USF Libraries
USF Digital Collections

The costs of sprawl--revisited


Material Information

The costs of sprawl--revisited TCRP project H-10 literature review
Running title:
Transit Cooperative Research Program (TCRP) H-10
Physical Description:
1 online resource (i, 193 p.) : ;
Burchell, Robert W
Rutgers University -- Center for Urban Policy Research
Brookings Institution
Parsons, Brinckerhoff, Quade & Douglas
ECO Northwest, Ltd
Transit Cooperative Research Program
Rutgers University, Center for Urban Policy Research
Place of Publication:
New Brunswick, NJ
Publication Date:


Subjects / Keywords:
Land use -- United States   ( lcsh )
Local transit -- United States   ( lcsh )
Municipal services -- Finance -- United States   ( lcsh )
Urban policy -- United States   ( lcsh )
Cities and towns -- Growth -- United States   ( lcsh )
Suburbs -- United States   ( lcsh )
Infrastructure (Economics) -- Costs -- United States   ( lcsh )
bibliography   ( marcgt )
technical report   ( marcgt )
non-fiction   ( marcgt )


Includes bibliographical references (p. 175-193).
Statement of Responsibility:
Rutgers University, The Brookings Institution, Parsons Brinckerhoff Quade and Douglas, ECONorthwest.
General Note:
"Principal investigators, Robert W. Burchell ... et al"--P. 2 of cover.
General Note:
"April 22, 1997."
General Note:
"Work in progress--not for quotation or attribution. April 23, 1997"--P. 2 of cover.

Record Information

Source Institution:
University of South Florida Library
Holding Location:
University of South Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 028898375
oclc - 754226379
usfldc doi - C01-00057
usfldc handle - c1.57
System ID:

This item is only available as the following downloads:

Full Text


LITERATURE REVIEW AND ASSESSMENT TCRP PROJECT H-10 "The Costs of Sprawl-Revisited" Research Team RUTGERS UNIVERSITY CENTER FOR URBAN POUCY REsEARCH 33 Livingston Avenue, Suite 400 New Brunswick, New Jersey 08901 1982 (908) 932 3133 and THE BROOI(INGS lNsmUTION 1775 Massachusetts Avenue, NW Washington, DC 20036 (202) 797-6000 and PARSONS BRINCKERHOFF QUADE AND DoUGLAS 400 6th Street Portland, OR 97204 (503) 274-9554 and ECONORTHWEST The Atrium, Suite 400 99 West Tenth Street Eugene, OR 97401 ( 541) 587-0051 Principal Investigators Robert W. Burchell David Listokin Distinguished Professors Hilary Phillips Research Associate (908) 932-3133 Anthony Downs Senior Fellow (202) 797-6132 Samue l Seskin Principa l Professional Associat e Judy S Davis Supervisory Planner (503) 274-9554 Terry Moore Vice President and Project Manager David Helton Michelle Gall Economists (541) 687-0051 WORK IN PROGRESs-NOT FOR QUOTATION OR ATTRIBUTION April 23, 1997


'"lbe Com of Sprawl-l!rAA!od" Uteral\lofe Review TABLE OF CONTENTS INTRODUTIO N AND BACKGROUND ......... .... .... .. . ............................ I PART ONE: HISTORICAL OVERVIEW AND ANALYSIS OF THE LITERATURE ON SPRAWL ......... 6 Historical Overview of the Literature on Sprawl ........................................ ..... 7 Topical Review of tbe Literature on tbe Costs and Benefits of Sprawl. .................. 20 PART TWO: LITERATURE SYNTHESIS OF THE COSTS AND BENEFITS OF S PRAWL .................. 43 Relating Sprawl's Defining Elements to it s Alleged Impacts ....... ..................... 44 Detailed Analysis o f Costs and Benefits o f Sprawl ..... . ............... ...... ......... . .54 PART THREE: APPENDIX ANNOTATION OF KEY STUDIES AND RESEARCH BIBLIOGRAPHY ............. . ... ............. 103 Annotation of Key Studies ................................................... ... ... ..... .. 104 Public-Private Capital and Operating Costs .................. ..... ..... . ... ....... I 05 Transportation and Travel Costs ...................................... .............. 117 Land/Natural Habitat Preservation .......................... .. ........ . ............ 138 Quality of Life ... .............. ... ..................... .......... .......... . ...... 142 Social Effects .. ....................................................................... 152 Research Bibliography . ............ .......... ............. .. . ......... .......... .......... 174 RUTGERS ta.OOK.1NGS PARSONS ECONOfthwest TRANSIT COOPERADVE RESEARCH PROGRAM (TCRP) fi.IO




INTRODUCTION The literature review that follows is an analysis of the writings and srudies con cerning a pattern o f land development in th. e United States tenned "sprawl." Sp rawl is the spread-out, skipped-over development that c h aracterizes the noncentr.U city metropolitan areas and non metropolitan areas of the United States. Sprawl is two-story. single-family development on one-quarter to one acre lots. accompanied by strip conunercial centers and industrial parks of a similar number of stories and amount of land takings (Ewing 1997). Sprawl occurs because local government in the United States encourages this fonn of development via zoning and sub division ordinances which, in tum, are reflective of the desires of their citizens. This type of development is sought by the public at large because it (as well as a combination of other factors):(!) man ages congestion while accommodating unlimited use of the automobile; (2) dis tances new development from fiscal and social problems of older core areas; (3) provides a heterogeneous economic mix; (4) ensures neighborhoods in which housing will appreciate and schools that provide both education and appropriate socialization for youth; and (5) generates property taxes that pay for local and school district operating expenses that are les s expensive per $1.000 of property value than locations closer in (Burchell !997a). Sprawl works so well in terms of acceptance by the public that the AAA rated locations for both residential and nonresidential development are increas ingly farther out rather than closer in and more rather than less segregated by type of land use (Gordon and Richardson 1997) Gated communities, fannettes, research parks, law offices medical groups, hardware and home improvement stores theatrical and comedy clubs, new R\lfGBtS &ROOKING$ PARSONS and used car lots and restaurants all now seek peripheral locations in pursuit of their markets. Starti11g with single-family subdiv i sions, then shopping centers and garden apartments. then research and industrial parks, then restaurants and entertainment facilities and fmally, discounters of every fonn, all of these enterprises have made their way increasingly far out into the metropolitan area. The unique aspect of all of this is that few entities have ever failed because their locational decisions were in the wrong direction. Occasionally a retailer or a residential development went under because an exit on the interstate or beltway wasn't developed as planned, but rarely has an economic entity failed in the United States because it was developed too far out. The newest and soon-to-be one of the most successful airports in the United States is 33 miles from the city of Denver-a $40 taxi ride from the airport baggage claim to the downtown Hyatt. I s the above an anomaly? No-Cincinnati's new airport is so far from the downtown that it is not even in the same state If all of tbe above is true, wby should the citizens of the United States accept anything else? The answer is that they no longer can pay for the infrastrucrure that is necessary to keep the current system of development going In the state of South Carolina, statewide infrastructure costs for the period 1995 to 2015 are estimated at more than $56 billion. This i s equiva lent to raising $1,000 from every citizen every year for the next twenty years. In addition to a massive infrastructure con servation program and the adoption of numerous technologic al cost savers, funding infrastrucrure in this state could require an increase in the gasoline tax of 2, an increase in the state sales tax o f 0.5%, an increase in property taxes of 12.5%, the tolling of all interstates at 31} mile intervals impact fees on residential and nonresidential development of $2,000 per unit and per 1,000 square TRANSIT COOI'/IATIVE RESEARCH PROGRAM (TCJ!P) H-10


feet, respectively, and a mandatory 10 percent set-aside for infrastructure in all state, county, municipal, and school dis trict general fund and intergovernmental transfer revenues (Burchell1997b). What is the big-ticket item for the above infrastructure? The answer: primarily roads. Roads will cost $25 billion of the more than $56 billion. Roads will cost 2.5 times what will be spent on primary/ secondary and higher education infrastructure, three times what will be spent on health infrastructure, including all hospitals and institutions and all water sewer treatment systems, ten times what will be spent on public safety, adminis tration and justice infrastructure, fifteen times what will be spent on environ mental proteCtion infrastructure, and twenty-five times what will be spent on all cultural and recreational infrastructure. Dually maintaining and underutilizing two systems of infrastructure-()ne that is being abandoned in and around central cities and close-in suburbs and one that is yet to be used fully in rural areas just begin ning to be developed-is causing govern ments to forego the maintenance of all infrastructure and not to provide other than growth-related infrastructure. The costs of infrastructure just cannot be paid for (Downs 1994) Employing new forms of technology. dedicating infrastructure funding from general fund budgets and from intergov ernmental transfers, and relying on new sources of revenue allow a state poten tially to meet its infrastructure needs. Doing with less infrastructure and devel oping differently further allow infrastruc ture costs to be met. But even these are not being done on a regular basis. And by no means is an alternative to the current pattern of land development the ultimate panacea If South Carolina were to go to compact development and manRUTGERS IROOIONGS PAitSONS ECONthd 2 aged growth measures to curtail spread development, the state would be able to save only about 10 percent of the $56 billion, or approximately $5.6 billion. This is because about 40 percent of public infrastructure costs are not growth related, one-third of the remainder is not new growth-related, and when develop ment pattern savings are applied to the appropriate portion of infrastructure costs, the saving is only 12-15 percent. Conversely, raising the gasoline tax by 2 raises only $56 million in new reve nues statewide-one-thousandth of total required infrastructure costs-and one hundredth of the amouni that potentially can be saved by altering land develop ment patterns (Burchelll997b) In sum, the current way of developing in metropolitan areas is not disliked by the majority of the American public-it simply can no longer be afforded. Thus, the primary concern about sprawl develop ment, at a time when the average Ameri can is satisfied with its outcome, is cost. In addition to financial concerns about capital facilities, other resource depletion is also causing public concern. Land is being consumed at triple the percentage of household formation rates, automobile use is growing at twice the increase in population, and prime agricultural land forests, and fragile lands encompassing natural habita ts are decreasing at com parable reciprocal rates (Landis 1995). The above situations have caused the pro fessional transportation and city planning communities to begin to look at sprawl to determine whether an altemati ve to this growth pattern can be conceived, and even more importantly, whether it makes sense to pursue this alternative. What is the alternative, if any. and does this alter native pose a viable option to current methods and forms of metropolitan development? A significant literature has developed in this area and is discussed in detail below. IY.NSIT COOPERATIVE RESEARCH PROGRAM (TCRP/ 11-10


BACKGROUND "Sprawl," in its broadest sense, has long been an American zeitgeist. Alexis de Tocquev i lle, touring the U nited States in the early 1 800s, observed "no urban growth boundaries," but rather marveled at "America ... where evefYthing is in constant motion ... and where no boundaries were set to the efforts of man." Today's spraw l is the "frontier" of long ago, or more recently the post-war suburb, both of which have been extolled as defining American influences. Yet values are subject to change While some view contemporary development patterns as a reflection of the invisible but sure hand of the market (Gordon and Richardson 1997), the unbridled movement outward of leapfrog, low density development is increasingly being viewed as an American ill. Sprawl has taken on both a pejorative as well as a descriptive connotation, an intermixing that makes a balanced discussion which disentangles the costs and of sprawl that much harder. To begin to set the context, criticisms of suburban sprawl are not necessarily criticisms of all suburbanization. The shift to the suburbs bas been manifest for mor e than half a century. In 1940, only 15 percent of the United States population was in suburbia-defined as metropolitan areas outside of central cities. As we approach the millennium, about 60 percent of the population is suburban. Even the most vehement critics of spraw l recognize that suburban growth has been and will continue to be inescapable in the United States With a recent population increase of some 20 million people per decade-a gain likely to continue for at least the next quarter..::entury-a great deal of that addition will continue to be in suburbs. It would be totally unrealistic to expect even a large share of growth to RUTGERS aROO:INGS PARSON$ ECONCII'ihw.11 3 literature Review occur solely in already built -up neighborhoods in cities or in close by communities. To call al l suburban settlements sprawl is a meanin g l ess exercise. Cross..::ultural and place-oriented differences factor into what is meant by sprawl. Density, or more specifically, low density, is one of the cardinal defining characteristics of sprawl. But density has to be set in context. Densities in the United States overall are roughly one-tenth that found in Western E urope. and in tum Western European density is much lower than that of Japan and a fraction of that found in such locations as Hong Kong and Indonesia (Jackson 1985). And in all of the above-named places, suburban densities are lower relative to the densities of central cities. Sprawl is not simply development that is at less than maximum density, but rather development that, given a national and regional framework (e.g., suburbs in the United States), is at a low density (specific threshold to follow) and one that may be too costly to maintain. Sprawl is often (yet probab l y should not be) equated with such loosely defmed terms as "social malaise This is not to say that sprawl does not af fect quality of life-it may-but sprawl has "taken it on the chin" in of its description as either "meaningless development" or more polemically, a "geography of no where" (Kunstler Like it or not sprawl is a purposefu l path en route to a specific place. Sprawl refers to a particular type of suburban peripheral growth. It involves very low-density developmen t that expands in an unlimited and noncontiguous way outward from the solidly built-up core of a metropolitan area. In terms of land-use coverage, such TRANSIT COOPERATM RESEARCH PROGRAM (JCRP) 11-10


"lllo Costs ol Sp!owi-Rovisi1od development contains primarily housing, including significant numbers of distant units scattered in outlying areas at extremely low densities. Sprawl also includes shopping centers, strip retail outlets along arterial roads, industrial and office parks, freestanding industrial and office buildings and other workplaces, as well as schools and other public build ings. These different types of land uses are, for the most part, spatially segregated from one another. The components of this development are individua!Jy located in small subdivisions and nonresidential tracts in zoning districts. Within each district, usually only one type of use is permitted. Examples are single-family residential districts, shopping center districts, strip commercial districts, and industrial or office park districts. Sprawl's other distinguishing include the consumption of exurban agricultural and other frail lands in abundance because at the periphery this is the cheapest land available. Under sprawl conditions, there is almost total reliance upon the automobile as a means of accessing the individual land uses. Some analysts also include the small developer and a lack of integrated development planning as important aspects of suburban sprawl. The result is relatively small residential subdivisions and nonresidential site plans created by individual developers operating independently of each other, within the zoning districts of the 10,000 local governments found in the United States, with almost no ability to control tempo and sequence of development. It is different, therefore, from the develop ment of large tracts of land each owned by a single developer, or under controlled develop ment conditions in a municipality wherein tempo and sequence of land use can be controlled by p hase. The legal R.UTGBtS BROOKINGS PARSONS literature Review framework within which sprawl occurs is fragmented into many relatively small units separately controlled by different local governments with differing rules and regulations. These localities have very different fiscal resources (assessed valuation of residential and nonresidential properties) per capita: some are quite wealthy, but others have very limited ability to pay for local services. These latter units of government are placed at a severe disadvantage. Sprawl is a complex phenomenon with attributes and consequences that have been both embraced and condemned. To aid in the study of what sprawl is and what its anendant costs and benefits are, first a historical overview of the literature on the subject for the past balf

alleged cost is the greater number of vehicle miles traveled (VMT) due to enhanced deconcentration. (That, in tum, raises the issue of what is more signifi cant-travel time or VMf). Each of these alleged costs and benefits is considered in tum below and relevant literature synthesized that bears on the subject. RUTGERS llOOKINGS PAISONS 5 literature R eview Both the historical overview of the literature and the topical synthesis are based on the research team's review o f hundreds of relevant studies. This full file is contained in a comprehensive bibliography with key studies annotated by topic area. The bibliography and annotations are found in an appendix, or third part of this literature search. TRANSIT COOPEI!ATIVE RESEARCH PROGRAM (TCRP/ 11-10




HISTORICAL OVERV IEW OF THE LIT ERATURE ON SPRAWL Sensitivity to the consequences of sprawl-like settlement long predates the coining of the tenn Tbe 1929 Regional Plan of the New York Metropolitan area, for instance, warned of a steady decrease in farms and open-space acreage in the region and underscored the need for settlement patterns that encouraged, rather than discouraged, "the face to face association that characterized the old v il lage community" (Regional Plan 1929, 23 and 216). At the same time, the Regional Plan spoke approvingly of ''many carefully planned outer subdivisions with good features" (Regional Plan 1929, 1). I t was not until roughly the late 1950s and early 1960s, however, that sprawl as a planning term entered the literature, with the pattern it depicted typically criticized. In 1956, a Canadian planning study described urban sprawl as "scattered building development" that led to "inconveniences in the placement of public and business facilities" (Lower Mainland Regional Planning Board 1956) A year later, William H. Whyte, describing urban sprawl as l eapfrog, scattered development spoke of it as a problem that had reached national proportions (Whyte 1957). Soon others entered the discussion; Marion Clawson, in 1962, described sprawl as a "lack of continuity in expansion" and noted it both was fostered by and contributed to land speculation (Clawson 1962) Somewhat more contemporary literature such as (1962), Harvey and Clark (1965) and Bahl ( 1968) viewed sprawl as charac terized by such features as low-density scattered and leapfrog development. Harvey and Clark (1965) identified the three cardinal traits of sprawl as: low BROOKINGS PARSONS ECONotfh.,.....e 7 density development ribbon development, and leapfrog development Even at thi s early stage, pundits acknowledged the difficulty in defining sprawl. Writing in 1972, David McKee and Gerald Smith observed that : Urban sprawl is rather d ifficult to define. In some circles the term is thought to be synonymous with suburbia. Certainly the problem exists in suburbia but suburbia itself is not the problem. Some equate sprawl with expansion. But this type of definition is not too helpful. (McKee and Smith 1972, 181-182) McKee and Smith went on to describe sprawl in four forms: I) very low-density development (e g two-to five-acre zoning) ; 2) ribbon-variety development extending along access routes; 3) leapfrog development; and 4) a "haphazard in t erming ling of developed and vacant land" (McKee and Smith 1972). The authors claimed that sprawl aggravated suburban problems (e.g. automobile dependence and the high cost of services and infrastructure) and also deleteriously affected cities by, among other things, depressing property values there Discussion of sprawl's effects transcend ed economics. Although the 1973 Rocke feller Brothers Task Force publication The Use of Land did not speak of sprawl per se, it concluded that tbe dominant pattern of "unrestrained, piecemeal urbanization" was leading citizens to ask how sucb growth affected their "quality oflife" (Reilly 1973, 33). In a similar vein, The lAnguage of Cities defined sprawl (Abrams 1971, 293-294): Sprawl, the awkward spreading out of the limbs of either a man or a community. The first is a product of bad manners, the second of bad planning TRANSIT COOPERAJIVE RESEARCH PROGRAM (TCRP} H-10


Sprawl is a by-product of the highway and automobile, which enabled the spread of development in all directions. As builders scramble for lots to build on, the journey to work is lengthened and green spaces are consumed by gas stations and elutrer. Professional work began to be undertalcen in numerous fields relevant to the study of sprawl. Examples include the 1963/nnovation Versus Tradition in Community Development (the effects of development patterns on road lengths); the 1967 Howard County Stutiy (comparative, county-wide costs for roads, utilities, schools and open space under sprawl versus more planned scenarios); the 1967 Urban Form and the Cost of Public Services (public service costs at different densities); the 1970 Planned Residential Environments, (different overall development patterns influence trip generation rates and distances); the 1972 Total Energy Demonstration (savings in energy consumption are likely in planned communities); and the 1972l.and Use Planning for Air Quality (development planning can affect air pollution on a regional basis). While not articulated, already the substantive foci in analyzing sprawl versus alternatives--namely issues of transportation, infrastructure, public service costs, and land and environmental issues--were being evidenced. Many of these studies were referenced by the bellwether The Costs of Sprawl authored by the Real Estate Research Corporation in 1974 (RERC 1974). As summarized by the RERC: This analysis presents a complete and inremally consistent set of estimates for direct costs and adverse effects RUTGERS &ROOKINGS PARSONS CONCH1h ..... et 8 resulting from prototypical housing types and land development patterns at neighborhood and community levels. Six neighborhood prototypes differing in housing type and density-are analyzed, along with six community prototypes which represent different degrees of community-wide planning .. .. Stated in the most general form, the major conclusion of this study is that, for a fixed number of households, sprawl is the most expensive fonn of residential development in terms of economic coStS, environmental costs, natural resource consumption, and many types of personal costs. (RERC 1974, 2-7) The Costs of Sprawl did not explicitly define the tenn "sp rawl." However, its analysis of six community level growth patterns implies that sprawl development bas at least two major traits: low average residential density (3 units per net residential acre or Jess), and a lack of overall planning at the regional or even community level. RERC did not define sprawl's gross density (population per square mile including all nonresidential uses) because RERC did not relate residential to nonresidential development in its analysis. Tbe RERC considered approximately 20 individual effects-e.g., "costs," "lands required," and "principal environmental impacts" (see Table 1). But as can be seen in Table 2, these can be grouped into four overall categories encompassing I. public -private capital and operating costs; 2. transportation and travel costs; 3. land and natural habitat preservation; 4. quality of life. Not considered in The Costs of Sprawl, and not part of its research charge, was any examination of sprawl's social effects, such as its impact on cities. TRANSJT COOPfl!ADVf RESEARCH PROGRAM (TCRP) ,.10


-o TABLE 1 The Costs of Spr aw l : Summary o f Find ings Communit IO 000 units Nei hborhood Low -dens ity Low..tJenslty Plonned High-density Single -family Single -family Townlu>use S raw/ PlilnMd Mix Plonned ConveniioiUll Clustered Clustered INFRASl"RUCfVRE Capital costs per unit Recreation $ 268 s 297 s 268 $ 297 $ 297 I $ 220 $ 274 $ 274 S ch ools 4 ,538 4,538 4 .538 4.538 4,538 5,354 5,354 4 538 Public Facilities 1,662 1,626 1,645 1,622 1,630 Roads/streets 3 ,797 3,377 3 235 2 708 2,286 3 080 2,661 2,111 U t ilities U21. Ull 2.W 2M2 l.W Infrastructure Subtotal 16,462 14,582 1 3 .556 1 2,487 10 995 14,137 11, 938 9,292 CONSTR.UcriONIOTHER1 31.J28 23.2Z6 12.211 31.2H l 1.32Q Tota l Unit Costs $ 51,456 $ 4 8 98 1 $ 37,283 $ 35,753 s 28,706 s 48,911 $ 46,258 $ 27,259 Public PropOrtion 1 9 % 12% 24% 16% 18% IS% 15% 20% Public Costs $ 9 777 $ 5,878 $ 8 9 48 s 5 720 $ 5 1 67 $ 7,337 $ 6 939 $ 5,452 01'\ATING Annual non-r e s idential opera1ing and mainttntu1ce costs per unit (in year 10) Opuating Costs $ 2.111 $ 2.067 $ 1,96 5 $ 1,937 s 1,873 I : 1,72 1 $ 1,720 Public Proportion 57% 51% 61$ 55% 55% 67% 67% Public Costs s 1,203 s 1,054 $ 1,1 99 $ 1,065 $ 1,030 1.153 $ 1.152 LAND Land required (for /0,000 units) Total Acres NA NA NA NA NA 5000 4000 Developed Acres 4,590 4,113 2 780 3,040 2,173 NA N A Vacant, Improved Acres 459 206 278 152 109 NA N A Vacant, Semi -improv ed 951 617 1.390 456 326 NA N A Acres Vacant . Unimproved I 1,064 1.552 2,352 3 392 I N A N A Acres Total Vacant Acres 1,410 1,887 3 220 2 960 3 ,827 I NA NA ENVIRONMENT Principal environmental impacts (jor 10.()()() unirs} Non-auto Air Pollutantstt 1,420 1,420 1 ,034 1,034 Sewage Eftluenf 4.5 4 5 4.5 4 5 Water U se d 1,170 1,100 910 9 1 0 Non-auto E.nerRY 2.355 2 355 1 750 1,750 No ul: AU figuru are per dwemn g u:ni t io 1973 d oOm N A =Not applicable Includes coMUUctiOfl c o s t of the. u n i tlltld Olhec such M land dedtcalion. b L b s per day. o: B illion llle11 per yell.r. 4 Mi:Ui on g.allon.s p y c v 0 B iUion BTU s per )'ell. RERC (1914), V ol. I, E.xu:wriw Swlt'IM4ry. 809 1,420 1,420 4. 5 4.5 4 5 760 1 ,205 1,059 1.400 2. 3 98 2.398 $ 1,388 72% s 999 3000 NA NA NA NA NA 951 4.5 913 1.595 1 000 units Walk up High rise A Ttment A rtm ent $ 252 $ 203 4,538 1,646 1,464 801 Ul2 2-ia 7 ,833 3 628 13.11 2 11.088 $ 21,282 $ 20,696 25% 13% $ 5,321 $ 2 690 $ 1 ,319 s 548 74% 51% s 976 s 312 2000 1000 NA NA N A NA NA NA NA NA NA NA 73 8 644 4 5 4 5 730 639 1,232 1.056


Litorature Review TABLE 2 The C osts o r Sprawl (RER C 1 9 74 ) S u b stanti ve Ar eas o r Inqui ry Topic s Co tuidered B y RBRC Public (1974) Prlvatt CapiJal and Operating Costs ''Costs" Infrastructure Recreation X Schools X Public Facilities X Utitilies X Road/suws Operating Costs X "und Requize d" T otal acres Developed acres improved/semi improved acres Improved acres Vacant unimproved acres "Principal Environmental impacts Nonauto air polt utants Sewage effluent Nonauto energy u se Water use The RE R C study evoked a flood of commentary-much praise as well as some criticism. Two of the better known criticisms were articulated by Altshuler (1977) and Windsor (1979). (These are considered subsequently in de t ail.) Among oth er po ints Altshuler argued !hat RERC underestimated the demand for service s by higher-density development and commingled the effects resulting from hi g h-density versus s maller-unit size W indsor, in paral l el criticized RE RC for not disentangling density from olher factors, and among other s h ortfalls, argued that RE R C RUTGEI.S aROOKINGS PARSONS ECONOifh_.... 1 0 Transportation Lllffd and Quality and Tra nl N atural of Lift C osts Habitat Pruervation X X X X X X X X X ignored !he benefit s of s prawl, su c h as its "response to consumer preference for single-family detached homes. Although The Costs of Spraw l dominated the literature for som e time, n e w analyses con tin ued to be pu b lis h e d. Ex amples include David Popenoe's (1979) depiction of sprawl as low-density, scattered strip development with adverse s ociol o g i cal im plications, and D avid Mills's (198 1 ) diseussion of how sp rawl-desc ribed by him as scattered, leapfrog development-both a betted and T R A NSIT COOPfRADVf RESEAR C H PROGRAM (TCRP} H-10


resulted from land speculation. Receiving much atXlaim was Kennelh Jackson's Crabgrass Frontier: The Suburbanizaticn of the United States, published in 1985. Although sprawl per se was not men tioned in this monograph, numerous traits attributed by Jackson to lhe "crabgrass frontier" were clearly "sprawl-like" in character. These attributes were: I) low residential densily wilb lhe absenoe of sharp divisions between town and country; 2) lhe socioeconomic distinction between the center and lbe periphery; and 3) a len g lby journey to work in terms of distance and time Jackson attributed lhe pennanence of the "crabgrass frontier" to p h ysical as well as socioeconomic factors (e.g., America was land rich and had fragmented local governments), and be noted its problems (e.g., high local public service costs and increased automobile dependence ) as well as its benefits (high housing amenily and individual open space). All in all with some exceptions (includ ing those studies cited above), in the decade follo wing The Costs of Sprawl, lhe literature on this subject was relatively quiescent This trend has reversed itself in lhe last decade; !here has been, as sbaJl be seen, an outpouring of studies. These are subsequently reviewed by substantive area. To give a sense of lbe current literature, with a focus on lbe definition of sprawl and its alleged costs and benefits, a sampling is discussed here. In 1993, a study conducted for lhe Chesapeake Bay Program defined sprawl as "residential development at a densily of less than three dwelling units per acre." This definition does not have a "locational component" and is a modified definition of one presented in an earlier draft: i. e., "developments having gross development llUTGUS IROOKINGS PAlSONS ECONCW'ihww 11 densities of less than three or four dwelling units per acre or minimum lot sizes of at least one-quarter of an acre, and frequently of at least one acre." The latter definition had been criticized by Uri P Avin (1993) as too high a density, since it would encompass many existing subdivisions in both Maryland and Virginia. Sprawl and more generally, suburbani zation, were condemned in a polemical book by James Kunstler ( 1993) The title of the book, The Rise and Decline of Man-Made Landscape, conveys his message The strident tone of lhe message is reflected by lhe following statemenc We have become accustomed to living in places where nolhing relates to anything else, where disorder. uncon sciousness, and the absence of respect reign unchecked. (Kunstler 1993) Peter Calthorpe's 199 3 book The Next American Metropolis offered a method for computing population den s itie s in an idealized form of modem settlement. He presented a s cheme for clustering housing and olher improvements around transit stops at specified densities which could be used to compute overall densities for ideal future metropolitan settlements His scheme essentially involved creating Transit Oriented Developments (TOOs) around the stations in a system of radial fixed-rail transit lines emanating from the region's major downtown. This approach was a metbod of quantifying aspects of a form of future growth alternative to sprawl. However, Calthorpe did not present any method of measuring lhe costs and benefits of sprawl or of lhe alternative form be suggested. Nor did be present any database to use in carrying out such measurements. TRANSIT COOPERAnVE RESE.'.RCII PltOGRAM (TCRP)If.l 0


In his 1994 book New Visions for Metropolilan America, Anthony Downs adopted a broader approach of defining sprawl that primarily related to density but included other characteristics as well. According to Downs, sprawl en compassed five major elements: (I) low-density, primarily single-family residential settlement (without any numerical density specified); (2) heavy dependence upon private automotive vehicles for all types of travel; (3) scatteration of job locations widel y across the landscape in mainly low-density establishments (also without any numerical density specified); (4) fragmentation of governance authority over land uses among many relatively small localities; and (5) widespread reliance on the flltering or "trickle down" process to provide housing for low income households. New Visions for Metropolitan America posed a basic method for analyzing sprawl by comparing its results to the results that might arise from alternative form s of metropolitan growth. Downs described a way of formulating alternative outcomes through an analysis of the basic traits of different growth strategies Downs's approach is incorporated later in this study and is described further there. RUTGERS BROOKINGS PARSONS ECOHotthWMI 12 As is apparent, even the most current literature on sprawl tends to describe its attributes rather than quantifying them. No quantified analyses of sprawl's relationship to other variables appears anywhere in the literature and, in tum, no one has mathematically or statistically linked sprawl to other conditions or metropolitan traits. The closest thing to quantification is the 1995 work of David Rusk in Cilies Without Suburbs He calculates an "index of elasticity" which measures the ability of c i ties to extend their boundaries to encompass surrounding urbanized development "Elasticity" is not the same as sprawl but comes the closest to a surrogate that has been found. Rusk claims that cities with high indices of elasticity are superio r to those with low indices of elasticity as they pertain to such traits as income distribution, racial integration, population growth, and economic development. The best cities are "elastic" cities because they have encompassed their suburbs. His index is applied both to the cities themselves and their metropolitan areas. TRANSIT COOPERATJY RESEARCH PROGRAM (TCRPJ Ji.l 0


Rusk did not perform any mathe m atical or statistical analyses rel a ting the variables just described, but three reviewers of hi s book

Richmond offers a wide-ranging critique of sprawl and includes numerous carefully culled statistics supporting his allegations against sprawl, many which are focused on the immediate subject of his analysis-the Chicago metropolitan area. Two aspects of his thinking, the components and critique of sprawl, are as one, since his criticisms are the ingredients he uses for his definition. Richmond's specification of sprawl is clear and most useful and is turned to elsewhere in this srudy. In defining sprawl, however, Richmond does not present specific alternative forms of growth, either conceprually or in terms of quantified analysis. Instead, he presents a long agenda of specific policy actions that would encourage a regional approach to managing furure growth. Therefore, his analysis does not provide either a method for measuring the costs of sprawl or a specific alternative development form that would provide a better outcome. Some of the most extensive quantitative work in formulating both the methods and data to address the costs and benefits of sprawl has been done at the Center for Urban Policy Research at Rutgers Uni versity and at the University of Cali fornia, Berkeley. Starting in the early 1990s, Rutgers University researchers, led by Robert W. Burchell, began to quantify the relativ e impacts of alternative patterns of development. One or two years later, under John D. Landis, similar efforts were being undertaken at the Institute of Urban and Regional Develop ment at Berkeley. Both research organiza tions have looked at the prospective impacts of alternative development patterns. Both research organizations have comprehensive land-use models to carry out these analyses (Burchell 1992a, J992b; Landis 1994, 1995). RUTGERS IR.OOKJNGS PAlSONS ECC>Northwtil tA Uterot\lre Review The Rutgers effort involved an analysis of the differing effects of "trend devel opment" (sprawl-like) versus "planned development" (with compact form and managed growth attributes) in New Jersey. The results obtained are shown in Table 3. This Rutgers srudy was fol lowed by other similar studies in Lexington Kentucky (1994), the Delaware Bsruary (1995), and the states of Michigan ( 1997) and South Carolina ( 1997). In all instances, polar develop ment patterns were contrasted-"current" or "trend," versus "compact," or "planned." Tbe exact nomenclarure is unimportant; what is important are the differing land-use configurations and their impacts, which are indicated below (Burchelll997, A-1): Current, or trend, development is his torical development in an area. The land use literature describes this type of de velopment as skipping over existing development; land-consumptive and inefficient use of available land at or near the core of the metropolitan area; and requiring significant accompanying infrastructure in the form of roads, water and sewer lines, public buildings. and the like. Compact, or a more managed type of development attempts to direct growth to already existing locations of develop ment while preserving yet-to-be devel oped areas. Nationally, the land-use literature portrays compact development as more efficient in its land-use patterns and thus less land-consumptive. Accord ingly, it often requires somewhat less development infrastructure. Compact development is also viewed as not limiting or restricting population or employment growth at tbe county, regional or state levels. TRANSIT COOPERATIVE RESEARCH PROGRAM (TCRP} H-10


