternative to thecurrent pattern of land development thepanacea. If South Carolina were to switchto compact development and managedgrowth measures to curtail spreaddevelopment, the state would be able tosave only about 10 percent of theprojected $56 billion infrastructure costs,or approximately $5.6 billion. This isbecause about 40 percent of publicinfrastructure costs are not growth-related,and only about two-thirds of theremainder is new growth-related. When



development pattern savings are appliedto the appropriate portion of new growth-related infrastructure costs, therefore, thesaving is only 12–15 percent.

On the other hand, increasing the gasolinetax by c2/ /gallon in South Carolina,would have raised only $56 million innew revenues statewide—one one-thousandth of the total requiredinfrastructure costs—and one one-hundredth of the amount that potentiallycould be saved by altering landdevelopment patterns (Burchell 1997b).

In sum, most of the American public isnot unhappy with the current pattern ofdevelopment in metropolitan areas—itsimply can no longer afford it. Thus, theprimary concern about sprawldevelopment, at a time when the averageAmerican is satisfied with its outcome, iscost. And costs need to be measured notjust in terms of capital improvement butalso in terms of resource depletion. Landin the United States is being consumed attriple the rate of household formation;

automobile use is growing twice as fast asthe population; and prime agriculturalland, forests, and fragile landsencompassing natural habitats aredecreasing at comparable reciprocal rates(Landis 1995).

In sum, most of the Americanpublic is not unhappy with thecurrent pattern of development inmetropolitan areas—it simply canno longer afford it.

As a result, the professionaltransportation and city planningcommunities are beginning to look atsprawl to determine whether analternative to this growth pattern can beconceived, and even more importantly,whether it makes sense to pursue analternative pattern of growth. Does anyalternative pose a viable option to currentmethods and forms of metropolitandevelopment? A significant literature hasdeveloped in this area and is overviewedin this section.

The Costs of Sprawl–Revisited Defining Sprawl


CHAPTER

DEFINING SPRAWL

Sprawl, in its broadest sense, has longbeen an American zeitgeist. Alexis deTocqueville, touring the United States inthe early 1800s, observed "no urbangrowth boundaries," but rather marveledat "America ... where everything is inconstant motion ... and where noboundaries were set to the efforts of man."Today's sprawl is the frontier of long ago;it is akin to the post-war suburb—both ofwhich have been extolled as definingAmerican influences.John Delafons, Fellow at theHarvard/MIT Joint Center in 1961, choseas a research topic a comparison of Britishand American land-use controls. Hiswork, Land Use Controls in America,provides an insightful look at the growthof the U.S. "system" of controls from1920 to 1960 by an outsider who camefrom a country with a very formal systemof land-use controls.

Delafons describes the U.S. system ofmaster planning, zoning, and subdivisioncontrol as heavily influenced by a "prairiepsychology." He explains that U.S.development patterns are characterizedby:

a) a supply of land which is viewed asvirtually unlimited;

b) land that is open to all and propertyownership rights that are encouragedand protected by the U.S.Constitution;

c) economic forces that are barelyunderstood and should not betampered with;

d) development professionals whoprepare land for development and donot question whether the land shouldbe developed (i.e., they make sureutilities are in place and feeder roadshave been planned for); and

e) a basic distrust of elected andappointed officials, so that allprocedures are codified anddevelopment that qualifies underthese procedures does so "as ofright," with minimal public review.(Delafons 1962)

U.S. development controls, he claims, are"static" and thus lack the ability to controltempo (timing) and sequence (whichlocation first) of development.Development is free to wander and to takeplace incrementally in jurisdictions in theUnited States because existing land use

1



controls allow this to happen (Delafons1962).

Many agree with Delafons' insight.Although some view contemporarydevelopment patterns as a reflection of theinvisible but sure hand of the market(Gordon and Richardson 1997a), theunbridled movement outward of leapfrog,low-density development is increasinglybeing viewed as an American ill(Richmond 1995). Sprawl has taken onboth a pejorative as well as a descriptiveconnotation, an intermixing that makes abalanced discussion, which attempts todisentangle the costs and benefits ofsprawl, difficult.

U.S. development controls are"static" and thus lack the ability tocontrol tempo (timing) andsequence (which location first) ofdevelopment.

The shift to the suburbs has, of course,been manifest for more than half acentury. In 1940, only 15 percent of theUnited States population resided in thesuburbs (defined as metropolitan areasoutside of central cities). As themillennium approaches, about 60 percentof the population is counted as suburban.Even the most vehement critics of sprawlrecognize that suburban and exurbangrowth patterns have been and willcontinue to be inescapable developmentforms in the United States. The recentpopulation increase of some 20 millionpeople per decade is likely to continue forat least the next quarter-century. As aresult, there will continue to be skipped-over development in rural andundeveloped areas. It would be totallyunrealistic to expect even a moderateshare of growth to occur solely in alreadybuilt-up neighborhoods in cities or inclose-by inner suburbs. Even the suburbsare being bypassed now by developmentseeking locations at the fringe of

metropolitan areas (Nelson and Sanchez1997).

A WORKING DEFINITION OFSPRAWL

Density, or more specifically, low density,is one of the cardinal definingcharacteristics of sprawl. But density hasto be set in context; cross-cultural andplace-oriented differences factor into thedefinition of sprawl. Densities in theUnited States overall are roughly one-tenth what they are in Western Europe; inturn, Western European density is muchlower than that of Japan and only afraction of what is found in such locationsas Hong Kong and Indonesia (Jackson1985). And in all of the above locations,suburban densities are lower than thedensities of central cities. Sprawl is notsimply development at less-than-maximum density; rather, it refers todevelopment that, given a national andregional framework (i.e., suburbs invarious locations of the United States), isat a low relative density, and one that maybe too costly to maintain.

Sprawl refers to a particular type ofsuburban peripheral growth. It refers todevelopment that expands in an unlimitedand noncontiguous (leapfrog) wayoutward from the solidly built-up core ofa metropolitan area. In terms of land-usetype, sprawl includes both residentialand nonresidential development.Residential development containsprimarily single-family housing,including significant numbers of distantunits scattered in outlying areas.Nonresidential development includesshopping centers, strip retail outlets alongarterial roads, industrial and office parks,and free-standing industrial and officebuildings, as well as schools and otherpublic buildings.



These different types of land uses are, forthe most part, spatially segregated fromone another. The components of thisdevelopment are individually located insmall subdivisions in zoning districts.Within each district, usually only one typeof use is permitted—e.g., single-familyresidential, shopping centers, stripcommercial, industrial, or office parks.

Another of sprawl's distinguishing traits isits consumption of exurban agriculturaland other frail lands in abundance; theseare the types of land found at theperiphery of development. The loss ofagricultural acreage takes place insignificant amounts because it often is thecheapest land available for development.Fragile environmental lands areswallowed up because they are part of theotherwise developable tracts. These tractswould not be developed if theenvironment was adequately protected.

Under sprawl conditions, there is almosttotal reliance upon the automobile as ameans of accessing the individual landuses. Seventy years ago, the streetcar wasthe most popular form of transportation tothe suburbs. Nowadays the automobile isthe most efficient means of accessingsprawl's outward extension and skipped-over development. For seven-day-a-weekbusiness and recreational use, includingboth at-peak and off-peak use, nothing canmatch the automobile for cost, efficiency,and versatility—at least in the short term.

Some analysts also include the smalldeveloper and a lack of integrated land-use planning as important aspects ofsuburban sprawl, and point to therelatively small residential subdivisionsand nonresidential site plans created byindividual developers operatingindependently of each other within thezoning districts of the 10,000 localgovernments found throughout the UnitedStates. The legal framework within which

sprawl occurs is fragmented intonumerous relatively small units,separately controlled by discrete localgovernments with unique rules andregulations. These localities have differentfiscal resources per capita (assessedvaluation of residential and nonresidentialproperties). Some are quite wealthy;others have limited ability to pay for localservices. The poorer localities are at asevere disadvantage when competing fordevelopment.

The automobile has replaced the streetcar,stimulating sprawl development.Source: Minnesota Historical Society.

Still, it must be stressed that sprawl isalmost impossible to separate from allconventional development. Even thoughone may be able to comprehend whatappears to be a better method ofdevelopment, it is difficult to translate thatmethod into practice.

Some components of sprawl are not easilymeasured. For example, although it ispossible to track residential single-familyand nonresidential commercial and retaildevelopment taking place at low densitiesin the United States, accessed byautomobiles in rural and undevelopedareas, this is the point at which almost alltracking stops. Measures of leapfrogdevelopment or development that isspatially segregated are virtuallyimpossible. Measures of how muchdevelopment is being delivered by small



developers in local jurisdictions isachievable but generally unproductive.

Finally, although a measure of grossresidential density (number of dwellingunits divided by area of jurisdiction) isavailable from several sources and canprovide some indication of land taken perdeveloped unit, the gross measure oftenmasks the actual land takings ofindividual new developments.

On the other hand, there is little evidenceto suggest that conventional developmentin a given location is anything other thanleapfrog, segregated, and land-consuming.Thus, sprawl development can becharacterized with some certainty as low-density residential and nonresidentialintrusions into rural and undevelopedareas, and with less certainty as leapfrog,segregated, and land-consuming in itstypical form.

A WORKING DEFINITION OF THECOSTS OF SPRAWL

The "costs" of sprawl have been talkedabout for decades, often without a fullunderstanding of what these costs are andto what level they should be assigned. Inthe original RERC (1974) Costs of Sprawlstudy, costs were calculated in sixdifferent substantive areas and assigned tothree different levels: infrastructure andtransportation costs were assigned to thecommunity, housing and quality-of-lifecosts to the individual, and energy andenvironmental costs to both thecommunity and to society as a whole(RERC 1974). This is a characteristic ofthe sprawl literature which is onlybeginning to be addressed at the end of atwenty-five-year observation period. Thework of Sam Seskin of ParsonsBrinckerhoff and Terry Moore ofECONorthwest on full-cost accounting oftransportation costs is breaking new

ground in viewing the totality of costs ofpublic policy decisions (ParsonsBrinckerhoff and ECONorthwest 1996).Their work is the exception. Most cost-accounting efforts assign sprawl costs toeither the easiest or the most commonlevel of measurement.

For definitional purposes, the "costs" ofsprawl are the resources expendedrelative to a type, density, and/or locationof development. These "costs" involvephysical, monetary, temporal, andsocial/psychological resources. Theyinvolve costs to the individual, to thecommunity, and to society. Most of thecosts specified to date are physical ormonetary, although occasionally socialcosts (e.g., the loss of upward mobility) orpsychological costs (e.g., the loss of senseof community) are documented.

There is little evidence to suggestthat conventional development in agiven location is anything otherthan leapfrog, segregated, andland-consuming.