Burchell developed a series of quantitative models relating to land consumption, road, transit, water/sewer infrastructure, fiscal impacts, hous ing cost, and quality of life to examine the relative effects of the alternative devel opment patterns. Application of these models across the aforementioned jurisdi ction s indicated compa.table order of-magnitude findings. For instance, a shift away from sprawl to compact growth was projected by Burchell to save water/sewer utility infrastrocrure costs by 8 percent in New Jersey, 7 percent in Lexington, 8 percent in the Delaware Estuary, 4 percent in Michigan, and 1 3 percent in South Carolina. Table 4 summarizes the array of findings from the various Burchell s tudies (1992 -1997 ) Table 5 groups the effects of sprawl, some dozen in all, into five overall categories. The Berkeley effort employs the California Urban Futures (CUF) model of the San Francisco Bay Area to tabulate land consumed under (a) "business as usual," (b) "maximum environmental protection," and (c) "compact cities" scenarios. These scenarios are differ entiated, by (a) not restrict ing development either within the city or within unincorporated areas, (b) applying a range of environmental restrictions to both locations, but not restricting growth per se, and (c) restricting growth to acknowledge some environmental limitations and countywide minimum population projections. The two latter alternatives showed consider.Wle overall and sensitive environmental land savings relative to the business as usual scenario. Total land saved in tbe final rwo scenarios (band c) were 15,000 and 46,000 acres, respectively. Scenario B saved nearly 60,000 acres of prime agricultural land, 10,500 acres of wetlands, and 3,000 acres of steep-sloped land; Scenario C saved 29,000 acres of prime agricultural RUTGER-S <OOKINGS PARSONS ECONorth.., 15 Literature Review land, I 0,500 acres of wetlands, and 8,000 acres of steep-sloped lands (Landis 1995). There are two final references in this his torical overview of sprawl. The first is what has been referred to as the Bank of America study (Bank of America et a!. 1995). The name that became associated with this study is more important than its position in alphabetical order because although four groups actually sponsored the study (Bank of America, Calif ornia Resources Agency. Greenbelt Alliance, and Low-Income Ho using Fund), only California's largest bank is ever associ ated with the study's results. This is because those who champion land devel opment approaches other than sprawl point to this study as the work of one of the private sector's most influential members. If the banks are finally recog nizing that sprawl can no longer be tolerated, the impacts of differing land development patterns on society's resources has indeed hit the big time. The Bank of America study summarized changes in population, demographics, and employment that had taken place over tbe two decades prior to 1990. It also referenced a land-use pattern lhat had taken place during this same period of time: the study termed it "sprawl." Sprawl was decentralized employment centers and residential tracts accessed almost exclusively by tbe automobile. These decentralized locations were safer and cheaper to locat e in and had plucked all fiscal and physical benefits from the central city. Further, this was aided and abeued by the federal subsidies given to the automobile. Tbe Bank of America report was criticized for its inability to adequately interpret lhe long-standing criticisms of the RERC Costs of Sprawl repon and the fact that its own study seemed to buy into every argument that TRANSIT COOPERATIVE RESEARCH PROGRAM (TCRP) Hoi 0


"The Costs ol Literat\lrt Rtview TABLE 3 Burchell (1992)-New Jersey Impact Assessment: Summary of Impacts for Trend versus Planned Development Growtlollhwlor"" Trtnd P/4toMd TrtMVmw Imp# 1 2 Pi4"Md DfwlfJpM#Itl I P0PUI.A1'10N GROwnt (persons) 520.012 520.012 0 0 II. HOVSEHOID GROwnt (bousebolds) 431,000 431.000 0 0 Ill. EMPI..oYM:E:m' GROwn! (employees) 653,600 653,600 0 0 IV. INFAAS'!RuctuR A. !lOADS ($ millions)3 Local $2,197 $1,630 $567 25.8 State = ill .l.l2. .LU Total !loads $2,924 $2,225 $699 23.9 8. lJiu.rnES-Walcr ($ millions) s 634 s 550 $84 13.2 c UTD..mEs-Sewer ($ millions) $6,790 $6,313 $4?7 7.0 -----Total Utilities $7,424 $6,863 $561 7.6 E. SCHOOLS ($ millions) $5,296 $5,123 $173 3.3 F ALL INFAAS'!RIJCIURE (sum of A-Bin $ millions) $15,644 $14,211 $1.433 9.2 v. LANDCONSUMFTJON A. Overall Land (acres) 292.079 117,607 174,472 59.7 B. Frail Lands (acres) 36,482 6,139 30.343 83.2 c Agricultural Lands (ac=) 108 ,000 66,000 42_,000 38.9 VI. HOUSEPiucE A. Median Cost per Unit SI72.S67 $162,162 $10,495 6.1 (1990 S) B. Housing Index 118 126 8 6 7 ibi2ber is more affordable\ Source : Roben W. Burchell et al.. 1992a. b I. n TABLE 4 Burchell (1992) Finding of Savings of Compact Growth ve rsus Current or Trend Development Lexington, KY and Area of Delaware Michioan Public Private Capital and Operating Costs I. lnfrasaucture Roads (local) 14. 8 .7% 12.4% 2. Utilities (wa ter/sew er) 6 7-8.2% 13.7% 3. Housing Costs 2.5 8 .4% 6.8% 4. Cost -Re venue Impacts 6.9% 3 .5% Land/Natural Habitat Preservation I. Developable Land 20.5.2% 1 5 5% 2 Agricultural Land 18--29% 17.4% 3. Frail Land 20-27% 20.9% Sourh New Carolina ,.,. .. 12% 26% 13% 8% 7% 6% S% 2% 1 5% 6% 18% 39% 22% 17% RUTGO.S 81.00KCNGS PAISONS ECC>Nof'thwtll 16 TRANSIT COOPERATIVE RESEARCH PROGRAM (TCRP) H-10


Literorure Review TABLE 5 Burcbell (1992-1997) ANALYSIS OF TREND VERSUS PLANNED DEVELOPMENT Substantive Areas o( Inquiry Public-Priww Capilill and Topics Omsidered By Operating Burchell {1992 -JWJ) Costs Warer/s..wer X infrastructure School capital X facilities Housing cost X Fiscal impacts X Roads Transil Land capacity Agricultural lands Frail lands Quality of life Intergovernmental coordination Effects to urban and t'W'al ceruers favored the anti-sprawl position without even looking at contrary evidence Those who championed the study as a sununary of the ills o f spraw l used the Bank of America's logo to promote the position that the business community, at long last, was calling for managed growth to con serve national resources The fmal reference in this historical overview of the literature on sprawl consists of two recent ( 1997) "point and counterpoint" articles i n the American Planning Association Journal.. The point article by Peter Gordon and Harry W. Richardson (1997) critiques the and evidence frequently to make the case for compact development (e.g., energy, transporta tion, and infrastructure efficiencies) and argues that the decentralized suburban pattern of development in fact offers UOOICINGS PARSONS ECON.,.,_ Traruporttllion Land and and Trove/ Narural Habital Quality Social Costs Presen.'alion Life Effects 17 X X X X X X X X many advantages, such as reduced travel times, higher consumer satisfaction, and reduced housing costs. In counterpoint, Reid Ewing (1997) claims a strong case for the adverse effects of sprawl (as opposed to the benefits of compactness) For the purposes of this review, their respective definitions o f terms bear note. For Ewing, sprawl is defined both by a series of three characteristics-(1) leapfrog or scattered development, (2) commercial strip development and (3) large expanses of low-density or sing l e -use developments -as well as: such sprawl indicators as low accessi bility and lack of functional open space (Ewing 1997, 108109). Gordon and Richardson do not specifically defme sprawl (or compactness for that matter); instead, they reference its various traits. Sprawl is alternatively denoted by Gordon and Richardson as l ow density, TRANSIT COOPERATIVE RESEARCH PROGRAM (TCRP/ 11-10


dispersed, or decentralized development whereas compactness is associated with higher densities and a downtown or centtal-<:ity spatial pattern versus a polycentric (or dispersed) spatial pattern (Gordon and Richardson 1997, 95). Although the point-<:ounterpoint authors address roughly 15 different subjects in discussing sprawl and its alternatives, the subjects can actually be grouped into five broader areas as shown in Table 6. In sum, since sprawl first entered the planning lexicon almost half a century ago, there has been an ourpouring of literature on the subject. The prior discussion bas presented an overview of an admittedly vast body of monographs, articles and reports. It has noted how sprawl has been defined and indicated the major topical areas of discussion with RUTGDS BJtOOI(INO$ PARSONS ECOHOI1h.,..ll 18 llhroturo Revl-ew respect to sprawl's costs and benefits. There are five such groupings: I) public private capital and operating costs, 2) transportation and travel costs, 3) land and natural habitat preservation, 4) quality of life, and 5) social effects. These individual categories obviously contain significant overlap. The objective is not to defme mutually exclusive groups but to begin to point out and synthesize the major concerns of the literature. This will be done in stages First, the Hterature on the five substantive areas by each of the categories will be overviewed. Second. this literature will be analyzed and deficiencies noted. Third, individual alleged costs and benefits subsumed under these five topical areas wiU be considered. Finally the pertinent literature will be reviewed and synthesized. TRANSIT COOPERATIVE RESEARCH PROGRAM (TCRP) fi.IO


Author Ewing (1997) Gordon and Richardson ( 1997) TAB L E 6 Ewing and GordonRicbardson ( 1 997) Su b stantive A reas of Inquiry Public Tl'(lltJportarlqn Lol!dand Topics Considered Private andTrt.lvtl Natural Habitat By AUlhors Capital ond COilS PreservatiO n Ope rating Costs Infrasuucwre costs X Public service X costs Transit X Vehicle miles X traveled Loss of X resowce lands Energy consumption Psychic and social costs Impac t o n central cities Infrastructure and Operating X Efficiency Transit X Economical resoorcc X allocation Congestion X Open space ard X agricultural land Energy glut Density p!eferenctS Downtown impaCLc: Equity Qutlliry SIX Ia/ lUI EJfects X X X X X X X RVfGERS BROOKINGS PARSONS ECONOI1h......t 1 9 TRANSIT COOPERATIVE RESEARCH PROGRAM (TCRP/ H-10


"'11--e Cosh of TOPI CAL REVIEW OF THE LITERATURE ON THE COSTS AND BENEFITS OF SPRAWL 1. Public-Private Capita l and Operating Costs Public capital and operating costs of sprawl refer to the construction of roads, water and sewer infrastructure, and public buildings. as weU as the annual expenditures to maintain them (in both smaU enclaves in remote locations of the metropolitan areas where population is growing, and in central cities from which some of the population growth is being drawn). Private capital and operating costs of sprawl refer to the construction and occupancy costs of private housing and commercial and industrial development. Most of the literature discusses bow metropolitan location and density/form of development cause these costs to vary. Subsets of this literature group"engineering per capita cost," "alternative growth," "regression," and "retrospective" studies-are described below The engineering per capita investigations examine the costs of different types of development by app l ying such factors as cost per linear foot of roadway, expense per gaUon of treated sewage, and municipal police cost per resident or per employee. The classic example of such a study is 'I'M Costs of Sprawl (RERC 1974). This analysis of neighborhood and community prototypes (Table I) applied a wide variety of engineering-per capita factors ranging from BTU consumption per square foot to schoo l and fire/police station capital standards per increment of new population. This was the first glimpse of capital cost variation by type of development and was weU received at a time when resource RUTGEf.S a.ROOKINGS PARSONS ECONorth.,.._. 20 literature Re'oie"W conservation was on every American citizen's mind. As noted, however, RERC's application of these simple engineering and per capita standards was criticized by a number of reviewers. Altshuler (1977) argued, for instance, that a uniform per capita public safety cost applied by RERC to all neighborho prototypes ignored the fact that demand for public safety services was affec .ted by density and would rise as density increased in the various prototypes. Despite such criticisms, many engineering per capita cost studies have been undertaken examining the capital and operating expense s of varying panerns of development. This is the mos t common method used in the literature from the 1950s to the 1980s and cited by Frank ( 1989) in The Costs of A/Jernative Development Pa"ems. Tbe consensus of that literature, according to Frank, was that capital costs were highest in situations of low density and for development located a considerable distance from central facilities Other e>:ampl es of stu

"The Cosio of Sprcrwi-ReviO!ed' evident; in The Costs of Sprawl, only a handful of hypothetical neighborhoods were examined The Burchell analyses range from large subregions of states to entire states They also are more inter related, with a series of land use, trans portation, and infrastructure models applied across a broad spectrum of substantive concerns to examine the effects of two differing development patterns. These models employ per capita averages but go beyond them. For instance water consumption is related not only to population growth, but also to housing type, density, and the intensity of occupation of structures; water pollu tion is related to population growth and influences such as density, housing type, and local soil conditions. Yet even Bur chell's multi layered modeling approach to specifying impacts has been criticized by some as oversimplifying a more com plex development reality (Gordon and Richardson 1997, 99) Regression analyses apply multivariate statistical tools to further refine the linkage between growth and public private capital and operating costs. Ladd ( 1992) used this approach to show how density would affect such public costs as public safety, utility delivery, and traffic management Ladd found these costs to increase initially as densities rose from very low levels (from 100 to 250 persons per square mile); the increase was due to development threshold costs. For instance, while septic systems could be relied upon at very low densities, density began to increase, more expensive sewage systems were needed. But at moderate l evels of density (200 to I 750 persons per square mile), Ladd found that costs would decrease as efficiericies of scale were realizedcentral sewage treatment facilities would replace individual subdivision-level package plants. Yet costs rose once again at higher densities (2,000 to 3,000 RUTGERS BROOKINGS PARSONS ECONOftfwmt 21 Literoture Review persons per square mile), as demand for such services as police and full-time fire protection grew. Ladd is cited by both the defenders and critics of sprawl, with the fonner referring to the cost declines at mid-range densities (Ewing 1997, 115), and the latter noting that costs rise with the initial increase of density and increase again at higher densities (Gordon and Richardson 1997, 99). A multiple regression analysis linking population growth and tax rates was perfolllled by the DuPage County Development Department ( 1991 ). This study was conducted in communities in the suburbs of Chicago, many of which were experiencing sprawl in the fonn of nonresidential strip development and industrial parks. The DuPage County study showed that property tax rates, entered as the dependent variable rose with nonresidential growth, entered as one independent variable-a fmding contrary to the conclusions of many other studies. Although pointing the field in new directions, the DuPage analysis was significanUy criticized. One reviewer concluded that there was some evidence for fmdings in the direction of the DuPage County results, but they were not nearly as strong as had been presented in the report (McDonald et a!. 1992). Other critics of the analysis included (a) those who believed both sides of the regression equation fonned an identity (whose intercorrelation prevented solution). (b) those who thought the research design should undergo significant alteration, and c) those who thought both dependent and independent variables should be recast (Mills eta!. 1991). Most believed that this study, as most of its genre, suffered from the inability to standardize for the quality and quantity of services delivered. A final group of studies bearing on the issue of how development patterns influence public-private capital and TRANSIT COOPfRADVf RESEARCH PI!OGRAM (TCRP) H-10


operating costs include a number of retrospective studies considering the effect of the overlay of regulations inherent in managed growth on the cost of housing. A number of investigations reveal that in the immediate areas where there are managed growth restrictions housing prices increase (Fischel 1990). Schwartz, Hansen, and Green ( 1 98 1 ), for instance, followed the effects over time of the Petaluma (California) growth control plan and found that after several years, Petaluma's housing prices had risen 8 percent above those of a control community with no growth restrictions. A similar study of the growth limitations in Davis, California (Schwartz, Zorn, and Hansen 1989) f ound housing prices t o be 9 percent higher there. Katz and Rosen (1987) analyzed 1,600 sales transactions of single-family houses in communities throughout the San Francisco Bay Area. Of these 1,600 transactions, almost 200 involved houses located in communities where a building permit moratorium or binding rationing system was recently or currently in effect, and housing prices in these areas were higher than in co m munities without such growth controls. According to Fischel (1990), this study is particularly valuable since, unlike other studies such as th e Petaluma and Davis investigations, the Katz-Rosen analysis did not focus on just a single community. This series of studies linked growth controls (as opposed to growth management) with increases in consumer housing costs 2. Transportation and Travel Costs Transportation is a discipline unto itself with a vast number of monographs, articles, and other publications devoted to it. The body of literature considered here includes key studies relevant to the current investigation of the costs and RUTGERS 1-.00IONGS PAlSONS ECON

TABLE 7 SUMMARY OF TRANSPORTATIO N S TUDIES REL EVANT T O THE EXAMINATION OF SPRAWL Author Setting Cambridge Systematics (1994) Los Angeles. CA suburban work sites Cervero (1989) 57 national "Suburban Empl oyment Centers" Cervero (1991) 83 buildings in 6 suburban activity centers Cervero (1996) Dispersed subccnters i n San Francisco Bay area Cervero and Gorham 1 4 matched-pair (transit (1995) oriented versus auto-oriented) neighborhoods in Los Angeles and San Francisco Davis (1993) Portland, Oregon Downs (1992) Hypothetical urban ar.a (s i mulation) and Page 1994 Stx Palm Beach County communities Frank and Pivo (1994) Gordon and Richardson (1997) 20 largest metropolitan areas RUTGUS 8lOOKINGS PAJSONS ECON"""- 23 Findinx Of the three site variables !.level of mixed use; 2. access to suvice; and 3 urban design features (e. g providing shade trees and sid ewalk), only urban des ign coupled with TDM materially increased commuting by ttansit. Center higher density and greater land use mix resulted in more commuting by transit ride-sharing and walking-but the denser, more mixed centers had slower travel speeds because of congestion. Strong r elationship be t ween density and transit use Significant (23 percenQ1980-1990 increase in the commuting VMT because of longer distances and a greater use of SOVs. with 80 pucent of the VMT gain attributed to the l onger distances between home and work. Urban Design-Hi;J>er transit use in San francisco ttansit-onented neighborhoods; no difference in Los Angeles Average ex urban home buyer had a commuting trip 6-7 minutes longer than his counterpart in suburbia-

TABLE 7 (continu ed) Author SeltlnJt Handy (1992) Match

(1989) (1993) The "facts" of the change in travel statis tics are themselves subject to dispute. Critics of sprawl indicate that subutban versus urban commuting times have been increasing-thus taking issue with the assertion that suburbanization means less congestion (Ewing 1997) Olhexs cite statistics that suburban vexsus utban commuting times now tend to be lower for suburbanites-purportedly an outgrowth of the freer flow of traffic with deconcentration (Gordon and Richardson 1997) In addition to the study and deciphering of gross travel statistics, lhe transpor tation literature looks at characteristics that both define deve l opment type and affect travel behavior. The characteristic most studied in this regard is density particu l arly how density affects trip length, mode choice, and other transpor tati on decisions. Critics of sprawl point out that residents of lower density residential environments use automobiles RUTGERS BROOkiNGS PARSONS ECONOI!fl_.., areas 25 Literature average increased. travel times only 40 seconds-alll'ibuted to an increase in vehicle trips (the fastest mode of travel). Other findings-a stabilizing in the number of people using b"ansit and average commuting times by late 1980s greater in suburbs than in cenual cities. the variation in use in percent in 1 970 times. Lower residential densities are associated with shorter commuting times by car o r transit and the clustering of commercial activities (e.g., CBD) produces congestion tha t increases commuting times. out areas inc.reased 0.5 to 14 percent with vecy significant increase: of 10 percent or more, in locations more and transit l ess, and have longer work Uips than residents in higher density areas. Tbe low-{!ensity areas are also more costly to serve with transit (Ewing 1997). Studies of this type include hypothetical simulations Downs (1992) showed that density of deve lopment at the urban fringe has a significant impact on com muting distances, with a move from very low to medium densities having the greatest impact on reducing commu ting d istance. Ex amination of density's influ ence also includes numerous empirical investigations of communities of varying residential densities (higher versus lower) and their travel behavior. Holtzclaw ( 1990) examined five communities of differing densities in the San F rancisco Bay area and found that those with higher residential densities had lower average vehicle miles traveled (VMTs). TRANSIT COOPfRAITVf RESEARCH PROGRAM (TCRP/ fi. 1 0


The hegemony of density is challenged, however, by those claiming that it is the "bottom line" effect of commuting time p lus congestion tha t matters, and that decentralized suburbs fare better on these measures (Richardson and Gordon 1989) To a much lesser extent, l and-use characteristics other than density are examined with respect to their travel influences. Davis ( 1993) considered how l eapfrog development increased commuting times in Oregon suburbs. Cervero (1989, 1996), Handy (1992, 1995), Cambridge Syste!Jlatics (1994), and Parsons ( 1996) examined how the integration (rather than the separation) of different land uses resulted in such travel effects as enhanced walking for internal trips and enhanced transit for external trips. A final component of the transportation literature considered here establishes baseline figures on the cost of travel. There are numerous site-specific investigations including in such places as Madison, WI (Apogee 1994) and Denver (Parsons 1996). User governmental, and societal costs of travel were identified in these locations, and the respective costs were then found to vary by travel mode (e.g., auto, transit, walking), type of trip (e.g commuting versus multipurpose shopping), time (of f -peak versus atpeak hours), the physical environment (higherversus lower density) and other factors (e.g., single-occupancy vehicle [SOV) versus high-occupancy vehicle [HOV) trips). Total automobile costs in the United States are provided by Voorhees (1992) and Delucchi ( 1996) and are estimated at $1.15 trillion and $2.36 trillion, respec tively. Voorhees includes direct expenses and some indirect costs, including wages lost from accidents. Delucchi includes a broader inventory of expenses such as the kUTGERS a-.OOI(INGS PAl .SONS E(0Northwed 26 literotvro R.-Mw cost of travel time, and such far reaching externalities as global warming. This broader accounting explains Delucchi's much higher cost estimate. Neither author includes the costs of sprawl in his tallies. In Delucchi's case this is because he views sprawl as a result of locational decisions, not motor vehicle use. Voorhees does not tally sprawl expenses because the cost is unknown and is at least partially factored indirectly in some of his calculations (e.g., higher fuel costs because of dispersed land uses). Mackenzie (1992) focuses on the subsidy to automobiles in the United Statesestimated at $254 billion yearly-and includes many externality costs in his tally (e. g U S armed forces providing security to Middle East oil shipments) Hanson ( 1992), who includes fewer externality costs than Mackenzie, estimates the automobile subsidy at $100 billion annually. Others rebut that, per user, it is transit. not the automobile, that is the most heavily subsidized mode of travel (Gordon and Richardson 1 997) Studies on the cost of travel are important to an analysis of the costs of sprawl for several reasons. First, they allow the analysis to make use of a ''full cost" framework, in which the researchers attempt to quantify and account for a much wider range of costs than may be typically included in discussions of public infrastructure costs or private housing costs. These include user costs govern mental costs, and costs to society at l arge. Second, the research indicates the magnitude of many of these costs and reveals that the estimates, particularly for social costs, vary widely. Third, the research indicates the incidence of the costs It emphasizes the need to TRANSIT COOPER/lOVE RESEARCH PROGRAM (TCRP) 11-10


"The Costs of Sprawl-4!.vi.ritod' consider both government and societal costs in the debate about the costs of sprawl. Fourth, the research deals with the costs of different modes of travel. Evidence presented in the transportation literature makes clear that tr.IVel behavior varies in different built environments, both with regard to the mode of travel and distances traveled. In order to estimate the costs of the set of travel behaviors associated with different spatial patternS of settlement and ultimately sprawl and its alterna tives--it is necessary to introduce the ftndings of the literature regarding the of different travel modes. 3. Land and Natural Habitat Preservation This subset of the literature includes, as a starting point, investigations on the overall trends on land consumption and on the threats to such frail lands as wetlands and prime agricultural acreage. lllustrative is a recent (1997) study by the American Fannland Trust (AFI') entitled Fanning on the Edge that noted that: Between 1982 and 1992, every state lost some of itS higher quality farmland, prime or unique, to urban development. Texas lost more prime and unique farmland than any other state (489,000 acres), accounting for 11.5 percent of the total loss in the United States. Other leading states with fannland lost to urban development were North Carolina, Ohio, Georgia, Louisiana, Florida, IUinois Tennessee, Indiana, and California (AFT 1997, 2) Numerous studies deal specifically with how different development patterns bear on land and natural habitat preservation. RVTGEI.S BROOKINGS PARSONS ECONcnh-st 27 The Costs of Sprawl (RERC 1974) ex amined land consumption in its different neighborhood prototypes (Table 1). Burchell (1992-1997) and Landis (1994, 1995) apply land consumption models to analyze compact versus trend develop ment scenarios in six locations across the United States and fourteen counties in California, respectively Burchell's analysis (Tables 3 and 4) and Landis's studies both show that the compact dwelling approach saves more land generally and especially targets agri cultural and fragile environmental lands for saving. A subject of the literature that cross -links the issues of land and natural habitat preservation and public-private costs is that of the fiscal effects of open space. Tischler & Associates (1989) find that improved open space (for recreational uses) does not pay its way. RKG Associates (1989) indicate that higher growth rates, given the medium-term introduction of marginal costs, are more costly to local government in the long term tban lower growth rates; however, growth rates can be attained by the purchase of developable land and holding this land off the market in perpetuity. The Vermont League of Cities and Towns and the Vermont Natural Resources Council ( 1990) caution against assuming that approving development is a fiscally positive act. Receiving considerable attention are AFT "cost of community service (COCS)" studies, which find that predominantly agricultural properties provide more revenues than costs--as opposed purportedly to all other land uses. The AFT studies have received considerable attention, but they bave been critiqued (Burchell and Listokin 1995) as understating farm-related costs (e.g., municipal and scbool costs engendered by farm families and municipal costs imposed by workers are not charged to TRANSIT COOPERAllVf RESEARCH PROGRAM (TCRP) H-10


!be fann sector) while overstating costs to other land uses. 4. Quality of Life This subset of !be literature, like the preceding, consists of general investigations on the concept of "Quality of Life" (QOL) and how can it be measured-as well as specific studies that attribute both positive and negative QOL qualities to sprawl and its alternatives. When Schopenhauer cynically stated that there are only two sources of human unhappiness-"not having what you want and having what you want"-he may have been alluding to the dilemma of quantity versus quality (Environmental Protection Agency 1 973, 1 -2). The aspiration of the past for quantities of things is rapidly giving way to a rising concern for quality of life. This shift is precipitating new attempts by individuals and by government to ascertain just what will bring a sense of well-being to people. As one would expect, tbere is no consensus on what QOL entails, so the review of the literature includes numerous studies that grapple w ith defining the concept and monitoring its attributes These studies come from various spectrums of the literature. A good place to start is with the QOL rankings of cities in the popular literature, including the Places Rated Almanac (Savageau and Boyer 1993), Money magazine's "Best Places to Live in America" (Fried eta!. 1996), and Fortune magazine's ''Best Cities: Where the Living is Easy" (Precourt and Faircloth 1996). These rankings all use similar categories of QOL life measures such as: 1. cost of living (cost of food, housing, utilities, taxes, insurance, and II.UTGRS IROOtONG$ PAASONS 28 literature Review transportation); 2. jobs and economic well-being (number of new jobs); 3. housing costs (cost of the average home); 4. transportation (average time commute, availability of public transportation, and number of airline flights); 5. crime (violent and property crime rates); and 6. climate (number of sunny days and average temperatures). One ranking with an environmental focus is Hall and Kerr's (1991) 1991-1992 Green Index. Drawing from a variety of public and private data sources the Green Index uses 256 indicators to measure and rank each state's environmental health. The indicators encompass a broad range of environmental conditions including air quality, water pollution. toxic waste, community health, forest s, and congressional leadership (Andrews 1996; Meyers 1987). In addition, many states, cities, and other jurisdictions have started to use indicators (also called report cards, benchmarks, or vital signs) to track changes Indicators are sets of specific measures for which data are available, grouped into categories. The specific measures are usually identified through a public process in which participants are asked to identify what aspects of their area (neighborhood, city, region, state) they care about most, and then to develop specific measures for those aspects. Ulustrative are the Oregon Benchmarks measurable indicators that the state of Oregon uses to assess its progress toward broad strategic goals (Oregon Progress Board 1994) Categories of Oregonian benchmark measures, under the heading Benchmarks for Quality of Life, are: Unspoiled natural environment: air, water, land, plants/fish/wildlife, and outdoor recreation. Developed communities that are convenient, affordable, accessible, TRANSIT COOPERAJIVE RESEARCH PROGRAM (TCRP/ H-10


and environmentally sensitive: community design, transportation, housing, access for persons with disabilities, access between communities, and emergency preparedness. Communities that are safe, enriching, and civic minded, with access to essential services: publ .ic safety, justice, access to cultural enrichment, sense of community, access to health care, and access to child care. The social sciences add to the literature on the quality of life. Maslow (1970) and Zinam (1989), for instance present a psychological hierarchy of needs ranging from the physical (safety of natural habitat) to the physiological (material well-being) to collective self-actualization (cultural heritage and consensus on values). Economists (Gabriel et al. 1996; Roback 1982 and 1992; and Rosen 1979) have developed QOL estimates and have entered into the discussion the concept of a "ftr st paycheck" (e.g., monetized wages and jobs) and "second paycheck" (QOL concepts of livability and urban and environmental amenity), with interactive effects between these "paychecks" and urban fon:n (Whites1aw and Niems 1989). This review of the popular, "benchmark," and social science literature on QOL provides a large number of potential QOL measures to be considered for incorpora tion in this study. These are listed in Table 8. (A subsequent section in the curren t investigation distills these mea sures for application to the sprawl and development alternatives.) While the vast majority of the literature on QOL deals generally with the overriding issue of what QOL is, a smaller number of studies specifically concern themselves with the QOJ..r.sprawl nexus. RUTGERS Bl()()I(JHGS !'ARSONS ECONOf'lh_, 29 U!erature Rmew The previous discussion about the "ftrst and second paychecks'' and their associa tion with urban development fon:n is i llustrative. Studies by Dowell Meyers (Meyers 1989, 1988, and 1987), also bring together issues of QOL and urban planning -urban form tbat are particularly germane to this project Another illustra tive article is one by Dav id Popenoe ( 1979) entitled ''Urban Sprawl: Some Neglected Sociological Considerations." Popenoe defmes sprawl as very low density urban development, oriented to the automobile, with detached single family houses on relatively large lots-a pattern which, as he sees it implies scatteration, a scarcity of open-green spaces, and a scarcity of commurtity focus in both the p hys ical and social sense Popenoe argues that sprawl leads to such negative QOL features as "inten sifying residential segregation by race and class which is "destructive to the central city," and leads to the "proliferation of fragmented and overlapping govern mental urtits. Some current studies often add as a QOL critique that contemporary, single-use residential subdivisions, accessible primarily by the automobile, bave lost their sense of ''place The idea of a multi-use village with people biking and walking has been replaced by row upon row of houses without sidewalks and bike paths. Instead they offer multiple off-street parking spaces to accommodate the automobile. According to the authors of these studies, most Americans would prefer to return to more traditional urban forms (Krieger 1991; Calthorpe !993; Nelessen 1994; Duany and Plater-Zyberk 1995). TRANSIT COOPERATIVE RESEARCH PROGRAM (TCRP) H-10


In opposition are those who claim that f o r the last twenty years American s have repeatedly indicated their preference forand satisfaction with-suburbs and suburban living (Gordon and Richardson 1997). The suburbs offer predictabiliry in the realms of enhanced public safery, public education, and housing invest ment. The suburbs also contain the single-family detached housing so desired by most Americans (Dyckman 1976; Goldberg and Mercer 1986; Audirac et al. 1990; Fannie Mae 1992, 1994). Compo nents of this debate are mirrored in an important subset of the literature location choice and decision making for households and businesses. 5. Social Issues There are many social issues related to sprawl; assembled for this r e view is literature focusing on how sprawl affects cities. Thi s includes as a starting point, studies on how to measure the "condition or health" of a ciry, especially relative to suburban communities. Examples are the "intrametropolitan hardship index" developed by Nathan and Adams (1976); the Bradbury, Dunne and Downs (1982) static and dynamic measures of "urban decline," "urban distress, and "city dispariry"; and various economic-social ratios of urban versus suburban cond i tion s in State of the Nation's Cities (Glickman, Labr, and Wyly, 1996). Next is a group of studies considering the historical development of suburbs w ith a recurring leit m otif of separation of, and exc l usion from, the older urban center. This theme is expounded upon by Fishman (1987) in Bourgeois Utopia (both hi. storical and more modem "technoburbs" were designed to house the more mobile elite and to keep the less RIJTGERS 8R00Kl'NGS PARSONS 30 Literature R.-..iew mobile poor and minorities in the cities) and in Jackson ( 1985) in Crabgrass Frontier (suburbanization led to the "polarization of the metropolis" [Jackson 1985, 274] that segregated the advantaged suburbs from distressed cities). Also considered are numerous recent studies that link the welfare of cities to the economic and social health of the overall metropolitan area (Downs 1994; Ihlanfeldt 1995; Ledebur and Barner 1992; Rusk 1995) and propose, in tum, that urban revitalization is futile without a closer integration of cities with their suburbs (Calthorpe 1993; Richmond 1995) Conditions in cities and the interconnec tions between cities and suburbs are cited by Ewing (1997) in his discussion of the costs of sprawl. According to Ewing, cities are not only important in their own right for retaining "higher order central place functions," but reflecting the work of Downs, Ihlanfeldt, Rusk and others, cities and suburbs are "inextricably linked within the metropolitan economy" (Ewing 1997, 117). Accordingly, Ewing asserts that sprawl, which encourages outward movement of population and functions from cities is a detriment both to cities and ultimately to suburbs as well. There is an alternative view, however An early expression of this was by Stemlieb and Beaton (1972). Sternlieb viewed cities as relics that bad lost much of their economic reason to exist as populations and emp l oyment suburbanized; this dysfunction was expressed by Sternlieb in his depiction of "cities as a sandbox" (Stemlieb 1964). Yet, for the urban poor and minorities there was hope in the form of "zones of emergence" (Stemlieb and Beaton 1972). These were inner-ring suburbs which, as their more mobile pop ulations left for the more vigorous outerlRANSfT COOPERATIVE RESEARCH PROGRAM (TCRP) H-10


titercture Review TABLE 8 Potential Quality or Life Measures Cost of Lhiae Cost of living index ReaJtb Care Doctors per capita Recreation coaomy Availability of advance

zation and dispersion. In a variation on the zone of emerg ence concept, Gordon and Richardson propose that minorities and those of modest income are, in fact. seeking to live in inner-ring s uburbs (Gordon and Richardson 1997, 102). Gordon and Richardson do not accept the conten tion that cities and their s uburbs are inextricably linked. In this opinion they are joined by numerous academics (Fish man 1987; Hartshorn and Muller 1989) as well as writers in the popular press (Turque and Washington 1991), wh o ar gue that suburbs are no longer dependent on central cities (authors cited by Thlan fel dt 199 7. Yet, as noted, there is a co n trary view by many authors (Downs 1994; Ewing 1997; Ru sk 1995) who argue that city-suburban fates are linked and that one of the most delet erious eff ects o f sprawl is harm first to tbe cities, and then ultimately to the entire metropolitan area. A NALYSI S O F THE LIT E RATUR E ON T H E C O S TS AND BE NEFITS OF S PRAW L Topic a l C o verage One logical starting point for analyzing this literature is to overview the distrib ution of the studies by topic and t o see where there is more or less topical coverage. This can be done quantitative ly, that is by examin ing b ib liographies on the topic-such as tha t assembled by the current author s (see appendix), as well as by other s (e.g., Ewing 1997; Gordon and Richardson 1997; the Growth Management and Research Clearinghouse 1993)-and t o count tbe numerical distribution of srudies by subjec t type This raises its own issues, however. such as bi as in the respective bibliographies (e.g., those emphasizi n g studies c ritical versus supportive of sprawl) and differences reflecting the 32 varying professional orientations of the bibliophiles (e.g., traffic engineer or historic preservationist). In doi ng a quantitative census of the literature on, and topical coverag e of. the literature on sprawl, there is also a question of how to count multiple studies by the same autho r b u t of a similar type. Should the Altshuler ( 1 977) criticisms of The Costs of Sprawl methodology, originally enunciated by him in 1977 an d then repeated in a jointly authored monograph b y Altshuler and Gomez Ibanez (1993) count as one or two entries in this cens us? Should Downs's pro lific publi cati ons invo lv ing the issu e of sub u rban exclusion of minoriti es be individually counted? (Downs 1970, 1972. 1973. 1981, 1985, and 1 994) The same issue of coun tin g arises in the parallel series of publications by Gordon and Richardson (1989, 1994, 1996 and 1997), Ewing ( 1994 1995, 1996, and 1997), and Burchell (1992a, 1992b, 1 994, 1 995, 1996 1997a 1997b). Y et another issue in doing a numeri cal census of the lite .rarure is wheth e r all entries are weighted similarly. Do 17re CostsofSprawl(RERC 1974) and the New Jersey impact assessment (Burchell eta! 1992a. 1992b}-both influential analyses of hundreds of pages each "weight" the same as bri efer and less subs t antive di sc ussions ? Over and above these questions. an attempt t o quantify the topical co v erage of the litera ture on sprawl is frustrated by such fundamental issue s as what is meant by sprawl and whal counts as llterature o n this subject. The former is no small maoer given that sprawl is often not defined nor its full elemental charac t eristics agreed upon. The latter in v olves the question of whether this literature consists only of materials on sprawl per se (e.g., examining spra wl' s TI!ANSlT COOPERAI!VE RESEARCH PltOGAAM (TOP)


"llte Com ol effects on infrastructure costs). or more broadly on topics relevant to a discussion on sprawl. The quality of life (QOL) subset is illustrative, for there is very little literarure directly dealing with QOL and sprawl, yet there are many studies more generally on QOL. For these many reasons, the topical Iiterarure coverage will not be analyzed quantitatively (e.g., so many studies of a certain type representing a given percentage of the literature) but rather qualitatively, that is, which subsets of the literarure are more or less frequently considered. By literarure is meant those studies relating sprawl with a substantive attribute of public-private capital or operating costs, transportation, land consumption/preservation, quality of life, and social issues. Without providing a statistical count, it is clear that the literature on sprawl focuses on the infrastrucrure and transportation issues---tlle "hard, physical" side of the subject. The classical and still repeated areas of greatest attention concern bow development patterns affect demand for arterial and local roads, sewer and water distribution lines, and transit and road systems. The measures focused on are the barometers of infrastructure and transportation, such as linear feet of roads, vehicle miles traveled, and vehicle hours of delay. The operating costs of the infrastructure, once installed, are given much Jess attention. Thus, although there are numerous studies relating density to the linear extension of roads and utility lines, there is scant attention paid to how density affects their operating expenses, such as the relative public safety (e.g., police patrol) expenses of police departments under sprawl versus more compact development patterns. RUTGERS tROOKINGS PARSONS 33 literature Review Also, at the core of the sprawl literature is examination of how this development pattern influences land consumption both total acreage and acreage of specific categories, such as agricultural or forest resources. This attention is the mirror of sprawl's defining characteristics. many of which are land-based traits: low density, l eapfrog development, and segregation of land uses The land consumption of sprawl is a topic commonly referred to, yet interestingly there aren't nearly as many analytical studies dealing with land as with the classic concerns of infrastructure and transportation. With respect to QOL and the social aspects of sprawl, there is a paucity of Iiterarure. Part of the difficulty is that it is more challenging to agree on the social parameters to be examined and agree to disagree on the appropriate social measures. In contradiction is the consensus among transportation engineers that the appropriate measures are roadway congestion (vehicle hours of delay), VMT, and the like. Although there is burgeoning recent interest in sprawl's psychological and social effects, such as its inability to insti ll a "sense of community" (Nelessen 1994; Etzioni 1993) and its consequence to cities (Downs 1994; Rusk 1993), this area of the literature is relatively lightly explored. Databases The five substantive literarure categories apply both common and subject distinctive data sources. Across categories, socioeconomic in formation from the decennial census (population and housing), the triennial American Housing Survey (AHS), and similar bases are frequently tapped. Both published (e.g., printed census reports) and computerized (e.g., Public Use Microdata Sample of the decennial TRANSIT COOPERAOVE RESEARCH PROGRAM (TCitP) H-l 0


census) sources are accessed. Land -use information of different types is also considered across all literature categories. This includes both deseriptors of a gross o r aggregate nature, suc h as population density derived from the City-County Data Book. and finer-grained land-use information, such as the neighborhood mix of land uses and urban design features, that are found as part of local records (e.g., zoning maps) and ind i vidualized study surveys. Supplem e nting this common socioeconomic and land -use information are data specific to the five literature categories. For the public-private capital and operating costs group, this includes both an array of engineering infrastructure as well as financial information. Examples of the former category include the Institute of Transponation Engineers (1984) Recommended Guidelines for Subdivision Streets; DeCitiara and Koppelman's (1975) Manual of Housing Planning and Design Criteria (1975); and the Urban Land Institute, National Association of Home Builders, and American Society of Civil Engineers' (1976) Residential Streets. Examples of the latter type of resource include the quinquennial Census of Governments as well as local operating budgets from municipalities, counties and school districts. The other litera!llre categories draw on in parallel sources penaining to their respective discip l ines and interests As examples, for transponation and travel costs, the National Personal Transport ation Survey is accessed; for landlna!llral habitat preservation, data from the U.S. Census of Agriculture; for quality of life, such guides as the Places Rated Almanac (Savageau and Boyer 1993); and f or city conditions examined under social issues, the City-County Data Book, and specific RUTGERS BlOOICINGS PAASONS 34 \.ittrature ii'Mw city distress measures from Downs et a!. (1982) to Rusk (1995). This overview does not convey the variety and richness o f the data sources that are tapped by the subsets of the Jiterdture. To convey that better, the information accessed by one of the categories-transportation and travel costs-is described in greater detail here. This body of literature draws upon database s relating to travel as well as information on social and land-use characteristics. Travel sources typically relied on include, as noted, the National Personal Transportation Survey (Pisarski 1992; Richardson and Gordon 1989, as examples); travel and commuting information from the decennial census and American Housing Survey (Gordon and Richardson 1991; Parsons 1996); and a variety of other sources such as Highway Sraristics (Dunphy and Fiseher 1994); travel diaries kept by households being surveyed (Kitamurea 1994); and automobile odometer readings from California smog inspections (Holtzclaw 1994) Household information such as that related to age, income, and occupation of residents is derived from the decennial census and the American Housing Survey (Gordon and Richardson 1989; Parson 1996; Pisarski 1992) Household surveys may supplement/update those national and regional data bases Land-use information comes from local planning and zoning records as well as from other sources. Gordon and Richardson (1989) for instance, measured density from the U.S. Geological Survey LANDSAT files. The costs of travel investigations incorporate a broad array of data sources on topics ranging from accident -rel ated medical expenses, to armed forces TRANSIT COOPERAOVE RIOSEAIICH PROGRAM (TCRP) HlO


spending (e g for providing security for overseas petroleum sources), to global warming. One fmal note with to the data in the sprawl literature is that much is of a secondary nature-that is, collected by one party and then reanalyzed or just cited by another This is exemplified by The Cost s of Sprawl neighborhood prototype data originally assembled by RERC in the early to mid-1970s (Table I) and still being relied upon some two decades later by today's commentators on the subject of sprawl (Altshu l er and Gomez Ibanez 1 993; Ewing 1997; Gordon and Richardso n 1 997) Study Type and Methodo l ogy There are various dimensions in considering methodology One is whether a study is empirical that is, examining something actual as opposed to a simulation, where events are modeled rather than observed. Both types are found. with the incidence varying by subject. In the public-private capital and operating costs literature, because development's effect on inf r astructure is often not readily observed, many such studies are simulations. Illustrative is The Costs of Sprawl (RERC 1974); Downing's (1977) capital extension supplement to RERC's original work; and Peiser's (1984) analysis of infrastructure costs in a hypothetical larg e subdivision. Operating costs, such as operating expenses per capita, are more readily observed (albeit not necessarily linked to development pattern), as are other financial parameters, such as tax rates, and these are often l ooked at mathematically. Examples are Ladd's ( 1991) regression analyses rel ating density to per cap ita government spending and the DuPage County ( 1992) regression of observed nonresidential development to observed tax rates. RUTGEtS BROOKINGS PAASONS CONol1hw..r 35 LQrotvro Review With respect to transportation and travel costs, because actual travel is routinely studied and indeed counted, much of this li t erature is empirical. Examples include Pushkarev and Zupan ( 1977), linking density and transit use in 100 urbanized areas; Cervero ( 1989), examining density and modal choice in 57 "suburban employment centers''; and Parsons et a!. ( 1996), examining the effects of density. urban design and mixed use on the demand for transit in various locations ranging from II metropolitan areas t o individual Chicago and San Francisco communities At the same time, the transponation literature reflecting the underlying discipline of transponation engineering with its modeling prowess, also incorporates some largescale si m ulations, such as the 50-year simulation by Metro (1994) and Downs s ( 1992) Stuck in Traffic modeling. The remaining literature categories, while applying some simulations, suc h as The Costs of Sprawl modeling of land consumption in alternative neighborhood prototypes are largely empirical. Tbe Green Index of l ocations (Hall and Kerr 1991), incorporated in quality-o fl ife measurement, thus encompasses more than 250 indicators related to environmental health (e.g., air and water pollution and community and workplace health statistics). City socioeconomic h ealth focuses on such observed characteristics as unemployment and property tax rates (Bradbury, Downs and Small 1 982) Much of the e m pirical work tends to be descriptive. A common style is a case study of what occurred at one or a limited universe oflocations. Examples include Duncan's (1989) analysis of infrastruc ture costs in a number of Aorida developments; Ewing's (1995) analysis of household travel partems in a Aorida county; and Ewing, Haliyur and Page TRANSIT COOPERATIVE RESEARCH PROGRAM (TCRP) fi.l 0


( 1994) and Cambridge Systematics (1994), examining travel in Palm Beac h County (FL) and Los Angeles (CA), respectively. Not coincidentally, the case analyses are focused on locations experiencing rapid growth-Qften sun belt locations. In fact, so many of the invesligations of travel proftle related to urban design have taken place in California, especially around San Francisco and Los Angeles, that questions about the replicability of the results observed to the rest of the country are beginning tO be raised by the research community at large. Various quantitative skills are inco rporated in the literature. The per capita infrastructure studies, for instance, are essentially arithmetic compilations, but higher-order applications are found. especially with the transportation and travel costs analyses. Here, for instance, are found such statistical tests as analysis ofvariance (e.g., comparing travel behavior in auto-oriented versus transit oriented neighborhoods [Handy, 1995] and multivariate regression (e.g., Pushkarev and Zupan [ 1977] using regression to show that much of the variation in transit use is explained by density). It must be remembered, however, that even the most "statistical" of studies are still cross-sectional They show in the travel literature for example, the correlation between current urban form (e.g., low to high densities, and segregated or mixed uses) and current travel behavior (e.g., mode choice or VMT) but do not show how changes in urban form have influenced changes in travel choices. This is one of numerous deficiencies in the literature noted below. RUTGERS &ROOKING$ PARSONS CONotth"t1 36 Deficiencies In the Literature on Sprawl Uterature Review I. Almost no analyses of sprawl adequately define it. Most commentators on sprawl do not defme it explicitly. This was the case, surprisingly in The Costs of Sprawl (RERC 1974), and the omission of a defmition continues throughout the literature. Where sprawl is defined, or at least charac terized, reference is often made to a limited number of traits such as low density or leapfrog scattered development (Ewing 1997, 108). Many studies, however, omit several other defining traits that cause many of the alleged negative impacts of sprawl such as dependence on the automobile and fragmentation of governmental land-use authority. This leads to the second major deficiency in the literature. 2. Most analyses of sprawl focus too narrowly on only a few of its key aspects. An adequate definition of sprawl must include the causal elements that underlie all of sprawl's many alleged negative impacts in order for subsequent analysis to respond to those impacts effectively. Therefore, a key part of this study is exploring the linkages between the many alleged negative and positive impacts of sprawl and its defining characteristics 3. There are other definitional cwn measurement questions. Take, for instance, "density." Numerous studies focus on the density of the region and relate certain characteristics, such as travel behavior or infrastructure costs, with the region's density. However, densities vary widely within regions and the real question is "How does the density of the specific places TRANSIT COOPERATIVE RESEARCH PI!OGRAM (TCRP) H-10


-n.. Com ol Sprawl-llomifod where people live and work affect, say, their travel choices?" The densities of these places may be substantially different from region wide averages. For instance, Gordon and Richardson (1989) use SMSAs as the unit of measurement in their analysis of densities and commuting times in 82 SMSAs. But is this meaningful given that no SMSA has uniform density throughout? And at an SMSA level perhaps density is a proxy for age of development, city size, or some other factors that affect travel behavior (e.g., transit use), as opposed to the variable density per se. In parallel are the definition and measurement of "segregation of uses But which uses, and separation or integration, at which geographic scale? Cervero (1996), for instance, found that the job-housing (JH) balance at the city level was not significantly associated with the variation in external (to the community) commuting. Does this mean that land-use integration as reflected in the JH ratio does not affect travel behavior, or that it really does-but that the measure of land use i ntegration is lost when the JH is scaled at the community-wide, as opposed to a neighborhood, level? There are other defmitional and measurement issues. The quality of life defmitional conundrum immediate ly comes to mind, but other seemingly easier-to-ascertain effects pose issues of their own Take, for instance, land consumption. While it is a tautology that development consumes land, does, for instance, a single-fami l y home built on a 50-acre farm "consume" all of those 50 acres, and if a fraction, on what basis is that fraction apportioned? RVTGD.S UOOJONGS PARSONS ECONOithwest 37 l.itoroture Review 4. Most critics of sprawl do not recog nize thar it provides substantial benefits to many households; hence, they do not take account of those benefits in their analyses. Several critics of sprawl, such as Kunstler (1993), engage in rhetorical exaggeration to emphasize their negative views of it. This polemic rhetoric cannot be classified as a accurate--observation about the reality of American suburbs. Yet, Kunstler and many other opponents of sprawl continually use at least one sided presentation-rather than balanced and facrually accurate descriptions of sprawl. Significant exaggeration is also employed by some defenders of sprawl, such as Gordon and Richardson (1997). 5 There are limited comprehensive empirical analyses. There is much discussion on sprawl but far fewer "facts" in the form of empirical, quantitative srudies. The paucity of data is illustrated by the frequency of srudies using "secondhand" or once removed information. This is exemplified, as noted, by the reanalysis of The Costs of Sprawl (RERC 1974) neighborhood and community prototypes some 20 to 30 years after the fact by Altshuler and GomezI banez (1993) Frank's 1994 review and reorganization of prior studies conducted over three decades in his Costs of Development is another example. It is not that the reanalysis or categorization is unimponant-on the contrary, it is quite valuable-but rather it points to the dearth of new empirical research. In a parallel vein is the tendency of the empirical research to be of a case srudy narure. Case studies provide valuab le insight, but as they are placeTRANSIT COOPERATIVE RESEARCH PROGRAM


specific, the ability to generalize from these areas--and, therefore, upon them is limited. society generally. But no quantitative analyses of sprawl have attempted to 6. Even when a quantitative analysis is estimate the size of these social costs, atrempred, rhe ropical coverage is and most analyses simply ignore them uneven, wilh nmch more attention paid 10 the "physical side" of conceptually. infrasrrucrure-rransportation and 8 The extant Uterarure also has Umiled Uuut-underemphasizing service and scope in the time frame of analysis social cosrs. The reason for s u ch a looking at effects over a few years focus is simple: far more complete rather than a longer span. The and reliab l e data are available for concatenation of limitation in analysis physical costs (e.g., developmentrelated to ge o graphic scale noted generated costs for roads, water earlier (i.e., focusing only on newly systems, and sewer systems) than for developing areas) and analysis of a serv i ce cos t s. An engineering manual, limited time span and time may very for instance, can inform the cos t per well lead to an overestimation by the linear mi l e of road, but there is seant literature of the costs of leapfrog literature on bow road mileage affects development. As noted by Altshuler police patro llin g c o sts. Th e re is an and Gomez-Ibanez (1993, 72): even larger gap in o ur know l edge concerning social costs. Discussions Estimat e s probably overstate the of sprawl's effects on quality of life added co sts of leapfrog are often superficial if not polemic, development in communities that and there are large g aps concerning expect continued growth and sprawl's effects on cities. What is the eventual i nfill development on the true social cos t of higher unemp l oy-vacant land. Compared with the me nt rates in inn e r dty areas, or the planned communities, the sprawl exclusion of low-income households communities contain subs tantially from outlying subu r b s-if s u ch more vacan t land that is i m proved u nemp l oym e nt or exclusion is or semi-improv e d by some road sprawl rela t ed? Measuring such costs and utilit y access. Developing is e x tremely difficult N everth e less, improved vacant land in the future some attempt at doing so must be presumably would cos t l ess than made in order t o inclu d e such costs in developin g unimproved land. If the overall analysis of sprawl. infill developmem is expected, the n a portion of the a dded costs 7. Most discussions of sprawl focus of leapfrogging eventually will be almost entirely on new growth areas. recouped-the costs of sprawl This may resu l t from the f act that wou ld be the costs of supplying sprawl itself occurs almost entirely in some infrastructure i n advance of new growth areas around the its eventual need and would be metropolitan pe ri p hery. True, recent lower the more rapidly infill was discus s ions of sprawl s uch as tho s e expected. described earlier by Downs (1994) and Rusk (1995), have begun to 9. Most commentators do not recognite recognize that draining valuable that two types of fragmented resources away from close-in areas governance-those over land uses has serious negative impacts upon and over fi scal resources-

fundo.mental causes of many of IM most widely a/tacked alleg ed nega tive results of sprawL 1be main reason for this failing is that the analyses are not com prehen s ive enougb. Th ey focus on a few of the most obvious elements of sprawl and the con se quences of thes e elements, rather than looking at the entire relevant spectrum of elements and con seque nc es. In addition they may besitale t o recommend changes m such fundamental American institutions as fragment ed control ove r land u ses and separation of e ach community's fiscal resources from those in other communities Several recent opponents of sprawl hav e recognized these connections quite explicitly, however, such as Anthony Down s (1994), Henry Richmond (1995 ) and David R usk (1993). 10. Mo s t op ponents of sprawl fail t o describe realistically feasible altemative forms of metropolilan settlement tha t would remedy tm negative conditions /My attribute to it. Like most social critics, tbey concentrate on describing what tbey dislilce, not o n how to remove those faults in a realistic manner. But tbe aspects of society they want to change are inextricably bound u p witb othe r fundamental elements-such as fragmented governance The changes they cal l for might have much b asic and widespread repercusstons than their analyses recognize. In fact, every cri tic's cau to radically change social con diti on X implies the following conditions: a) X is soci ally undes irabl e because it imposes unacceptabl e costs on some people. 39 b) The refo re, soc iety should gre atly alter or e limin ate X in the future c) X resulted from forces that coul d have been chann eled differently; it was not an inevitab l e out com e of irrev ersibl e or uncontroUable forces d) W e kno w what those chan geabl e forces are, and w e know how to handle them differently to avoid X in the f uture. e) It is politieally feasible to adopt those other methods of handling the force s c oncerned in the future f) W e kno w wbat the disa dvan tages of u s ing those otber metbods are. and bow large the costs are, even if the disadvantage s and cos ts are whoUy unrelated to X. g) The future benefits of eli minatin g, or sub stan tially altering, X are significantly larger than tbe costs and disadvanta ges of u sing those other methods of handling tbe forces that generated X. All critics of sprawl postulate conditions I and 2 but many stop tbere Some proc eed through con dition s 3 and 4 as part of their prescriptions for change But few deal witb condition 5, and almost none address conditions 6 an d 7-which require tbe ability to measure botb the costs and ben efit s of X and its elimination. Yet realisti c social analysis requires meeting aU seen conditions ll!ANSIT COOPERADVf RESEAI!CH PROGRAM (rCRI>J /1. I 0


Even the most detailed quantitative analyses of sprawl's costs tend to define only one alternative to it, and then compare the costs of future development in just those two future scenarios. That is true, for example, in the series of analyses d.ireeted by Robert W. Burchell (1992-1997). Sprawl is a complex phenomenon containing multiple future development scenarios, not just two; therefore, an adequate analysis of sprawl's costs compafed to the costs of alternative forms of settlement must allow for more than two alternatives. Conversely, no analysis can be useful if it presents dozens or hundreds of alternatives as equally plausible The best approach is to define three or more major alternative settlement patterns (but less than ten), and to conduct multiple sensitivity analyses concerning key elements in each of those patterns. II. The modeling of the analysis is often overly simplistic For instance, the per capita engineering studies have been criticized (rightly) as relating capital/operating costs linearly to lane miles of roads and related factors, which tend to be greater under sprawl, but not incorporating how costs can increase with thresholds of density because of congestion, public safety needs, and the like-a dimension that would increase the cost of compactness (Altshuler and Gomez-Ibanez 1993). Oversimplification also underlies the critique that comparisons of sprawl and its alternatives have not suffi ciently included the qualitative differences in housing amenity allegedly superior for the detached units that characterize sprawl (Gordon and Richardson 1997; Windsor 1979). RUTGERS 8ROOkiNGS PARSONS ECONOf'f!WMI 40 literature Review Limited depth also incorporates ques tionable associations that are drawn. This is in part a matter of an inade quately specified model or controls. For instance. Newman and Kentworthy ( 1989) applied only a single variable-urban density-to explain automobile use, whereas other factors are clearly involved. These two authors (Newman and Kentworthy 1989), in analyzing per capita auto mobile dependence, used gasoline consumption per capita as a proxy measure of automobile dependence. That equivalence is questionable given the fact that many factors, such as gas prices and fuel efficiency char acteristics, affect per capita gasoline consumption-not automobile dependence alone HoltzClaw (1990) related density to VMT without con trolling for income levels or other characteristics of households that influence VMT. Cervero' s (1989) analysis of 57 suburban employment centers did not control for the centers' ttansit availability and the quality of the pedestrian environment. Similar ly, tbe Cambridge Systematics (1994) study of suburban work sites did not control for these sites' levels of tran sit service. The difficulty in extrapolating the factors that influence these dynamics, and to that end incorporating controls, is illustrated by the examination of the effect of urban design Many re searchers are interested in whether neotraditional design features (com bined with a greater mix of uses) will result in travel behavior different from patterns observed in typi cal suburban deve l opment. To date, there has been too little experience with these new types of suburban development to answer the question. The refore, studies look at older neighborhoods TRANSIT COOPERATIVE RESEARCH PROGRAM (TCRP) H-10


Com of Sprcwi-Rovisited"' that have a more pedestrian-friendly environment and a fmer-grain mix of uses. But it is not clear whether behavior of residents of long-standing older neighborhoods accurately predicts the behavior of residents of new neighborhoods who in all likelihood are more accustomed to using cars. Holding aside that, the matched pairing of existing neigh borhoods into "transit versus autooriented" or "traditional versus suburban" to test the effects of alternate design patterns on travel runs bead long into the practical difficulty of coming up with these pairings First, neighborhoods often don't slot that neatly into two polar categories. Sec ond, even if this demarcation can be realized, there are variables other than overall design that can affect the travel behavior equation; resident income, occupation. and age, for example. This matching is a difficult exercise to accomplish, since design preferences and household profiles often inter relate. The problem of understanding the relationships is underscored by the cross-sectional nature of many of the studies. Infrastrucrure costs rise as development is effected in a sprawl pattern; thus, sprawl is tagged with the heightened capital expenses. Clearly, howev er, many other fac tors, from rising income to changing amenity levels, are at work (Altshuler and Gomez-Ibanez 1993). Gordon, Richardson, and Jun (1991) link decreasing conunuting time to the suburban deconcentration of job and residences that bas occurred at the same time; but does the fonner cause the latter, or is it merely coterminous? Similarly. Richardson and Gordon (1989) hypothesize that increases in nonwork trips are due to suburban decentralization occurring at the same RUTGER"S UOOKINGS PARSONS 4t time. Again, that could be true or could be unrelated to the spatial pattern but rather fostered by such influences as rising incomes, greater participation of women in the work force, and societal changes in leisure activities In short there is much peril in drawing conclusions from cross sectional research; yet that charac terizes many sprawl studies. The obverse of these deficiencies helps guide the current research. As detailed elsewhere, sprawl and its alternatives are explicitly and formally defmed. This effort, in fact, builds from the literature As noted in the overview to the literature some of the more recent studies on sprawl have differentiated it from othe r types of development on numerous fronts. In New Visions for Metropolitan America, Anthony Downs (1994) indi cates five elements from Jow-{fensity, primarily single-family development to widespread reliance on flltering to provide low income hous ing. Henry Richmond's Regionalism: Chicago as An American Region (1995) brought forth eight components of sprawl (listed earlier). T o Richmond's sound base, this study adds two more-(!) the commercial strip development described by Richard Moe (1994), and (2) a dependence on the filtering process to provide housing for low-income households as indicated by Downs (1994). Altogether, then, sprawl is a form of urban development that contains the following ten elements: I. Low residential density. (This is detailed elsewhere in this study at 3.0 units per net residential acre or less.) 2. Unlimited outward extension of new development. 3. "Leapfrog" development. 4 Spatial segregation of different types of land u ses through zoning regulations. TRANSIT COOPERAnVE RESEARCH PROGRAM (TCRP) fl. I 0


5. No centralized ownership of l and or planning of development. 6 All transportation dominated by private l y own e d motor vehicles. 7 Fragmentation of governance authority over land uses among many l ocal governments 8. Great variations in the fiScal capacity of local governments because the revenue-raising capabilities of each are strOngly tied to the property values and economic activities occurring within their own borders. 9. Widespread commercial strip development along major roadways. 10. Major dependence upon the filtering or "trickle-down" process to provide ho u sing for low income households. This definition both builds from the literature on sprawl and stands in marked contrast to prior studies that either do not define sprawl or else characterize it too simply (e.g., "lack of continuity in expansion" [Clawson 1962) or "low density ribbon or leapfrog development" [Harvey and Clark 1965)) and/or pejoratively (e.g "awkward spreading out of a community" [Abrams 1971)). In addition to defining sprawl, detai l ed elsewhere in this study is specification of alternatives to sprawl. These include "loosely bounded growth" and "tightly bounded growth" (see "Definition of RUTGU,$ U;OOKINGS PARSON S A2 lileroture Review Sprawl and Its Alternatives," Task I of the present study, for details on the attri butes of spraw l and alternative develop ment patterns.) The literature review underscores the need for a comprehensive l ook at the effects of sprawl. To this end, I) a full menu of benefits as well as costs of the different development scenarios must be considered; 2) these bene. fits and costs must span the range of phys i cal as well as soc ial consequences; 3) the benefits and costs analysis mus t be territorially com plete--encompassing urban, suburban, and ex urban locations and developing and developed areas; 4) the span of analysis must be long enough to encompass the dynamic of shifts over time, such as areas initially leapfrogged under sprawl subsequently being "filled-in" by development; 5) analysis of costs and benefits must incorporate the complexity of influences (e.g., varying thresho l d influences of density on capital-operating costs, and recognition that varying density thresholds, as well as other factors affect travel); and 6) caution must be exercised in not ascribing causality when the underlying evidence is merely cross-sectional To further this e x pansive consideration of the costs and benefits of sprawl and its alternatives the literature on the subject with its acknowledged defic i encies---ean be tapped. This is accomplished in the follow ing ponion of the literature discussion TRANSIT COOPERATIVE RESEARCH PROGRAM (rCRP) H-10




1'he Co#s ol Sprowf..-Revisited.,. RELATING SPRAWL'S DEFINING ELEMENTS TO ITS ALLEGED IMPACTS A search of the literature reveals that various commentators have attributed nearly two dozen negative and about one half this level of positive impacts to sprawl. These impacts are set forth in Table 9. The list is not a scientific taxonomy; it does not include al. l the alleged effects of sprawl bul rather is inclusive, in the judgment of the research team of some of the most significant impacts. Further, not a ll of the allegations are correct, nor are all those that are correct of equal importance. In fact, deciding which ones are valid, and estimating the importance of each, are major purposes of the study that will follow this literature search However, this inventory presents a comprehensive set of allegations based on the relevant literature for discussion here. The allegations have been classified into the five substantive categories of the literature: I) public-private capital and operating costs; 2) transportation and travel costs; 3) land and natural habitat preservation; 4) quality of life; and 5) social issues Collecting and Sorting the Literature Before considering what the literature informs us on each of those alleged negative and positive effects, it is insuuc tive to link these consequences to sprawl's defining traits. Tables 10 and II evaluate the causal importance of each of sprawl's ten alleged defining in relation to each of its 25 negative and 12 positive impaCts. In eac h m atrix, sprawl's ten defming traits are set forth from left to right as vertical columns. The alleged i mpacts are set forth from top to bottom as horizontal rows, grouped RUTGERS llOOKINGS PARSONS ECONonh'"" Literature ReW.w into the five categories The Tab le 10 matrix contains only alleged negative impacts; Table I I contains only alleged positive impacts. Each cell in the matrix indicates a "score" that represents the authors' subjective judgments concerning the degree of influence each defmed trait (at the top of the coluDlll) has upon producing each alleged impact (at the left of the row). The "scores" arc reflected by the following symbols: + 2 Indicates that the trait has a major inftuence in cau1ing the alleged impact. + 1 Indicates that the trait has a moderate o r minor inftuence in causing the alleged impact 0 Indicates that the trait ha no inftuenoe in causing the alleged impact. -2 Indicates that the trail has a negative inRuence in causing the alleged impact; that is, tho trait tends to re duce the incidence of tho impact. As noted above, the "scores" reflect subjective judgments made before most of the research on the study has been carried out. Therefore, these "scores" are subject to futur e revision. However, the matrices h ave been designed and pre sented in a manner to make it relatively easy for other observers to substitute their own subjective judgments for those of the authors. Determining, in a rough manner, the relative overall significance of each trait to all of its alleged i mpacts can be accom plished by examining the total scores of each trait in the matrix. For example, the coluDlll labeled "Low Density" in the TRANSIT COOPERATIVE RESEARCH PROGRAM (TCRP) H-10


e; TABLE 9 Alleged Costs and Benefits of Sprawl Substantive Concern I Alleged Costs/Negatives I Public Private Capiral and Operating Costs 1 1. More infrastructure costs U. Tran.sponation and Trnel III. Land/Natural Habita t Preservation IV. Quality or Ufe V. S ocial Issues 2 Higher public operating costs 3. More expensive private residentiaVnonresidential costs 4. Worse pub l ic fiscal impac.ts 5. Higher aggregate land costs 6. MoreVMT 7. More automob ile use 8 Longer travel/oommuting times 9. More personal transportation spending 10. Loss of agricultural land II Reduced farmland productivity 12. Reduced fannland viabilietically displeasing 16. Le. ssened sense of community I 7. Greater stress 18. Higher energy consumption 19. More air pollution 20. Lessened historic preservation 21. Fosters suburban exclusion 22. fosters spatial mismatch 2 3. F osters residential segregation 24. Worsens city fiscal stress 25. Worsens deterioration Alleged Benefits/Positives 1. LoweT public operAting oosts 2. Less expensive priva t e residentiaVnonresidential 3. FostetS efficient development of "leapfrogged" areas I 4 LessVMT j 5. Shorter travelloommuting times 6 Reduced ttansit subsidies 7. Enhauccd petSOnal and public open space 8. Preference for lowdensity living 9. Less concentration-associated crime I 0. Reduced costs of good.< and service.< 11. Fos t ers home rule 1 2. Enhances municjpal diversity and choice


TABLE 10 Bttwun Cbaracterislits of Spr aw l ond Its Ntcau.. Effects DEFIHIHG CHAAACTERI8Ttc8 OF SPRAWL 1lWlf Flllt!UNGFOR NEGATHEIIPACTSJl$) lOW OUlWAAO SPAlW.lY lEN'fROG S'lloi()TOR J.AND.USE lOCAl FISCAl LOW.atCOIIE TOTAI.CAUSALPOJNTS -> ElmNSlOH SEo _.tiog 2 2 I 2 0 2 0 0 2 1 .. L..ANOINATUAAL HAinAT PRESERVATION -_ _. . -t. _!!';;. = -10. Leu ohgiotlU'alln:l I 2 0 2 1 0 0 0 1 0 1 2 0 2 0 0 0 0 0 0 12. __ _, 0 0 0 1 0 0 0 0 0 0 13.Louofbgle............,.lands 2 2 0 2 1 0 0 0 1 0 1<.lmened reglcnol open._ I 0 0 0 1 0 1 0 0 0 IY.QUAUTYOFLFE ,. ;, > > oA ., > > !, To .... ,, 1$. dopleasing 0 0 0 0 0 0 0 0 2 0 16. Lenened sense or c:ommuru1y 1 1 0 2 o 1 1 1 1 1 0 0 0 0 0 0 0 0 2 0 I a. Hlgi'IM energy consumpfon 2 2 0 2 0 1 0 0 1 t rD. Moreo

TABLE 11 Relationships Between Characteristics or Sprawl and Its Positive Elfe OPERAnNG COSTS 1. lowOI' publie opnting cosh 2 leu privat. midential/nonresidetlfal costs 3. Fosters effiden.t <1evelopme.11t of teapfroggOO" areas II. TRANSI'ORTAnON AND TRAVEL COSTS "Leu VMr 5. Short 6. Recllc:ed transH subsidies I ILLAND/MATURAL HABITAT PRESERVATION 1. Enhanced pei'SOI\81 Jnd f)Ublie: OP< $paoe f\1. QUAUTY Of' LIFE 8. Preference lor low-de.,;ty IMng 9. Leuens conc:entrJtioo-assoclated i'ne 10. Reduces or goods and serkes 11. F0$1et$ greater ceonomle weD being V. SOCIAL ISSUES 12. Fos19f'S homo rule Ky: 2 Mojor r lol'ion 1 Modrot. or min cousol rtlolion 0 No eousol relation -2 N.goli'le causal relation U NUMITEO OUTWARD


negative impact matrix (fable 10) con tains eleve n "major influence ratings," seven "moderate or minor influence ratings," and seven zero scores. Similar observations have been carried out for the other nine defining traits. Based on these calculations, three of sprawl's defining traits appear to be especially important in causing alleged negative impacts. These are low density, unlimited outward extension, and leapfrog development. Two others--the spatial segregation of land uses and vari ances in local fiscal capacity-seem to be of relatively weak significance. Among the remaining traits, widespread conuner cial strip development and the use of filtering for low-income housing seem slightly more significant than lack of central ownership or planning, transport dominance by motor vehicles, and highly fragmented governance over land uses The analysis suggests (fable II) that the same three defining traits-low density, unlimited outward extension, and leapfrog deve/cpment-appear to be the most importan t in causing alleged positive impacts as well. Next in relative importance are transportation dominance by private motor vehicles and highly fragmented land-use governance. Least significant are the use of filtering for low income housing, the spatial segregation of land uses, and widespread strip conunercial development. These relative evaluations are merely suggestive. Since they do not apply in the same manner to specific alleged negative impacts, not too much significance should be placed upon this overall evaluation. RUTG.fRS PAASONS ECONonhwNt 48 \.Wotvre Review More important conclusions can be drawn from these matrices concerning vital causal linkages between certain categories of alleged impacts of spr awl and particular elements of its definition Certain categories seem to be more heavily influenced by certain defming traits than others. The linkages for both the alleged negative and positive impacts of sprawl follow in Table 12. Clearly, there is a great deal of similarity between the positive and negative matrices in Table 12. Paradoxically, the traits that seem key causes of many of sprawl's negative impacts also appear to be key causes of many of its positive impacts-in such categories as public private capital and operating costs, transportation and travel costs, land and natural habitat preservation, and the quality of life The fact that sprawl can simultaneously be associated with both costs and benefits in relatively narrowly defined fields shows how complex the phenomenon is, and how difficult it will be to design policies that reduce its negative impacts without also reducing its positive effects. To begin the process of better under standing the numerous negative and positive impacts of sprawl, these are considered on an item-by-i tem basis drawing on the discussion in the literature. The review of costs and benefits is grouped by the five sub stantive categories of public-private capital, transportation and travel costs, land and natural habitat preservation and so on. TRANSIT COOPERATIVE RESEARCH PROGRAM (TCRP) Hl 0