The "benefits" of sprawl are mirrorimages of costs. They involve resourcegains due to type of development patternand include categories of gain similar tothose of losses stated above. This mightinvolve a temporal gain in suburb-to-suburb travel time because mostresidences and jobs are now bothsuburban, or monetary gains due toreduced housing costs also from buildingfarther out, or social gains such as theability to achieve homeownership, againdue to location in more distant places.

Costs and benefits are reported in the formthat the primary research provides. Inalmost all cases, these are costs at thecommunity level as opposed to costs at theindividual or societal levels, or benefits atany level.

The Costs of Sprawl–Revisited Historical Overview


CHAPTER

HISTORICAL OVERVIEW

Sensitivity to the consequences ofsprawl-like settlement predates thecoining of the term. The 1929 RegionalPlan of the New York Metropolitan area,for instance, warned of a steady decreasein farms and open-space acreage in theregion and underscored the need forsettlement patterns that encouraged "theface to face association that characterizedthe old village community" (RegionalPlan 1929, 23 and 216). At the same time,the Regional Plan spoke approvingly of"many carefully planned outersubdivisions with good features"(Regional Plan 1929, 1).

Concern about sprawl-like patterns ofdevelopment was appropriate at this time.The Standard Zoning Enabling Act(1922), drafted under the aegis ofSecretary of Commerce Herbert Hoover,the Standard City Planning Enabling Act(1928), and the legalization of zoning thatresulted from the 1926 Supreme Courtdecision (Euclid v. Amber Realty)unleashed a barrage of "model" zoningand planning-enabling legislation acrossthe United States. Euclidean zoning ofsegregated land uses and the emergence ofthe automobile began to establish the first

distant "suburbs" throughout the UnitedStates.

It was not until roughly the late 1950s andearly 1960s, however, that sprawl as aplanning term entered the literature. Theland development pattern it depicted wastypically criticized. Herbert Gans in TheLevittowners described Levittowndevelopment of the 1950s as "residentsliving in a sea of cell-like structures on a

Levittown, Pennsylvania: post-World War IIsuburbia.Source: Carl Byoir and Associates (New York).

Courtesy American Planning Association.

remote potato farm with cars spilling outof every street" (Gans 1967). In 1956, aCanadian planning study described urbansprawl as "scattered building

2



development" that had led to"inconveniences in the placement ofpublic and business facilities" (LowerMainland Regional Planning Board 1956).A year later, William H. Whyte,describing urban sprawl as leapfrog,scattered development, spoke of it as "aproblem that had reached nationalproportions" (Whyte 1957).

The political and social climate of theperiod, however, provided definitefinancial incentives for building homes inthe suburbs in the form of federallyinsured low-cost mortgages. This periodalso witnessed the massive federallysubsidized expansion of U.S. highways(1956 Interstate Highway Act), includingthe establishment of the interstate system.The new roadway system, together withthe growth in accessible, low-costmortgages, helped push development farbeyond the nation's central cities (APA1997). Relatively few people seriouslychallenged this new pattern of growth inthe outlying areas or questioned thechanges in central cities brought about bymulti-lane freeways.

Others soon entered the discussion,however. Marion Clawson, in 1962,described sprawl as a "lack of continuityin expansion," and noted it was bothfostered by, and contributed to, landspeculation (Clawson 1962). Similarliterature of the period, includingLessinger (1962), Harvey and Clark(1965), and Bahl (1968) viewed sprawl ascharacterized by such features aslowdensity, scattered, and leapfrogpatterns. Harvey and Clark (1965)identified the three cardinal traits ofsprawl as lowdensity, ribbon, and leapfrogdevelopment.

Even at this early stage, punditsacknowledged the difficulty in definingthe term sprawl. Writing in 1972, DavidMcKee and Gerald Smith observed that:

Urban sprawl is rather difficult to define.In some circles the term is thought to besynonymous with suburbia. Certainly theproblem exists in suburbia but suburbiaitself is not the problem. Some equatesprawl with expansion. But this type ofdefinition is not too helpful. (McKee andSmith 1972, 181-182)

McKee and Smith went on to describesprawl in four forms: 1) very low-densitydevelopment (i.e., two- to five-acrezoning); 2) ribbon-variety developmentextending along access routes; 3) leapfrogdevelopment; and 4) a "haphazardintermingling of developed and vacantland" (McKee and Smith 1972). Theauthors claimed that sprawl aggravatedsuburban problems (e.g., automobiledependence and the high cost of servicesand infrastructure) and also deleteriouslyaffected cities by depressing real estatevalues, among other things.

The political and social climate of theperiod provided definite financialincentives for building homes in thesuburbs.

Discussion of sprawl's effects transcendedeconomics. Although the 1973Rockefeller Brothers Task Forcepublication, The Use of Land, did notspeak of sprawl per se, it concluded thatthe dominant pattern of "unrestrained,piecemeal urbanization" was leadingcitizens to ask how such growth affectedtheir "quality of life" (Reilly 1973, 33). Ina similar vein, The Language of Cities andthe Encyclopedia of Community Planningand Environmental Management definedsprawl, respectively, as:

the awkward spreading out of the limbsof either a man or a community. Thefirst is a product of bad manners, thesecond of bad planning. Sprawl is a by-product of the highway andautomobile, which enabled the spreadof development in all directions. As



builders scramble for lots to build on, thejourney to work is lengthened and greenspaces are consumed by gas stations andclutter. (Abrams 1971, 293-294)

the uncontrolled growth of urbandevelopment into previously rural areas.Sprawl refers to a mixture of land usesoccurring in an unplanned pattern. Urbansprawl has been strongly criticized as anunattractive and inefficient use of land andresources, causing excessive infrastructurecosts related to extending utilities toremote areas. It has also been accused ofeliminating environmentally importantopen space while leapfrogging developableparcels. (Schultz and Kasen 1984, 378-379)

THE FIRST STUDIES ON THECOSTS OF SPRAWL

In the 1960s, professional research beganto be undertaken in numerous areasrelevant to the study of sprawl. Examplesof this early research include InnovationVersus Tradition in CommunityDevelopment (ULI 1963), which looked atthe effects of development patterns onroad lengths; Howard County Study(Howard County 1967), which consideredcomparative, countywide costs of roads,utilities, schools, and open space undersprawl versus more planned scenarios;Urban Form and the Cost of PublicServices (Kain 1967), which consideredpublic service costs at varying densities;Planned Residential Environments(Lansing 1970), which looked at howdifferent overall development patternsinfluence trip generation rates anddistances; Total Energy Demonstration(HUD 1972), which measured likelysavings in energy consumption in plannedcommunities; and The Relationship ofLand Use and Transportation Planning toAir Quality Management (Hagevik 1972),

which examined how developmentplanning affects air pollution on a regionalbasis. Although not articulated, thesubstantive foci in analyzing sprawlversus alternatives—namely, the issues oftransportation, infrastructure, publicservice costs, and land and environmentalissues—were already being formulated.

Many of these early studies werereferenced by the bellwether study, TheCosts of Sprawl, authored by the RealEstate Research Corporation in 1974. Assummarized by RERC:

This analysis presents a complete andinternally consistent set of estimates fordirect costs and adverse effects resultingfrom prototypical housing types and landdevelopment patterns at neighborhood andcommunity levels. Six neighborhoodprototypes—differing in housing type anddensity—are analyzed, along with sixcommunity prototypes which representdifferent degrees of community-wideplanning. ... Stated in the most generalform, the major conclusion of this study isthat, for a fixed number of households,sprawl is the most expensive form ofresidential development in terms ofeconomic costs, environmental costs,natural resource consumption, and manytypes of personal costs. (RERC 1974, 2-7)

The Costs of Sprawl did not explicitlydefine the term "sprawl." As a matter offact, those close to the study indicate thatthe term appeared as an afterthought in thetitle and summary of findings and was notused explicitly elsewhere in the study. Theanalysis of six community-level growthpatterns within the study implied thatsprawl development had at least twomajor traits: low average residentialdensity (3 units or less per net residentialacre), and a lack of overall planning ateither the regional or community level.RERC did not define sprawl's specific



density characteristics, nor did it define itsresidential and nonresidentialcomponents.

RERC considered approximately 20individual effects (see Table 1). As seenin Table 2, these costs can be grouped intofour overall categories encompassing:

1) public-private capital and operatingcosts;

2) transportation and travel costs;3) land and natural habitat preservation;

and4) quality of life.

Not considered in The Costs of Sprawl,and not part of its research charge, wasany examination of sprawl's social effects,such as its impacts on cities.

The RERC study evoked a flood ofcommentary—much praise as well assome criticism. Two of the better knowncriticisms were articulated by Altshuler(1977) and Windsor (1979). Among otherpoints, Altshuler argued that RERCunderestimated the demand for servicesby higher-density development andcommingled the effects resulting fromhigh density and smaller-unit size.Windsor, in parallel, criticized RERC fornot disentangling density from otherfactors, and among other shortfalls,argued that RERC ignored the benefits ofsprawl, such as its "response to consumerpreference" for single-family detachedhomes. These early points of oppositionon the costs/benefits of sprawl are stillpresent twenty years later and can be seenin the recent exchanges betweenGordon/Richardson and Ewing on thesubject (Gordon and Richardson 1997a;Ewing 1997).

Although the findings of The Costs ofSprawl dominated the literature for sometime, new analyses continued to bepublished. Examples include DavidPopenoe's (1979) depiction of sprawl as

low-density, scattered strip development,which focused on its adverse sociologicalimplications. In 1981, David Millsdescribed sprawl as scattered, leapfrogdevelopment, and discussed how it bothabetted and resulted from landspeculation.

Not considered in The Costs ofSprawl, and not part of itsresearch charge, was anyexamination of sprawl's socialeffects, such as its impacts oncities.

BURCHELL/LISTOKIN ANDTISCHLER ON FISCAL IMPACTS

During the time period between the firstand interim studies on the capital costs ofgrowth, the national work of Robert W.Burchell/David Listokin of RutgersUniversity and Paul Tischler in fiscalimpact analysis, or the examination of theoperating costs of growth, came to thefore. From the early 1970s to the late1980s, numerous studies were undertakenon the municipal and school district costsof growth. Burchell and Listokin wereparticipating authors in HousingDevelopment and Municipal Costs(Sternlieb 1975) and coauthored TheFiscal Impact Handbook (1978) and ThePractitioner's Guide to Fiscal ImpactAnalysis series (1980, 1985). PaulTischler, a private consultant, undertookstudies throughout the country using theMUNIES and FISCALS models developedby him and others.

The fiscal impact studies sought topreview for a community, county, orschool district the impact of projecteddevelopment on future educational andnoneducational public service demands.Burchell and Listokin offered an averagecosting approach built on regional andstatewide demographic multipliers for thedemand for public services, and average



historical costs for the costs of publicservices. Burchell and Listokin balancedthe calculation of costs with revenuecalculations in three categories: propertytax, non-tax, and intergovernmentaltransfers. This was termed the Per CapitaMultiplier fiscal impact technique, whichbecame the method used in creating theirfiscal impact hierarchy and the basis oftheir Development Impact AssessmentHandbook (Burchell, Listokin, andDolphin 1994). Burchell and Listokinfound that most conventional residentialdevelopment negatively impacted the hostservice provider, whereas open spacedevelopment and nonresidentialdevelopment broke even or positivelyimpacted the host service provider. Thesestudies paid little attention to explicitcapital costs except that ongoing debtservice was a component of operatingcosts.

Paul Tischler used a marginal costingapproach in most of his fiscal impactanalyses. In MUNIES and FISCALS, agreat deal of time was spent gatheringboth site-specific data and information onexcess or deficient service capacitylocally. Tischler actually termed acomponent of his overall fiscal impactanalysis a "level of service analysis." TheTischler studies involved detailedcalculations of how a specific communitywith a particular set of financialconditions would respond if growth wereto take place immediately.

Paul Tischler headed the economiccommittee of the American PlanningAssociation from 1980 to 1990. Tischlergenerally reached the same conclusions onthe fiscal impacts of residentialdevelopment, open space, andnonresidential development as didBurchell/Listokin. Conventionalresidential development was generallyfound to be fiscally negative, open spaceor undeveloped land to be break-even, and

nonresidential development to producepositive fiscal impacts. Tischler andAssociates was involved in costs ofgrowth studies in numerous locationsnationally and has also been involved inalternative development and impact feestudies.

These two groups, with differentapproaches and different audiences, foundgenerally the same conclusions on thefiscal attributes of various types of landuses. They established for the planningand land development fields asolidification of opinion on the futurepublic costs of residential andnonresidential development.

THE INTERIM STUDIES:MANAGED GROWTH COSTS INCALIFORNIA; THE COSTS OFSPRAWL IN FLORIDA (DUNCANAND FRANK)

In the early 1980s, in response to therampant development of the 1970s,growth control ordinances beganspringing up in California and Floridacities. These included Davis (CA),Petaluma (CA), and Boca Raton (FL).Before one or more of these ordinanceswere challenged and set aside, initialinquiry concerned their potential impacton local housing costs. If growth werecurtailed through building permit orpopulation caps or through adequatepublic facilities ordinances, would thesefactors contribute to increased housingcosts? Almost everyone looking at theseissues concluded that growth controlordinances did increase local housingcosts (Katz and Rosen 1987; Schwartz etal. 1981, 1989). Further, excessive growthmanagement through protractedpermitting processes, including fiscalimpact analysis, coastal zone managementprocedures, natural resource inventories,and other mechanisms, was also found toincrease housing costs (Parsons 1992).

TABLE 1REAL ESTATE RESEARCH CORPORATION (RERC 1974a)

THE COSTS OF SPRAWL: SUMMARY OF FINDINGS

Notes: All dollar figures are per dwelling unit in 1973 dollars.NA = Not applicablea Includes construction cost of the unit and other expenses such as land

dedication.Source: RERC (1974), Vol. 1, Executive Summary.

b Lbs. per day.c Billion liters per year.d Million gallons per year.e Billion BTU's per year.

Rutge

rs • Broo

kings • Pa

rsons Brinc

kerho

ff • ECO

No

rthwe

st14

TRAN

SIT CO

OPERA

TIVE RESEA

RCH

PROG

RAM

(TCRP) H

-10

The C

osts o

f Spra

wl–Re

visited

Histo

rica

l Ove

rview



TABLE 2REAL ESTATE RESEARCH CORPORATION (RERC 1974) THE COSTS OF SPRAWL:

SUBSTANTIVE AREAS OF INQUIRY

By the late 1980s, two important costs ofsprawl studies were undertaken in Florida.James Duncan, a consultant working forthe Florida Department of CommunityAffairs, studied the capital infrastructurerequirements of sprawl (scattered) versuscompact development forms. Duncanfound that various forms of scattereddevelopment could be as much as 70percent more costly than equivalent formsof compact development (Duncan et al.1989).

A colleague, James Frank of Florida StateUniversity, in research conducted for theUrban Land Institute, updated severalearly (1950s and 1960s) isolated costs of

Various forms of scattereddevelopment could be as much as70 percent more costly thanequivalent forms of compactdevelopment.

sprawl studies with 1987 data and prices,and assembled their results. His findingswere similar to Duncan's: "contiguous"development was 45 percent lessexpensive for roads, water, and sewer than"leapfrog, far-out" development (Frank1989). The Duncan and Frank studies arecited throughout the costs of sprawlliterature.



CHARACTERIZING SPRAWL:CRABGRASS FRONTIERS ANDEDGE CITIES

Kenneth Jackson's Crabgrass Frontier:The Suburbanization of the United States,published in 1985, received muchacclaim. Although sprawl per se was notmentioned in this monograph, numeroustraits attributed by Jackson to the"crabgrass frontier" were clearly sprawl-like in character. These attributes were:

1) low residential density and theabsence of sharp divisions betweentown and country2) the socioeconomic distinctionbetween the center and the periphery3) a lengthy journey to work in termsof distance and time.

Jackson attributed the permanence of thecrabgrass frontier to physical as well aspolitical factors (e.g., that America wasland-rich and had fragmented localgovernments). He also noted its problems(e.g., high local public service costs andincreased automobile dependence) as wellas its benefits (high level of housingamenity and individual open space).

Approximately six years after thepublication of Crabgrass Frontier,journalist Joel Garreau published EdgeCity: Life on the New Frontier (portionsof the book were actually in print beforethis time). Unique to Garreau's work wasthe concentration on peripheralnonresidential clusters brought together atsuburban junctures of major beltways andaxial interstate roads. These "edge cities"formed a new kind of metropolis becausenonresidential development was soonjoined by high-density residentialdevelopment to form relatively self-sustaining urban clusters at edges of built-up areas. These clusters were unique; nomore than fifty existed in the UnitedStates, and they represented sprawl at anurban scale (Garreau 1991).

Tyson's Corner in Fairfax County, Virginia, theprototypical "edge city."Source: County of Fairfax (Virginia), Office of

Comprehensive Planning.

During the early part of the 1980s, in acountry with a newly refound admirationfor capitalism, and in the latter part ofthat decade, in a recession that paid theprice for earlier deficit spending, theliterature on sprawl was relativelyquiescent. The trend has reversed itself inthe 1990s; as will be seen, there has beenan outpouring of studies. These studiesare reviewed in Section II of this reportby substantive area. To give a sense ofthe current literature—and the currentdefinition of sprawl and its alleged costsand benefits—a sampling is discussedhere.

These "edge cities" formed a new kind ofmetropolis.

SPRAWL AND CITIES: DOWNS,RUSK AND BARNETT

In his 1994 book, New Visions forMetropolitan America, Anthony Downsadopted a broader approach for definingsprawl that primarily referred to densitybut included some other characteristics aswell. Downs, building on an earlier work,Stuck in Traffic (Downs 1992), definedsprawl as encompassing five majorelements:



1) low-density, primarily single-family residential settlement (withoutany numerical density specified)2) heavy dependence upon privateautomotive vehicles for all types oftravel3) scatteration of job locationswidely across the landscape inmainly low-density establishments(also without any numerical densityspecified)4) fragmentation of governanceauthority over land uses among manyrelatively small localities5) widespread reliance on thefiltering or "trickle down" process toprovide housing for low-incomehouseholds.

New Visions for Metropolitan Americaproposed a basic method for analyzingsprawl—i.e., comparing its results to theresults that might arise from alternativeforms of metropolitan growth. Downsdescribed a way of formulating alternativeoutcomes through an analysis of the basictraits of different growth strategies.Downs's approach is incorporated anddescribed in more detail later in SectionII.

As is apparent, even the most currentliterature on sprawl tends to describe itsattributes rather than quantify them. Veryfew quantified analyses of sprawl'simpacts or relationships to other variablesappear anywhere in the literature. As aresult, few studies have mathematically orstatistically linked sprawl to otherconditions or metropolitan traits.

A limited attempt at quantification wasput forth in the 1993 work by David Ruskin Cities Without Suburbs. He calculatedan "index of elasticity" that measured theability of cities to extend their boundariesto encompass surrounding urbanizeddevelopment. "Elasticity" is essentiallythe same as annexation, i.e., movement

outward from the city center (sprawl)without the creation of new politicaljurisdictions. Rusk claims that cities withhigh indices of elasticity are superior tothose with low indices of elasticity, interms of income distribution, racialintegration, population growth, andeconomic development. The best cities are"elastic" cities, he claims, and applies hisindex both to cities themselves as well astheir metropolitan areas.

Rusk himself did not performmathematical or statistical analysesrelating the variables just described, butthree reviewers of his book did. John P.Blair, Samuel R. Staley, and ZhongcaiZhang (1996) used multiple regressionemploying measures of growth andeconomic welfare over the period 1980-1990 as independent variables, againstRusk's index of elasticity as the dependentvariable. These reviewers concluded thatRusk's index of elasticity had statisticallysignificant effects of the expected typeson city employment,

The most current literature tendsto describe sprawl's attributesrather than quantify them.

population, poverty, and per capitaincome growth and significant effects ofthe expected types on metropolitan-areapopulation and employment growth—butnot of the expected types on metropolitan-area per capita income or poverty growth.However, even where the regressionequations identified statisticallysignificant effects, they had low R2s (lowexplanatory power), an outcome thatindicated that other unspecified variableswere possibly not included in theregression equation. An implication ofthis analysis was that either Rusk's indexof elasticity is not a useful indicator ofsprawl or the indicator itself, due to itsconstruction, inherently produced lowlevels of explanation.



City-suburban relationships were alsoconsidered by Jonathan Barnett in his1995 book, The Fractured Metropolis.This analysis of metropolitan area trendswas strictly narrative and advanced thethesis that U.S. metropolitan areas weresplitting into "old cities" and "new cities."Barnett proposed that future growth beredirected into the "old cities." Much ofhis work was skewed toward physicaldesign and planning; it favored compactdevelopment over sprawl and encouragedcommercial development within, and thecreation of urban growth boundariesaround, older metropolitan cities.

SECOND GENERATIONSTUDIES ONTHE COSTS OF SPRAWL

Research into methods to address thecosts of sprawl and a study of theunderlying data have been undertaken atboth the Center for Urban Policy Researchat Rutgers University and at theUniversity of California—Berkeley.Starting in the early 1990s, RutgersUniversity researchers, led by Robert W.Burchell, began to quantify the relativeimpacts of alternative patterns ofdevelopment. One or two years later,under John D. Landis, similar efforts wereundertaken at the Institute of Urban andRegional Development at Berkeley. Bothresearch organizations have looked at theprospective impacts of alternativedevelopment patterns. Both researchorganizations developed comprehensiveland-use models to carry out theseanalyses (Burchell 1992a, 1992b; Landis1994, 1995).