"Th Com ol Sprawl-Revisifed" literature Review TABLE 12 Categories of Alleged Negative Impacts I. Public-Private Capital and Operating Costs II. Transportation and Travel Costs m. Land/Natural Habitat Preservation IV. Quality of Life V. Social issues Categories qf Alleged Positive Impacts I. PublicPrivate Capital and Operating Costs ll. Transportation and Travel COSIS m Land/Natural Habitat Preservation IV. Quality of Ufe v. Social RUTGERS BROOt(IN.GS PARSONS ECONorlh-st Key Defining Traits Underlying Those Impacts Low Density Leapfrog Development Unlimited Outward Extension Low Density Leapfrog DevelOpment Unlimited Outward Extension Transport Dominance by Motor Vehicles Commercial Strip Development Leapfrog Development Unlin:tited Outward Extension Low Density Widespread Commercial Development Leapfrog Development Transport Dominance by Mo t or Vehicles Unlimited Outward Extension Use of Filtering for Low-Income Housing Highly r-ragmented Land-Use Governance Great Veriances in Local Fiscal Capacity No Central Ownership or Planning Key Defining Trails Underlying Those Imoacts Leapfrog Development Low Density Unlimited Outward Extension Low Density Leapfrog Development Transport Dominance by Motor Vehicles Unlimited Outward Extension Low Unlimited Outward Extension DevelOPment Low Density Unlimited Outward Extension Leapfrog Development Transport Dominance by Motor Vehicles Highly Ftagmented Land-Use Go,.ernance No Central Ownership or Planning Great Variances in Local Fisca l Capacity A9 TRANSIT COOPERATIVE RESEARCH PROGRAM (TCRP) H-10


Each of the alleged negative and positive impacts under these five substantive groupings is individually conside red following a common presentation forma t as follows : I. Topic. What i s the specific subj ect matter of the alleged cost or benefit? 2. Allegation/Basis. Synopsis of the alleged cost or benefit and the basis or logic of the supposed effect. 3.l.ireracure Synthesis. Pertinent s tudi es on the allegation are cited. either sup port ing or rebutting it. The presentatio n of the lit era ture synthesis i s accomplished through both tex t and a matrix The matrix distinguishes between whethe r "the literature agrees" about I. whether or not the alleged factual condition exists under conditions of sprawl (or more generally whether development pattern affects the item in question), and 2. whether or not the alleged factual condition-if it exists h as been significantly linked to sprawl. For example, concerning the allegation that "spraw l generat es more t o tal trave l than higher den sity forms of development," it is first noted whether there is agreement among observers who comment on this subject that low-density senlernents in fact generate more total travel (in person miles traveled) than higher-de ns ity senlements. There is mostly agreement in t his regard. The next observation concerns whether there is agreement in the lite rature tha t the presence of greater travel time in low density senlements is significantly linked to sprawl. There is, again. mostly agreement on the second count; however, there remains the question whether the amount of travel is as significant an impact as the time of travel. For simplification, these judgments in the above examp le would be sho wn in the form of a simple matrix, as show n : R VTGfU P.OOIONGS ,AitSOHS ECONcd!.west so e .'! .. ... s-,.,;;_ ...;;:,., a.-. I "Doe!,!"'' coMiiliofl J thtl X r ,, tron$1y I ::d t o wll X An "x" placed in a matrix cell indicates that it contains the appropriate artswer to the question on that line. The matrix is not a scientific measuring instrument. It could have been organized in multiple alternatives ways. Even as c u rrently structu red, there wa s not always consensus among the research t eam on bow to "sco r e an item, suc h as whether there is "mostly agreement" or "some agreement" in the literature, or for that matter bow convincing the literature is on any given subject The purpose of the matrix and the accompanying discussion is to synthesize in a systematic way. at the onset of the research, the important studies on spra wl. This effort informs us as to wh at prior researchers on the su b ject have con sidered and debated, what data have been used and bow data have been analyzed, where there are gaps in the state of knowledge, and so on The literature synthesis is summarized in Table 13. It indicates the following: Extent of the Literature. As noted earlier, the lit era ture on sprawl clusters in the topical areas of public privat e capital and operating costs, transport ation and travel costs, and to a Jesser extent, land and natural resources preservation. There are far fewer studies specifically relating sprawl to quality of life or social effects Within these respective clusters certain topics have received more attention than others In the public pri vate capital and operating costs group, the issue of development patterns and infrastruc ture expenses bas been f ar more frequently TRANSIT COOPERATIVE RESEARCH PROGRAM (TCRP) H-10


Subs tantl v t Cooc.ern lll Land/NahJraJ H.abitat P r eservation L ife I mpacts V. Social l 11ues E xt

'7ho Com of Sprawi-Rrmiled" studied than the issue of how sprawl influences operating costs or housing expenses, and especial ly the subject of development patterns and pubUc fiscal impacts (public service costs Jess public revenues). Similarly, much more attention has been paid to how development affects the amount, and mode, of travel than the translation of that to household spending or, for that matter, national transportation spending. Also evident from Table 13 is that the Uterature tends to the more prolific on the criticism leveled at sprawl rather than its def ens e. Acr oss the topics, there are more studies alleging negative impacts on costs rather than positive impacts on benefits. "Agreement" in the Literature. As noted earlier, this evaluation is a judgment call by the research team. Having said that, there are discernible areas of greater or Jesser consensus in the literature. Tbere is greater agreement that certain "physical" characteristics exist under sprawl (e.g., "Does condition notably exist?"), than quality of Ufe and social effects. These physical character istics are more travel, especially by automobile; greater consumption of land; and high er infrastructure costs. By contrast there is much less agreement on such matters as the quality of life attributes of whether there is "greater stress" or "lessened sense of community" un der spraw I. With respect to the issue of whether certain alleged and benefits are "strongly linked to sprawl" (holding aside the issue of causality), there are few areas of high con sensus. There are, as examples, sprawl's link to greater automobile travel and consumption of more farmland and frail lands By contrast, there is "substantial disagreement" on many fronts, such as whether housing is more expensive under sprawl and, relatedly, whether land is lUTGERS BROOKINGS PAASONS ECONorthw.Q 53 Literoturt Rovltw more costly and if traveVcornrnuting times are longer or shorter. The literature points to what might be some of the research thrusts of the current investigation. On the physical engineering side of sprawl that is, the issues of infrastructure, transportation, and land-the studies to date point to many appropriate measures to be considered (e.g., vehicle miles traveled [VMT] and congestion), and the relationships to be examined (e g., density's effect on modal choice and travel time). However, these must be brought together more definitively by the research team and areas of outstanding disagreement from prior work (e.g., are commuting times shorter or longer under sprawl) empirically examined so that answers can be had. This and more need to be done where there are lingering gaps in our knowledge concerning the effect of development patterns on operating costs, productivity of farmland, and so on. On remaining topical concerns quality of life and social effects-the challenge to the research team is even more fonnidabl e, because here interrelationships are more complicated, yardsticks of measurement less agreed upon, and association with development pattern whether spraw l or otherwisemore obtuse. The research design being developed addresses these issues. It builds from the expenise of the research team and is informed by the literature review, such as the overview that has preceded this discussion and the costs and benefits itemized in the discussion that follows. This item-by item analysis is organized by the five topical areas of the literature. For each topic, first the relevant costs on negatives of sprawl are prese nted, followed immediately by sprawl's alleged benefits. The discussion builds from the annotation of key studies and the compre hensive bibliography found in the appendix. TRANSif COOPERATIVE RESEARCH PROGRAM (TCRP) H-10


DETAILED ANALYSIS OF COSTS AND BENEFITS OF SPRAWL The Costs and Benefits of Sprawl: Alleged Negative Public-Private Capital and Operating Costs Topic: More In frastructure Costs AllegationJBasis lnfrasrructure of a wide s cope-local and regional roads utilities (e.g .. water and sewer systems), schools, and other purposes-is more ex:ptmsive under spra wl than under compact development. ThJs allegation alludes to Infrastructure that is primarily publi c (e.g., state, county. and local government roads ; public utility systems; and public schools) and occasionally private ( e .g., privately owned utility systems and subdivision level roads that are not dedicated to the public sector). The effect of sprawl on the cost of infrastructure allegedly occurs for several reasons. At sprawl's lower development densities, various components of infrastructur e thai are linearly related (e.g., sidewalks, curbs, subdivision-level roadways, and water and sewer mams) serve a lesser increment of development tban this infrastructure would serve at higher levels of density The segregation of land uses associated with sprawl further increases infrastructure costs Segregatio n of land uses by residential and nonresidential types often means that parallel infrastructure systems have to be provided to individual residential and nonresidential locations Further s prawl's leapfrog development. which locates growth away from existing development. does not capitalize on pockets of surplus infrastructure capac1ty that may already be present in and around existing development. Finally, RUTGEU tROOIONC$ PAlSONS ECONodh-u fragmented govemanoe, a seeming natural accompaniment of sprawl. often leads to duplicative city halls, police stations, courts, fire houses, schools, water/sewer treatment facilities, and so on Lite rature Synthesis As shown earlier in Table I The C o sts of Sprawl (RERC 1974 ) found that c apital costs per unit were higher in the "low diversity sprawl" and "sprawl mix" neighborhood prototypes than they were in the "planned mix" and "high-density planned mix" prototypes. The Costs of Sprawl also found that capital expenses per unJt were higher in deta ched housing (more pronounced under sprawl) than they were in attached housing (more pronounced under compact development) The first finding of The Costs of Sprawl. although criticized has basically stood the test of time (Altshuler 1977); the second fmding proved to be the undoing of the study (Windsor 1979). Frank ( 1989) reanalyzed (includinj1 using current cost numbers) several stud1es conducted from the 1950s to the 1980s that examined relationships between land use and infrastructure costs ( including The C o sts o f Sprawl). Accounting for the limitat i ons of the Costs of Sprawl study, be concluded that infrastructure c osts were highest in situations of low density and for development located a conside rable distance from centralized public servic es (conditions of sprawl). Infrastructure costs were low est in situations of higher density and for development that was centraJIY and/or contiguously located (condJuons of compact development). Duncan ( 1989) analyzed the infrastructure costs of multiple Aorida residential and . nonresidential developments wtth varymg patterns of development. Costs were higher for those with sprawl characteristics than for those w ith TRANSIT COOPERATIVE RESEARCH PROGRAM (ICRP) H-10


literOh.lre Revltw TABLE 14 Duncan ( 1989)-Fiori da Growth Pattern Study : Capital Facility Costs under Sprawl versus Compact Development (per dwelling unit; 1990 dollars) Category of Average of Case Average or Case Sp r awl Versus Capital Costs Studies under Studies under Compact D eve l opment Spraw l Developmeot1 Compact Dev e lopmeot2 Difference #I% Roads $ 7,014 $ 2,784 (+) $4,23{) 60.3 Schools 6,079 5,625 (+) 454 7.4 Utilities 2,187 1,320 ( + ) 867 39.6 Olher 612 () II 1,7 TOTAL $15 941 $10,401 (+) $5,540 36.7 Notes : I Sprawl deveJopmem 3$ defined hete include the following p,auems of 'urban form" analyzed by 1he florida study: .. scattered,'' "linear." and ''satellite. The capilal cost figures shown in thi s table are averages of the Florida case studies characterized by tbe scattered, linear. and satellite pat terns (e. g .. Kendall Drive. Tampa Palms, University Boulevard and Cantonmcni). 2 Compact development as defined here includes the following panems of ''urban form" analyzed by the Florida study: .. contiguous .. and 'compact." The capital cost figures shown in this table are averages of the FJorida case studies characterized by the oontiguous and oompact patte-rns (e.g Countryside, Downtown Orlando, and Southpoint .) Source: Memorandum from Jan\es Duncan and Associates to Robert W. B urchelt and David Listokin. May 8, and James Ouncan et al., The Starch/()r f!ficitnl Urban Grow1h Pautrns Repon prepared for the Go\'ernor' s Task Force on Urban Growth Patterns and the Aorida Depanment of Community Affairs (TallahasS, Jul y 1989). compac t development characteristics (see Table 14). The l ongest run modeling of infrastructure costs under different development scenarios has been performed by Burchell et al. (1992-1997) both in New Jersey and in other locations The inf rastructure models applied by BurcheU rel ated development density and housing type to the demand for local/state roads and water/sewer infrastructure. The studies found that the amcunt o f land consumed for developme nt was directly related t o lane miles of road required for two-lane (local) and four-lane (state) roads. Thus, ltUTGUS UOOKINGS PARSONS ECONonhWNt 55 density of development was found to be inversely related to lane miles of local and state roads and their attendant infrastructure costs Housing type, and to a Jesser extent density, was rel ated to the amount of water and sewer services consumed (in gallons) by development. Almost all of the difference in residential water usage related to whether or not occupants of residential and nonresidential facilities watered their lawns. Lawn watering takes place primarily in single-family detached residences and high-value research and headquarters co mme rcial uses. In the latter, the difference in water usage between various commercial and TRANSIT COOPfRADVf RSfARCH PROGRAM (TCRP) If. I 0


industrial uses is also related to the service or product that is generated by the facility. Larger and more significant than water/sewer usage are differences observed in water/sewer infrastructure, particularly as related to the number of feeder hookups from the trunk line that an individual land use requires. Higher density. the clustering of land uses. and attached housing and linked nonresidential uses all contributed to a reduced number of infrastructure feeder lines and reduced costs. A model sensitive to these differences, applied in New Jersey to alternative growth scenarios differentiated by a sprawl-like versus more compact development patterns, showed the former's infrastructure costs to be considerably higher. The fmdings were basically similar in order of magnitude across most subsequent locations analyzed by Burchell et. a!. (Burchell and 1995) (see earlier Tables 3 and 4), and were also very comparable to the findings reached by Frank and Duncan in their studies (see Table IS below). TABLE IS Re-lativeInfrastructure Costs Of Sprawl Versus Compact Development From Three Major Studies Compact Development Costs as Percent of Sprawl Development Costs: Findings from Three Major Studies InfrastJUcwre Cost Category Sprawl Development Duncan Study (1989) Frank Study (1989) Burchell Studie.< (1992 1997) Compact Development Costs: Synthesis from Three Major Srud.ies Roads (local) I 00% 40% Schools 100% 93% Utilities 100% 60% Source: Burchell and Listokin (1995) and Table4. Other relevant research indicating higher infrastructure costs under conditions of sprawl include Archer (1973) and Duensing (1977). Base data on infrastructure and its costs, not related to development pattern such as the average capital outlays per single-family house, or costs per linear foot of roadway. are provided by Fodor [1995], Nichols et a!. [1991], Nelson [1988], FACIR [1986], OP&R [1982]). The above body of research reflecting, in part, an approach dating to The Costs of Sprawl, has been criticized on several counts by the same author (Altshuler RUTGEit.S BROOKINGS P.AtSONS ECONMhwt 73% 99% 66% 56 74-88% 97% 86-93% (in percent, relative to sprawl) -75% -.95% 1977 and Altshuler and Gomez Ibanez 1993) as follows: I. The higher infrastructure costs found in instances of lower versus higher density (i.e., sprawl versus compact development) is not meaningful because the housing units and their attendant scale found under the different development alternatives (i.e., more detached housing under sprawl and more attached housing under compact development) are not comparable. 2. The higher infrastructure. costs attributed to sprawl due to its leapfrog patterns will essentially be neutralized TRANSIT COOPERATIVE RESEARCH PROGRAM (TCRP) HlO


"The Cosls of Sprawi-RevW!od" as areas that were initially passed over are u l timately developed. The nel{t wave of growth will capitalize on the infraslructure in place. Thus, the higher in itial cost will be recouped "the cost of sprawl is the cost of supply i ng some infraslnlcture in advance of its eventual need and will ultimately be lower the more rapidly that inflll takes place" (Altshuler and Gomez-Ibanez 1993, 72-73). 3. The higher i nfraslnlcture costs (under sprawl) attributed to the distance of development from central facilities does no t consider potential economies of scale that could be realized in regionalized over sized lnlnk lines or similarly l ocated water/sewer treatment plants (Altshuler and Gomez-Ibanez 1993, 73). In other words, the added "costs of d i stance" because feeder lines are longer under sprawl are not significant if these feeder lines are attached to regionally located (and oversized) trunk lines and wa t er/sewer plants. Holding aside the above considerations which relate to whether significant cost differences are valid, the cost difference in infraslructure between sprawl and compact development patterns is found by at least one res e arch e r to be quite TABLE Literature Review slight. Peiser (1984) is often cited in this regard (Gordon and Richardson I 997) Peiser ( 1984) el{amined infraslnlcture costs for new residential development in two Texas "prototype" c o mmunities, one p l anned, the other unplanned The planned and unplanned developments were located on 7,500-acre sites in Houston. The planned commun ity was designed to accommodate a population of about 80,000 r esidents in 26,500 dwelling units and a workforce o f 72,000 in 24 million square feet of office and industrial space. T he development was largely self-contained and near existing development in the form of a large center. The unplanned development was located in a primary growth corridor at the urban fringe, typical of Houston s spraw l pattern (100 to 500 acre subdivisions and strip malls and shopping centers), and scaled to accommodate a similar number of residents (80,000) and workers (72,000) as the p l anned deve l opment. In Peiser's model, the diff e rence in capita l expenses for the planned and unplanned scenarios was about 5 percent in favor o f the plann e d development. The finding in the Peiser study that contradicts other fmdings in the field is the inclusion in overall planned development infraslnlcture savings of higher road cos t s associated with planned as opposed to unplanned development (Table 16). 16 Peiser Model Planned Development Costs (for 80.000 Component residents) ($ in millions) Roads $t0. 0 Sewer 4.3 Water 9 2 Drainage 16 3 $39.8 RUTGERS BlOOKINOS PAJ:SONS 57 Unplanned Deve lopment (for 80, 000 residents) ($in millions) ss. o 4 7 11. 8 17 4 $41.9 TRANSIT COOPERATIVE RESEARCH PROGRAM fTCRP/ H-10


"!!.t Com ol Sprowi-Rty;sit.d" In sum, whi l e there is general agreement that development density is Iinlced to infrastructure costs, there is less agreement concerning the inter rel ationship between sprawl (as a l ess carefully defined development form) and infrastructure costs. Literature Synt hes i s Mat r ix I C:! .. ., Nod.or """-con

complex and more individualized than those delivered in smaller, less-dense jurisdictions. Foot patrol or two-person automobile police pattol versus one person automobile police pattol. fully paid fire deparunents versus volunteers, significant numbers of special education teachers versus contJ:acted-out special education services, a ll complicate differentiating the costs encountered in more intensely versus less intensely populated jurisdictions Local government costs nationally average about $700 per capita; school district costs average about $7,000 per pupil (Census of Governments 1992). Of the former, about 60 percent is for salaries and wages, 35 percent for other expenses, and 5 percen t for cap i tal purposes. For the latte r, 70 percent is for salary and wages, 20 percent for other expenses, and 10 percent for capital purposes The oppos i te of sprawl development. compact or managed growth, impacts on operational costs primarily by encouraging more regionalism in school systems and more sharing of non-police local public resources and finally by reducing the amount of local roads and water/sewer utility lines and hook-ups that are constructed and paid for by local debt service and mainrained and paid for out of annual operating budgets. Burchell, in his analysis of the growth alternatives in the Impact Assessment of the New Jersey State Development and Redevelopment Plan, found that combined municipal and schoo l district operational costs could be reduced by 2 percent annually under planned (compact) as opposed to trend (sprawl) growth (Burcbelll992a) While the percentage seems smal l these are annual l y occurring as opposed to one-time savings, and these could be potentially applied to local RUTGE'R.S BROOKINGS PAASONS ECONORTHWESl 59 Lilero.ture RMew budgets national l y that sum to $175 billion per year and school di s trict nationally budgets that sum to $500 billion annually. In similar type studies in the Delaware Estuary. and in the state of Michigan, municipal costs were found to be 5-6 percent less annually due to compact as opposed to spraw l development. Basically equivalent findings have been found by James Duncan in Florida (Duncan 1989). Conflicting findings have been suggested but not empirically tested, by Altshuler and Gomez-Ibanez (1993) and Gordon and Richardson ( 1 997). Altshuler and Gomez-Ibanez indicate that the i n ability to contto1 for the quality and quantity of services under comparison renders most of these studies at best "ti m e and locational bound" by who is providing what types of public services when, and at worst unable to be used t o draw appropriate conclusions from, given their inability to differentiate between levels of service provided (Altshuler and Gomez-Ibanez 1993) Gordon and Richardson indicate that Burchell's prospective alternative development scenarios allow no ability for the trend scenario (sprawl) to improve over time and similarly no ability for the plan scenario (compact growth) to be worse than envisioned due to Jack of full compliance with this alternative (Gordon and Richardson 1997). Literature Synthesis Matrix .. :.... ,_ o..... i COfldlfiOn notobly exiltf X I !..' t !'tong y li nl.d to .. X 1RANSIT COOPERATIVE RESEARCH PROGRAM (TCRP) Il-l 0


"'The Costs ol Sprawl-Revisited" The Costs and Benefits of Sprawl: Alleged Negative Public/Private Capital and Operating Costs Topic: Resident ial and nonresidential development costs Sprawl causes residential and Mnresidential building and occupancy costs to rise due to larger lot sizes and strucrure sizes in locations where land i s less expensive Literature Synthesis Development costs are the costs to develop residential and nonresidential properties. They involve land and improvement costs and are impacted by the scale of each. Spacious single-family dwellings on large lots are usually the most expensive types of housing; spread out,low ri se nonresidential development on large parcels of land are the most expensive type of commercial and industrial development. They are both low density examples o f their respective development forms. To the degree that density increases for residential development and floor-area ra.tios increase for nonresidential development, holding all other strucrurelenvironmental amenities constant, residential and nonresidential development costs will decrease. Similarly, to the degree that structure s ize is less, holding all other structure/ environmental amenities constant, residential and nonresidential development costs will also be less. Other factors affecting the costs of residential and nonresidential development are: I) the amount of zoned land available for development as determined by the local zoning ordinance, and 2) the time it takes development to engage and clear the permitting processRVTGERS BROOKINGS PARSONS ECONOitTHWE$T 60 litoroture Review also largely determined by local land u se regulations. If land is limited or inappropriately zoned, residential and nonresidential development costs will rise ; if government regulations are excessive, permitting time will increase, and the costs of development will also rise. In the Impact Assessment of the New Jersey State Development and Redevelcpmenr Plan ( 1992a), Burchell found that if development under the plan alternative is contained around existing development and is also increased somewhat in density and floor-area-ratio, that even with significant decreases in density to preserve lands at the periphery, overall residential and nonresidential deve lopment costs will be approximately I 0 percent less per unit and per 1,000 square feet under this scenario. Somewhat less savings (6-8%) emerged from studies conducted by Burchell in Lexington, Kentucky, (1993), the Delaware Estuary (1995), and the State of Michigan ( 1997). Other studies of residential development have produced essentially parallel finding s as they relate to the effects of increased lot and structure size on housing costs Seidel ( 1980), Downs ( 1 973), Schafer (1975). and others have found that large lot zoning and minim u m building size increase the costs o f new housing. This same rype of analysis applied to nonresidential development, although not often looked at by researchers in the field, has produced similar ty p e s o f findings. Others have found that large Jot single family zoning and minimum building size are associated with sprawl development, and smaller lot sizes (zero lot line) and different types and intensities of development (single-family attached and multifamily development) are associated TRANSIT COOPERADVE RESEARCH PROGRAM (TCRP) /1.10


with compact development (CH2M Hill 1 994 and Avin 1996). Linking the above two sets of findings, the savings noted by housing type would extend to these two polar development fonns One important caveat is well worth notin g. One cannot assume that housing preference changes will accompany development pattern shifts. In other words, if compact development is opted for mo!"'! dense fonns of housing compnse this type of development, it cannot be assumed that market preferences will correspondingly shift and families previously occupy ing less dense types of housing under sprawl will opt for the more intense development fonns under compact development. Funher, if there is a cross over between housing types, one mus t carry the occupancy profile of the fonner to the new type of housing unit. Otherwise false conclusions could be drawn with regard to. development cost savings associated wtth the often -small er, and less intensely occupied housing of compact dev e l opmen t as well as with the annual fiscal impact savings resulting from this devel opmen t form. A critical error was discovered by Windsor in his review of the Costs of Sprawl (Windsor 1979). The Costs of Sprawl study failed to account for the fact that the change in characteri s tics of new townhouse occupants switching from single -famil y occupancy (if they could be assumed t o do so) would he closer to the characteristics of occupants of the units of housin g that they left than they would he to the historical charact eristics of units that they were seeking. This lack of realization led 10 the erroneous conclusion that compact development (containing a larger percentage of townhouses) was less expensive to servic e than sprawl devel opment (containing a larger percentage of single-family homes) when the same households that occupied the RUTGERS U.OOIONGS fCOtiOimmpensate for these costs Further, fragmented governments co mpete for land uses according to these land uses' fiscal superiority Most economic uses are withdrawn from central cities but only the choicest suffice for suburban jurisdictions. There are not eoough "good" land uses to go around and only a few jurisdictions truly benefit fiscally from their presence. Literatur e Synth esis In aoalyzing the impacts of land uses, it bas become accepted that. generally speaking. some types of land uses are beuer fiscally than others Nonresiden tial land uses, for the most part, have been shown to be more profitable ; most standard forms of residential land uses less profitable (Table 1 7). Further, within the restdential and nonresidential sectors, there are varying degrees of profitability. Profitab ility means that some land uses produce more revenues than costs, i.e., if TIIIINSIT COOPERAnVf RESEARCH l'tOGRAM (TCRP) H-10


Table 17 The Hierarchy of Land Uses and Fiscal Impacts (+) MUNICIPAL BREAK-EVEN REsEARCH 0mCE PARKS OFFICE PARKS INDUSTRIAL DEVELOPMENT HICII -RISEIGAADEN APARTMENTS (STUDIO/I BEDROOM) AGB-RESTIUCl'BD HOUSING GARDE:N CONDOMINIUMS (I -2 BEDROOMS) OPEN SPACE RETAIL fACILmES TOWNHOUSES (2-3 BEDROOMS) ExPENSIVE StNGLE FAMlLY HOMES (3-4 BEDROOMS) TOWNHOUSES (3 -4 BEDROOMS) INEXPENSIVE SINGLE-FAMILY HOMES (34 BEDROOMS) GARDEN APARTMeNTS (3+ BEDROOMS) MOBI LE 110M$ (UNRESTRICTED AS TO OCCUP,WCY LOCIUY) (+) SCHOOL DISTRICT BREAK-EVEN ( ) Note: The above list contains too many disclaimers to i .nclude here. Suffice il to say that specific fiscal impacts of a land usc must always be viewed in the context of other land uses' impacts and within the fiscal parameters of the jurisdiction jn which the land use is being developed. lUTGEJ.S 8(00tCINGS PAASONS E<:ONORTHWEST 62 TRANSIT COOPERATIVE RESEARCH PROGRAM (TCRP) H-1 0


service levels are maintained at lhe same level after development, taxes could be decreased On !he olher hand, lhe reverse is also true. In some cases, costs exceed revenues and, all things being equal, taxes might have to be increased (Burchell and Listokin 1994a). Position on the fiscal impact hierarchy depends on type of unit (reflecting size or intensity of use) wilhin bolh residential and nonresidential classifications. Fiscal position also depends on the service district in which it is being viewed. Often. for instance, a small condominium or age-restricted housing may break-even or be just positive or negative in the municipal service jurisdiction, yet both may be very positive fiscal ratables in the school district. On !he other hand, larger townhouses may be just below break even in lhe school district yet significantly negative in lhe municipal jurisdiction. Fiscal impacts and observed differences under sprawl versus compact growlh are depe nd ent upon two different influences from development pauerns. The first is t he ability to influence type of development by compact versus sprawl growth. To the degree !hat dwelling type can be changed by compact development in sub-state settings, the demographics and, resultantly,lhe pub lic service of development will change. The second is lhe ability of compact development to influence the intensity of development and geographic spread of new neighborhoods. If compact development can provide tighter development patterns, infrastructure provision wilJ be less. So too will !he annual debt service on capital costs for roads, water/sewer lines, and so on as well as the annual costs of maintenance associated with these new facilities. Related to this is the location where development takes place. If located RUTGERS aROOIONGS PARSONS fCONO'tTHWIST 63 near existing development, excess service capacity may be drawn upon. If development is skipped over, public service infrastructure will almost always have to be provided at costs greater than extending existing facilities. Burchell's Impact Assessm,.,.t of the New Jersey State Development and Redevelopment Plan (Burchell eta!. 1992a) employed a fiscal model to view lhe effects of trend versus plan development. The Rutgers fiscal impact model estimated lhe number of people employees, and students !hat were generated by development under each of lhe development scenarios and projected !heir future costs and to host public service jurisdictions. While at the regional and state levels, population and employment projections did not vary between alternatives, at lhe muni cipal level there were significant differences. In the compact development c ase, urban communities wilh slack service capacity receive more growlh !han rural areas with lesser amounts of public service infrastru cture. Wilh reduced infr..structure provision and potentially reduced annual maintenance on this infrastructure. Ibis lead to dimi. nished fiscal impacts for this alternative. Burchell's study in New Jersey found !hat: By containing population and jobs in already developed areas and by c.eoting or expanding centers in newly developing areas. the State Plan offers an annual $112 miltion [or 2 percent] fiscal advantage to municipalities. This advantage reflects the ability under plan to draw on usable excess operating capacity in already developed areas as welt as efficiencies of service deli very. For instance. fewer l311e-rniles oflocal roads will have to be built under plan, thus saving municipal public works maintenance and debt service costs. Public school districts will realize a TRANSIT COOPfRATIVf RESEARCH PROGRAM (TCRP) H-10


"Tho Costs of $286 million (or 2 percent ) annual fonancial advantage under the Stat e Plan, again a reflection of drawing on usable excess public school operating capacity and other service and fisca l efficiencies realized due to the redirection of population under the plan a1temative. Thus, municipal and school district providers of public scr\'ices could be ahead fiscally by cl0$e to S400 million annually under plan compared to trend, while me c.:!o. ..... No a.,, sut..lonlio l """"'-uoe!. eondi6on I\Of;biV -.JI.iltf X il ftrOI\g l y lillkd to _...,, X The Costs and B enefits of Sprawl: Alleged Negative Public/Private Capital and Operating Costs T opic: Higher Aggregate Land Costs Allegation/Basis Total/and costs of urban senlements are higher under sprawl This occurs even RUTGERS 8ROOICINGS PARSONS ECONOR.TKWEST 64 literoture A:.view though the average price of land per acre may be lower because a given total population occupies more suburban land than under higher density urban forms of growth. Literature Synthesis Most of the modeling effons to date that involve prospective development futures bave found that alternatives to "status quo" development pauern which is usually sprawl, consume less overall land than this deve lo pment pattern does. In New Jersey Lexington, Kentucky, the Delaware Estuary and Michigan, alternatives to sprawl consumed 20-40 percent l ess overal l land (Burchell 1 9921997). In the San Francisco Bay area, alternatives to sprawl consumed 10-25 percent less overall land than did sprawl (Landis 1995). Thus, land consumed under sprawl bas almost always been shown to be more than land consumed under compact growth patterns. Funber, in the Burchell ( 1992-1997) studies because densities were increased to design levels under compact growth, housing costs were less due to the reductio n in land costs associated with this alternative In other words, in situations where there were no growth restrictions, h ousing costs were more under sprawl because land costs were more. Thus in the above four Burchell study locations, housing costs under sprawl development were more due to the land component of the se costs. T his was true because under compact development the majority of deve l opment taking place closer-in was subject to density increases of 10 to 30 percent. Total land costs of urban seulements have been found t o be generally higher under the sprawl alternative. (See Negative Land/Natural Habitat section.) TRANSIT COOPERATIVE RfSEARCH PROGRAM (Tct!P) Hl 0


Literature Synthesis Matrix The Cost and Benefits of Sprawl: Alleged Positive Public-Private Capital and Operating Cost Topic: Lower Public Operating Costs Allegation/Basis Local land school district operating costs are lower under sprawl development because service demands and the costs of meeting these demands increase with higher densities (such as those associated with compact development). Literature Synthesis Gordon and Richardson ( 1997) express this argument, citing the research of Ladd (1992). Ladd (1992) argued that except within a range of very 1ow densities, per capita public service costs for traffic management, waste collection and disposal, and crime control. increase with higher densities (Gordon and Richardson 1997, 99}. Again, this is the type of research that has not standardized for the quality and quantity of public services delivered in jurisdictions of varying densities. What the above research indicates is that not taking into account what services are delivered or who delivers them in a service district. oper.1ti ng costs, whatever they are comprised of; appear to be less in jurisdictions of low density than in jurisdictions of high density. RUTGERS BROOK.ING$ PARSONS ECONORTHWEST 65 literature Review Further, the comparisons that are made are usually between locations of rural suburban (I to 3 units per acre) density and those of urban density ( 16 to 30 or more units per acre). These stud ies may well be measuring the differences in range and complexity of public services delivered in densely populated u rban areas versus the very limited and much simpler public services delivered in rural suburban areas. (See Operating Costs in the Negative Impacts Section.) Literature Synthesis Matrix The Cost and Benefits of Sprawl: Alleged Lower Costs of Private Housing Topic: Less Expensive Private Residential/Nonresidential Costs Allegation/Basis Sprawl has /ewer housi11g costs because it does 1101 limit tlze amount of development. Many managed approaches to growth seek also to control growth. Various forms of growth control limit housing produced and drive up the costs of housing. Literature Synthesis Does the overlay of regulations inherent in managed growth, drive up the cost of housing ? There are number of studies that reveal that in the immediate area where there are growth restrictions, housing prices increase (Fischel 1990). For instance, Schwartz, Hansen and Green ( 1981} followed the effects over time of TRANSrr COOPERADVE RESEARCH PROGRAM (TCRP} H-10