Costs were defined primarily in terms ofresource consumption at the communitylevel. Sprawl was defined as skipped-over, low-density residential andnonresidential development.

The Rutgers effort involved an analysis ofthe differing effects of "trenddevelopment" (sprawl-like) and "planneddevelopment" (compact form withmanaged growth attributes) in NewJersey. The results obtained are shown inTable 3. This Rutgers study was precededby similar work for the State of Marylandas part of its original attempt at a GrowthManagement Act. Significant efforts toconfine sprawl to the Baltimore-Washington corridor have beenundertaken in Maryland.

Sprawl is defined as skipped-over,low-density residential andnonresidential development. —Burchell 1992a; Landis 1994

The New Jersey and Maryland analyseswere followed by similar studies forLexington, Kentucky (Burchell andListokin 1994b), the Delaware Estuary(Burchell and Moskowitz 1995), and theStates of Michigan (Burchell 1997a) andSouth Carolina (Burchell 1997b).Research is also currently underway, atRutgers, for the State of Florida as part ofits Eastward Ho! initiative, a developmentplan aimed at keeping a large share offuture development east of Route I-95 infive southern counties. In all instances,

Florida's Eastward Ho! initiative hopes to avertthis potential future.Source: Tim Reilly, Sunshine: The Magazine of

South Florida.



TABLE 3BURCHELL (1992)—NEW JERSEY IMPACT ASSESSMENT:

SUMMARY OF IMPACTS OF TREND VERSUS PLANNED DEVELOPMENT

Source: Robert W. Burchell 1992a, b

TABLE 4

BURCHELL (1992-1997) FINDINGS OF SAVINGS OF COMPACT GROWTHVERSUS CURRENT OR TREND DEVELOPMENT

Source: Robert W. Burchell 1992-1997



polar development patterns arecontrasted—i.e., "current" or "trend"growth is measured against "compact," or"planned" growth. The exactnomenclature in the studies isunimportant; what is important are thediffering land-use configurations and theirimpacts, which are related below:

Current, or trend, development is historicaldevelopment in an area. The land-useliterature describes this type ofdevelopment as skipping over existingdevelopment; land-consumptive andinefficient use of available land at or nearthe core of the metropolitan area; andrequiring significant accompanyinginfrastructure in the form of roads, waterand sewer lines, public buildings, and thelike. Compact, or a more managed, type ofdevelopment attempts to direct growth toalready existing locations of developmentwhile preserving yet-to-be developedareas. Nationally, the land-use literatureportrays compact development as moreefficient in its land-use patterns and thusless land-consumptive. Accordingly, itoften requires somewhat less developmentinfrastructure. Compact development isalso viewed as not limiting or restrictingpopulation or employment growth at thecounty, regional, or state levels. (Burchell1997a, A-1)

Burchell developed a series of quantitativemodels relating to land consumption,road, transit, water/sewer infrastructure,fiscal impacts, housing cost, and qualityof life to examine the relative effects ofalternative development patterns.Application of these models across theaforementioned jurisdictions indicatedcomparable order-of-magnitude findings.For instance, a shift away from sprawl tocompact growth was projected byBurchell to reduce water/sewer utilityinfrastructure costs by 8 percent in NewJersey, 7 percent in Lexington, 8 percentin the Delaware Estuary, 14 percent in

Michigan, and 13 percent in SouthCarolina. Table 4 summarizes the array offindings from the various Burchell studies(1992-1997). Table 5 groups the effects ofsprawl, some dozen in all, into five overallcategories.

The Berkeley effort employed theCalifornia Urban Futures (CUF) model ofthe San Francisco Bay Area to tabulateland consumed under three scenarios: (a)"business as usual"; (b) "maximumenvironmental protection"; and (c)"compact cities." These scenarios weredifferentiated, respectively, by (a) notrestricting development either within thecity or within unincorporated areas; (b)applying a range of environmentalrestrictions to both locations, but notrestricting growth per se; and (c)restricting growth to acknowledge someenvironmental limitations and countywideminimum population projections. The twolatter alternatives showed considerableoverall land savings, particularly sensitiveenvironmental land savings relative to thebusiness-as-usual scenario. Total landsaved in scenarios b and c was 15,000 and46,000 acres, respectively. Scenario bsaved nearly 60,000 acres of primeagricultural land, 10,400 acres ofwetlands, and 2,800 acres of steep-slopedland; Scenario c saved 28,000 acres ofprime agricultural land, 10,400 acres ofwetlands, and 8,000 acres of steep-slopedlands (Landis 1995).

In a series of relatively current articles inEnvironment and Planning Behavior,Landis discussed the development and useof the second generation of the CaliforniaUrban Futures Model. These articles wereless about sprawl and land savings andmore about urban modeling; still theysuggested a framework for understandingand predicting the land-and habitat-takingeffects of sprawl.



STUDIED REACTIONS TOSPRAWL—LUTRAQ (OREGON)AND CONCURRENCY (FLORIDA)

In the late 1980s and early 1990s, sprawlgrowth on the northwestern andsoutheastern coasts of the United Statesresulted in two different reactions—bothsupported by so-called "friends"organizations. In the first case, theorganization was the 1000 Friends ofOregon, in the second case, the 1000Friends of Florida.

In the early 1990s, growth in the Portlandregion was believed to hinge on theconstruction of a Western Bypass aroundthe city. An alternative plan was sought totry to accommodate growth without theneed for more highways. Sam Seskin ofParsons Brinckerhoff, leading a team ofresearchers in the Land UseTransportation Air Quality simulation(LUTRAQ), compared the transportation

impacts of a transit-oriented development(TOD) plan to the impacts of a preferredBypass alternative. The LUTRAQalternative shifted the location of 65% ofnew residential units and 78% of new jobsto locations within walking distance oflight rail or bus transit lines byreconfiguring expected development intoa series of mixed-use centers. Thealternative showed a reduction in vehiclemiles traveled and a reduction of the useof the automobile (Davis and Seskin1997). Portland voters responded byapproving a $1 billion rail line alongwhich TOD will occur, and Seskinreceived an American PlanningAssociation award for the research effort.

Subsequent analyses produced byGenevieve Giuliano, however, found onlysmall gains associated with non-automobile mode shares and very smallreductions in vehicular travel. Equally

TABLE 5

BURCHELL (1992-1997) ANALYSIS OF TREND VERSUS PLANNEDDEVELOPMENT: SUBSTANTIVE AREAS OF INQUIRY

Source: Robert W. Burchell 1992-1997



distressing, the magnitude of investmentin transit services needed to be quite largeto achieve the resulting changes in modeshares. The LUTRAQ studyunintentionally demonstrated the limits ofmaking large investments in transit toinfluence travel patterns (Giuliano 1995b).

In Florida, meanwhile, the reaction tosprawl was to limit development if itcould not be shown that sufficient publicfacilities would be in place at the time thatdevelopment occurred (Florida GrowthManagement Act 1985). This procedure,termed "concurrency," included bothmandatory (transportation) and voluntary(schools) components. At first, thosedistant from the scene thought that theprocedure was responsible for shuttingdown growth in the state. After the dustfrom the housing recession of the late1980s settled, however, those originallyopposed to concurrency reluctantly agreedthat it had channeled growth effectively.In the meantime, those who originallyfavored concurrency vehemently opposedit because roads were being built andwidened and new schools were beingconstructed (albeit at developer cost) toofar from the locus of existingdevelopment. Growth was slowed, but italso was accommodated in locationswhere it should not have been (Mofson1997).

AT WHAT SCALE ISMEASUREMENT TO TAKE PLACE?URBAN FORM ANDTRANSPORTATION

At about the same time that Burchell andLandis were looking at development formand its effect on resource consumption,two other important considerations beganto emerge. The first was the scale at whichtransportation impacts were being viewed;the second was the effect of transportationon urban form, and vice versa. In other

words, while attempting to define theindicators of sprawl and more compactforms of development and their resultingimpacts, it became apparent that oneneeded to specify at what level impactswere being measured—individual,community, or societal. Almost all studiesto date have been undertaken with impactsspecific to the community level. But SamSeskin from Parsons Brinckerhoff, andTerry Moore from ECONorthwest, beganpursuing the issue of "full" costs oftransportation, attempting to view thecosts of transportation decisions at theindividual and societal scales as well as atthe community level. They determined,for instance, that although using anautomobile was efficient at the individualand community scales, it was expensive ata societal scale (air pollution). Althoughtransit was efficient at individual andsocietal scales, it was expensive at acommunity scale (the cost to delivertransit). And walking, although efficient atcommunity and societal scales, wasexpensive at an individual scale (the costof the individual's time) (ParsonsBrinckerhoff 1996; Moore and Thorsnes1994).

Seskin and Moore shifted the inquiry toissues of the impact of urban form ontransportation, and vice versa. The urbanform impacts on transportation were muchas expected. Seskin and Mooredetermined that sprawl development couldbe served well only by the automobile;much more compact development led totransit solutions. Mixed-use developmentenabled walking and biking.Transportation impacts on urban formwere not quite a mirror image of the first,however. Significant use of theautomobile led to unlimited spreaddevelopment. Transit presence broughtusers who also needed an automobile;mixed-use development promoted footand bicycle use, but an automobile wasstill required. Land use can affect



transportation mode and vice versa, butAmerican society today remains heavilydependent upon the automobile (ParsonsBrinckerhoff and ECONorthwest 1996).

CERVERO ANDTRANSPORTATIONACCESSIBILITY MEASURES

One of the most widely publishedacademics in the field of transportationplanning is Robert Cervero, from theUniversity of California at Berkeley. Eversince his first book, Suburban Gridlock,was published in 1986, Cervero has beensolidly represented in the land-use/transportation literature. His latestbook, The Transit Metropolis (1998),deals with transit-oriented cities. Cerverohas done important sprawl work relating ajobs-housing "imbalance" to expandingcommutes (Cervero 1996), and Bay Areagrowth trends of job decentralization toincreased VMT per worker (Cervero andWu 1996). Other aspects of his workinvolve (1) suburban congestion as well asmeasures for its relief (Cervero 1986,1991a); (2) the role of suburban activitycenters as alternatives to sprawl, andcommuting patterns within these centers(Cervero 1989, 1991b, 1996); and (3) thefeasibility of transit in suburbanlocations—i.e., the required density andimplementation costs (Cervero 1994a,1994b).