"Tht Com ol the Petaluma (California) Plan. This plan severely limited building permits favoring dwellings with costly design features and developer-provided amenities and services to the community. Using a statistical (i.e., hedonic) pricing technique, the authors compared the price of a standard bu ndle of housing characteristics to the corresponding price in nearby Santa Rosa. which had not adopted growth controls during the period. The authors found that after several years. Petaluma's housing prices had risen 8 percent above those of Santa Rosa. Schwartz, Zorn, and Hansen (1989) did a similar study of the growth controls in Davis, California. comparing house prices in Davis to those in a control sample of other Sacramento suburbs. They found that growth controls caused house prices in Davis to be nine percent higher in 1980 than they would have been without them. In Petaluma (Schwartz, Hansen, and Green I 98 I) and in Davis (Zorn, Hansen, and Schwartz 1986), the effects on the housing stock affordable to low and moderate-income households relative to control areas were also monitored. In Petaluma, the authors found that the percentage of the housing stock that was affordable to low-and moderate-income households had dropped significantly below that of a control group (Fischel 1990). In Davis on the other hand, growth controls required that those who received building permits to construct some units earmarked for low-income occupants. Thus, the limited growth that did occur in Davis contained both low-income aod high-income housing According to Fischel (1990), however, an unanticipated offset to this apparent success occurred. The authors noted that ll.VTGERS B.ROOIONGS PARSONS fCONORTHWEST 66 Literature Review existing housing in Davis increased not only in price but in quality. Fischel's interpretation of this outcome is that older housing was filtering up rather than down Katz and Rosen ( 1 987) analyzed 1,600 sales transactions of single-family houses during 1979 in 64 communities in the San Francisco Bay Area. Of these transactions, 179 involved houses located in communities where a building permit moratorium or binding rat i oning system was recently or currently in effect. According to Fischel ( 1990). this study is particularly valuable since, unlike the above California studies, it does not focus on just a single community. The authors found that the price of houses sold in the growth-controlled communities was higher than those sold in other communities. Where growth is conzrolled as opposed to managed, housing costs are higher. Literature Synthesis Matrix The Costs and Benefits of Sprawl: Alleged Positive Public-Private Capital and Operating Costs Topic: Fosters Efficient Development of Leapfrogged Areas Allegation/Basis Sprawl fosters efficient in-fill development. Sprawl permits appropriate, relatively high-density development of still-vacant, close-in sites late in the development period of a metropolitan area, without having either TRANSIT COOPERATIVE RESEARCH PROGRAM (TCRP/ H.JO


to d emolish existing impr ovem ems on tho se sites at great cost, or to expend publi c funds buying such sites in advance and reserving them for later development. This occurs because the "leapfrogging" aspect of spraw l l eave s sizable tracts of land vacant and undeveloped Parcels remai n vacan t long after the wave of current gro wth has pass ed them by. Tbese parcels can be developed later as "in-f111" sites at relatively high densities, which arc more approp riate to their more central locations. This process of defe rred developm ent is more efficient than first deve loping all periphera l land at low densities, and then tearing dow n the existing structures when the development market, reflecting the preferences of structure occupants, shifts to higher den sities. Literature Synthesis This point is considered by Peiser (1984) and is also discussed by Altshuler and (1993). It is often a highly negl ected compo nent of the analysis of infrastructure costs rel ated to sprawl. Just as those who call for full costing methods to expand and account for costs of sprawl to the private sector and to society as a whole, there are those who believe tha t the secondary costs of spraw l must be adequ ately tabulated in any accounting scheme related to this and ot her devel opment alt ern atives In an accounting system, those land areas that are skipped o ver and initially n ot u sed, become reasonably inexpe nsive to access and serv ice secondarily. F urther the po!Cntial of u ltimately using these skipped-over lands for inner ring open space also becomes apparenl Only Altshuler and Gomevlbanez (1993) have begun to addres s these issues. uTG

'1?1e Com of Sprowi-Revisited" opposed to sprawL These studies encompass both simulation models and empirical investiga tions. In Stuck in Traffic ( 1992), Downs developed a "what if' scenario testing model to evaluate how changes in the density of development would affect commuting distances. Downs found that the density of growth at the urban fringe has a significant impact on commuting distances; further, shifts from very low to medium densities have the greatest impacl. Increasing exurban densities from 886 to 2,800 persons per square mile decreases commuting distances by 8 percent; an increase from 886 to 4,353 persons per square mile decreases commuting distances by 14 percent. Beyond this, further increases in density only shorten trips by a small amount. In Metro (1994), a simulation was conducted for the Portland Oregon metropolitan area to examine how trip behavior would be affected by alternative development patterns. Among others the patterns included "Growing Out" al1owing lower density single -family growth beyond the current urban growth boundary (UGB) and "Growing Up" keeping all growth within the UGB, increasing densities of single-family housing and introducing more multifamily housing. The Metro simulation found that the concentrated development of the "Growing Up" scenario produced the greatest reduction in vehicle mil es trdveled over the base case (16.7 percent). Several empirical studies have also shown that VMf decreases with higher density. In both a 1990 analysis of commuting in the San Francisco Bay area, and a 1994 study of 28 California communities, Holtzclaw found that neighborhood density was negatively related to VMT. (Holtzclaw also noted RUTGEI.S BlOOKINGS PARSONS E<:ONORTHWEST 68 literature Rview that other fac tors, such as transit access, affect VMf). There is evidence that the mixing of uses shortens trip lengths because of the greater interspersing of residence, employment, shopping and other functions. FrdDk and Pivo (1994) and the Middlesex Somerset, Mercer Regional Planning Board (1990) (central New Jersey) found that with greater land use mixes (and with a higher jobs housing balance), trip distances decreased. Segregation of uses and a dispersed, leapfrog development pattern-both characteristics of sprawl-were linked to increased travel in a recent Cervero (1996) study of dispersed subcenters in the San Francisco Bay area. Between 1980 and 1990 these centers experienced a significant (23 percent) increase in the commuting VMf-with 80 percent of the increase attributed by Cervero to the longer distances between home and work Gordon and Richardson, however, citing research by Crane (1996) hypothesize that the mixing of land uses (and other features of planned unit development and/or neotraditional neighborhoods) make trips cheaper because origin destination distances are reduced. Therefore, a broad mix of uses "mean(s) more vehicle trips, and it is conceivable, perhaps more probable than not, that total VMf may increase" (Gordon and Richardson 1997, 987). This is a minority viewpoint, however, and is clearly put forth by Gordon and Richardson as a hypothesis, as opposed to an empirical finding. (See also "Amount of Travel" and "Travel Time" sections under the positive impacts.) By contrast, much of the transportation literature associates mixing of uses with lessened rather than heightened, VMf. TRANSIT COOPERATIVE RESEARCH PROGRAM (TCRP) Hl 0


In short, both simulations and empirical studies show that sp(awl's defining characteristics-low-density, leap-frog development and spatially segregated land uses-are associated with increased travel. Literature Synthesis Matrix .. s .. Go.wvl ..... """"" o; j Oo!.!t!is condition notably xitf X 11 if 1frongl'y ......... 1,.,_., X The Costs and Benefits of Sprawl: Alleged Negative Transportationtrravel Cost Topic: More Automobile U se Allegation/Basis Under sprawl a higher fraction of total travel must be made in automotive vehicles as opposed to using transit or walking This assertion is almost true by definition, since one of the defining characteristics of sprawl is that motor vehicles are the dominant mode of transportation. Sprawl, with its low densities and spatial segregation of uses requires that virtually all trips be made in an automobile while residents of areas with higher densities and a greater mix of uses h ave the option of riding transit, biking, or walking Literature Sy n thesis The relevant literature relat es modal choice (i.e. election to use an automo bile versus transit, walking, biking, and other alternatives) to development density and development pattern. Density bas a major impact on travel behavior. Residents oflower -den sity RUTGEJtS U.OOKINGS PAASONS ECONORTHWEST 69 literotvre Review residential environments use automobiles more and transit less (and often have longer work trips) Low density environments are also more costly to serve with transit. Numerous researchers have identified threshold levels of residential density for various types of trans i t to be viable for work trips Frank and Pivo (1994) put the base leve l at 9 to 13 persons per acre. Employment density also matters. Compact downtowns support higher use of transit for work trips. Employment densities at work destinations, for example, need to be around 50-75 employees per acre for bus transit to begin to be feasible. Both light rail and commuter rail transit are more cost effective and efficient with dense central business districts which assemble numerous employment destinations very close together Development pattern outside the central business district also affects travel. Unlimited expansion of the urban fringe leads to lower densities and an increase in the number of locations that require automobile access Segregation of uses also results in great e r usc of the automobile (and more travel). On a stretch of Route I in New Jersey, between the Woodbridge and Menlo Park Malls, there are 100 parcels of strictly commercial office or retai l development. Only the malls are a mixed use. To go to lunch for any of the office users requires a vehicular trip to e i ther mall or a retai l facility virtually none are conducive to walking and most are not within walking distance. Lunch hour on this section of Route 1 is as busy as prime commuting time. Simulations of growth in regions have demonstrated that mixing uses in transit corridors results in less dependence upon the automobile and greater use of transit. At the activity center or neighborhood level, a mix of uses encourages walking TRANSIT COOPERATIVE RESEARCH PROGRAM (rCRP) H-10


"'b o Costs of Sprawl-Rov;sm.d' for short trips and trans it use for longer trips. (It is not clear however, whether walking trips replace automobile trips or are in addition to them). Not all observers agree with the abo ve statements linking lower density and land use s egregation with he ightened travel. Gordon and Richardson (1997) argue tha t mixing of uses and other neotraditional design motives may rather than reduce automobile use (see Table 16). Yet there is an abundance of literature to support the initial view. Examples include the classic study of Pushkarev and Zupan (1977) which claims that population density explains much of the variation in transit use; a more recent investigation such as Metro (1994), which arg u es that a development scenario of higher density and containment within an UGB has the highest transit use; Cervero (1989), which shows that suburban centers with higher density and greater land use mix have higher commuting by non-auto modes; Kitamura (1994) which c laims that density is the most important variable in influen cin g non-motorized travel; and Parsons ( 1 996), which argues that h i g he r density and mixed use-not urban design are significantly related to other than-automobile modal choice Other studies reaching similar conclusions include Parsons (1996), Cervero and G o rham ( 1 995), Handy (1994 and 1995), Holt zclaw (1990 and 1994), Cervero 1991) and Newman and Kentworthy (1989). In short, the preponderance of the literature associ a tes sp ra wl's characteris t ics of low densi ty, segregation of uses, and other traits with increased automobile use RUTGERS BlOOKINGS PAR.SON S ECONORTHWEST 70 literatur e Review Literature Synthesis Matrix No Cr.-or """"" -I o..._ "' 1 eondlrton notob_k_ ..u. n X I ,, l'ro"GG r r""lc.d ro -""""" X The Costs and Benefits of Sprawl: Alleged Negative Transportation/Travel Cost Topic : Longer TraveUCommuting Times Allegation/Basis Residents in sprawl locations spend more time commuting a.s well as in other foms of travel. Trave l commuti n g time is greater under sprawl because its low density,leapfrog pattern of devel opment, and segregation of uses enhance the separation of residence, place of emp loyment, retail, and other functions. Literature Synthesis Gordon and Richardson ( 1991, 1997) have argued that the suburbanization of jobs has reduced commuting times In their words: "suburbanization has been th e dominan t and successfu l mechanism for reducing congestion. It has shifted road and highway demand to less congest ed routes and away from core areas. All of the available recent data from national surveys on self-reported trip lengths and/or duration support this v i ew" (Gordon and Richardson, 1997. 98). Other researchers have reached similar conclusions. Levinson and Kumar (1994), for instance, found that the TRANSIT COOPERAnVE RESEARCH PROGRAM (TCRP) I< l 0


"Tht Cosls of Sprowl-Rovisiled" average commuting times in the Washington, D.C. metropolitan area remained stable from 1968 to 1988-despite this region's significant growthbecause a greater dispersion of activities helped keep travel time constant There is contrary evidence, however. Vincent et al. (1994) analyzed the National Personal Transportation Survey Data for 1990 and found that commuting times for residents of urbanized areas outside of central cities were longer than those of central city residents. The average peak period commute for the suburbanites was 21 minutes compared to 19 minutes for central city residents. Likewise, off-peak commutes of suburbanites were 19.7 minutes long compared to 17.2 minutes for central city residents. Pisarski (1992) further reports that based on census data, suburbanites had larger increases in commute times between 1980 and 1990 than central city residents. The average travel time for suburban residents that commuted either to suburban or central city locations increased by 14 percent, while the average commute time for a central city resident increased by only 5 to 7 percent Ewing (1995) argues that it is regional accessibility, not density or mix of uses, that detennines total travel times. He found in a Florida study that residents in neighborhoods where jobs, schools, shopping, and other services were most accessible spent 40 minutes less per day in vehicular travel than residents of neighborhoods where these activities were least accessible Shorter automobile trips, not use of other modes, makes the difference here. Some simulations show less congestion (measured in vehicle hours of delay) with sprawl than with more dense development. Others show the reverse. Cambridge Systematics (1994), for RUTGERS U.OOKINGS PARSONS ECONORTHWEST 71 l.iteratuJe Revitw instance, found almost a 20 percent reduction in vehicle hours of delay under a "highway only" scenario versus an alternative which clustered jobs, housing and shopping near transit li nes. Literature Synthesis Matrix NoCr.or sw;!nliof s-. o---o ... ,..

conclude that residents of denser, more transit-friendly neighborhoods (as opposed to sprawl locations) should be able to spend a smaller share of their budgets on travel. However, because household characteristics, such as family size and life style can affect both travel behavior and budget allocations, it is unclear whether households in sprawl development configurations spend a higher fraction of their budgets on transportation due to sprawl or for other reasons. Literature Synthesis Matrix NoO.or Sub;:Oidiol "-"' ..... """"'-o; 1 Doe! . !hl col'ldi"on Mltobly JCidt X I it IITOI'ISiy ,., .. ........ X The Costs and Benefits of Sprawl: Alleged Positive Transportation/Travel Cost Topic: Less Vehicular Miles of Travel Allegation/Basis Sprawl generates less travel versus other land use development patterns because residences and businesses ultimately "self correct" (Gordon and Richardson 1997) to be in proximity to one another. Other factors (see below) also contribute to sprawl's lessened travel. Literature Synthesis Gordon and Richardson ( 1997) argue that the market forces embodied in sprawl may realize reductions in travel as residences and businesses ultimately spatially locate near one another. (See "Travel Time" under positive impacts.) In addition they hypothesize that the mixing of uses and other features of RUTGERS BROOKINGS PARSONS ECONORllfWfST 72 literature Review planned unit developments and/or neotraditional neighborhoods as compact development forms may increase VMf. From another perspective, while there may be more non-automobile travel in areas with mixed use-this travel may not replace but may be in addition to necessary automobile trips. Literature Synthesis Matrix ..:!, S11b;!.lioJ ,_ ""'- I ':"" xlatt X !f ttronsl)' X The Costs and Benefits of Sprawl: Alleged Positive Trans portation/Travel Cost Topic: Shorter TraveUCommuting Times Allegation/Basis Sprawl shortens travel/commuting time, because suburban-to-suburban work trips which characterize sprawl are sh orter in average duration than commuter trips between central cities and suburbs. Moreover automobile vehicle commuting trips are much shorter in average duration than public transit commuter trips and the former predominate in sprawl. Thus, even though sprawl may cause average commuting trips to be longer in velriclc miles traveled (and even this is arguable), they are shorter in the amount of time consumed. Under sprawl there are also more single-occupant vehicle trips undertaken-the fastest and most direct mode of travel. TRANSIT COOPERAD\If RESEARCH PROGRAM (TCRP) H-10


"'f>e Cods of Sp10wf-Revisifed" Liter a ture Synthesis See "Amount of Travel" in the negative impacts section This is the same statement in reverse. Gordon and Richardson (1997, 98) argue as follows: "Industry moves [0 the suburbs. following the labor force, which allows many work-ers to enjoy a shoncr work rrip in time if not in distance and reduces congestion pressures in traditional centers." Although the adjustment is not instantaneous and there are inevitable short -t erm disequilibria, the important point of the Gordon and Richardson argument is that the self-corrections are made relatively fast. Most available recent data from national surveys on self reported trip lengths and/or durations corroborate the view that suburbanization has been a dominant force in shifting road and highway demand to less congested routes and away from central areas. The findings from all seven recen t large-scale national household surveys present a consistent story of the containment of metropolitan area commuting times (Gordon and Richardson 1994b) Evidence from NPTS reports (Nationwide Personal Transportation Study) ( 1977, 1983, 1 990); a commuting questionnaire included in the American Housing Surveys (1985, 1989); and the two decennial Census reports ( 1980 and 1990) make the same po i nt. See also "Amount of Travel" and "TraveVCommuting Time" in the negative impacts and "Amount of Travel" in the positive impacts sec t io n s. Literature Synthes i s Matri x c.::ror Ho0-r ,_ o.-.. .. noto xhrt X l!nkecl to ......... X RUTGERS BROOKINGS PARSONS ECONOI:THWEST 73 literature Review The Cos t s and Benefits or S p rawl: A lleged Positive Transportation/Travel Cost T opic: Reduced Transit Subsidies Allegation/Basi s Sprawl is with less capital spending and generates fewer operating losses for transit because a much higher percentage of all travel under sprawl is undenaken by the private automobile. Alternative forms of development that rely more on public transit, especially fixed-rail transit must rely on large amounts of public subsidy to build transit facilities and to operate them since such systems almost never recoup their costs from the farebox L ite rature Synthesi s Gordon and Richardson ( 1997) argue that the auto subs i dy is less than that of the transit subsidy (Pucher 1995 and Ewing 1997) in turn rebut that the auto subsidy is in fact quite significant when a full cos t accounting i s done. There is general agreement that the automobile dominance of sprawled regions results in less extensive, bus-only transit systems that primarily serve a small share of transi t -dependent households (Although some regions typical l y associated with spraw l such as Los Angeles, have rail transit) T here is also general agreement that bus-only systems require less capital outlay than rai l systems but higher operating costs. There is little disagreement about whether sprawl results in Jess or effective transit systems. Sprawl is associated with very cost-effective commuter i ntercept lots l ocated on or near major interstates or freeways. Less sprawl appears to enhance the use of local TRANStr COOPERATIVE RESEARCH PROGRAM (TCRP) H IO


bus systems and reduce the use o f regional bus systems (Burchell J992a) For agreement with the fanner statement, see Ewing 1997. Both historical and recent research shows that residential and emp l oyment d ensities are positively related to light rail and c ommuter rail cos t--efficiency (annual operating costs plus depreciation per vehicle mile) and effectiveness (passenger mile per linemile). Similar studies have not been done for bus systems. Literature Synt hesis Matrix ....... ...... ... a-w.;: ... Mf o--I)( 1 condiliofl M>toblv e.w.l.,t X h i t ttongl y IWE.eodto ........ X The Costs and Be n efits o f Sprawl : Allege d Negative Land/Natural Habitat Impa c t T o pi c: Loss of Agricultural Land A llegation/Basi s Sprawl removes more prime agricultural land from fanning use than othe r more compact fonns of development. This happens f or three reasons. First, low density uses i nherent in spraw l's resident ial development patterns require more space for t h e direct placement of dwe lling units than higher density uses und e r compact development. Second, the scatterati o n o f dwelling units across the landscape far from the edges of built-up settlements renders the agricultural use of much of the land adjacent to the scatte red dwellings inefficien t and in competition. Third, the prospect o f obtaining h i gh prices for land motivates farmers and land speculators to assemble large parcels from prior farm use because these l an d s are cont i guous and can be bought in bulle ,, RUTGERS BROOKlNGS PARSONS fCONORTHWEST 7 .t l.iterofuro Review Literature S ynthesis Mul tiple studies have documented the significant losses of agricultural lands to th e current d evel opmen t process These studies range from national reviews of the Joss of farmlands and fanns over time, such as the National Agr i cultural Lands study (198 1 ) and the American Fannland Trust's Fanning at the Edge (1997), t o regional/state investigations of a similar type (e.g., N e lson [1992] in Oregon and Adelaga [1991] in New Jersey). There is substantial disagreement, however, about whether this loss of agricultural land creates significant social costs. To some observers it appears t hat there is no shortage of prime agricultural land in the United States, since the nation has often produced crop surpluses (Gordon and Ri chards o n 1997). Y e t demands for food are rising sharply throughout the world, as living standards increase in once-poor nations and the world's total population expands Price s of major agricultural crops bave already increased substantial l y in the last few years. Hence, in the long run, the world will need all the food produc tion capacity it can muster (Ewing 1997) Further, there are widespread policy initiatives to try to preserve fannland M any states, (e.g Maryland, New Jersey and Vermont) and other levels of government (e.g., Lancas t er County, PA) have adopted progr ams in recen t years, ranging from the purchase of development rights to the enactmen t of "right to farm" laws, in order to foster land an d farmland preservation ( N elson 1992) Land development patterns are related to fannland loss. N u merous g r owth management p l ans---attempting t o reverse sprawl-include farmland preservation as TRANSIT COOPERATIVE RESEARCH PRO GRAM (1CRP)


an objective (New Jersey 1 991; Vermont 1 988; Maine 1988). These are couched in goals of planned developmen t as opposed to of sprawl's taking of farmland. Th e limited empirical investigations of sprawl's impact on "consuming'' farmland--and in opposition, the i mpact of alternatives t o spraw l on farmland we.:e performed by Burche ll et al (1992-1997) in New Jersey, Lexington Kentucky the Delaware Estuary Michigan, and So uth Carolina, and by Landis (1995) in t be San Francisco Bay area These analyses employed land consumption models at the minor c i vil subdivision level to look a t differen c es between tre n d development or "b u siness as usual" scenarios, and more environmentally conscious land d evelopme nt ap p roaches. The former embodied sprawl -lilce cbar.!Cieristics; the latte r more compact and planned development characteristics These models allowed future projections of households and jobs to be c o nverted to the de m and f or residential and nonresidential structures, and ultima t ely to d emand for r esidential and nonresidential land, with rul es for spillover to adjacent municipalities and to unincorporated areas. In both the Burch e ll and Land i s studies historical rJtes of farmland takings were applied t o land consum e d under existing de v elopment patterns, and goals of farmland retention were applied under the a l ternatives. (A similar procedure was used for environmental land consumption comparisons.) In the Burchell study, agricultural lands i ncluded s uch categories as cropland that is harvested pastured lands in permanent past u re, and woodlands that could be used for agricultura l purpo s es Fragile environm ental lands enco m passed floodplains and wetlands, acreage with steep slopes o r with critical ha bitat des i gnation, aquifer recharge areas and RUTGERS &A.OOKtNGS PARSONS E<:ONOIIHW$1' 75 literaturt Review critical sensitive watersheds, and steam bu f fers. The models emp l oying different densities development locations and occasional l y different housing types under the alternat i ve futures, calculated the total agricultural (and fragi le environmen tal lands) that would be consumed. Burchell's res u lts showed s avings in agricultural acreage consumed of roughly 20 percent in South Caroli na, Michigan and Lexington under trend v ersus plan; abou t 30 percent in the Delaware Estuary; and 40 percent in New Jersey. (See tables 2 and 3 for details ) Landis results in the San Francisco Bay Area were even more pronounced. Scenario C (compact g r owth) saved nearly 50% in farmlands and ste e p s l oped areas and c l ose to I 00% in wetland areas (Landis, 1995, 449). Lite ratur e S yn t h es i s Matr i x .. ;;.., NoO.w Wt:;: nl>of ...... """""'" OiiO!Mi e-nl 1 . ':h' COf'ldi'lion notobl y exi ll f X il atrong y 111'1\:d to 5Dtowff X The Costs a nd of Sprawl : Allege d Neg a tive Land/Natural Habita t Impact Topic: Reduced F armland Pr od u ctivit y A ll ega tion/Basis The productivity of land beingfamted near s cattered sprawl settlements is reduced by t he difficu!Jy of conducting efficient farming operatioiiS near residential subdivisions. Subdividing land into small lotS for resid e ntial purposes inhibits farmers ability to operate on large c ontiguo u s land parcels, TRANSITCOOPERATM RESEARC H PROGR AM (rCRP) /1.10


"!he Costs of Sprowi-R..nm.a thereby reducing the effic ienc y of mechanized agricultural operations. Furthermore, under sprawl development, subdivisions and fanns are typically interspersed and residents often object to the smell, noise. truck traffic, and other l ocal conditions associated with active agricultural uses. Further, when this contiguous development s i tuation occurs, local governments often impose restrictions on farming. These conditions bring about an "impermanence syndrome," (see below) that is antithet ical to sustained farmland productivity. Literature Synthesis There is an extensive literature on constraints to farming in urbanizing locations (Lisansky 1986; Lopez et al. 1988; and Nelson 1992). In rural areas that can be readily developed, high land values shift the far mers' "objective function" from agricultural operations to capital gains from real estate sales Real estate sales, in turn, reduce average farm size, thus limiting the realization of economies o f scale-a characteristic of U.S. agriculture To this are added a variety of restraints ranging from restcicti ve regulations to recurring vandal ism. All of these factors foster an "impermanence syndrome" -a reluctance of the farmer to invest in new techno logy and farm infrastructure with land idle, awaiting conversion to real estate use. Numerous studies related to sprawl al. lege that the impermanence syndrome is deleterious to farmland productivity (AFT 1997) The direct rel ationship of sprawl development patterns to farmland co nsumption was examined by Burchell et al. (1992a) for the state of New Jersey. In addition to projecting the total farmland that would be lost under trend versus plan development, the New Jersey analysis identified the quality of farmland that -.urGEJtS U.OOKINGS PARSONS ECONORTHWfST 76 Liltratur Review would be consumed-"prime," "marginal" and "poor." The New Jersey analysis showed that not only would trend developmen t draw down more farmland, but since the better quality farmland is the most amenable for development (in that it is flatter, drains better, and so on) trend's farmland lo ss was concentrated in the "prime" and "marginal" catego ries while plan's farmland consumption would be overall less and wholly contained in the "poor' farmland category. The Burchell et al. New Jersey study thus considered the association of farmland quality and development patterns-but on l y from a farmland c o nsumption perspective. No analysis to date bas examined how development pattern (i.e ., sprawl versus compact) would affect the productivity of farmland that remains in agricultural use. Literature Synthesis Matrix G.:!rol No0.or ,_ o....._ !"'s e.xisl1 X hill tr ongly linked to spr awft X The Costs and Benefits of Sprawl: Alleged Negative Sprawl Effects Topic : Reduced Farmland Viability (water constraints) Allegation/Basis Growth through sprawl causes great expansion in the demand for water for urban uses, thereby reducing the amount of water available for agriculture. This is e specially significant in those southwestern regions where sustained shortages of water exist. Agriculture uses much more water than urban settlements JRANSCf COOPEAATIVE RESEARCH PROGRAM (TCRP) H-10


in many states where farming depends upon irrigation, such as Texas, Arizona, California, and Colorado. Therefore, as urban settlements expand in these areas, more water must be diverted from agriculture to supply the basic human needs of the residen t population. This restricts the operation of farming in such areas. Furthermore, single-family property owners and corporate commercial facilities use water for lawn sprinkling, which appears to be an excessive use of this natural resource. Literature Synthesis Multiple stud i es have examined generally how development in more arid location s, especially in the West and Southwest United States, is drawing down the water supply, with potential conflict with the irrigation needs of agriculture The literature, however, has not examined the specific association of sprawl and farmland viability with respect to water supply. This would involve a multi linked analysis of: I) how development affects water demand; 2) whether development's consumption of water would differ under sprawl versus other forms of development in these areas; and 3) the relationship of steps (I) and (2) to the amounts of water supply for agricultural and residential settlements in given locations compared to the total supply available there Although a fully linked analysis such as the one described above has not been undertaken, an extensive literature exists on water demand relevant to steps (I) and (2). For instance, the Army Corps of Engineers incorporates in its water demand forecasting model, among other RUTGEI.S PAR.SONS ECONOIUHWEST n literature Re'liew factors the magnitude of lawn sprinkling, which is likely to be higher under sprawl versus compact development (NJDEP 1980). The Hittman water demand model includes housing density as one factor-a variable clearly different under sprawl versus more forms of compact development (NJDEP I 980). In a similar vein, the multi variable IWR-Main water forecasting model (Bauman and Dziegelewsk:i I 990) incorporates in its multiple coefficients development density and the number of housing units by type (detached versus attach ed)-variables different under sprawl versus compact development. The Burchell et a!. ( 1992) analysis of trend versus plan development in New Jersey considered how water demand influenced water consumption under these two scenarios and incorporated some of the variables (e.g., housing type) noted above. Burchell found only small differences in water demand by development scenario; from I 990 to 20 I 0 the increase in water demand statewide was projected to be 60.1 million gallons per day (MGPD) for trend versus 58 0 MGPD for plan. This analysis did not, however, relate this 2 MGPD variation finding to the demands on water supply for residential development versus agricultural uses in New Jersey. Water supply is not a development constraining issue in New Jersey -as it is i n more arid regions of the United States. Literature Synthesis Matrix TRANSfT COOPERATIVE RESEARCH PROGRAM (TCRP/ 11-10


T h e C os ts a n d Be nefits o f Sprawl: Alleged Negative Land/Natura l Hab ita t Impact T o p ic: Loss of F ragile Environmental Lands (e.g., acreage in wetlands, forests flood plains and steep slopes) A llegati on/Bas i s More frail lands are destroyed by sprawl than wuler more compact settlement patterns. Because sprawl spreads urban development over a much larger area than more compact sett l ement it inherently consumes more land Because lan.d development under sprawl is not centmlly planned or supervised, there is a greater probability that fragile environmental l ands will be converted to res i dential and other uses. Local governments are likel y to misjudge the consequences o f environmental degradation because they are not concerned w ith the overall balance between environmental l y sensitive l ands and develop i ng l and uses in the region as a whole. L i terature Synthesis Several studies document losses of and threats to fragile lands. Dahl (1990) estimateS that since colonial times the United States (48lower states) has lost about II 0 million acres of wetlandsabout 55 percent of the starting wetlands inventory. The Michigan Society of Planning Officials (MSPO) estimates that 20 percent of Michigan s forested, wetland and steep l y sloped areas were lost between 1970 to 1990 (MSPO 1990). Numerous growth management plansattempting to reverse sprawl-have evaluated how managed versus traditional RUTGERS BR.OOKINGS PAlSONS ECONORTKW'EST 78 devel opment patterns affect fragile lands. The se include the Orlando, Florida Urban Area Growth Management Program (Orlando FL 1981), the Evaluation of City of San Diego Growth Management Program, and the Report of the Year 2020 Panel of Experts (Chesapeake Bay Executive Counc il 1988). The Orlando study examined how managed growth versus a "continuation of past trends" would affect t h e preservation of wetlands and flood plains. It projected under managed growth a saving of almost 20 percent in the inventory of these fragile environ mental lands, (i.e., 20 percen t less acreage l ost). Analyses of sprawl's impact on fragile has been conducted by Burchell et al. (1992 -1997) in New Jersey, Lexington KY, Del aware E stuary Michigan, and South Carolina, and by Landis in the San Franc i sco Bay area. Burchell et al. found that plan (compact) versus trend sprawl like) development woul d reduce consumption of fragile environmental lands by alm ost one-fifth T he range of the saving was from 12 to 27 perc ent, depending on the s tarting level and location of wetlands, forests and lands of steep slope in these jurisdictions (See Tables 2 and 3). Land i s' findings were even more pronounced, favoring the c ompact growth scenario and were calculated separately for steep slopes and wetland areas (Landis 1 995). Literature Synthesis M a t r i x c.::!. Nob.w Sub;;!.,,;o, ...,_, 1_.:-...,. """" OitoQII'Hflltlnl u o!.':"ll eondihon nolobl y i$lf X it ifi'Oflgty lin k d l o KII'Owlf X T RANSIT COOPERATIVE RESEARCH PROGRAM (TCRP) H-!0


The Costs and Benefits of Sprawl: Alleged N egative Land/Natura l Habitat I mpact Topic: Lessened Regional Open Space Allegation/Basis The sening aside of open spaces for public uses by residents of the entire region may be "wuler-financed" in sprawl-dominated areas, compaxed to those with more regionally oriented governance structures. Municipal governments motivated by fiscal pressures to provide benefitS only for their own residentS may he unwilling to devote resources to creating facilities for use by persons throughout the region. Literature Synthes i s There is scant literature dealing with this issue explicitly to detennine whether a substantial cons ensus existS. The only literature that does exist is that very large scale developmentS and conservation developmentS, both often of a nonsprawl nature, frequently have significant set asides for contiguous open space. Arendt (1994) points to a movement from golf course communities t o open space communities so that the private and public sectors have a greater chance to share in the l and resources. The Sterling Forest Corporation. developing a 12,000acre development in Tuxedo NY, prior to land buyout by the federal government, pledged 75% of the land to remain in some definition of private/public open space (St erling Forest Corporation 1993). Literature Syn thesis Matrix I """""' -O.ot I ...,_ 0.._ _, I condlti011 e:ilitf X l l it tlrongty linkd lo I JOrawlt X RUTGERS artOOKINGS PARSON. $ ECONOI.TliWfST 79 literature Review The Costs and Be n efits o f S p rawl: A ll eged P o sitive Land/Natu r al H abitat Impact Top i c: Enhanced Personal and Public Open Space A ll egation Basis (a) Sprawl provides more open space directly accessible to individual in the fonn of larger private yards anached to their dwellings than are possible in more compact forms of senlement. The average lot size in sprawl settlement patterns is much larger than in more compact fonns of sertlement, and a higher fraction of dwellings have separate individual yards. Therefore. more households have direct access to their own private open space, and that space is larger, on average, than equivalents found in more compact sett lementS. (b) Sprawl provides both larger amount s of and more accessible open space withaut large public expenditures for buying land by leaving large unserrled sites "inboard" ofthefanhest ow urban subdivisions because of irs leapfrog development .. This provides aesthetic and recreational benefitS to the public without requiring use of taxpayers' funds Literature Synthes i s Personal open space continues to be high on the list of the desires of most Americans (Fannie Mae 1995). In surveys conducted by the Federal National Mortgage Agency, not only are yards desired by the buying public prospective homebuycrs want yards on all s i des. In the mid-l990s according t o most current surveys of buying preference, single-family detached housing is more popular than it was a decade ago. Much of this is related to TRANSIT COOPERATIVE RESEARCH PROGRAM (TCRP) H-10


"The Com o f occupants not wanting to pay community association maintenance f ees, but at least some of this is related to the desire for more rather than less personal open space (NAHB 1995). There i s a very small literature that indicates that the skipped-over development patterns of sprawl create parcels of land that can be used for inner suburban or urban open space as this becomes a local priority Except in the wealthiest and most resilient of inner suburbs. this almost never is a choice or option of local government. Most governments in these localities feel pressed for fiscal resources and dispose of these land parce ls to the highest bidder. Thus, the oppos i te to what is popu l arly suggested often takes place. Through the local variance process these lands frequently are given a higher intensity residential or nonresidential use. Those abutting properties rather than getting permanently improved open space receive more intensive and occasionally neighborhood disruptive land uses. This is because th ese uses can pay more i n taxes than existing neighboring uses and the previously undeveloped vacant land. Thus. although there appears potential for inner open space t o be the result of skipped-over lands, rarely does this happen (Downs 1994). L"t S tb M t 1 e rature s:vn es t S a rtx ., No Cr.or ,_ ,.;_,, .,_ eondilicon -.xis t f X h it 'trong1y IW!ked to ,.....11 X Tbe Costs and Benefits of Sprawl: A ll eged Negative Quality Of Life Impact Top ic: Aesthetically Displeasing RUTGEI!S PARSONS ECONORIHWEST 80 Utero.ture Review Allegation/Basis Lowdensity patterns are less pleasing aesthetically wui provide fewer cultural oppartunities An important element of the quality of life of any community is the aesthetic and cultural satisfaction of its residents in daily life. If the environment they normally encounter is dominated by the homogeneous architecture of subdivisions and '"strip" malls. the absence of quality civic spaces and landmark buildings, and a lack of pedestrian-scale amenities, the aesthe t ic satisfaction people derive from their surroundings is reduced. Moreover, sprawl does not easily lend itself to the formation of communities that have a feeling of cohesiveness and can organize to support the arts or other cultural inst itutions. Literature Syn t hesis The aesthetically less pleasing aspect s of sprawl, such as visual uniformity, are often cited as a cost of this form of development (Nelessen 1994). Critics of sprawl often decry i t s ugliness. For example, Sh. ore (1995) maintains that "spread city" is inherently ugly because the settlement pattern has no clear form; retail businesses loca t ed along highways must use "raucous" signs to attract passing motorists; and a significant portion of tbe land is given over t o the automobile. James Kunstler in a public presentation in Lansing, Ml in 1996 described U.S. suburbs as "use l ess and without purpose and occupied by people of the same make-up (Kunstler 1996). Low-density developments, however are not necessarily l ess pleasing aesthet icall y than more compact forms of development. The aesthetics vary from development to development. Some low density residential developments particu l ar l y high-in come ones, may have TRANSIT COOPERATIVE RESEARCH PIIOGRAM (TCIIP) H l 0


much more open space and elaborate landscape designs than high-density residential areas In fact, defenders of sprawl often contend that the open spaces of sprawl make it more attractive than compact forms of development. The literature reflects these two conflicting opinions. There is little evidence within the literature, however to suggest that Americans find sprawl less attractive than more compact forms of develo pment or that low-density living provides them with fewer cultural opportunities. Visual preference surveys have been used t o gauge the reaction of Americans to sprawl, but such studies are often criticized for failing to make a distinction between sprawl and factofli not typically associated with that form of development, (e.g., architectural design). Moreover, survey research does not consistently indicate that Americans overwhelmingly find sprawl to be aesthetically less pleasing than compact forms of development. While some surveys have revealed that when shown images of both sprawl and traditional communities, individuals favor the latter by a wide margin (Neuman 1991), some aspects of sprawl appear to appeal to Americans. Individuals were found to favor homogeneous neighborhoods over mixed neighborhoods by a margin of two to one (Bookout 1992). Survey research in Florida has suggested that individuals there have a strong preference for low d ensity or ex urban living (Audirac and Zifou 1989). On the subject of cultural activities, Shore (1995) co n tends that sprawl does not allow for the formation of communities that easily organize to support activities such as the arts. As a result, low-density residential communities may have fewer and lower quality c .uJtural activities than urban areas. Shore argues that a movement away from "spread city" and lUTGO.S BROOKINGS PARSONS ECONORTHWEST 81 literoture Review toward the restoration of downtown areas would result in more cultural activities and other services that are supported by large communities. In general, few will argue with the belief that an attractive and atl>1hetically pleasing community increases its overall quality of life. Within the economics and migration literature, it has been well documented that a community viewed as having a high quality of life will attract and retain individuals. Studies of migration patterns find that a community's s cenery, natural environment, and outdoor recreational opportunities are important factors in attracting and retaining individuals. A survey study of migrants to and residents of 15 wilderness counties found that scenety and environmental quality were more important factors in attracting individuals than employment opportunities or cost of living (von Reichert and Rudzitis 1992 ) Two of the most important conditions that "lone eagles" (individuals who are able to liv e anywhere and telecommute to work) cited as influencing their decision to move to the state of Washington were the quality of the natural environment and th e outdoor recreational opportunities there (Salant et al. 1996). Cushing (1987) demonstrated that proximity to mountains and coastlines influen ced population migration due to the aesthetic qualities and additional recreational opportunities that such features provided. Empirical results indicated that inteJ:litate migrants are attracted to hilly terrain and major coastlines. As n o ted however, there is only some agreement over whether low-density developments are aesthetically less pleasing than more compact development patterns. In particular, the literature fails to indicate a significant causal relationship between sprawl and aesthetically less pleasing low-

"The Com ol Sprawl-Rtviritod" the literature does indicate, however. is that the aesthetics of low-density areas will vary from place to place and that the preferences of individuals will vary from person to person Literature Synt hesis Matrix ... ._ , .......... "'*-ooMiM ....... X "" ........ X The Costs and B e nefit s of Sprawl: Alleged Negative Quality Of L ife Impact Topic: Lessened Sense Of Community Allegation/Basis Low-den.rity dev e lopm ent weakens households' connections to both their immediate neighbors and to the l arger metropolitan commu nity and e n courages unsocial values Sprawl weakens the linkages of residents both between nearby neighbors, and among all ot h er residents of their metropolitan area. Linkages with neighbors are reduced because the low residential density, the heavy orientation t owards moving by car rather than on foot. and the lack of neighborhood retail outlets and other meeting places diminish interpersonal contacts. Linkages with other residents throughout the metropolitan area are also diminished by the fragmentation of governance and fiscal resouroes that prevent a commonality of purpose, and by the extreme diffusion of households and jobs throughout an area. The resultant loss of a sense of community makes it difficult to generate support for region-wide anacks on social and other problems which cannot be solved by purely local policies and actions Finally, ltUTGUS llOOKINOS PAitSON$ ECONOlfHWEST 82 Literature Revitw because spraw l in its most pejorative manifestations is wasteful, unaesthetic, and antisocial i t is not an environment that nurtures the important social values of ecology, sustainability, and community. Literature Synthesis Critics of sprawl often claim that a loss of "sense of community" is one of its greatest social costs (Ewing 1997). Defenders of low-density settlements however, deny that residents experience any less "sense of community" than residents in big cities or more compact settlements (Go r don and Richardson 1997). In fact, tbe evidence from as far hack as Her bert Gans (as cited in Jacobs 1961) indicates that some dense areas lack community, while some s uburban areas have it. Much of the controversy arises because sense of community" is difficult to define and even more difficult to measure. In his review of the literature on "sense of community" Cochrun (1994) finds that the term has been used to describe a number of things but that the most comprehensive definition was developed by McMillan and Chavis (1986 ) McMillan and Chavis identified four factors that contribute t o sense of community: (I) membership; (2) influence; (3) integr ation and fulfillment of needs; and ( 4) shared emotional connection. Cochruo (p. 93) offers a definition of "sense of community" that incorporates the four factors identified by McMillan and O!avis: "People who bave a stroog oommuni!y feel fikelhey belong in their nc:ighbo

collective of their neighbori!oods." In Edge City: Life on the New Frontier, Garreau ( 1991) searches for a defmition of community, particularly within Edge Cities, and reaches the conclusion that community and neighborhood no longer mean the same thing. Instead, Garreau maintains that "mobility" and "voluntary" are two important terms that help to define community-individuals want to be able to both join and leave communities at their choosing. Moreover, Garreau contends that a community should be a social grouping that is readily available to individuals and does not interfere with individual freedoms. In partial contradiction, in an article examining changes in suburban lllinois, Lemann (1989) found that community building efforts in Naperville, a fast growing suburb of Chicago, were hindered by the high rate of turnover of residents. Critics of sprawl argue that residents in mixed-used neighborhoods have more sense of community and social interaction than do residents living in low-density developments because they are more likely to walk from place to place and, consequently, they are more likely to have contact and interaction with others. Residents in low-density areas, on the other hand, rely more on their cars for shopping and recreation trips and, hence, are less likely to develop conlacts and friendships with their neighbors (Nasar and Julian 1995). Drawing on the work of Glynn (1981) Nasar and Julian assessed the psychological sense of community across different neighborhoods and housing conditions in northwest Columbus, Ohio. They found that residents of mixed-used areas had significantly more sense of community RUTGERS BAOOIONGS PARSONS ECONORTHWfSJ 83 than residents of single-use neighborhoods Literature RevMw Opponents of sprawl also maintain that low-density development weakens a "sense of community" by segregating residents (Duany and Plater-Zyberk 1995; Kelbaugh 1993). According to Kelbaugh (p. 24), suburban insularity breeds "ignorance, misunderstanding, and ultimately builds tension'' among residents. Kelbaugh prefers high-density, mixed socioeconomic, racial, and ethnic neighborhoods because they allow individuals to "rub shoulders" with fellow residents on a daily basis and work out differentes. Similarly, Duany and Plater-Zyberk contend that suburban housing fosters a breakdown of the larger community because it segregates residents by income into enclaves. Kunstler (I 996) attacks suburban sprawl and the zoning laws that have created it. The allegation that lowdensity residential living lowers "sense of community" may be inferred from his remark that "The model of human habitat dictated by zoning is a formless. soul-less, centerless, demoralizing mess .... it corrupts and deadens our spirit." Like Duany and Plater Zyberk (1995}, Kunstler argues for development patterns that are mixed-use and provide housing for people with different incomes. Sprawl may weaken not only households' connections to neighbors and the larger community, but it may also weaken connections between family members who occupy the same residence. Some contend, for example, that sprawl reduces the amount of time parents spend with their children because more ho useholds must have two people working outside the home in order to pay for the multiple automobiles required by TRANSIT COOPERATIVE RESEARCH PROGRAM (TCRP) HIO


daily life.' This need t o support the household's transponation f acilities may, i n fact, even reduce the quality of chi l d care provided by parents. Some contend that mothers working outside the h ome provide lower quality child care than tho s e who stay at home. The subject is fraught with controversy (J oseph 1992). Meanwhile, Kelbaugh (1993) examines another potential non-economic social cost associa t ed with sprawl-the t ensions that result from parents spendi ng l ong hours commuting i nstead of with the i r children or each other The literature does not readily provide support for the opposite allegation: i.e., low-density development strengthens households connections to both their neighbors and larger community. Ewing (1994a) suggests. howe v er, that low density development does not provide residents with any l ess "sense of community" than higher-density d evelopment. After reviewing extensive literature on sprawl, he concludes that the re i s not eno ugh evidence to determine whethe r a lack of an identifiab l e commun ity is associated with sprawl. One further issue related to a lack of "sense of community" is the "throw -away mentality" or more elegantly, the l ack of value for eco l ogy and sustainable life styles. Some argue that sprawl encourages th e "throw away" mentality among households In a sense sprawl deve lopment may be seen as a continuation of the "fronti e r mentality" of early American s e ttlers who believed the y could change their current situation by l eaving existing homes and problems behind and moving west onto I Some ht.v e toft\ltd tbat i r o nly per Mn worklnt

alleged effects. In this case, the overlap is with transportation effects, which include allegations about traffic congestion and travel times. The reader is referred to those sections for more details on points discussed here. Much of the debate about commute time has been based on data that compare travel times for residents of suburbs and central cities. There is little data on the travel times associated with the density of development. Ewing ( 1997) in his analysis of household travel patterns in a sprawling Florida county purpons to show that households living in the most accessible areas spend about 40 minutes less per day traveling by vehicle than do households living in the least accessible locations (Ewing 1995; Ewing eta!. 1994) Ewing states that this savings in travel time is due almost entirely to shoner auto trips, and that the significant land-use variable affecting travel times is regional accessibility, not local density (Ewing 1997). There is evidence that increased commuting time increases the stress of commuters. Novaco et at. (1990) found that increased travel impedance as measured by commuting distance and time, is associated with increased measures of stress. T ravel impedance was also found to have stat isti cally significant effects on job satisfaction, work absences due to illness, and overal l incide nce of colds or flu. Subjective or perceived conditions o f travel impedance were found to have statistically significant effects on mood at home in the evening and chest pain. Consequently, the study found that job change in the sample was primarily related to commuting satisfaction This study validated resu lt s from the authors' previous work, which found that impedance characteristics of commuting are stressful, as measured by effects on blood pressure, tolerance for RUTGERS BROOKINGS PARSONS ECONORTKWEST 85 literature Review frustration, negative mood, and overall life satisfaction. This earlier work also found that the desire to change residence because of transponation conditions was related strongly to high impedance (Novaco et al. 1979; Stokols and Novaco 1981; Stokols eta!. 1978). The physical stress effects of impedance have been corroborated by a study of the effects of average commuting speed on blood pressure and proofreading measures (Schaeffer et a!. 1988). Koslowsky and Krausz (1994) directly addressed the links among commuting time, stress, and workers' attitudes toward their jobs, based on a statistical analysis of survey responses from the over 600 nurses that participated in their study. The researchers found that commuting is a possible source of recurrent stress that can lead to undesirable organizational consequences. This study also found that the correlation between commuting time and stress was stronger for those who drove to work compared to those using public transit, but the authors do not rigorously explore the reasons for this difference Koslowsky and Krausz ( 1 994) also cite prior literature that found a relation between the commuting experience and such organizational outcomes as: absenteeism (Taylor and Pocock 1972), lateness (Gaffuri and Costa 1986), and turnover (Seyfarth and Bost 1986). The literature on the stress effects of commuting does not rigorously address the link between commuting stress and the density of development or urban form. Novaco et. al. (1990) address this link by stating their finding that stress effects are strongly associated with freeway travel and with road exchanges; they also assen that freeway travel in southern California has become increas ing ly congested because roadway TRANSIT COOPERAnVE RESEARCH PROGRAM (TCRP) H-10


"The Costs of Sp!owi-ReviJited" capacity has not kept pace with continued growth. It has been alleged that commuting thro ugh th e aesthetically unattractive env ironments provi ded by the commercial strip development typical of sprawl produces more psych o l ogical stress on commuters than does commuting through environments dominated by trees and open space Very little literature pertaining to this allegation exists One study, however, claims to have tes t ed commuters psychologically and arrived at a finding that supports this claim (Ulrich e ta!. 1991). Other sections o f this report comment in more de t ail on the evidence regarding sprawl and travel time. No conclusion is made here. The professional literature suggests. however, that commuting can be shown statistically to contribute to stress-a happy coincidence o f sc i ence and common sense. L iterature Synthesi s M atrix No0.or -o;, s .. f:uiOI>tioJ J.a- o-. --I . condillion notob l y it.lf X b i l strongly linlc.d to tprowft X The C o sts and Be n efits o f S p rawl: A lleged N e gative Quali ty o f Life Im p a ct T o p i c: Higher Energy Consumption A llegation/B a sis Society consumes more scarce energy under sprawl, especially imported oil. Sprawl requires more travel overall and more of this travel is by energy-inefficient automobiles instead of more efficient modes of transit. RVTGUS IROOKINGS PAlSONS ECONORTHWEST 86 liotaroture Review Li t e r a ture Synt h es i s Ewing ( 1997) and many other researchers, contend that the ev i dence consisten tl y d em onstrates that automobile use, and hence energy use, is higher with sprawl. Yet, Gordon and Richardson ( 1997) are no t convinced that the link between vehicle miles of travel, energy use, and density is fumly established. Coloring this argument is the perspective on energy scarci t y and availability Gordon and Richardson speak of an energy glut and an OPEC cartel that has lost its c l out (Gord o n an d Richardson 1997. 97) while Ewing cautions that energy sources are not unlimited and reliance on foreign energy supplies is a continued con c em for the United Sta t es foreign po licy (Ewing 1997, 114). Literatur e Synthesi s M atrix c.:"' NoC.tor S vb;;!...tiol ...... ""'""' econdition .. tstl X I s it "ronetr X T he Costs a nd B enefits o f Spr awl: A ll ege d Negative Q uality of Life I mpact T opic: More Air Pollution All e g a tion/Basis Sprawl worsens the overall air pollution in a metropolilan area. Sprawl probably generates rnore vehicle miles of travel than other forms of development; so total vehicle emissions are much larger. Under many local climatic condi t ions. this can generate a greater total amount of air pollution even thoug h i t may resu l t in l es s intense local pollution than would TRANSIT CO OPERATIVE RESEARCH PROGRAM (ICRP) H-10 f


"The Com of Spruwi-RoviW.d" occur in some very high-density portions of more compact regions. Literature Synthesis Most, but f ar from all, observers agree that low-density settlements generate more total automotive travel than more compact settlements, other things being equal (see prior discussion); therefore, l ow density settlements generate more auto-oriented emissions per I 00,000 residents. However, the intensity of air pollution in each metropolitan area is affected by many other factors, such as location s of mar urban centers, prevailing winds, mountain barrie(S, t emperature inversions, and general climate. Hence there is substantive disagreement whether sprawl is a key factor detennining the degree of air pollution in each metropolitan area. Burchell in the New Jersey Impact Assessment of the State Development and Redevelopment Plan found that air pollution would be very similarly reduced i n both future alternative development scenarios (Burchell et a!. 1992a). Most of the reduction would be due to more stringent emission controls that would affect the entire motor vehic l e fleet of New J ersey as opposed to differing l ocations of wherein a region this fleet would be replaced In other words, development pattern, at leas t in this instance, did no t significantly influence air pollution levels. (The New Jersey Impact Assessment also considered effects on water pollution under trend and p l an conditions Plan was found to generate abo u t one third less water pollution than trend, although heavy metals in urban stormwater runoff were increased under the plan development scenario ) RUTGS.S 8a.OOKINGS PAR.SONS ECONOitTHWESl 87 l.iter it more difficull to preserve historically significant older structures than would othenvi s e be the ca s e. Sprawl encourages economically viable firms and households to leave inneHity neighborhoods by pennitting them to move to the s uburbs without pay i ng the full marginal costs of their doing so In particular, those who move are not required to compensate those who are excluded from suburban communities for losses the residents that remain behind suffer. Therefore, the economic ba.<>e supporting o l der structures of historical significance is weakened. Neighborhood conditions in the vicinity of such structures also worsen because of the i ncreased concentration of poverty and historic structures located there are consumed by these forces Literature Synt hes i s This allegation has been put forward mainly by the N a tional Trust for Historic Preservation in its various publications attacking sprdwl. The following argument (Beaumon t 1996) summarizes the reasoning behind the professed association between sprawl and preserva t ion Not many oth e r TRANSrr COOPfRADVf RfS fARCH PROGRAM /TCRP) fl. I 0 ,,


observers of sprawl have commented on whether or not !hi s association is valid Sprawl affects historic preservation in five major ways (Beaumont 1996, 264): Sprawl ad v e r sely affecls older downtown and neighborhoods where historic: buiJding.s are concentrated When lhc:: economic vitality of a historic area suffer s the buiJdings in it oflen become underused or empty. Over time. many of them are "demolished by' or torn down to make way for surface parking toes. Sprawl destroys community character and the countryside Cohesive Main Streets, o J d stone fences historic trees, country roads-these and other of the American landscape are rapidly being destroyed by sprawl dtwelopment and the vast of asphalt required co accommoda t e it. Spraw l r educes opportuni ties for face-to face interaction among people thereby making it more difficul t to create or ret ain a sense of community By scattering the clemen t s of a community ac. ross the landK'ape in a haphauud way, spraw l provides no town centers and reduces the sense of ownership-and therefore a lso t he commit ment-that people ha v e coward their community. SprawJ foreclose$ alternative s to the a u tomobile as a means of transport the rcby add ing to press ures to create o r widen roads that often result in the demolition of historic r esou,ces or the degradat ion of their Set t ing s Sprawl leaves o lder cities and towns with cxccs$ivc l y high concentration s of pOOr people witl) social p ,oblems. making these places a very difficult env i ronment in whi c h to rcvita.li7.C comm u ni t ic$ Literature Synthesis Matrix c.:... No Crtor ,_ .,;_, """---uo!.!"" noto uislt X l 1 it stro:9'Y X RUTGERS 8ROOICINGS PAISONS CONOI.lHWEST 88 literature Review T he Cos t s a nd Benefits of Sprawl: Alleged Positive Quality Of Life Impact Topic: Preference for Low-Density Living AllegationfBasis Many households prefer low-density residential living Many consumer preference surveys reveal that a key pan of the "American dream" is ownership of a detached, single -family home w ith attached private open space in the form of a backyard More important !han !he stated preference, however, is the revealed preference: for !he l ast 50 years, suburban development has been the primary form of metropolitan residential growth, and single family housing units have been the dominant residential form. Consumers clear l y c hoose low density suburban living given existing alternatives and prices. Most housing developers consistently build low density subdivisions because they are easier to market !han higher-density devel opments. If low-density, single-family, detached housi n g is what many people want, lower land prices at the urban fringe make it affordable to most people (Downs 1994) Literature Synthesis The suburbanization of population and jobs in the United States has been well documented. In 1950, almost 70 percent of the population of 168 metropolitan areas lived in central cities ; by 1 990 over 60 percent of !he population of 320 metropolitan areas lived in the suburbs and a majority of jobs in metropo litan areas were in tbe suburbs as well (Rusk 1993). The process of suburbanization lowered average population densities in urban areas Between 1950 and 1990, the number of residents in urbanized areas with population over one million in TRANSIT COOPER AliVE RESEARCH PROGRAM (TCRP) H-10


"The Costs of Sprrrwi-Rovisitod" 1990 increased 92 percent, while average population density decreased 44 percent (Wendell Cox Consultancy 1996). The fact that Americans choose to live in low density areas has been cited as strong evidence that Americans prefer that lifestyle. A preference for suburban living bas also been shown in other studies. The most recent annual survey by Fannie Mae ( 1996) shows that home ownership is a top priority for 69 percent of Americans, and that 73 percent desire a single-family detached house with a yard on all sides. Another study that generated quality-{)f life rankings for the fifty U .S. states over the period 1981-1990 found that sparsely populated, mountainous western states such as Montana and Wyoming had a higher quality of life ranking than more densely populated midwestern and eastern states (Gabriel et a!. 1996). Urban congestion has been cited by "lone eagles" (individuals wbo are able to Jive anywhere and teleconunute to work) as a factor influencing their decision to move (Salant et a!. 1996). The most recent Journal of the American Planning Association (Winter 1997) has two articles dealing with alternative views on sprawl that sununarize many of these arguments Gordon and Richardson (1997) revisit several issues relevant to the compact cities discussion, including residential density preferences. They maintain that consumers, given the choice between low density suburban living and high-density urban living, over whelmingly choose the former: "Bu t tha t suburbanizat ion itself should be an object of attack is amazing, given the expressed preferences of the majority of Americans for suburban lifestyles and tbe supposed sanctity of consumer sovereignty" (p. 99). Drawing on the literature, they attempt to dispel the belief that the choice RUTGERS 6ROOIONG$ PARSONS ECONOII'HWEST 89 li!eraturo Reviow for low-density residential living is a constrained choice, strongly influenced by government policies that promote suburbanization, including subsidized automobile use and zoning laws that restrict high-density development. Gordon and Richardson argue that more subsidies are given to public transit than to auto travel, and hence, government policies do not necessarily promote low density living over high-density living. In response to the argument that developers are prevented by zoning and land-use regulations from building at higher densities, Gordon and Richardson maintain that developers are just offering what the market demands: ''The risks of building an unacceptable product are very high, and builders are well aware of the strong consumer preference for the single-family detached home .... (p. 97). Though Ewing (1997) agrees with Gordon and Richardson (J 997) that the recent choice of U.S. households bas been for low-density suburban living over high-density urban living, he contends that given a larger set of residential living choices consumers do not necessarily favo r the former: ''There is strong consumer preference for new single-family detached housing a housing type in the suburbs. But most people could do without the rest of the suburban package" (Ewing 1997, Ill). Ewing maintains that compact development is capable of holding itS own in the marketplace. He cites evidence from the literature on consumer preferences. According to Ewing, the literature reveals sever.!! things: (I) the suburbs often rank below small towns villages, and rural settings; (2) home buyers, given a choice, are evenly divided between low and medium-to-high density residential settings ; (3) home buyers in high-priced housing markets often prefer small-lot houses; and (4) the TRANSIT COOPERATrVE RESEARCH PROGRAM (TCRP) H-10


public, given a choice, is a l most evenly divided between mixed and single use areas (Ewing 1997, Ill). In his earlier "Characte ri stics, Causes, and E ffects of Sprawl: A Lit erature Review" Ewing (1994) offers additional evidence to bolster his con t ention that consumer preference surveys do not clearly support low-density living over more compact forms of settlement. Surveys whe r e people are shown images of sprawl and traditional communities reveal that for the mos t part, the latter are f avored by wide margins (Neuman 1991, 74). Some surveys, however, have found tha t peop l e f avor homogene ous neighborhoods over mixed-use neighborhoods by a margin of about two to one (Bookou t 1992, 128), and that people prefer low-density suburban or exurban living (Audirac and Zifou 1989). Other surveys of consumer pr e ferences have also shown mixed resu l ts. A Sep t ember 1995 survey of people who shopped and ultimately bought in planned communities indicated that 57 percent of the respondents agreed w i th the statement "I'm tired of living in the steril e uniformity of most subur b s Yet, more than three-fourths of the respondents believed in the American dream of a big yard and a house set back from the street (Bradford 1996). There may be som e thing approaching universal agreement that U .S residential patterns in metropolitan areas have become increasingly suburbanized (i.e. have l ower density or sprawl) There is probably close to general agreement that many, if not a majority, of U.S. prefer single-family detached h ousin g g iven current options and prices albeit observers raise the issue whether ho u seholds would move in significant numbers to other options if they were available. RIJTGERS aROOKINGS PARSONS ECONORTHWEST 90 literol\l r e R.-view The question a bout whether spraw l is strongly linked to these residential choices is a matter of interpretation. At one extreme, the choice of low-density hous i ng is. i n ess en ce the definition of spraw l so the question of whether it is caused by sprawl is circular Another interpretation is that the mere existence of the pattern (spraw l ) and its accompanying low-density hous i ng influences peoples' preferences, like th e advertising of any product. Literature Syn th esis Matri x .. ;;,, Nr> a.(ll' I o.!!:'lollliol ..,_ ""'-' __.. :::;;:: X h it ttrongl'y l inbd to I aprowfl X The Costs a n d B enefi t s of S prawl : All eged P osi t ive Qua li ty O f Li fe Impa ct T opi c: Low Concentrations Associated Crime Allegation/Basi s ww-densiry development patterns have lower crime rates. H ouseho l ds move out of central citi es to escape the high rates of crime they have encountered there. Relatively high crime rates are statistical l y associated with very l ow-income areas especially within large cities. Such areas also often have much higher population d e nsities than the typical in spraw l development. L ite r a ture Synthes i s Statistics appear to indicate that urban residents expe ri ence higher rates of crime t han their suburban or rural counterparts. In 1994, the est ima t ed rate (per I ,000 TRANSIT COOPERAnVf RESEARCH PROGRAM (TCRP) H-10


"The Costs of Sprowi-Rtvisiltd" persons aged 12 and older) of personal victimization, which includes robbery, assault, rape, and personal theft, was highest for inhabitants of urban at 67.6. Suburban areas experienced a rate of personal victimization of 51.8 and rural areas had a rate of 39.8 (Pastore and Maguire 1996). 1995 crime statistics released by the Federal Bureau of Investigation (FBI) indicate that the Crime Index (comprised of selected violent and property offenses) was higher in Metropolitan Statistical Areas (MSAs) (5,761 per 100,000 inhabitants) than cities outside MSAs (5,315 per 100,000). Rural counties had the lowest index number of 2,083 per I 00,000 inhabitants (Federal Bureau oflnvestigation 1996). Research does not strongly indicate that the higher density living commonly found in urban areas is associated w ith higher crime rates Using 1974 census data, Newman and Kentworthy (1989a) correlate density with crime statistics for 26 major U.S. cities. Simple linear correlations suggest that there is no significant relationship between crime and density. Similarly correlational studies within the environmental psychology liter ature find no consistent relationship between population density and social pathologies (Sherrod and Cohen 1979). Several studies indicate that communities with high quality-of-life rankings exhibit low crime rates (Roback 1982, 1988; Rosca 1979). The amount of crime in a community may also affect migration patterns for both workers and firms. Salant et al. (1996) and von Reichert and Rudzitis (1992) found that crime was a factor that influenced individuals' decisions to migrate to a community that they perceived would provide a better quality of life. Results from the Salant et al. study also indicated that individuals were attracted to RliTGERS UOOKJNGS PARSONS ECONORTHWI:ST 91 literature Review locations that provide a safe place to live. A study using data from the 1983 Annual Housing Survey, however, found that few individuals moved to a particular neighborhood for greater safety (Spain 1988). The main reasons f or moving that survey respondents reported were t o find a less expensive place to live and to reduce their commuting times. A study by Gottlieb (1995) concludes that firms in the high-tech sector are less willing to locate in characterized by high levels of violent crime. Studies have found that perceptions of personal safety differ between residents of high-density urban areas and low density suburban areas. A 1995 nationwide telephone survey of over I ,400 adults attempted to discern how safe ind ividuals felt in their communities. When asked, "In the past year do you feel safer, not as safe or about the same on the streets in your neighborhood?" 14 percent of suburban residents felt less safe compared to 22 percent of urban residents. On the other hand, 12 percent of urban residents compared to 9 percent of suburban residents felt their safety bad increased over the past year (Pastore and Maguire 1 996). Through interviews Hummon (1990) determined that rural residents view danger as both an integral part of city life. and an indicator of social pr oblems Urban residents, however, considered crime and danger to be more a factor of socioeconomic conditions and location than an integral part of city life. Using surveys of low-income single parent women, Cook ( 1988) found that urban women were two times more likely than suburban women to indicate they felt u nsafe in their apartments and neighborhoods. Researchers within the criminal justice field conclude that perceptions of crime and security vary with site characteristics TRANSIT COOPERATM RESEARCH PROGRAM (TCRP) H-10


"The Com of and socioeconomic conditions and thus, fear of crime does not always accurately reflect actual crime rates. Instead, fear of crime is often derived from in complete knowledge of crime rates, observable evidence of disorder, and prejudices arising from neighborhood change (Skogan 1986) Other studies conclude that the direct effects of the physical environment on crime rates range from small to moderate (Taylor and Gott f redson 1986). Within the popular literature, there appears to be agreement that crime reduces a community's overall quality of life. Studies from popular literature commonly use crime as one measure of a community's quality of life . F or example, quality-of-life rankings of cities in the popu l ar literature, including the PU>ces Rated Almanac (Savageau and Boyer 1993) Money magazine's "Best Places to Live in America" (Fried et al. 1996), and Fortune magazine's "Bes t Cities: Where the Living is Easy" (Precourt and Faircloth 1996), all include some measure of crime as a component of a community's overall quality of life. In particular, Fried et al. found that a low crime rate is one of the top 10 quality-of life characteristics desired by Money magazine subscribers Selected crime statistics obtained from t h e Federal Bureau of Investigation (FBI) indicate that lower-density developments, such as suburban and rural areas, have lower crime rates than high-density urban areas. Empirical studies that have examined the relationship between crime and density however, have found mixed results-increased density does not necessarily result in higher crime rates The mixed results may be a factor of how individual studies define and measure crime and crime rates. There appears to be agreement that suburban residents RUTGERS U.OOKINGS PAASON S ECONORTHWfSJ 92 literature Review perceive themselves to be more safe than their urban counterpans. Altho ugh the literature appears to demo n strate, at best, correlation between density and crime, it does not demonstrate causality between sprawl and low crime rates. Studies have found that the effect of physi cal environment on crime rates ranges from minimal to m oderate and that crime is more a factor of socioeconomic conditions than density. An argument might be made that sprawl reduces crime rates in a round about way-sprawl is correlated with higher incomes whi ch, i n tum, are often correlated with spending more on home protection and public safety. This argument however, does not demonstrate that spraw l causes lower crime rates. Li terature Synthes i s Matrix c.::.ot s:.!. S..b;;:.,,;ot """'"' 01 (IOofldilion notablY e xi1tf X I it l i n d to JPtOwtf X The Costs and Be n efits o f Sprawl: A ll eged P o sitive Quality Of Life Impact Topic: Reduces Costs Of Goods And Services A ll egation/Bas i s Many lwuselwlds find the cost of public services and other good.! and services in suburban locations a better value. Retail development in suburban, l ower-density locations reduces the costs of many goods for consumers. For the public sector, suburban locations often provide better services (especially schools) for an equ i valent or lower tax burden For TRANSIT COOPERATIVE RESEARCH PROGRAM (rCRP) H-10 ,,


"'The Costs of private sector goods and services, particularly retail sales, the lower land values in suburban areas allow land intensive development formats, which include eltpansive ("big box") floor space and parking. The fonnats in tum attract high-volume, low-cost retailers The parking is necessary because of the type of goods being purchased (electronics, appliances, home improvements) and the vol umes of the purch ases (super sizes and quantities for super discounts) Literature Synthesis The alleged benefit for publi c services substantially overlaps the alleged benefits reviewed under the beading Social I ssues in this literature revi ew Two of the alleged benefits diseussed there are gennane here: The ability of jurisdictions to define a relatively homogeneous population with relatively similar service needs (which also provides opportunities for both economi e s of scal e and concentration, and for dropping services not needed by the homogeneous population (e.g . social services for lowincome households). The ability to have different tax levels and service qualities. There is an on-going professional debate about the institutional structures by which public services are most efficiently and fairly provided and a large body of literature on the subject. Not surprising l y, the poles of the deba t e are occupied by those who believe in the efficiency of markets, and those who believe such markets operate imperfectly without government intervention In the 1950s, Tiebout (1956) laid out the basic arguments for market c h oice (which, when applied to gov ernment is sometimes R.UTGER.S IROOtCIN_GS PAI.SONS ECONORTHWEST 93 literotvrelteview referred to as "pub li c choice"). He argued in favor of mul tiple small governments that allow households to "vote with their feet," choosing to live where the combination of public serv i ces, quality, and cost best met their preferences In contras t are those who argue (see, for example, Foster ( 1996) that typical market failures in the provision of public goods require larger units of government so that external costs can be internalized, increasing the odds tha t sufficient public goods will be provided. Arguments are made for the improvement of both efficiency and equity. Because this topic is being treated elsewhere in litis report, it is simply noted here that there is an elltensive literature arguing both positions There is certainly no agreement on litis subject, nor is any likely, since to come to a conclusion would require, among other thing s, agreement on two issues where people's opinions derive as much from underlying philosophies as from the results of social science: the proper scope of government intervention, and the tradeoffs between effi c iency and equity. On the private s ide, the re is ample anecdotal evidence that big box retailers make th eir money by high volumes on low margins, which for consumers means low cost. The growth of these retailers (e.g Wa l Mart. Horne Depot, Costco) is evidence of demand, which suggests that they are giving consumers more of what they want. Additional anecdo t al evidence suggests that many people who would oppose such retailers in their neighborhoods are some of the same ones who drive often substantial distances, to shop at these stores in other parts of a region. Tbe next question for this research, however is: To what extent are lowTRANSIT COOPERATIVE RESEARCH PROGRAM (TCRP/ H-J 0