Cervero's latest contributions from asprawl perspective are two papers he co-authored on suburban accessibility: (1) a1997 paper co-authored by Timothy Roodand Bruce Appleyard, entitled "JobAccessibility as a Performance Indicator:An Analysis of Trends and their SocialPolicy Implications in the Bay Area"; and(2) a 1996 paper co-authored by KaraKockelman, entitled "Travel Demand andthe 3Ds: Density, Diversity, and Design."In these papers, Cervero and hiscolleagues show through factor andregression analyses the effect of currentdevelopment patterns on employment

accessibility. They try to document, inother words, how well transportationserves employment markets. In the firstarticle, Cervero finds that current sprawldevelopment patterns have the largestimpact on severely poor neighborhoodsbecause they separate jobs from jobseekers. Minorities are particularlydisadvantaged, because even with equaleducation, vehicle availability, andaccessibility, blacks still haddisproportionately high unemploymentrates.

Current sprawl developmentpatterns have the largest impacton severely poor neighborhoods.

In Cervero's second article, he looks atwhat can be done. He measures the effectsof density, diversity, and design onaccessibility, and finds that compactmixed-use, pedestrian-friendly designscan reduce vehicle trips, vehicle milestraveled, and the use of the automobile.Density, he concludes, affects businesstrips; diversity affects both work and non-work trips, but has less of an effect thandensity; and design affects primarily non-work trips. He upholds the views of thenew urbanists—somewhat, because heshows that sensitive land design andbuilding arrangements can reduce traveldistances and alter modes of travel.

THE BANK OF AMERICA STUDY:BUSINESS EMBRACES THE ANTI-SPRAWL MOVEMENT

In 1995, four groups—Bank of America,California Resources Agency, GreenbeltAlliance, and Low-Income HousingFund—published a study on sprawl thatquickly came to be known as the Bank ofAmerica Study. Those who champion landdevelopment alternatives to sprawl pointto this study, the work of one of theprivate sector's most influential members,as a landmark. If the banks finally realizethat sprawl can no longer be tolerated,recognition of the impacts of differing



land development patterns on society'sresources has indeed hit the big time.

The Bank of America study summarizedchanges in population, demographics, andemployment that had taken place over thetwo decades prior to 1990. It alsoreferenced a land-use pattern that hadtaken place during this same period oftime and termed it "sprawl." Sprawl wascharacterized by decentralizedemployment centers and residential tractsaccessed almost exclusively by theautomobile. These decentralized locationswere safe and cheap places in which tolocate and had plucked all fiscal andphysical benefits from the central city.Further, the study noted that the trendtoward sprawl was aided and abetted bythe federal subsidies given to theautomobile.

If the banks finally realize thatsprawl can no longer be tolerated,recognition of the impacts ofdiffering land developmentpatterns on society's resourceshas indeed hit the big time.

The Bank of America report was criticizedfor its inability to adequately interpret thelong-standing criticisms of RERC's (1974)The Costs of Sprawl report. The Bank ofAmerica study seemed to buy into manyof the arguments that favored the anti-sprawl position without an adequate lookat contrary evidence. Nonetheless, thosewho championed the study as a summaryof the ills of sprawl used the Bank ofAmerica imprimatur to promote theposition that the business community, atlong last, was calling for managed growthto conserve national resources.

IS SPRAWL LIKED OR DISLIKEDBY THE GENERAL PUBLIC?FANNIE MAE VERSUS "VISIONPREFERENCING" SURVEYS

A question discussed and debated in anumber of circles is whether Americanslike their current development patterns.Often, those responding have difficultymaking the distinction between shelterand location, and between both of theseand way of life.

There is a popular literature that ratesplaces on such indices as cost of living,public safety, climate, job growth,transportation accessibility, and access tocultural and recreational amenities(Savageau and Boyer 1993). Clearly,suburbs in the Southeast and Southwestfare better on this rating scale than citiesin the Northeast and North Central regionsof the country, or, for that matter, ruralareas in any location. An economicsliterature looks at the determinants ofworker migration, identified as jobavailability, good climate, and lowerhousing costs (Duffy 1994; Greenwood etal. 1991; Roback 1988; Rosen 1979).Psychological reasons for moving oftenparallel the economic determinants:physical (safety), physiological(economic), belongingness (sense ofplace), and personal satisfaction (culturaland recreational amenities) (Zinam 1989).Again, suburban locations appear to dobetter than urban locations on both of theabove sets of criteria.

Americans are asked about theirenvironments through two basic devices: anational, annual, in-person, in-homeFannie Mae survey of owners and renterson their housing (Lang and Hornburg1997) or an occasional, professionallyadministered "visual preferencing" surveyon their environments (Nelessen 1994).

Eighty percent of Americans contacted inthe first survey identified the traditionalsingle-family home with a yard as theideal place to live. To afford it, theywould rather live farther out than take a



second job, tie up savings, put children inday care, or incur heavier debt. Finally,they would rather occupy an averagehouse in a good neighborhood than a goodhouse in an average neighborhood (FannieMae 1994).

Respondents often have difficultymaking the distinction between shelterand location, and between both of theseand way of life.

Visual preferencing surveys are typicallyemployed by planners and architects totest sentiment for a redirection in currentdevelopment patterns and forms (Nelessen1994). These surveys contrast the currentversus an alternative development patternand architecture and ask those surveyed topick between the two. Often it is hoped bythose who administer these surveys thatthe alternative development pattern willbe chosen and, accordingly, localities willdevelop residential and residential areas ina different way (Calthorpe 1993). Most ofthose who experience this exercise ofchoice opt for the alternative, whichtypically shows a denser, more traditionalresidential village center and less spread-out residential subdivisions and stripcommercial developments (Nelessen1994).

The results of most of the two surveys onconsumer preference and sprawl indicatethat people feel comfortable with theircurrent housing and its suburban locationbut also think that sprawl has an ugly lookand that suburbs are becomingincreasingly congested. Whether peoplewould change their housing type (single-family), form (single-lot subdivision), orlocation (suburbs) to achieve a different"look" or "feel," or to be free fromcongestion, remains a crucial question.

AN UNUSUAL FINDING: THE CITYIS IMPORTANT TO THE REGION;THE USUAL FINDING: PEOPLEDON'T CARE

The United States has had a love–haterelationship with its cities for at least fiftyyears. This has taken two forms. The firstis inquiry into the continued importanceof the central city; the second is whetheror not people will choose to live and workthere.

In the mid-1990s, two articles rekindledinterest in, and attempted to quantify theimportance of, the central city to itssurrounding area. One was written byElliot Sclar and Walter Hook in 1993,"The Importance of Cities to the NationalEconomy"; the other was written by KeithIhlanfeldt in 1995 and entitled "TheImportance of the Central City to theRegional and National Economy." At atime when most scholars viewed thecentral city's role in the region and nationas not critical and one of declining value,Sclar/Hook and Ihlanfeldt breathed newlife into the debate on the role and futureof the central city with the followingarguments:• In most metro areas, the higherpaying

jobs are found in the central city.• In the metro areas of the 100 largest

U.S. cities, half of suburban familieshad at least one worker employed inthe central city.

• Sixty-seven percent of suburbanresidents surrounding the 100 largestU.S. cities depend on the city for majormedical care; 43 percent have a familymember attending an institution ofhigher learning there.

• Cities provide low-cost housing forlow-wage workers employed in—andnecessary for—the activities ofsuburbs.

• The overall appeal of a region isinfluenced by conditions prevailingwithin its central city.



Sclar and Hooks argued that the UnitedStates subsidizes suburbs throughhomeownership income tax deductionsand by federal/state cost-sharing ofhighway construction. Continuedsubsidization will cause increasing autodependence, and a further channeling ofmost infrastructure expenditures to roadbuilding, at the expense of educationoperating costs. According to the authors,the United States ranked lowest amongthe seven most industrialized nations inpercent of GNP that supported education.

Ihlanfeldt found that central cities possesscertain "agglomeration economies" (thebenefits of scale) that will sustain theirprimacy in a region. These includecommunications, labor, and producerconcentrations. Moreover, financialservices such as investment banking,commercial banking, legal auditing, andactuarial services were provided primarilyby central city firms to suburban markets,and in many cases to world markets.According to Ihlanfeldt, these activitieswere not likely to be taken on by suburbanfirms, because few suburban firms havethe appropriate scale to conduct them.

The United States subsidizes suburbsthrough homeownership income taxdeductions and by federal/state cost-sharing of highway construction.

The second issue regarding urban areaswas whether upwardly mobile householdswill continue to reside there. In the 1970s,the United States experienced significantmovement of jobs and residents toexurban or rural areas. During this periodof time, non-metropolitan areas were thelocations of the fastest relativeemployment and household growth(Sternlieb and Hughes 1983). During the1980s, there was stabilization, if notgrowth, of metropolitan areas. Buoyed bysignificant immigration and a slowing ofmetropolitan to non-metropolitan

outmigration, metropolitan areas werebeginning to grow (Gordon, Richardson,and Yu 1997; Nelson et al. 1995, 1997).According to Peter Gordon, recent Bureauof Economic Analysis (BEA) RegionalEconomic Information System (REIS)data indicate that the trend is once againtoward outer areas; indeed, over the lastsix years, outward metropolitanmovement is almost as pronounced as itwas during the 1970s. Gordon et al. findsthat the one constant in all of this has beenstrong suburban growth, with parallelrural growth tilting the scale to outwardmovement, and even stronger suburbangrowth with reduced declines of urbanareas tilting the scale toward inwardmovement. The consistency of thesuburban component of this trend andrenewed non-metropolitan growth (theoutward movement) do not bode well forthe future of the central city. Gordon andhis colleagues conclude, citing additionaldata from the Economic Census CBD file,that:

The location decisions of households areinfluenced less by workplaceaccessibility than by availability ofamenities, recreational opportunities, andpublic safety. In addition, the locationsof firms are clearly becoming morefootloose under the influence of theinformation revolution, just at a timewhen core agglomeration diseconomies(pollution, congestion, crime, fiscalinstability, etc.) appear to beoutweighing the original agglomerationeconomies that pulled people andeconomic activities together. In thisview, the central cities are not comingback any time soon (Gordon,Richardson, and Yu 1997)

The suburban component and renewednon-metropolitan growth (the outwardmovement) do not bode well for thefuture of the central city.



THE VALUE OF OPEN SPACE ANDFARMLAND: THE FARMER ANDCONSERVATIONIST AS PLAYERSIN THE SPRAWL ARGUMENT

In the latter part of the 1980s and the early1990s, the American Farmland Trust(AFT) began a series of studies todiscourage the conversion of farm tractsto sprawled residential subdivisions. Notonly was farmland ideal for developersbecause it was flat, it also was, for themost part, the cheapest land available. Thepercentage of farmland being lost in theUnited States was many times thepercentage growth of householdformation. The analyses of the AFT,called "Cost of Community Services,"presented detailed case studies of thecost/revenue superiority of farmland toother types of land uses. Studies wereundertaken in Massachusetts, Connecticut,Pennsylvania, Virginia, and the Midwest,and are heavily cited today. Theconclusions drawn always demonstratethis group's advocacy and point tofarmlands as a fiscal benefit tocommunities in which they are located.Regardless of methodology, the studiesachieve their goal of representingfarmland not merely as fiscally neutral butas fiscally positive. "Smart" communitiesshould not want to lose this net revenueproducer to other forms of development(especially residential), which would bemore costly (AFT 1992b).