"'!!!o Com ol Sprowi--Rovind" density development patterns essential for those cost savings? Recent work done to help evaluate the impacts of plan s for metropolitan Portland. Oregon (ECONonhwest 1996) sheds so me light on this issue. After quantifying vacant land supply, researchers conducted focus groups and work sessions with retail developers and brokers. The ir opinions were that to sa tisfy today s consumers most retail development had to a ceo mmodate the automobile, and that vacant, low-priced land in sizable parcels was critical to Jelail development that accommodated the automobile especially big box Jetail High density areas are likely to have higher land values, less vacant land smaller parcels, and existing residents to oppose the new Jetail developm ent. A few central cities have seen new discount Jetailing. In most cases, however, the development has occumd on underulilized i ndusaial p arcels whose zoning either defmes the retail uses as eompatible or makes variances easy to receive. In that case, low value land is still the primary factor allowing the development to proceed is JeasOnable evidence ro eonclude that people want are to be had at lower prices in lower-density pans of metropolitan areas. As with other effe cts. whether sprawl causes this eff ec t is a matter of inteJ}lJetation. On one band sprawl is the effect: the low-density retail pattern is what enables retailers to reduce prices. On the other hand, a pattern of sprawl may be causal if it implies more retail of the same type is desirable and allowable, and if it creates a pattern tha t allows more low-<:ost land to be developed more easily. Sprawl probably doe s both. lUTGEl $ 8lOOICJNGS 'AISONS ECOHOoT! Literature Synthes i s Matrix ..::.., .:.:. ... a... w.;;"'"' o..-DitOgn-nt 1=.:: . ,, X i ::_:o ""'' 9'Y X The Costs a n d Benefits or Sprawl: Alleged Positive Quality Of Life Impact T opic: Fos te rs Greater Economic Well Being Allegat.lon/Basls As an outcome of a free market, sprawl from decisions made by individual h o usehold and [inns to their welfare (as measuJed by utility or profit). By resaicting these individual choices, effons to limit spraw l will reduce the overall st andard of living. A central tenet of free-market economics is that individual households and finns act in ways to maximize their welfare and the result of these individual d ecisions is to maximize welfare for society as a whole. In this eon text. sprawl is considered to maximize welfare for society because it represents the outcome of individual cho ices by households and frrms about where to locate and bow 1 0 bu ild homes and businesses. Critics of free-market economics point out that decisions are based solely on the costs and benefits faced by the individual h ouseho l d or finn, and so do not consider the eosts or benefits to others that may result from their decision (the costs and benefits to others are referred to as extemolities) Critics of sprawl poinl out the negative externalities-traffic TRANSIT RESEARCH PROGRAM (rCRP) 11-lO


"'Tht Com o' SprowJ...Revisiftcl" congestion, increased public infrastructure costs, and accelerated development of fann land and open space, for example-and argue !hat these ellternalities reduce social welfare. Critics of sprawl often suggest policies to address the negative externalities of sprawl. It is !he debate over these policies !hat the alleged impact on economic well being is most often discussed. There is also extensive debate about !he level of negative Cllternalities; whether these externalities are caused by sprawl and the effectiveness of policies to address these externalities. This debate occasionally touches on whether !he policies will affect !he costs and benefits faced by individual households and firms. A primary concern is whether policies to limit sprawl will increase !he cos t of housing-this impact is address elsewhere in this study. An argument !hat is also occasionally brought up is whether policies to limit sprawl will in turn limit job growth in an area, and !hereby reduce income for area residents and limit economic development opportunities. These are the impacts that are focused on in this section. Literature Synthesis In the New Jersey impact assessment Burchell et al. ( 1992a ) found !hat New Jersey could accommodate simi lar magnitudes of population and employment growth under both trend and plan developmen t patterns. Distributional patterns would diffe r, however; plan development would direct more jobs to urban and rural centers and fewer to suburban areas !han trend (Burchell 1992a, 19). Sheppard (1988) relates sprawl to the economic well-being of residents. RUTGERS &ROOKING$ PARSON$ ECONORTliWEST 95 lil'erolvre Review Sheppard found !hat an increase in space available to a particular class of residents results in lower rents at all locations, increased "suburbanization" for all classes, and increased utility for all classes. Sheppard cautions the reader. however, that !he results consider neither externalities nor the public good associated with !he exercise of development controls. Most authors argue simply that s prawl must maximize welfare bceause it results from free-market decisions; Gordon and Richardson (1997, 99). The economic well-being benefits of sprawl are most often addressed in arguments against policies to limit sprawl. These arguments are based on !he considerable literature that shows !hat measures that increase density increase the cost of land. It is argued that an increase in density will reduce job growth and economic development opportunity by increasing !he cost or limiting the number of s ites available for commercial development, and by increasing the cost of housing which in tum will limit the supply of labor (ECONorthwest 1994). There is considerable evidence that measures to control growth cause the price of land to increase. Shilling et a!. ( 1 991) found that state land-use controls both restrict !he supply and i ncrease demand for residential land, driving up its price. Brueckner (1990) cites a large empirical literature documenting !he effects of growth controls on housing and land markets. His evidence conclusively establishes !hat growth controls raise housing prices in communities where !hey are established (Dow all and Landis 1982; Elliot 1981; Katz and Rosen 1987; Schwartz et al. 1981; Schwartz et al. 1989) Most of the literature that addresses the impact of growth controls on land prices TRANSIT COOPERATIVE RESEARCH PROGRAM {TCRP) Il-l 0

PAGE 100

foc.uses on the residential land market. There appear to be very few articles that address the impac t of sprawl, or measures to control sprawl, on conunercialland marketS. the level of employment growth, or wage income. While there are logical reasons to suspect that uninhibited growth fosters more employment and wage growth than lirnited growth, the literature does not document this at all. Literature Synthesis Matrix .. :.. Ho0.0. $\IO;,!.,..of -,__ condiliotl Motobi; e xillf X n il 51fol'jfy l'nked to X The Costs and Benefits of Sprawl: Alleged Negative Social Impact Topic: Fosters Suburban Exclusion Allegation/Basis Suburban exclusio!Ulry zoning increases the concentration of low-income households in cerrain neighborhoods. Most lowand moderate-income households cannot afford to Jive in suburbs where exclusionary zoning raises housing costs; thus, such households become disproportionately concentrated within central cities and older inner-ring suburbs. Hous ing in many parts of these c.onununities is generally older, smaller, more functio nally obsolete, less well maintained, and much Jess costly to occupy than housi ng in newer suburbs. Moreover. subsidized housing unitsespecially those in public housing projects-are heav ily concentrated within older neighborhoods in central cities and inner-ring suburbs, because residents of other areas-including most suburbs refuse to perrnit them within their RUTGERS BROOI
PAGE 101

'The C..,. ol Sprowi-Rtitittd" who occupy affordable housing in more suburban locations take on the employment characteristics, ambition levels, and success rates of the pop u lation of those jurisdictions (Wi sh and Eisdorfer 1996; Davis 1993; Fischer 1991). In New Je rsey, close to 1 5,000 affordable housing units have been built and occupied as a result oflegislation emanating from the series of Mt. Laurel, New Jersey Supreme Court cases in tha t state. Occupants of this housing are employed, doing we ll at local schools, and integrated withou t incident in neighborhoods they would not have had access to without these court decisions. In the Gautreaux and Special Mobility Program studies, residents moving from the central city to the s u burbs using housing vouchers have hig her rates of employment, higher salaries, and ch ildren with better school attendance and higher grades than families who did not choo s e to move. While the confounding i ssue of self -se lection is clearly presen t here, i.e the successful and ambitious families are the ones that opted to participate in the moves there is a growing body of lite rature tha t indicates tha t "p l ace" ma tt ers. T here is a "rub-off" effect of place wherein success patterns can be communicated by r esidents to newcomers who specifical l y wish to improve their current economic and social positions (Poismao and Botcin 1 993) Literature Synthesis Matrix ., N o d.,. ....... ...... .,....,. 1)'10w--nt CCH'Idition notoblv exilf X b it sl'fongl'y lilnbd to sprowfl X RUTGERS Ba.OOKINGS PARSONS ECONOR:THWfST 97 Literature Review The Costs and Benefits of Sprawl: Alleged Ne gative Social Impact Topi c: Fosters Spatial Mismatch Allegation/Basis The resu/Jin. g s patial mismatch" between where most new jobs are being created (far-out suburbs) and where many low skilled workers must live (inner-city neighborhoods) aggravates high rates of unemployment in those neighborhoods. The unlimited expansion of urbanized uses on the periphery of the metropolitan area permits many employer s to move to locations that are very far from inner-city neighborhoods Consequently, unemployed workers living in those neighborhoods cannot eilher readily learn abou t job opportunities in far-out locations or afford to commute to such jobs even if they learned abo u t and qualified f o r them. This aggravates both high rates of unemployment in i nner-ci t y neighborhoods and suburban shortages of unski ll ed labor. Literature Synthesis John Kain was one of the first to examine tbe evidence that a mismatch exists between lower skilled and otherw i se attainable jobs that are increasing in the suburbs and high levels of unemployment of residents who live in central ci.ties should be ab le to a c cess these jobs (!Cain 1 992). Spatial mismatch has also been examined by sociologi st John Kasarda (1990) and William Wilson (1987) and economists Keith lhlao f e ldt and David Sjoquist (1990). Although the original literature related the mismatch to black workers o f all ages, later studies focused on the spatial mismatch as it affected young black workers. Race as the causative agent is the main focus of inquiry 1RANSIT COOPERATIVE RESEARCH PROGRAM (TCRP) fi.IO

PAGE 102

throughout most of the studies mentioned above. In other studies by Bennett Harrison (1974a) and John Kasarda ( 1990), causes of the mismatch (which according to them may not be spatial) are extended to inadequate skills and education, and limited transportation or access to transportation. Findings on spatial mismatch, while not always consistent in unearthing a spatial component [see Bennett Harrison (1974a), David Ellwood (1986), and Jonathan Leonard (I 987)], are persistent in their specification of a mismatch of some type. The reality of "mismatch" is a population desiring to be employed in one location and available jobs going unfilled in another. Often, the unfilled jobs are lower order jobs that are not worth accessing by public ttansit if the prospective worker must also pay for child daycare services in order to access the job Other jobs similarly located in the suburb s may require skills that applicants, even after training, cannot meet. Or that casual workers available during the summer or during college breaks can easily meet wit bout training. This is such a complex area that sprawl versus more compact development patterns probably play only a small role here This will grow to be a big issue with significant consequences as workfare replaces welfare. The relationship between sprawl and cenlral city unemployment rates, the bottomline issue of the above discussion, is even more complex than relationships between sprawl and spatial mismatch. Literature Synthesis Matrix RUTGERS BROOKINGS PAASONS ECONORTHWST 98 literowre Review The Costs and Benefits of Sprawl: Alleged Negative Social Impact Topic: Fosters Residential Segregation Allegation/Basis Residential segregation by race and income is greater under sprawl than where less fragmented govef1Ulnce over land uses exists. Exclusionary zoning by many outlying suburban communities inhibits the consttuction of relatively lowcost housing for low-and moderate income households. This occurs because residents of each community control1and use decisions therein. They usually take into account only their own interests in making such decisions-not the interests of the region as a whole or of citizens in other parts of it. They have major economic motives for ttying to minimize the number of low-cost housing units within their own communities. These include maintaining hou sin g prices as high as possible and excluding households whose need for public services-especially schools-will cost the community more than the taxes these households will contribute to the community. Because blacks and Hispanics have m uch lower incomes, on average, than other major groups in American society, such income segregation is also an effective means o f achieving ethnic segregation in many areas. Literature Synthesis There is only partial agreement about this allegation, partly because it is difficult to fmd American metropolitan areas without fragmented governance over land uses, and partly because such a small number of them exist in the United States that statistical testing of conditions in them versus conditions elsewhere probably are not valid. TRANSIT COOPERATIVE RESEARCH PROGRAM (TCRP) fi.IO

PAGE 103

Yet coming at this issue from another direction, those states and regions that have made oven effons to provide affordable housing in locat ions where it has not before existed are achieving integration in those location s. In New Jersey, where a municipality must provide its fair share of affordable housing or lose its right to zone, racial and ethnic integration is taking place in what were predominantly white outer ring neighborhoods. New Jersey' s affordable housing progr3lll requires that those who fill munidpal quotas come from outside the municipality's boundaries but inside its commuting region. There are strict advertising and queuing requirements that ensure that minority households in central cities have an equal chance of occupying af fordable housing in the suburbs. With these kinds of mandateS, integration of neighborhoods moves quickly and directly (Wish and Eisdorf er 1996). Literature Synthesis Matrix The Costs and Ben efits of Sprawl : A ll eged Negative Social Impact Topic: Worsens City Fiscal Stress A llegation/Basis Central city governments become fiscally strapped or because they must provide costly services to large numbers of very poor households, while the properties owned, occupied, or patronized by such households produce relatively low tax revenue s. Low-incom e t.UTGERS lltOOIONGS PARSONS ECONORlliWEST 99 neighborhoods in particular have higher costs of crime and ftre prevention, street cleaning. and public health and welfare services than middleand upper-in come neighborhoods. Yet the former produce lower property and sales tax rev enues per capita than the latter. This forces city governments serving such communities either to raise taxes above those in surrounding commuruties o r to provide lower quality and quantities of key public services to their residents, or both. Literature Synthesis Only limited agreement exists on the extent to which sprawl is regarded as a major cause of fiscal stress. The concentration of very poor households within inner-city neighborhoods is not caused solely by suburban s praw l; other causal factors largely unrelated to the specific form of growth within a metropolitan areas also contribute t o this result Unfortunately, it is probably impossible to decide scientifically how to allocate "respons ibility" for this outcome among these causal factors; a fact which presents an obstacle to "proving" that sprawl contributes sigruficantly to this outcome The ability of households and employment to shift locations in a metropolitan area is virtually unrestrained. To the degree that househ olds and emp l oyers seek safer and more aesthetically pleasing locations, and these are found distant from the core, they will move there. And if taxes are lower or there are tax iocenti vcs to relocate, core to peripheral relocation will take place IJ high income residential and nonresidential properties are either footloose locationally or are being bid out of central locations to more distant location s. only those who choose not to be or are not bid-Qut will remain. These are often poorer households and TAANSIT COOPEI!I\TlVE RESEARCH PROGRAM (TCRP) H.IO

PAGE 104

"!h. Com ol Sprowi-Revisite
PAGE 105

Li terature Synth esis Matrix -.,.:.., s:!, ... St411ofl._., "'-1""!..""' ultlf X t. if llto"fly 10 owl! X T h e Cos t and Benefits of Sprawl : Alle ge d P os i tiv e Social Impact Topic : Fost ers Hom e Rule A llegation/Basis Sprowl keeps govemtMnt decisions about land use at the where individual citizens have much more chance of influe ncin g the results than they do where regional dec ision m aking predominates Because spraw l involves fragmentation of government powers among many relatively small localities, it keeps land -use decision making closer to the people most directly affected by it. Thi s satisfies the strong American desire for local soverei gnty. Like-minded citize ns can then pass zoning and other regulations that exclude types of development from their communities they do not like. This in tum allows them to prev ent "s ocially undesirable" influences in their neighbor hood s and scho ols Such negative influence s include potentially dangerous households with characteristics markedly differen t from their own, as well as region serv ing lan d uses with negative l ocal spillovers like airportS o r incinerator p l ants. Lite .rature Synthesis The literature dealing with the merit of borne rule praises its "small govemment" dem ocratic responsiveness, as is illus trated in the following quotati ons: Others came to suburbs for better schools. This bas been due. at least in port, to the R UTGU$ aa()()f(INGS PAUON S ECONO!tlHWUI I 0 I Uleroture Review responsiveness of these schools to parental expectations. rooted in tum in the smaller size of many suburban school districts. Indeed, in an ag_e primarily gjve.n over to Slate cenaaliutioo. tbe suburbs have encouraged a coontervailing decentralization governance, forcing a healthy kind of competitiveness onto local governments (Carlson 19%, 34-S). The tmld in many places has been for cities to incO
PAGE 106

'"The Costs: ol S,orow!:Revisifed" The Cost and Benefits of Sprawl: Alleged Positive Social Impact Topic: Enhanced Diversity and Choice Allegation/Basis: Sprawl provides citizens with a great variety of localities with differing tax levels, public service qualities, and housing costs, thereby increasing the range of choices available. The many individual localities in a metropolitan area function like suppliers of "bundles" of tax levels, public services, and local amenities in a market. Competition among them provides households with many more choices of living environments than would exist if all key fiscal and land-use decisions were made centtally and applied similarly throughout the metropolitan area. This process, fJtSt conceptualized by Charles Tiebout, ( 1956) is widely praised by economists for bringing many of the virtues of a free market to the public sector, thereby benefiting potential residents of suburban communities by expanding their choices. Literature Synthesis There is reasonable agreement that housing costs, public services (primarily education), tax levels, and housing stock/aesthetics of community form the bundle of goods that is bid for in community selection. Within a m etropolitan area citizens have significant choices of communities, and within a fragmented metropolitan area, they have even more choices. Those who "shop" for communities take all of the above into account before making a locational decision. Sprawl's conttibution to the above is the massive amount of reasonable alternatives (not best or worst) that it offers the Jocational consumer. Rut GElS a U.OOKINGS PARSONS ECONORTHWI:ST 102 literature Review Fragmented governments, primarily supporting residential housing, offer infmite variations of the bundles of housing, public services, and tax structure described above. Most of the variations found at the periphery of metropolitan areas are superior in housing value, school systems, property tax levels, and housing amenities to locations found closer in. As such, these are the locations most often sought; the closer-in communities, the locations most often left behind. The most significant variable appears to be housing cost and housing appreciation The combination of the two seems to be maximized in locations more distant from, as opposed to closer to, the urban core (Downs 1994). (See also Home Rule section.) Literature Synthesis Matrix o.:=..t Noa.ow-,_ """""'' d.l::io con 1 n notobly eAi"t X b illkongly '"k..d to sprowtf X 111ANSIT COOPERATIVE RESEARCH PROGRAM ITCRP) HlO ,,

PAGE 107


PAGE 108

Annotation of Key Studies I Public-Private Capital and Operating Costs 2. T ransportation and Travel Costs 3 Land/Natural Habitat Preservation 4. Quality of Li fe 5. Soc ia l Effects RUTGlS BI.OOKINGS PAlSONS ECONorth.,.st 104 TRANSIT COOPERAnVf RfSfARCH PROGRAM (TCRP) H-10

PAGE 109


PAGE 110

Altshu ler, Alan. April 1977. ''Review of The Costs of Sprawl," Joumal of the Ameriam Planning Association, VoL 33, No. 2. pp. 207 209. This is a short review article of the RERC study, published about two years after the study itself. For purposes of simplicity, Altshuler focuses on the two extreme cases analyzed by RERC-high-density housing and low-density single-family housing. He begins by summarizing the major findings of the RERC study. He then asks three questions: (I) Have the results of the theoretical analysis been calibrated against actual community experiences? (2) Does the report itself fully support the conclusions stated in its summary? and (3) Are the reported advantages of high density over low density mainly due to differences in density or to the fact that the high density community is more fully planned? His answer to all three questions: "No." One key issue he raises is: Does density per se affect the demand for community services? RERC explicitly assumed it did not, but he questions that assumption. Low-density areas often have no sidewalks, above-ground utility lines, and infrequent street lights, compared to high-density areas. There is also likely to be lower demand for professional security services. And RERC did not include any estimates of mass transit spending in its study though such spending would surely be higher in high density communities that relied more on mass transit. Therefore high-density settlements are likely to have higher community service costs than low-density ones, which would offset much of any savings in private costs. The savings for bigh density living versus low-density living are only $238 per year in operating costs, for a density rise from 3.5 to 19.0 units per acre plus more intensive planning. Four-fifths of J.UTG9.S BI.OOIONGS PARSONS ECONORTHWEST 106 Literature Rtview the savings are attributed to density alone He believes, given the omissions mentioned above, that this amount is so small it may vanish if the analysis is done correctly. He also makes the point that the dwelling units in the high-density settlement are 34 percent smaller than those in the low density settlement, and this accounts for a large part of the differences in capital costs and in energy costs. Also, the report examines energy components of settlements that account for only about 20 percent of urban energy consu mption omitting the other 80 percent Five-sixths of the heating and air conditioning savings from high density are attnbutable to smaller housing unit sizes. RERC also assumes that average travel per household in high density would be 9,891 miles versus 19,673 miles per household in low density. Only local trips would vary by density, but RERC attributes costs savings to the entire travel mileage-which is an error. According to Altshuler, correcting this error eliminates four fifths of the claimed savings in auto energy consumption. With proper analysis, the total energy savings of high density versus low density is about 3 percent, of which only I percent is attribu t able to density alone Another issue is whether higher residential density would lead to higher density in other types of land uses. He thinks not. Since he claims residential land uses account for only 30 percent of all urban land use, changing residential densities would leave most of the density of a region unaffected. Altshuler therefore claims that 'The case that low-density living is a highly expensive luxury remains to be made" (p. 209). TRANSIT COOPERATIVE RESEARCH PI!OGAAM (TCRP) H-10

PAGE 111

Even so, the author commends RERC for having put forth a systematic analysis that can serve as a starting point for other studies that do not contain this study's errors. Altshuler, Alan A., and Jose A. Gomez. Ibanez. 1993. Regu/QJum for Revenue: The Political &onomy of Land Use Exactions Washington, D.C.: 'The Brookings lostitut.ion. 175 pages. The central issues and themes of this text relate to govenunent-mandated exactions paid by real estate developers. Exactions may be in-kind or financial. The legal theory underlying development exactions is that governments, having reasonably detennined that certain public needs are "attributable" to new development, may require that their costs be "internalized as part of the development process. A key premise of the argument for exactions is that land development is a major cause of escalating local infrastructure demands and costs. The costs of growth are then studied in already built-up communities. Alternative estimates of revenues and expenditures for the city of San Francisco are discussed, as are approache s for allocating Montgomery County, Maryland, study expenses and revenues among county businesses and residents. In this text, the Real Estate Research Corporation's Costs of Sprawl study is critiqued According to Altshuler and Gomez-Ibanez, the principal problem with the RERC study is the meaning associated with the cost differences. They argue that the degree of variation between the quality of housing units from one community to another does not allow costs to be compared Therefore, if the same conditions are not replicated in later studies, community estimates will be difficult to compare meaningfully. RUTGERS BROOKINGS PAASONS ECONORTHWEST 107 Liletature RevMw American Farmland Trust. 1986 DensilyRelaled Public Costs. Northamp ton, MA: American Farmland Trust. This study estimates the net fiscal impact (the difference between pubic costs and revenues) of new residential development in Loudoun County and determines whether these impacts vary significantly with the density of development (from one unit per acre to 4.5 per acre). It found that public costs were highest and the net fiscal impacts the lowest for dwelling units at the lowest densities. American Farmland Trost. 1992. Does Farmland Protection Pay? The Cokt of Communily Services in Three Massachusetts Towns. Northampton, MA: American Farmland Trust. This report summarizes the American Farmland Trust's fmdings in a study of three Pioneer Valley, Connecticut towns (Agawan, Deerfield, and Gill). It is organized into four main sections: methodology, town reports, findings, and discussions. Five basic steps were taken for the cost of community services studies discussed in the repon: I) discussions with local sponsors to defme land use categories (residential, industrial, commercial, and farm/forest/open land); 2) data was collected for each town; 3) public revenues were reviewed and allocated by land use; 4) public expenditures were allocated by land use; 5) the data was analyzed and ratios were calculated The study found farmland to be greatly desirable--in contradistinction to other land use. TRANSIT COOPERATIVE RESEARCH PROGRAM (TCRP) H-10

PAGE 112

Buchanan, Shepard and Bruce Weber. 1982. "Growth and Residential Property Taxes: A Model for Estimating Dire<:t and lndind Population Impacts.'' umd Economics, Vol. 58, No. 3 (August): 324-337. This paper attempts to determine the extent to which population affects single family residential property taxes and the mechanisms through which such impacts are transmitted. This paper extends the population-local fiscal behavior model in several ways. For instance, it broadens the scope of such models. Whereas previous studies have geoerally focused on the determinants of local government expenditure behavior, this paper develops a model of the tax -le vying and assess ment behavior of local governments. Burchell, Robert W., et al. 1992&. Impact Assessment of 1M Jusey Interim State Develop-nt tuld Redevelopment Plan. Report II: Findings. Trenlon, NJ: New Jersey Oftic:e of State Planning. This analysis compared the impacts of development in New Jersey over the period 1990 to 20 I 0 under two develop ment scenarios-trend versus plan A series of models was developed to examine the relative effects. The analysis found that there was more than enough land statewide to accommodate the projected twenty-year development (1990-2010) of persons (520,000), households (431,000), and employees (654,000) under both traditional (TREND) and managed (PLAN) growth. Development between 1990 and 2010 under TREND would consume 292,079 acres, whereas PLAN would accommodate the same level of growth as TREND in terms of persons, households, and jobs indicated earlier yet RUTGERS laOOIONGS PARSONS ECONORTKWEST I 08 would consume only 117,607 acres-175,000 fewer acres than TREND (Burchell et al. 1992b). Thus, PLAN's overall land drawdown was 60 percent less than TREND. The impact assessment further found that managed growth would have the environmental advantages of preserving greater levels of fr:ail and agricultural lands. Reflecting historical rates of loss, under TREND 36,482 acres of fr:ail lands would be consumed for development. By contrast, under PLAN the consumption of these lands would drop to 7,150 acres, or 20 percent. Thus, managed growth in New Jersey could accommodate future development without spoiling more than 30,000 acres of frail environmental lands. In a similar vein, while 1990-20 I 0 development under TREND would consume 108,000 agricultural acres, under PLAN, 66,000 agricultural acres would be drawn down, representing a savings of 42,000, or 40 percent of prime agricultural land. Findings in this report also indicated that the state of New Jersey could save $1.3 billion in infrastructure costs for roads, utilities, and schools over a twenty-year period if a state plan managing growth were followed, as opposed to the sprawl patterns of development at that time. Other advantages of planned development were also indicated with respect to the environmental quality of life and so on. Burchell, Robert W., et al. 1. Economic Impacts of Trend Versus Vision Development in the Lexington MetropoliUur A"'a. 1995. 2. lmJHJCI of DELEP CCMP Vu:sus St4tus Quo on Twelve Municipalities in tlu DELEP Region 1995. 3. Fiuol Impacts of Allernative lAnd Development Pa&rns in Michig1111: T1u Com of eur...nt Develop-nt Versus CompDCI Growth. 1997. TRANSITCOOI'fRAIM' RESEARCH PROGAAM (TCRP)

PAGE 113

These three studies extend and broaden the application Burchell's New Jersey modeling of development alternatives (sprawl versus compact growth) to different geographic settings. This enables refinement and testing of the models under different taxing structures, means of providing and funding infrastrucrure, and geographic levels of investigation. Each of the srudies-Lexington, Kentucky, the Delaware Estuary, and the State of Michigan--looked at the land, infrastructure, housing, and ftscal costs of sprawl versus compact development Compact development was differently defined in each srudy; sprawl which WI>$ historical development. varied only marginally from place to place. There W&$ much more consistency in the definition of sprawl across these studies than there was in the alternative to sprawl. Findings from the Studies are as follows: SPRAWL FINDINGS OF MULTIPLE STUDII!S Land Land Roads Utilities Coscs Compact Om Sprawl Development (Lexington, KY and Delaware 18!1> 20!1> 14.8% 19.7% 6 7.2% 2.5-8.2% Compact Growth Ove< Sprawl Development of 17.4!1> 13.1!1> 12.4% 13.7% 13.7% Burchell, Robert W. el al. 1997 &ulh Corolintl Infrastructure Study. New Brunswick, NJ: Center for Urban Policy Research. This srudy involved a 20-year projection of infrastructure need in the State of South Carolina. It encompassed all public and quasi-public infrastructure required in the lUTGBS llOOIQNGS ,ARSONS fCOHOlliiWEST I 09 state including developmental (roads, bridges, water/sewer), educational, commerce, public safety, public heal t h recreational/culture and environmental. Twenty-eight individual categories of infrastrucrure were contained in the above seven groupings. Findings for the state were as follows: $56.7 bill ion in required infrastructure costs from 19952015 $16.7 billion in potential infrastructure savings due to technology, differing means of provision, and costs of sprawl savings. Much of the above savings carne from technology and differing means of provision as opposed to costs of sprawl savings. Three quarters of the remaining $40 billion ($2 billion per year for 20 years) or an average of $1.5 b illion annually could be raised via I 0 percent infrastrucrure set-asides in all state, county, municipal, and school district general fund and intergovernmental transfer revenues. The remaining $500 million annually could be taised from a variety of revenues including property tax, sales tax the tolling of roads, development impact fees. water/sewer charges and the like A 2 gasoline tax incr ease will raise only $56 million annually in revenues for infrastructure puposes. Dougbarty, Laurence, Sandra Tapella, and Gerald Sumner. 1975. MunU:ipal ServU:e Pricing: Impact on Fi#al Posilion. Santa Monica, CA: Rand. This report analyzes the impact of alternative municipal service pricing policies on urban structure and flllance. A pricing policy is basically a method of allocating the cost of a service to various components of the community (e.g., new residents, service users, the entire community). Primary attention is paid to pricing policies for the capital recovecy of services delivered to new development. TIU.NSIT COOPERAnVE RESEAAO! PROGRAM (TCRP)H-10

PAGE 114

The pricing stiUctures estimated in this report were developed from case examples in San Jose and Gilroy California. San Jose, a fairly large city thai bad already experienced much of its growth, and Gilroy, a city thai could yet undergo explosive growth, are prototypical of a large number of American cities. Downing, Paul. 1977. IAcal S.rvit:n Pril:ing and Tlulr Effects on Urban Spali41 Strverure. Vaucouver, Canada: University of British Columbia Press. This book is based on the argument thai wben urban users are charged an avernge price for services such as water supply, sanit.acy sewage, stoim sewage, and refuse disposal, the inner city effectively subsidizes tbe subu!bs and encourages urban '"'sprawl," or "leapfrog" development. Part of tbe argument or discussion suggests charging users the fuJI marginal costs of providing the services in question and thereby encourages a more efficient density and distribution of wban development. Duncan, James E., et. al. 1989. The SeJJN:h for Effieient Urban Growth Pattems. Tallahassee, FL: Department of Community Affairs. 56 pages. This study, conducted in Florida, encompassed detailed case studies of the actual costs (and revenues) incurred by several completed residential and nonresidential projects throughout the state. The projects were chosen as representative of five different development patterns ranging from '"scattered" to "compact t, RUTGERS IROOIONGS rARSONS ECONOtiiiWUT 110 DuPage County Development Depart ment. 1992./"'JHUtS of Development on DuPagt County Property Ttu:es. January. This study applies an econometric regression analysis to examine the relationship between types of devel opment in DuPage County from 1974 to 1989 and property tax levies over this time period. Property tax increases were associated with residential and non residential growth. ECONorthwest. 1994. Bvawalil>n of No Growth and Slow Growth Polit:ies for tlu Porlland Region. Portland, OR: Metropolitan Portlaud Government. This report is one of several prepared as part of Metro's (a Northwest-based planning agency) Region 2040-a regional planning process to evaluate future development panems and policies for the Northwest region. The report is to provide information pertinent to the question of whether the region could and should adopt policies to reduce the amount of growth anticipated over the next half century. Fodor, Eben. 1995. The Real Cost of Growth in Oregon. Eugene, OR: Energy & Environmental Planning Associates. Fodor estimates public infrastiUcture costs associated with the consiiUction of a "typical" single-family house The estimate includes costs for public facilities for sehoo l s, sewer, storm drainage, roads, water service, parks and recreation, and fire protection. A total capital expenditure of about $20,000 is estimated for a "typical" three-bedroom, single-family detached house on a 5,500square-foot Jot, in a "typical" Oregon city It is assumed that the house is part of a larger development or subdivision TRANSir COOPfiiATIVf RESEAJIC/f PROGRAM (TaP) 11-10

PAGE 115

1hal is located in ao urban area on previously undeveloped land with neart>y utilities. Frank, James. 1989. Tilt Costs of AlUmative Development Patterns: A R.,uw of tilt Lit.rtJ/un. Washiugton, D.C.: Urban Land Institute. This study reviewed the national lite!II!Ure cond ucted over roughl y a four-decade period coooeming development costs. Frank ordered the findings of the various reports and expressed them in equivalent dollar tenns (1987 dollars). He concluded from the national literature 1hal multiple factors affected development costs including density contiguity of development. di stance to central public facili ties (i.e., sewage and water plants), as well as other cbarac!Cristics, such as municipal improvement standards. In brief, capital costs are highest in situations of low density, sprawl, and for development located a considerable distance from central fllcilities. By contrast. costs can be dramatically reduced in situations of higher-density development 1hal i s centrally and contiguously located As described by Frank: When all capital costs an: touled . the total cost for low-density ... s pra wl ... is sligh t ly more than $35,000 per dweUio g unit. Further, i f that development is l ocated 10 miles from the sewage treatment plant. the central water source, the receiving body of wat er, and the major concentration of em ployment, almost $15,000 per dwelling unit is add ed to the cost, for a toul of $48,000 per dwe lling unit. . The cost can be reduced to less than $ 1 8,000 ... by choosing a central l ocation, using a mix of housmg typeS in which sing l e-family units constirute 30 percent of the total and apattments 70 percen t, and by p lanning conti guous development instead of leapfrogging. (Frank 1989, 39). T o the extent 1hal planned or managed growth fosters the more efficient patterns described above---<:entrally located, contiguous development 1hal includes units at somewhat higher density-it can a.c!Ueve infrastructure savings relative to tr.lditional developmeol Katz, Lawrence, aocl Kenneth Rosen. 1987 lnt.erjurisdlctional Elfecu of Growth Controls on Houslne Prices." Joumol of Law & &:onomks (April): 149160. KaiZ and Rosen analyzed 1,600 sales transactions of single-family b ouses during 1979 in 64 communities in the San Francisco Bay Area. Of these traosac tions, 179 involved houses that were located in communities where a building permit moratorium or binding rationing system was in effect. Findings show 1hal the price of houses sold in the growth controlled communities was higher than those sold in other communities without control s. Lade!, Helen, and William Wbeato11. 1991 "Causes and CGnaqueiiC
PAGE 116

lAgan, John, and Mark Schneider. 1981. "Suburban Municipal Expenditures: 1be FJretiS of Business Activity, Functional Responsibility, and Regional Context." Policy Studies Journal (9): 1039-SO. A wide variety of models exist seeking to explain variation io the level of suburban government activity. Alternative models stress concepts of local stratification and discrimination, the struc!Ure. of local decision making, ecological position, and public of which suggests varying hypotheses about which suburbs spend more and which spend Jess. In this paper, the relative strength of each of these alternative models is summarized and evaluated McKee, David and Gerald Smith. 197Z. "Enviroomental Diseconomies io Sub urban Expansion." TM American Journal of Economics and Sociology 31, 2: 181-88. Various types of urban sprawl are defined. Sprawl's causes are discussed with an analysis of effects on urban agglomerations, people, and the economy io general. The causes and effects of sprawl are part of a discussion on the development of suitab l e policy recommendations for dealing with problems in the area of suburban expansion Maryland-National Capital Park and Planning Commission and Montgomery County Planning Department. 1989. A Polky Vision: Ceniers and TnUJs. Silver Spring, MD: Comprehensive Growth Policy Study. This document summarizes how growth in Montgomery County will affect congestion, affordability, policy, and management lUTGBS llOOIONGS PARSONS ECONotTHWEST 112 MIUs, David. 1981. ''Growth Speculation and Sprawl in a Monocentric City.'' Joumal of Urban Economics (10): 201-226. Mills presents an economic theory of sprawl in a growing, rnonocentric city. Where decision makers have perfect knowledge, leapfrog development and discontinuous land rent functions may occur and be efficient io both an ex-post and ex-ante sense Where the extent of fu!Ure growth is uncertain, decision makers become speculators, and the spatial pattern of development is more complicated. Ex post inefficiency generally occurs. In the context of a formal monocentric city model, there are three land-use patterns that qualify as examples of spraw l. Leapfrog development occurs when a von Thunen ring of undeveloped land separates rings of developed land. This form of sprawl involves radical discontinuity. Scattered developrtU!nt, the second form of spraw I, occurs when there are annuli with both developed (homogeneously) and undeveloped land in them. Mixed development occurs when there are annuli with more than one developed use. Scattered development and mixed development forms of sprawl involve circumferential discontinuity. Mills gives theoretical explanations for each form of sprawl. Leapfrog de v elopment can be explained by interternporal planning on the part of decision makers who anticipate future growth with certainty. The essential idea here is similar to Ohls and Pines (1975), that i s, that land inside of the urban fringe is sometimes w i thheld from early development and preserved for more remunerative future options. Theoretical explanations for scattered and mixed development forms require that decision makers are uncertain about future growth and make speculative decisions. TRANSIT COOI'ftADV RESEARCH I'WGIIAM (TCRP) H-10