Growing out of this new attention tofarmland was the recognition that farmers,as owners of this land, were often opposedto growth management (and thus pro-sprawl) and needed to be brought into thenegotiation process. Otherwise, theywould sell their land to developers beforeit could be acquired via public purchase orthrough some type of transfer ofdevelopment rights. Farmers preventedpassage of the original Maryland GrowthManagement Act and threatened to do the

same to the New Jersey State Plan if theirreal estate interests could not be protected.In Maryland, it appeared that the farmerscould not be assuaged, and the GrowthManagement Act1 failed. In New Jersey,farmers were appeased at the eleventhhour with a promise from the New JerseyState Planning Commission that theirdevelopment rights would be purchased ata price somewhere between crop and realestate value, and the planning statutepassed.

Randall Arendt, influenced by living inboth walkable and planned open spacecommunities in New Jersey as a child, andseeing these concepts implemented inEngland as an adult, built upon IanMcHarg's Design with Nature (1969) inan attempt to make current developmentpatterns greener. In three of his latestbooks, Rural by Design (1994b),Conservation Design for Subdivisions(1996), and Growing Greener (1997), heprovides convincing evidence that openspace adds to the value of surroundingreal estate and to the quality of life ofthose who live within it. Arendt sees thecombination of compact development andopen space as leading to interconnectednetworks of green space (Arendt 1994b).An area-wide, interconnected greenwaycan extend open space and wildlifebenefits to the larger region. Further,successful control of sprawl will retain the"traditional character" of communities(Arendt 1996).

The Sierra Club, among otherconservationist groups, is activelycampaigning against sprawl. Its 65chapters and 450 groups are challengingsprawl at the grassroots level incommunities across America (Sierra Club1998).

1 Maryland ultimately passed a diluted version ofthe original act and has adopted a variety of "smartgrowth" procedures.



THE MECHANICS OF PAYING FORSPRAWL: IMPACT FEES, TAKINGS,AND PROPERTY RIGHTS

In order to pay for sprawl and not impactcurrent residents, local governments haveturned to economists and land-useattorneys to devise a system of assigning ashare of new required public serviceinfrastructure to new owners of developedproperty. These mechanisms are termedimpact fees, developer exactions, orproffers and are based on the rationale ofcharging development costs to those whohave caused them. Impact fees arecalculated by determining the specificcosts that one new unit of residentialdevelopment or 1,000 square feet ofnonresidential development will cause inroads, water/sewer, public buildings(schools and municipal), and other capitalinfrastructure. Impact fees, developercharges, or whatever moniker they areknown by, are currently the fastest-growing source of municipal revenues.Principal players in this group are JamesNicholas of the University of Florida andChristopher Nelson of the GeorgiaInstitute of Technology (Nelson 1988;Nicholas et al. 1991). Nicholas hasconstructed impact fee schedules innumerous counties and municipaljurisdictions; both Nicholas and Nelsonhave significant academic andprofessional publications in this area.

In order to pay for sprawl and notimpact current residents, localgovernments have turned to asystem of assigning a share of newrequired public serviceinfrastructure to new owners ofdeveloped property. Thesemechanisms are termed impact fees.

The issue with impact fees specifically,and growth management strategiesgenerally, is that these mechanismspresuppose government capacity to

regulate land. This amounts to a takingand thereby affects individual propertyrights. Although most of these techniqueshave been upheld, when they becomeoverly aggressive, they are subject tojudicial review.

This gets to what land-use attorneysdescribe as the "black hole" of takingsjurisprudence. Until recently, a severe testof a taking has been applied. A land-useregulation is a taking if it: (1) does notsubstantially advance a legitimate stateinterest; or (2) denies an owner alleconomically viable use of his or herproperty. Post-1990, there appears to bean easing of this test that favors propertyowners. Charles Siemon (Siemon 1997),Robert Freilich (Freilich and Peshoff1997), and Jerold Kayden (Young 1995)are recurringly involved in litigationconcerning these issues or in designingland-use regulations to avoid suchlitigation. Suburban developmentordinances that require payment for costsor link "social" objectives to thedevelopment of real property will betested by the courts. To pay for sprawl,local governments have become quiteinventive at both deriving fee schedulesand in locating property owners to whomthe costs can be assigned. Much as otherforms of payment for sprawl are dryingup, if governments are not careful, so toowill these mechanisms.

SPRAWL'S CRITICS ANDTHE NEW URBANISTS

In 1993, a study conducted for theChesapeake Bay Program defined sprawlas "residential development at a densityof less than three dwelling units per acre"(CH2M Hill 1993). This definition didnot have a "locational component" andwas a modification of a definitionpresented in an earlier draft—i.e.,"developments having gross development



densities of less than three or fourdwelling units per acre or minimum lotsizes of at least one-quarter of an acre, andfrequently of at least one acre." The latterdefinition had been criticized by Uri Avin(1993) for including properties with toohigh a density; it could be applied to manyexisting, close-in subdivisions in bothMaryland and Virginia. On the otherhand, in California, sprawl is currentlytaking place on 9,000-square-foot lots;obviously the upper-level density cutoffvaries considerably by region.

Sprawl, and more generally,suburbanization, were condemned in apolemical book by James Kunstler (1993).The title of the book, The Geography ofNowhere: The Rise and Decline ofAmerica's Man-Made Landscape, conveyshis message. The strident tone of themessage is reflected by the followingstatement:

We have become accustomed to living inplaces where nothing relates to anythingelse, where disorder, unconsciousness, andthe absence of respect reign unchecked.(Kunstler 1993)

Peter Calthorpe's book The Next AmericanMetropolis, published in 1993, offered amethod for determining populationdensities in an idealized form of modernsettlement. He presented a scheme forclustering housing and otherimprovements around transit stops atspecified densities which could, in turn,be used to compute overall densities forideal future metropolitan settlements. Hisscheme involved creating Transit OrientedDevelopments (TODs) around stations ina system of radial fixed-rail transit linesemanating from a region's majordowntown. This approach quantifiedaspects of an alternative form of futuregrowth. However, Calthorpe did notpresent any method of measuring the costsand benefits of sprawl, nor of the

alternative form he suggested. Neither didhe present any database to use in carryingout such measurements.

Street patterns of neo-traditional versus typicalsuburban neighborhoods.Source: Florida Department of Community Affairs.

Calthorpe is a "new urbanist," part of anurban design movement called "neo-traditionalism." Neo-traditionalism callsfor the development of neighborhoods thatresemble those of the past—i.e., with gridstreet patterns, fronted by proximatesingle-family houses with porches,sidewalks, alleys, and other traditionalfeatures. The elements returned toneighborhood design include mixed uses,



the grid-based street structure, higherdensities, pedestrian circulation, andtransit use. The elements removed includesingle uses, cul-de-sacs, low densities, andautomobile-dominated neighborhoodaccess.

The neo-traditionalists, led by AndresDuany, and joined by Elizabeth Plater-Zyberk (1995), Anton Nelessen (1994),Peter Calthorpe (1993), and others, viewcurrent development patterns (sprawl) asdriven by engineering standards and,accordingly, devoid of the capacity forhuman interaction. Neo-traditionalism isoften proposed as a design alternative tosprawl, even though developmentsincorporating this type of design can befound in sprawl locations. Nelessen'svision preferencing analyses aresometimes cited by those who opposesprawl as evidence that the Americanpublic is ready for this type of design. TheDuany-led new urbanists propose that thenew urban-like grids replace the currentsprawl-like suburban networks.

MORE SPRAWL CRITICS—"TRUSTS" AND "OLD FRIENDS"

A critique of strip commercialdevelopment, and sprawl in general,permeates the current literature of theNational Trust for Historic Preservation(NTHP) and its leadership (NTHP 1993;Moe 1996). At a 1995 conference on"Alternatives to Sprawl," Richard Moe,president of the National Trust, definedsprawl as low-density developmentlocated on the outer fringes of cities andtowns that is "poorly planned, land-consumptive, automobile-dependentdevelopment designed without regard toits surroundings." He described two types:

"sellscape" retail development frequentlyspurred by major discount chains such asWal-Mart and K-Mart, occurring along

major arteries and at highwayinterchanges; and "spread out" residentialdevelopment, usually consisting primarilyof single-family detached houses, locatedon the edges of existing communities or"leapfrogging" into previouslyundeveloped areas. (Moe 1996, 3)

In a later work, Changing Places:Rebuilding Community in the Age ofSprawl, Moe and Carter Wilkie (1997)indicated that sprawl was causingcommunities to be dysfunctional anddiminishing a sense of connectionsbetween people. The authors suggestedthat if sprawl were tested by a truly "free"market, far less sprawl would occur onprivate financing alone. They profferedthat sprawl developers captured benefitsfor themselves while everyone else in thecommunity bore the costs. Both authorscalled for better land-use planning andmore creative reuse of older urban andsuburban areas.

Commercial s t r ip development i sa manifestat ion ofnonresident ia l sprawl .

A more comprehensive view of thecomponents of sprawl was offered inHenry Richmond's 1995 book,Regionalism: Chicago As An AmericanRegion. Richmond's conceptualization ofsprawl included eight components:

1) low residential density;2) unlimited outward extension of

new development;3) leapfrog development;4) spatial segregation of different

land uses;5) decentralized land ownership;6) primacy of automobile

transportation;7) fragmentation of governmental

land use authority; and8) disparity in the fiscal capacity of

local government.



Richmond, former director of 1000Friends of Oregon and a participant in theLUTRAQ simulation study, offered awide-ranging critique of sprawl andincluded numerous carefully culledstatistics supporting his allegations. Manyof his criticisms are drawn from thesubject of his continued research—theChicago metropolitan area. His criticismsform the basis for his definition of sprawl.In defining sprawl, however, Richmonddoes not present specific alternative formsof growth, either conceptually or in termsof quantified analysis. Instead, hecontinues to propose an agenda of specificpolicy actions that would encourage aregional approach to managing futuregrowth. His analysis, therefore, does notprovide either a method for measuring thecosts of sprawl or a specific alternativedevelopment form that would provide abetter outcome.

SPRAWL EVENTS: LINCOLNINSTITUTE/GEORGIACONSERVANCY CONFERENCES

In the spring of 1995, the Lincoln Instituteof Land Policy hosted two importantconferences on sprawl. The first tookplace in Washington, DC, and was co-sponsored by the National Trust forHistoric Preservation and The BrookingsInstitution. This conference brought allthe national actors on sprawl together in adebate format. Sprawl's good and badattributes were debated before a nationalaudience. This was the first appearance ofthe defenders of sprawl. Peter Linnemanfrom the University of Pennsylvania andPeter Gordon from USC proved to bestrong supporters of the free-marketmerits of continued suburbanization.

So successful was the conference indrawing national attention to the sprawlissue, as well as in drawing attention tothe institutions that sponsored theconference, that the Lincoln Institute held

derivative conferences in two locations—Florida and California. Even though nodebate was scheduled, again the issue wasraised: How bad is sprawl? Gordon,joined by colleague Genevieve Giuliano,provided a strong and cogent argument infavor of sprawl and presented findingscontrary to the research of Seskin(LUTRAQ), Landis (California FuturesStudies), Burchell (Rutgers ModelingStudies), and Downs (New Visions forMetropolitan America). The savingsgleaned from LUTRAQ were described asminimal, and the land/infrastructuresavings of the California Futures andRutgers studies were trivialized. Downswas also criticized for assigning causes ofcentral city decline to sprawl that couldnot be defended.

In 1996 and 1997, at the annual meetingsof the Georgia Conservancy, sprawl wasagain the topic of consideration. Like theNational Trust, the Georgia Conservancyshifted its focus slightly from historicpreservation and was making a majorsubstantive thrust at curbing urban sprawl.These conferences, which again attractednational spokespersons on themanifestations and costs of sprawl, werenot a debate, but rather represented asummation on the ills of sprawl. TheAtlanta region was growing at a rate of55,000 jobs per year, and the economywas in such a boom period that growthwas flooding the arterials in and aroundthe city. Sprawl needed to be contained,and the conferences were the beginningsteps in an attempt to create a mood forregional growth management. However,even though some sentiment for growthwas apparent, the consensus was thatpolitical jurisdictions in Georgia were along way from being able to implement,even on a regional scale, the mostelemental of growth managementtechniques (a growth boundary).



THE SPRAWL DEBATE: EWINGVERSUS GORDON—IN PRINT ANDIN PERSON

The debate over sprawl was brought frontand center in two "point" and"counterpoint" articles in the Journal ofthe American Planning Association. Thepoint article by Peter Gordon and HarryW. Richardson (1997a) critiqued thearguments and evidence frequentlypresented in favor of compactdevelopment (i.e., energy, transportation,and infrastructure efficiencies) and arguedthat the decentralized suburban pattern ofdevelopment, in fact, offered manyadvantages, including reduced travel timesand lower housing costs, as well as higherconsumer satisfaction. In counterpoint,Reid Ewing (1997) made a strong case forthe adverse effects of sprawl (as opposedto the benefits of compactness). Ewingpointed to increased infrastructure costs,increasing travel distances, and significantamounts of developable and lost fragilelands as the adverse effects of sprawl.

For the purposes of this review, theauthors' respective definitions of termsbear note. For Ewing, sprawl was definedboth by a series of three characteristics—(1) leapfrog or scattered development; (2)commercial strip development; and (3)large expanses of low-density or single-use developments—as well as by suchindicators as low accessibility and lack offunctional open space (Ewing 1997).Gordon and Richardson did notspecifically define sprawl (norcompactness, for that matter). Instead,they referenced sprawl's various traits.Sprawl was alternatively denoted byGordon and Richardson as low-density,dispersed, or decentralized development,whereas compactness was associated withhigher densities and a downtown orcentral-city spatial pattern versus a

polycentric (or dispersed) spatial pattern(Gordon and Richardson 1997a, 95).

Adverse effects of sprawl includeincreased infrastructure costs,increasing travel distances, andsignificant amounts ofdevelopable and lost fragile lands.

Although the point-counterpoint authorsaddressed more than 15 different subjectsin discussing sprawl and its alternatives,the subjects can be grouped into fivebroad areas, as shown in Table 6.

The debate moved from print to person ina forum held at the University ofCalifornia—Berkeley in late November1997. Both Ewing and Gordon hadsignificantly increased the weaponry usedto support their individual positions.

Ewing began the session with points ofmutual agreement and spun out a longerlist than most expected. These includedthat: (1) the market for transit was limited;(2) infrastructure costs were higher forsprawl development initially but coulddiminish over time with infill; and (3)automobile costs as a function ofsuburban residence were high, but fewalternatives to this mode of travel and itscosts existed. Ewing and Gordoncontinued to disagree about whetherresource consumption (energy, land)differences under sprawl and compactdevelopment in light of national andglobal resources were sufficientlysignificant to cause concern, and whetherthe traffic consequences of sprawl(excessive travel and roadway congestion)could be argued away in terms of eithercurrent or future methods of resolution(higher travel speeds, congestion pricing).The session was narrowly focused onprimarily transportation issues and neverreally dealt with social or quality of lifeissues of sprawl.



TABLE 6

EWING AND GORDON-RICHARDSON IN PRINTSUBSTANTIVE AREAS OF INQUIRY



CONTINUATION OF SPRAWLPRINT—HOUSING POLICY DEBATEAND THE URBAN LAWYERSYMPOSIA

One of the leading housing journals,Fannie Mae's Housing Policy and Debate,and a respected legal journal, The UrbanLawyer, both published symposia onsprawl. Several of the individual articlesbear mentioning, but an important firstpoint is that both housing and urban legaljournals have come to recognize thatsuburban sprawl is an important topic forinclusion in their journals. This issignificant. Both of the journals have hadspecial issues on homelessness,exclusionary zoning, affordable housing,the economies of cities, the spatialmismatch of the poor in cities andavailable jobs in suburbs, and so on.Neither journal strays far from housingand urban problems. Thus, implicit in thepublication of the two special issues onsprawl is the notion that at least somecomponent of sprawl impacts on housingissues and quality of life. Sprawl does notonly potentially cause excess resources tobe expended in providing publicinfrastructure or, similarly, contribute tothe loss of special lands and habitats.Sprawl does not only chain users to asingle source of transportation for accessto residential and employmentopportunities. Sprawl has significantsocial and quality-of-life effects as well.

Suburban sprawl is an importanttopic for inclusion in housing andurban legal journals.

In The Urban Lawyer compilation ofarticles, Robert Freilich traced significantsuburbanizing periods and methodicallyviewed their impact on central cities.Sprawl, he noted, is the force that distillsthe city's economic base, and it isorchestrated by suburban land-usecontrols that promote exclusion (Freilich

and Peshoff 1997). Charles Siemonpointed to the very limited number oftechniques available to implement growthmanagement and the difficulty of usingthem without encroaching upon propertyrights (Siemon 1997).

In the Housing Policy Debate articles,Robert Lang pointed to the voracity ofsprawl and terms it suburbanization thatwas thriving and would not be shelved.Lang further commented that it was notproductive to refer generically tononresidential sprawl as "edge cities," avery limited phenomenon whose time maybe past. To Lang, sprawl epitomizescurrent market preference, and itsdirection is clear—a continuing outwardthrust from its urban core (Lang andHornburg 1997). William Fischel ofDartmouth also proffered in the specialissue of Housing Policy Debate that toomuch growth management could causehousing markets to diminish. Accordingto Fischel, if you continue to castigatesprawl, you may turn around and not findany growth (Fischel 1997).

The upshot of this debate was thatwhereas at one time sprawl had only asolid line of inquiry detailing its costs;there was now a growing line of inquirydetailing its benefits.

If you continue to castigate sprawl, youmay turn around and not find anygrowth.

YET ANOTHER CONFERENCE:CONTROLLING SPRAWL IN THELAND OF BARRY GOLDWATER

In the summer of 1998, in Phoenix,Arizona, the sprawl debate continued, thistime in a conference sponsored by theDrachman Institute of the University ofArizona and the Fannie Mae Foundation.



By now, the agenda and faculty hadpredictable topics and representations:Chris Nelson (Georgia Institute ofTechnology), Gary Pivo (University ofArizona), and John Holtzclaw (SierraClub) were there to plead the case againstsprawl. Peter Gordon (University ofSouthern California), Genevieve Giuliano(University of Southern California), andRobert Lang (Fannie Mae) countered withthe benefits and normalcy of market-driven development.

The polarized positions of forum speakersleft little room for anything other thanagreeing to disagree. Armed with data tobolster their cases, speakers clungsteadfastly to their agendas. In the fewinstances where the data were similar,these data were interpreted as consistentwith diametrically opposed positions. Thesprawl-anti sprawl positions hardened.

While established players echoed now-familiar refrains, several new playersentered the debate. The strongest of thesefavored the pro-sprawl position. RobertBruegmann, an urban historian(University of Illinois), spent considerabletime debunking the myth that sprawldevelopment patterns are either uniquelyAmerican or associated with the growth ofthe automobile. According to Bruegmann,sprawl was spawned in the nineteenthcentury by the horse and buggy and laterby streetcars. Suburban-typeneighborhoods were actually foundthroughout Europe in the nineteenthcentury, having nothing to do with eitherAmerican cultural norms or theappearance of its automobile. Bruegmannsees the new urbanists as adding little butdesign innovations to sprawled locations,while mouthing the platitudes of the newcommunity advocates of the 1970s(Bruegmann 1998).

The negative side of the Portland, Oregongrowth boundary was clearly articulated

by Jerald Johnson, an economicconsultant from that city. According toJohnson, the Portland urban growthboundary has succeeded in bothincreasing density and containing growth,but even more so, it has caused housingprices to rise. Johnson presentedinformation indicating that housing pricesin the city of Portland had increased atmultiples of the level of density increases.Portland is becoming a victim of its ownsuccess. Housing demand and prices arehigh in a community noted for outwardgrowth restraints and attention to qualityof life (Johnson 1998).

Controlling free market development is adifficult sell in the Southwest, yet theimportance of desert lands preservationwas clearly articulated at this conference.In a state known for its creativity insiphoning off Colorado's water to reclaimthe desert, there was recognition thatdevelopment had to be contained and themore valuable parts of the desertinventoried and preserved.

RESPONDING TO THECHARGE: REGIONALCOOPERATION ANDREGIONAL/STATEWIDEPLANNING

A one-man crusade against factionalizedgovernment, because it creates sprawl, hasbeen waged by Myron Orfield, staterepresentative for the City of Minneapolisin the Minnesota House ofRepresentatives. Orfield believes that thebest way to control sprawl is to get localgovernments to cooperate in developingregional strategies, land-use policies, andregulatory mechanisms. In his bookMetropolitics, Orfield composed anaggressive regional strategy that links taxbase sharing to affordable housingprovision, farmland protection, andurban/inner-suburb redevelopment



(Orfield 1997). Orfield is a realist,however, and acknowledges that regionalgovernments are not growing nationallybut regional cooperation is. Currently,there is increased willingness to shareselected municipal service deliverysystems; there is virtually no interest informing new regional governments(Petersen 1996).

At another level, there is an ongoing effortto promote planning at state and regionallevels and to coordinate planning withinfrastructure provision. State plans andgrowth management initiatives have beensuccessfully put in place for the entirestates of Colorado, Florida, Georgia,Maine, Maryland, New Jersey, Oregon,Rhode Island, and Washington, and forspecific areas (e.g., the Coastal Zone, etc.)in California and North Carolina(DeGrove 1990). The guru of statewideplanning, who has followed it for most ofhis career and has testified as an expertwitness in most state house hearings, isJohn DeGrove of Florida AtlanticUniversity. DeGrove is also politicallyastute and realizes that even the mostencompassing state plan or growthmanagement act will either be voluntaryfor compliance by subunits ofgovernment, or non-punitive for non-compliance by these same subunits.

No discussion of growth managementwould be complete without discussingthe work of Douglas Porter of the GrowthManagement Institute and Arthur C.Nelson of Georgia Tech. For a decade,Porter has been a focal point of theliterature on growth management. Porterparticipated in Sam Seskin's "Transit andUrban Form" study (ParsonsBrinckerhoff 1996d), Reid Ewing's BestDevelopment Practices (Ewing 1995a),and the LUTRAQ study (Davis andSeskin 1997). From State and RegionalInitiatives for Managing Development to

The New Jersey Transportation PlanningAuthority, Inc. (NJTPA) serves as a forum forcooperative decision-making in the 13-county,northern New Jersey regional area.Source: New Jersey Institute of Technology and

Rutgers University. TELUS: TransportationEconomic and Land Use System—State-of-the-Art Information System for the 21stCentury (April 1998).

Managing Growth in America'sCommunities (Porter 1992; Porter 1997),Porter has been involved in implementingmanaged growth alternatives. Thisincludes model regulatory andprogrammatic techniques and pairingthese specific techniques with a particulargrowth management issue or problem.

Nelson has similarly authored TheRegulated Landscape (Knapp and Nelson1992) and Growth ManagementPrinciples and Practices (Nelson andDuncan 1995) and has been a principal inmultiple regional costs of growth studies.One of the most difficult tasks in land use



is to effect meaningful regionwide growthmanagement. Both Porter and Nelsonhave been involved with many regionalgrowth management implementationefforts.

RESPONDING TO THE CHARGE:SUSTAINABLE DEVELOPMENTAND SMART GROWTH

As yet another response for a newdirection in land use, the sustainabledevelopment and smart growthmovements have emerged in the UnitedStates. The U.S. sustainable developmentmovement is a direct outflow of the WorldCongress on Sustainable Developmentheld in Rio de Janeiro in 1992. Thisphilosophy of development reflects adesire to "develop today withoutcompromising available resources forfuture generations." For the most part,overburdened U.S. communities in theSouth, Southwest, and West have justifiedgrowth management programs under theguise of compliance with this norm(Krizek and Power 1996).

Currently, twenty-one communities in theUnited States have adopted sustainabledevelopment ordinances that essentiallylimit growth to the degree that publicfacilities and services are in place to

This philosophy of developmentreflects a desire to "develop todaywithout compromising availableresources for future generations."

accommodate this growth. Counties andregions are preparing developmentpolicies consistent with the goals ofsustainability. In Florida, the Governor'sCommission for a Sustainable SouthFlorida in December 1997 enacted anenergy conservation policy for thesouthern portion of the state. Amongenergy-conserving ideas, the Commissionrequired utility companies to derive

measures other than expansion of the userbase as appropriate indices of companyperformance. Further, this Commission isdeciding how improved transportation,education, and employment opportunitieseither add to or possibly detract from thegoals of sustainability.

Precursors to current sustainabilityregulations were the 1970s growth controlefforts of California and Florida cities,and the concurrency requirement of theFlorida Growth Management Act of 1985.In the United States, the President'sCommission on Sustainable Development,the U.S. Department of Commerce, theU.S. Economic DevelopmentAdministration (EDA), and the U.S.Environmental Protection Agency (EPA)have implemented sustainabledevelopment objectives that their fundedprojects must observe. For the most part,the emphasis on sustainable growthensures that capital projects respect theenvironment of which they are a part anddo not unnecessarily spur growth inlocations where existing infrastructurecannot support the growth.

Smart growth was an initiative of theAmerican Planning Association (APA),the U.S. Department of Housing andUrban Development (HUD), and theHenry M. Jackson Foundation on the onehand, and the National Resource DefenseCouncil (NRDC) and the SurfaceTransportation Policy Project (STPP) onthe other. The APA/HUD initiative calledfor an updating of land-use controls tomake them more sensitive to the ongoingproblems of lack of housing diversity,traffic congestion, and environmentaldegradation. The initiative also called forland-use controls that emphasizedcompact development to conserveresources; that limited development inundeveloped areas while encouragedinvestment in older central cities; thatpromoted social equity in the face of



economic and spatial separation; and thatwere sensitive to the role of the privatemarket and the need for simplicity andpredictability in land use (APA 1997).

The NRDC/STPP Smart Growth effortconsisted of a "Toolkit" for policymakersthat attempts to promote growth that is"compact, walkable, and transitaccessible" and will ultimately "competebetter with sprawl in policy forums and inthe marketplace." The Toolkit contains:(1) three policy reports on sprawl'senvironmental, economic, and socialimpacts; (2) research reports on sprawl-induced fiscal impacts and infrastructurerequirements (including utilities androads); and (3) a "Smart GrowthGuidebook" (NRDC/STPP 1997).

Maryland adopted smart growthlegislation at the state level in 1997. Thislegislation withholds, or at least sharplylimits, any subsidies for new roads,sewers, or schools for politicaljurisdictions outside state-targeted smartgrowth areas (Maryland Office ofPlanning 1997). Rhode Island andColorado have also adopted similarinitiatives (ULI 1998).

The Smart Growth initiative callsfor an updating of land-usecontrols to make them moresensitive to the ongoing problemsof lack of housing diversity,traffic congestion, andenvironmental degradation.

Each of the above techniques has as itsbasis the better management of growthand more compact development for thepurpose of resource conservation.In a September 1998 speech inChattanooga, Tennessee, Vice PresidentAl Gore recommended a "renewed federalcommitment to the policies of smartgrowth (Gore 1998).

A GROWING CONCERN—THEEQUITY ISSUE IN SUSTAINABLEDEVELOPMENT

One of the under-researched sides of thesustainable development movement iswhether there is a dark side to the goal ofnot compromising the physicalenvironment for future generations. At theregional level, this dark side might takethe form of freezing the movement ofminority and ethnic populations to theouter reaches of the metropolitan area bymaking inner cities and inner suburbs"more attractive" to all, and especially tothese groups. Thus, exurban resourceswould be "sustained" by reduced access tothese resources by those with the leasteconomic wealth (Lake 1997). The centralthesis of the equity issue is that betterenvironments for some will mean worseenvironments for others. Even if there is asolution that improves conditions forsome without hurting others, the benefitsof better environments will still beunevenly distributed (Marcuse 1998).

New urbanists take a hit in this literaturein that their new environments for themost part continue to promote new spaceconsumption: suburban-bound, affluenthousing seekers (few "new urban"environments accommodate the poor inurban areas). Resultantly, many of thesenew environments do little to improve thephysical sustainability of urban areas.

In a four-day National Science Foundation(NSF) workshop at Rutgers University inthe spring of 1998, Robert W. Lake ofRutgers and Susan O. Hanson of ClarkUniversity brought togetherenvironmental and first/third worldresearchers from the United States,Canada, and the United Kingdom topropose an integrated agenda for studyingurban sustainability. This involveddifferentiating between such terms as



urban sustainability (developed areasfunction to minimize the consumption ofresources and manage equity) andsustainable development (an increment ofland is developed to minimize theconsumption of resources), anddetermining the effect of scale (localversus global) on both definitions andissues. The results of this workshop willbe used by NSF to formulate a researchprogram on urban sustainability.

The workshop focused on four issues, anda research agenda will be preparedregarding each. These are:• economy-environment,• local-global,• urbanization as process,• and governmental and institutional

intervention.

SUMMARY

According to Robert G. Healy of theNicholas School of the Environment atDuke University, the time may be rightfor an "alignment of the stars" on land usepolicy affecting sprawl. Healy points outthe following signs: (1) the transit bikepath and urban trail initiatives of the 1998Transportation Equity Act for the 21stCentury (TEA-21); (2) states moving tosustainable development, smart growth, oropen space land acquisition initiatives; (3)citizens approving park and recreationbond issues of $1.37 billion in 1997; and(4) private industry initiatives such as theSilicon Valley Manufacturers Group'sattempts to support public transportation,affordable housing, and environmentalprotection to ensure that sought-afterworkers will continue to be attracted tothe San Francisco region (Healy 1998).

Healy notes that the situation is differentfrom federal land use initiatives of the1970s that failed to get out of Congressbecause:

1) There is agreement in theenvironment and developmentcommunities that growth isinevitable but must be carefullymonitored;

2) Sprawl is understood as anoutcome of current conditions, andpositions—both positive andnegative—have been taken aboutit; and

3) Federal, state, and localgovernments are moving in similardirections in land use, and theirsprawl-abetting and sprawl-controlling roles are being carefullyexamined.

Sprawl is a type of growth in the UnitedStates that even the most unenlightenedrealize needs rethinking. Yet sprawl is soendemic to the culture of the United Statesthat it is almost impossible to change.Americans like its outcome. It providessafe and economically heterogeneousneighborhoods that are removed from theproblems of the central city. In low-density, middle-class environments, lifetakes place with relative ease, and whenresidents wish to relocate, they typicallyleave in better financial condition—theresult of almost certain housingappreciation in these locations.

The public services available to residentsin sprawl locations are more thanadequate—and their cost, until recently,has been relatively inexpensive. But costsare beginning to increase. Americans arelooking, albeit halfheartedly, for analternative to current developmentpatterns. There is a general sentiment thatcommunities and individuals specifically,and society as a whole, cannot continue topay for the costs of sprawl. Costs havebeen held at a manageable level onlybecause overall infrastructure is under-provided and developmental infrastructureis not repaired adequately or replaced.



Over time, sprawl has garnered a long listof detractors, but increasingly observersare asking that the issues be discussedfairly. Most of the early literaturecriticized sprawl, but much of the recentliterature asks for an analysis thatdeliberately isolates both the costs andbenefits of sprawl. This is the emphasis ofSection II of this study: to break down thephenomenon of sprawl into its basicalleged impacts, both positive andnegative, and to detail deliberately thestrengths and weaknesses of each impactstatement with specific citations from theliterature. Impacts are categorized in fivegroupings. These are:

1) public-private capital and operatingcosts;

2) transportation and travel costs;3) land and natural habitat

preservation;4) quality of life; and5) social issues.

The above categories obviously containsignificant overlap. The objective is not todefine mutually exclusive groups but tobegin to point out and synthesize themajor concerns of the literature.

tcrp report 39 - transportation research...

Documents