PAGE 117

Several criticisms of sprawl are cited and addressed with evidence generated from the monoceutric city model constructed for this analysis. Obis, James and David Pines. 1975. "Discontinuous Urban Development and Economic Efficiency." ljmd Economics (August): 224-234. Many observers argue that discontinuous development (land closer to Uiban centers is skipped over in favor of land further away) is inefficient for several reasons. First, it fails to make use of the most accessible land. Second. the expense of providing public services, such as roads and sewage systems, to new development is high. In contradiction, Obis and Pines argue that discontinuous development may be desirable and efficient in certain cases For instance, the development of retail and co!IIIDelCial services near the urban fringe must often wait for the maturation of critical scale. In rapidly expanding urban areas, there are contexts in which it may thus be efficient to skip over land for a period of time in order to reserve it for commercial uses after market scale increases. This strategy has been implemented in some planned communities. In Columbia, Maryland, for examp l e, the planners of this "new town" have explic itly reserved vacant land in residential areas for development of shopping clusters in the fu!lll'e-Mier increased residential den sities make such shopping enterprises economically feasible. Parsons, George. 1992. ..The Elfeds of C<>astaJ Land use RestricUons on Housing Prices.'' Journal of Environmental Economics and MtuUZgement. (Febroary): 25-27. RUTGW MOOKI"fGS P.AISONS ECOHOl11iWlSJ 113 In 1986, the Critical Areas Commission of the state of Maryland established a set of land-use restrictions thai limits residential development on land abutting the Chesapeake Bay In this paper Parsons estimates the effect of these restrictions on housing prices in one county on the Bay. Parsons uses a repeat sale analysis to infer the effect of restrictions by observing price changes on houses that sold both before and after the restrictions were enacted. He found thai housing prices in the Critical Areas w i th water frontage increased by 46-62 percent due to restrictions. Housing prices in the Critical Areas without frontage increased by 14-27 percent, and prices near but not in the Critical Areas increased by 13-21 percenL Peiser, Richard. 1984. "Does It Pay to Plan Suburban Growth?" American PIDnning Association (Autumn): 419-433. This article compares the economic results of planned community devel opment with those of unplanned development for a 7 ,500-acre site in Houston, Texas. The comparison indicates that planned development produces higher net benefits than unplanned development for all three cost areas investigated-land development costs, transportation costs, and social costs......utd that the difference in benefits is greatest in the area of transportation costs. The magnitude of the difference is small, however-on the order of I percent to 3 percent of total costs. Peiser, Rlcbard. 1989. "Density and Urban Sprawl." Land Economics, Vol. 65, No. 3 (August): 193-20 4. Peiser empirically tests several theoretical argwnents relaled to density and urban sprawl. Lot sizes and residential densities TAANSrT COOPERAJM' RESEARCH /'I!OGRAM (CCRP) H-10

PAGE 118

are examined over time along major growth corridors emanating from central Washington, D.C. and cen!ral Dallas. Tilis discussion suggests that lots should be smaller (i.e. densities should increase) the later they are developed after controlling for house size and distance from the central business district (CBD). The discussion notes that if higher densities closer 10 the CBD are desired, then cities should avoid policies that require sequential development. Furthermore, they should let the land market seek its narural level of densities, at least within the limits that existing road, utilities, and other infrastructure will support. The findings also assume that a competitive land market will achieve the desired result of higher density succeeding Uiban sprawl, that is, discontinuous development foUowed by l ater inflll. Propst, Luther, and Mary Scbmld. 1993. The Fiscal an4 Economic l"'JJ(Uts of l..oC
PAGE 119

Cou Analysis of Growth King Counly, Washington. Tischler & Asso dates, lne.: Bethesda, Maryland. This study, coaunissioned by the King County Growth Management Pl81llliog Council (GMPC), evaluated options for future development, focusing on projected public costs for roads, transit, wa1er and sewage utilities, and general government. It identifies marginal costs of new development for each of five ai!Cmalive land-use scenarios as well as revenues. The study found thai the altemalives with greater development in the urban centers and cities are more fiscally beneficial when roads, utilities, and general fund activities are analyzed However, this did not consider the impacts of transit, for which more analysis is needed. Windsor, Duane. 1979. "A Critique of The Costs of Sprawl," Journal of the Anwoiean Planning Associmion, Volume 45, No. 2 (July), pp. 279-292. This article is a critical review of the RERC study on sprawl. The first part of the article summarizes the fmdings from the RERC study. The author then criticizes RERC for not disentangling density from other factors. One criticism is that RERC uses different average-sized units for different densities, with smaller units for higher densities. Because smaller units have lower areas, they are less costly to build. This is a major reason why low-density, single-family units are considered more costly than higher-density units. It also affects utility costs. Total floor area is 44% lower in high-rise than single-family neighbor hoods. Windsor recomputes, costs as suming all housing units are the same size of 1 ,200 square feet. This greatly reduces the cost advantage of high density, though it still e:tists. However, it is believed thai this is equally unrealislic. In RUTGERS U()()t(INGS PAW)NS ECONOtKWm 115 reality, higher-i!ensity units are indeed smaller, on the average, so RERC is not entirely wrong. Probably the only way to avoid this problem is to do it both ways and compare results. Another criticism is thai RERC assumes structure costs are highest for single family homes, with high-rises in the middle, and walk-ups the lowest. Windsor believes high-rise costs should be the highest per square foot. RERC also assumes that developers will have to contribute more land to the public sector under single-family settlements than under high-rise settlements. RERC assumes thai clustering in higher density pa1terns results in the saving of vacant land. This is a source of savings, RERC claims. A given amount of population consumes more land under low-density than high density. But if the model assumes a given amount of land is fully developed at different densities, then different total populations are accomm<>dated. This is another case where more than one alternative must be analyzed The author criticizes RERC for "the underlying assumption thai cost mini mization is an appropriate principle for the planning and coordination of development patterns. Cost minimization is not a planning principle unless benefits are constant .... [RERq ignores the benefit side and is thus not necessarily an example of improving economic efficiency." Yet the author thinks it is not necessary to measure benefits. He claims the prevalence of low-density settlements reflects the benefit side because thai is what consumer have wanted. "Consumers choose to Jive at high densities only where land costs are very high, as in central cities." But where land costs are lower, as in suburbs, they prefer low density YRANSrr COOI'fRADVE RfSWCH PROGRAM (TOIP) K-10

PAGE 120

The author claims that suburbs resist high density on several grounds, not just costS "Voters are opposed to rapid population growth, the possible characteristics of new residents, the fiscal implications, and the Joss of suburban amenities lilce open space, semi-rural ambience, low densities, etc. E xclusionary land use controls are intended. in part to force low-density development; they function as a form of growth managemenc" (p. 291) By ignoring these facetS, says the author, RERC "does not properly evalua!C the relative efficiency of alternative develop m ent patterns." (p. 291) York, Marie. 1989. Eru:olll"tlging Compaet!Hvelopmenl in Florida. 311 pages. York analyzes three areas: growth management programs from around the country (Maine, Massachusetts, New Jersey, Oregon, Vermont); innovative strategies by region; and problems with redevelopmenc The analysis of the state wide programs consistS of descriptive summaries of state policies designed to address increasing growth management problems and the situations that prompted the development of these policies The innovative strategies section of this document examines urban growth boundaries (UGBs)-proactive growth management tools used to contain, control, direct, or phase growth to promote more compact, contiguous urban development. Basic to this strategy is the delineation of perimeter around urban development areas, effective for a specified period of time, within which urban densities are normally encouraged, while outSide the perimeter urban uses and densities are discouraged. A secondary analysis of several state studies is effected to identify the impact of UGSs on land prices However, the data are neither complete or consistent enough to draw firm conclusions concerning land price impactS. TRANSIT RESEAIIOI PROGRAM (ll:RP/ H.IO

PAGE 121


PAGE 122

1. CHANGES IN AUTOMOBILE TRAVEL These descriptive, aggregate studies reach inconsistent conclusi ons on whether com muting times have become, on average, longer or sboner. The studies do show thai commuting travel times have been fairly stable, even though conunuting distances have often increased. In addition, the studies show an increase in the amount of travel for non-work purposes. The authors who fmd evidence of stable or shoner commuting times suggest that this is a result of more decentralized employment and residences that allow people to travel on less congested highways. Likewise, authors attribute increase in non-wort< travel to more dispersed shopping opponunities. Although it is true that dispersion of jobs and residences bas been the IU!e during the periods studied, these descriptive studies d o not provide clear evidence that changes note
PAGE 123

Rkbardsoa, H W. and P Gordon. 1989. "Counting Nonwork Trips: 1'be Mlsslng Link In Transpor1al.lon Land Use, and Urban Policy." Urban umd pp. Ooe n:oent phenomenon bas been the growth in non -work travel, both during and off-peale. Using dala from the Nationwide Personal Transportation Srudy for 1977 and 1983, the authors find thai the numbers Of nOnWOrk trips increased three co four times fasler than work trips io all sizes of SMSAs. Non work travel even increased f aster than work travel io the peale periods They contend thai subwbanization, especially in the largest metropolitan was a principal cause of the inaease in nonwork travel although acknowledging thai demographic and work force changes are probably also involved. Subwbanization of businesses means the suburbanites have more close-by shopping and recreational opportunities and, therefore, may make more trips to satisfy immediate needs rather than wait until they bave a list of needs. The study, however, does not demonstrate either that shopping and recreational opportunities have increased io suburbia or thai households take more trips because of such an increase. Nor does the study rule out the effects of other factors such as rising incomes, greater participation of women io the workforce, and changes in leisure activities on nonwork travel choices Rossetti, M. A., aDd B S. E-.enolo. 1993. JounreJ /o W ork Trends ill tho Unlud SullO and Its MQjor Metropqliuut AHare about whether developiog more densely will reduce automobile dependency. Recem studies control for average income levels of the areas studied, but few srud.ies use d.isaggregate analysis to control fo r household iocome, number of children, number of workers, and other factors thai ioflueoce travel. TH/str COOPfV.JJVE IIESEAIICH PROGRAM

PAGE 124

Studies on density take a variety of forms: simulations for hypothetical pat terns of growth for specific regions, simulations of alternate patterns of growth for specific regions, and empirical analysis of regions, activity centers, and neighborhoods. Empirical studies tend to be cross-sectional, showing relationships between density and travel behavior, but unable to show bow changes in density would change travel behavior . 2.1 Simulations Simulations may be of entirely hypo thetical places or of specific regions. Many regions have evaluated the transportation outcomes of potential wban forms using their travel demand models. Only one, using one of the most advanced travel demand models, is described here. For a review, see R. Cervero and S. Seskin, "An Evaluation of the Relationship Between Transit and Urban Form," Transit Cooperative Research Program Research Results Digest, No. 7 (June 1995). Downs, A. tm. Stuck in Traffu;: Coping with Peak Hour Congestion. Washington, D.C.: The Brookings Institution. Downs developed a hypothetical urban area model to test how mucb changes in the location and density of development would change average commuting dis tances. The basic model uses values for the proportion of jobs in the CBD, central city, and suburbs and e"urbs and com muting distances similar t o the averages for large metropolitan areas. Different densities are created by varying the size of the suburbs and exurbs (and adjusting the proportion of population and jobs in each area, as needed to ma!Ch the size). The density of growth at the wban fringe has a significant impact on commuting distances, with a move from very low to lii!GERS BaOOI(JNGS PARSONS EOONOlmWEST 120 medium densities having the greatest impact. Increasing e"urban densities from 886 persons per square mile to 2,800 reduces commuting distances by 8 per cent. An increase from 886 persons to 4,363 persons per square miles decreases commuting trip lengths by 14 percent. Beyond that, large increases in density only shorten trips by a small amount. Metro. 1994. Region 2040 Recommenthd Allemative Dttision Kit. Portland, OR. This analysis of alternative urban forms of growth for the Portland, Oregon metropolitan area showed that more concentrated development, in conjunction with expansion of transit service, would reduce vehicle miles of travel and use of the automobile. This study used one of the most advanced travel demand models in the United States to simulate the trans portation outcomes. The study considered a base case that continues current devel opment patterns. This alternative requires that the urban area expand by more than half of its current size over the next SO years. The study also tested three different scenarios that concentrate various amounts of growth in transit corridors, centers, or in neighboring cities. "Grow ing Out" builds a larger share of single family housing than the region has 8l present, with more than one-fourth of growth outside the current urban growth boundary. ''Growing Up" keeps all urban growth inside the urban growth boundary by increasing densities and building a higher share of multifamily housing. "Neighboring Cities" moves about one third of expected growth to cities within commuting distance of the urban area. The most concentrated development of the Growing Up scenario produces the highest transit use (6 percent of all trips) and greatest reduction in VMf over base case levels (16.7 percent). The more TRANSrT COOPfl!ATIVf RESfARCII PROGRAM (TCRP/11-10

PAGE 125

dispersed patterns have lower levels of congestion, however. Like other simulations, this study had limited ability to change travel behavior because much of the built environment for the next 50 years is already in place. Some of the proposed changes in the way the regions develop also may nD! be politically or economically feasible. 2.2 Empirical Studies Cervero, R. 1991. "Land Use and Travel at Suburban Activity Centers.'' Transpor141Wn Qruuterly VoL 45, No. 4. pp. 479-491. This repon analyzes the effects of den sity, land use mix, and parking character istics on commuting behavior in suburban activity centers. 1be study uses data on 83 randomly selected buildings in six suburban activity centers. 1be data was collected as pan of the National Cooperative Highway Research Program project called Travel Characteristics of LArge-Scale Suburban Activiry Centers. The strongest relationship was between density (measured as height of building) and transit use. Having retail in the building has only modest effects on mode choice, primarily increasing transit and walking mode shares. Parking supply also has little effect, probably because most of the office building have generous supplies. Because the analysis uses buildings as the unit of analysis, it fails to consider center characteristics, such as distances between buildings and oppor tunities to shop and conduct personal business at other locations within the center. RUTGERS llOOtCINGS PAASONS 121 Dunphy, R. T., and K. M. Fisher. 1994. "I'ransportatlon, Congestion and Densi ty: New Insights." Paper presented at tbe Transportation Research Board 73rd Annual Meeting, Washington, D.C. (Jan uary). Using data on urbanized areas from Highway Slatistics, 1990, the authors investigate the relationships (using graphs) between density and vehicle miles of travel and travel use. They find some correlation between urbanized area population density and transit use (26 percent). but little between vehicle miles of travel and density (8 percent). Then, using data from the 1990 Nationwide Personal Transponation Survey, they find that people in denser areas make nearly the same number of daily trips as people at lower densities, but drive less. At most densities the average number of person trips per day is just below 4.0; only at 30,000 persons/square mile or more do trip numbers dip to 3 4 trips per day. However, the average number of trips by car drops from about 3.5 at densities below 30,000 persons/square mile to 1.9 at 30,000 persons/square mile People living at lower densities drive less because both transit and walking/biking become more viable options. The authors, however, also fmd a strong correlation between density and life cycle stage. They contend that demo graphics, rather than density, may be a primary reason for differences in travel behavior. This study is descriptive, suggesting relationships that need further analysis with multi-variate techniques to son out the relative effects of household charac teristics versus land-use density. The data analyzed in this study are aggregate, comparing whole regions rather than the specific places within regions where people live and work. JRANSrr COOPEIIADVE RESEARCH PROGRAM (TCRP) H-10

PAGE 126

Dzurik, Andrew. 1993. "l'ransportatlon Costs of Urban Sprawl: A Review of the Literature." Stau Tramporlll/Wn Polky lnitialiYe. Center for Urban Transporta tion November. This article cites nine important studies dealing with lranSpOrtation costs and sprawl. These include the classic RERC study, reviews by Altshuler and Windsor, and several others. The first part of the article cites a 1960s study of sprawl-"'The Nature and Economics of Urban Sprawl" by Harvey and Clark (1965). That study defines three types of sprawl: low-density development, ribbon development, and leapfrog development. Transportation is seen as a catalyst for urban sprawl. John Kain later argues that any savings from developing high density areas may be offset by higher construction costs per unit. RERC's report is cited extensively. It estimated that a low-density sprawl community would require more than six times the amount of minor streets in a planned high-density community. But only road length costs are considered as direct costs in the analysis of trans portation variations among these com munities. Two indirect costs are consid ered: travel time and air pollution. The report assumes twice as much VMf in a l ow-density community, and that causes the differences in such costs. This article then cites Altshuler's criticisms of the RERC report. Altshuler's book The Urban Transportaticn System (1979) argues that the American public has strong preferences for auto transportation and low-density settlements. Therefore, Americans will refuse to live in densities high enough to bring about any changes in the problems associated with sprawl, which be believes have been exaggerated anyway Bowler did a 1977 study showing that "user operated transportation" accounted for RUTGERS atOOKJNGS PARSONS ECONOlTHWEST t22 about one-seventh of consumer spending and was roughly constant from 1950 through 1973. He argues that suburban living results in higher use of energy and land resources for transportation than higher-density living would. When the urban environment is modeled as polycentric, then it is no longer true that a rising peroentage of suburban dwellers increases travel distances for jowneys to work, since work places also become decentralized. Yet many models assume that rising commuting costs are a major transportation cost o f suburbani zation. Gordon also argues that because work trips are declining as a peroentage of all trips, accessibility to workplaces is falling as a motive for choosing bolb places to work and places to live. Gordon and Richardson argue that decentraliza tion is an antidote to traffic congestion by scattering both origin and destination points and making suburb-to-suburb trips shorter than any other types. BART failed to replace the auto as the preferred means of commuting in the Bay Area, in spite of its enormous cost. Where light rail systems have been created, cities experienced small gains in public transit ridership over pure bus systems, plus major cost increases. Light rail tended to replace bus travel more than auto travel. The argument over compact development is discussed. One advantage it is sup posed to have is the use of excess capacity in existing i nfrastructures, rather than the need to build new infra. structures. This is not always the case but it was a major source of the econo mies found in the New Jersey sprawl studies. There is a discussion of bow much subsidy from local governments goes into highways and mass transit. In Mil waukee, Wisconsin, the local burden of TRANSIT COOPERAnvt RESfAIICH I'IIOGAAM (lOP) fl. I 0

PAGE 127

highway costs equals 59 percent of the local property tax levy. User fees do not pay the entire cost of auto travel, and that causes more than would otherwise occur. This study claims that the transportation costs associated with urban sprawl have not been studied in quantitative terms. Therefore, most questions about this issue are unanswered. The article cites 12 unfavorable views of sprawl's transporta tion costs in different articles and studies, with one -sentence summaries of their major complaints. Nothing very new is stated in this analysis. This article contains almost no quantita tive analysis, cites studies expressing all types of viewpoints-many contradictory -and arrives at no conclusions. It is not very useful, though it does mention some of the key issues involved. FraJik, L. D., and G. Plvo. 1994. Rellltio11Ships Betwun Liuul Use and Travel Behavwr in the hget Sound Regwn. Washington State Transportadon Center. Prepared for the Washington State Transportation Commission, Seat tie, Washington. September. The authors use data from the 1989 transportation panel survey for the central Puget Sound region, along with bouse hold characteristics from the census, employment data from the state employ ment agency, and land use data from the county assessor to identify the factors affecting travel behavior. They fmd that density, mix, and jobs/housing balance are all related to travel behavior, with employment density and jobs/housing balance having the strongest relation ships. At higher densities, trips are sboner but take more time. More trips are by alternatives to the single-occupant vehicle. As land use mix increases, trip distances times, and shares RVTGUS llOO<''GS PARSONS ECONORTHWEST 123 decrease. As jobs and housing are more balanced. trips distance and travel time go down. The relationships between density and mode split are not linear. The authors identify thresholds at which there is a substantial increase in transit use. These are 50-75 employees and 9-13 persons per gross acre for work trips, and 75 employees and 18 persons per gross acre for shopping trips. The use of carpool ing, however, seems unrelated to urban densities or other land use attributes. The study controls for household character istics, such as income and vehicle avail ability. Gordon, P., A. Kumar, and H. W. Rich ardson. 1989. '"lbe Innuence or Metro politan Spatial Structure on Commuting Times." }ounud of Urban Economia, Vol. 26. pp. 138-151. The authors combined data on residential and employment densities (residents or workers per acre of land zoned for that purpose) for 82 SMSAs from twelve states (from the U.S. Geological Survey LANDSAT file) with census data to identify factors that influence commuting times by auto and transit Lower resi dential densities are associated with shoner commuting times either by car or transit. For auto trips, concentration of industrial employment leads to shoner times, whereas concentration of commer cial employment increases trip times. The clustering of manufacturing produces economies in driving, but the clustering of commercial activities (such as in the CBD) produces congestion that reduces times. Other variables (land area, income, economic structure) have the expected signs, and the equations are fairly robust, explaining 61 to 87 percent of the varia bility in mean uavel times. Tbe authors conclude that polycentric or dispersed spatial structures reduce commuting times. !RANSIT COOI'fRAnV RESEARCH PltOGAAM (TCRP) fl. I 0

PAGE 128

The authors' use of SMSAs as the unit of analysis raises questions about what density means. No SMSA has uniform density througoout. Perhaps lower re gional density is a proxy for age of development, city siz.e, or some other factor that influences transit use. Holtzclaw, J. 1990. Explaining Urban DeiiSity and Transit lmJl"IS on Auto Use. Paper presented to the State of California Energy Resources Conservation and Development Commission by Natural Resources Defense Council and the Sierra Club. April 19. Holtzclaw compared the annual vehicle miles of travel of five communities with various densities in San Francisco Bay Area to test whether higher residential densities combined with better transit service and neighborhood shopping result in less driving. The study found that doubling residential density reduced annual vehicle miles by 20 to 30 percent. Better transit access also reduced vehicle travel. This was a cross-sectional study that demonstrated only correlation between density and vehicle miles of travel. It did not show that increasing density in a particular neighborhood would reduce vehicle miles of travel. The study did not control for income levels or other charac teristics of households that influence vehicle miles of travel. Holtzclaw, J. 1994. Using Residential Patterns and Trtmsil to Decrease Auto Dependenc e and Costs. San Francisco, CA: Natural Resources Defense Council. Holtzclaw used smog check odometer readings for 28 communities in Sao Fran cisco, Los Angeles, San Diego, and Sacramento with at least 20,000 residents each to evaluate the relationship between l
PAGE 129

From this the authors conclude that more compact cities would reduce automobile use. Reviewers have questioned the validity of using gasoline consumption as the measwe of automobile dependence, noting that many factors, such as gas prices and fleet characteristics, influence gasoline consumption. The analysis of automobile dependence also fails to make full use of the data collected, using only a single variable-urban density-to ex plain automobile use when other factors are clearly involved. This may overstate the role of density. Parsons Brinckerhoff Quade and Doug las, R. Cervero, Howard/Stein-Hudson Associates, and J. Zupan. 1996. "Com muter and Ugbt Rail Transit Corridors: The Land Use Connection." In Tr-ansiJ and UrlNua Form, VoL 1. Transit Coop erative Research Program, Transporta lion Research Board, Washington, D.C. October. This study updates the work of Push karev and Zupan by analyzing the effects of residential densities and CBD employ men t sizes and densities on light rail and commuter rail hoardings. The data is from eleven light rail cities in the United States with nineteen Jines and six commuter rail cities with forty-seven lines Boardings and transit service char acteristics data were provided by transit agencies Employment and populalion characteristics are from the 1990 Census. The data is used to develop models of light rail and corrunuter rail hoardings and costs. The empirical results are then used to estimate hoardings and costs for hypo thetical light rail and commuter rail corridors. Residential densities have a significant influence on light rail hoardings. A 1 0 percent increase in residential density tllrGElS llOOICINGS PAlSONS ECONORmWm 125 within two miles of stations increases station area hoardings by 5.9 percent, holding constant other factors affecting ridership, such as income. Residential densities matter less for corrunuter rail because it is a high fare mode, and many of its high-income riders come from low density suburban areas some distance from the city center Both the size and density of the CBD influence light rail ridership. A 10 percent increase in CBD employment density raises light rail hoardings per station by about 4.0 percent, holding constant the number of CBD employees, the residential density of stations and other factors affecting ridership. For commuter rail, a 10 percent increase in CBD employment densities increases station hoardings outside the CBD by 7 1 percent. The study concludes that light rail is most cost-effective and efficient in cities with larger CBDs and denser corridors. Commuter rail works best with dense CBDs. Other factors within the control of transit agencies, such as the availability of feeder bus service and park-and-ride lots, also influence ridership and costs. Pusbkarev, and J, Zupan. 1977. Transpo11111i4n an4 lAnd Use Policy. Bloomington, IN: Indiana University l'N$$. From census data, the authors estimated the effects of population density on transit use using areawide populalion densities and transit use data from I OS urbanized areas. Population density explained 55 percent of the variation in transit use in 1960 and 66 percent in 1970. They also estimated the effects of residential den sity, downtown floor space, and the presence or absence of rail transit for 27 urbanized areas. This increased the explanatory power of the equations, but the new variables were more significant than residential density in explaining TRANSrrCOOPERAVE IIESEARCH PROGRAM (TCRP) H-10

PAGE 130

transit use. They attribute this result to greater variability in office floor space than in residential densities among the areas studied 3. UNLIMITEDOUTWARD EXPANSION Although sprawl is characterized by unlimited outward expansion and leap frog development, there are few studies that specifically address the transportation effects of this son of development. The authors in Pan I who attributed stable commuting times and increasing non work travel to dispersed activities within metropolitan areas seem to assume unJim.. ited expansion. The simulations dis cussed in Pan 2 generally include an option of mlimited outward expansion at lower densities. They generally fmd that this type of development encourages greater use of the automobile but may be less congested than denser development. The following study finds that bouse holds who move outside a contained urban area generally have longer com mutes, measured in time, than their counterpam in the suburbs. Davis, J 1993 ''The Commuting or Exurban Residents.'' Urban Geography, Vol. 14, No. 1. pp. 7-29. A comparison of the commuting times of workers who bought home in the suburbs and exurbs of Ponland, Oregon, shows that the average exurban home buyer bad a commuting trip six to seven minutes longer than his counterpan in suburbia, controlling for occupation, income, and other household and job characteristics Data are from a survey conducted by the author of about 750 households that bought and occupied homes in 1987. Although some exurban households bad commutes similar to those of suburban RUTGERS IROOKINGS ECONOITHWEST 126 households the average exurbanite ap peared to trade off longe r travel times for more space, a rural environment, lower housing prices or better places to raise their children. At least in this metropolitan area. where there is an urban growth boundary to limit the outward expansion of the urban area, most people who move to rural residences outside this boundary spend more time commuting. However, exurban residents seem to sort themselves out so that those who live close to the urban area have central city and suburban jobs, whereas those who live farthest out most likely work in exurban towns 4. SPATIAL SEGREGATION OF USES (LAND-USE MIX AND URBAN DESIGN) Land-use mix and urban designs that encourage walking increase local accessi bility. That is, they offer more options for places to go that are close to borne or work. Theoretically, this could shorten trips and encourage use of nonautomotive modes of travel, or increase the number of trips made. Employment center studies compare a large number of centers while neighborhood studies nearly always com pare selected neighborhoods. Most of the neighborhood studies were done in Cali fornia, leaving open the question of whether the same results would be found in other areas, especially older cities in other parts of country. All of these studies have struggled with the issues of how to defme mix of use and urban design, and how to separate the effects of these characteristics from those of density. Many researchers are interested in whether a gieater mix of uses combined with neo-traditional design features will result in travel behavior different from TRANSIT COOPERATIVE RESEARCH PltOGRAM (rCRP) fi.JO

PAGE 131

that in typical suburban development. To date, there bas been too little experience with these new types of suburban devel opment to answer the question. TbeJ:e fore, studies look a1 older neighborhoods that have a fmer-grain mix of uses and a more pedestrian-friendly envirorunent But it is not clear whether behavior of residents of long-standing older neigh borhoods accurately predicts the behavior o(residents of new neighborhoods, who in all likelihood are more accustomed to using cars. The studies are organized from the most general to the most specific. The fii"St section compares two panems of deve l opment with the same densities in a suburban county. The second has studies on mixes of uses in employment centers. The third discusses studies that compare neighborhoods with different mixes of uses and designs. Overall, the studies find that mixing uses results in more trips by ttansit, wallcing, and bicycling, but the evidence is inconclusive on whether this actually results in fewer vehicle miles of travel. 4.1 Suburbs (Employment and Residential Areas) 1000 Friends of Oregon. 1996. Making the Land Use Transporl4tion Air QIUJ!Uy ConnecliiJn: Analysis of Alttrnalives. Vol. S. Portland, OR. May. This study compares auto-oriented versus transit -o riented patterns of land use and transportation in suburban Washington County, in the Ponland, Oregon metro politan area. The alternatives utilize the same land area and have the same overall densities. In the au to-oriented alterna tives, most new multifamily housing and jobs are at the urban fringe. The "no build" version includes fewer transpor tation improvements, whereas the "high ways" version includes a bypass freeway RUIGW llOOO<"'GS PARSONS EOONOllHWEST 127 and other highway improvements. With transit-oriented development, the majority of new multifamily housing and jobs occurs on vacant land near transit routes. 'This alternative also includes transit investments, retrofitting of pedestrian improvements, selected highway im provements, and a demand management program that includes parking charges for work trips. The region's travel demand model, which was enhanced to increase its sensitivity to density and design, was used to stimulate the transportation outcomes in 20 I 0 of each of the alternatives. A package of transit-oriented develol?' ment and transportation improvements that focus on non-automotive modes has the following effects within the study area: Reduces auto ownership rates by 5 percent from auto-oriented rates. Reduces single-occupant auto use for work trips to 58 percent compared to 76 percent in auto-oriented alterna tives. More than doubles share of work trips by transit over auto-oriented alterna tives (18.2 percent vs. 8.8 percent) Reduces daily vehicle trips per house hold from 7 5 to 7 2 trips Reduces vehicle hours of delay over "no build" levels more than highway building alternative (53 percent reduc tion versus 43 percent reduction). Reduces peak period vehicle hours of tra vel by three times the rate of high way building alternative (15.7 percent reduction versus 5.6 percent). Reduces daily vehicle miles of travel by 6.4 percent, whereas the highway alternative increases them by I. 6 percent. The results apply to the study area, the fastest-growing part of the Portland metropolilan area. Impacts would likely be less if transit-oriented land uses and ttansportation improvements were built TRANSIT COOI'fAATIYf RESEARCH PROGRAM (rCRP) H-10

PAGE 132

throughout the metropolitan area, as the remainder of tbe region has less growth to focus in transit-oriented developments 4.2 Activity Centers The following studies examine the effects of mixed use on uavel behavior. Cervero (1996) takes a broad view of mixed use (community jobs -housing balance), whereas the others examine a fine-grained mix of uses within centers. Cambridge Systematics. 1994. The Ef of Land Use IUid l'rrlvel Demond Strlllegies on Commuling Belutvibr. Washington, D.C.: U.S. Department of Transportation, Federal Higbway Ad ministration. This study tested tbe influence of employment site design characteristics on commuting mode choice at suburban work sites in the Los Angeles area. The research involved on-site data collection of specific urbat design and land use attributes to enstre a cardul calilration of the independent variables. The results that the presence of land-use mix and urb111 design features, such as shade trees and sidewalks, in cootdinatim with dem111d management progams, are responsible for incteaSing the percentage of work trips made by transit by three to fourpero:ntage points. An attractive urb111 envrorunent was the only fact>r that infilenced mode choi::e in the absence of a ttaYel demllld pro gam. Thus, mixed uses and access to services within the em ployment were not strmg enough by therrselves to generate more com mutilg by transit This study did not con1rol for fact>rs such as the lev d of transit service to the site. RUIGDS &ROOKINGS PAISONS fCOI'IOC1ltWtST 128 Uterature R.eview Cervero, R. 1989. Ameril:a's Suburban The lAnd Use-Trtltlspol'lalion Link. Boston: Unwin Hyman. Cervero compares the commuting char acteristics of workers in 57 Suburban Employment Centers. These centers have at least one million square of office space. 2,000 or more workers, and are at least five miles from the CBD. He uses cluster analysis to identify six typeS of centers office park, office center, large mixed-use center, moderate mixed-use center, sub city, and large conidor. He uses analysis of variance techniques to detenni:ne whether the center types differ in commuting characteristics. He concludes that the higher densities and greater land use mix do result in more commuting by transit, ride-sharing, and walking Ride sharing is greatest in the centers whereas walking is greatest in centers with significant retail and nearby multifamily housing. These denser, more mixed cen ters also have slower speeds of uavel because of greater congestion within the centers. This study did not control for ttansit availability and the quality of the pedestrian environment. Cervero, R. 1996. "Jobs-Housing Bal ance Revisited." Joumal of the Amtril:an Planning Association, Vol. 62, No. 4. pp. 492-511. Using data from the 1980 and 1990 censuses, Cervero compares tbe jobs housing balance of the 23 largest cities in the Sao Francisco Bay Area. The evidence shows that the jobs-housing balance generally improved, particularly as jobs increased in fonnerly housing rich areas. However, ho u sing did not grow significantly in job-rich areas largely because zoning and growth controls prevented housing growth. Fifteen of the communities showed small increases in the ratio of internal com muting to external commuting. Nonel'liANSIT COOPERAJIV RESEARCH PltOGRAM 10

PAGE 133

theless, about twice as many people commute in and out of the average community as commute within it. Thus. despite less segregation of uses (measured at 8 gross city-wide scale), many people continue to commute considerable distances in pan because of mismatches between the jobs available in the community and the type of housing available there. -1bis descriptive study demonstrates that cons i derations of the transportation consequences of the spatial segregation of uses need to look more closely than just at numbers of residences and jobs. The mismatches between the income of employees and housing prices and between new jobs and housing availa bility also must be considered. 4.3 Neighborhoods Studies in this section are oq:aniz.ed around the types of trips analyzed. Collectively the studies fmd that a greater mix of use and wban environments that are more friendly to the pedestrian encourage more walking and transit use. 1be non-work trips by walking and transit may. however. be additions to automobile travel, not substitutes for it. Cervero, R., and R. Gorham. 1995. "Commuting in Transit Versus Auto mobile Neighborhoods." Journal of tlu American Planning AssociaiUm, Vol. 61, No. Z. pp. ZIO-Z25. 1bis study compares work trip mode shares and trip generation rates between matched pairs of the transit-oriented and auto-oriented neighborhoods. Seven of the pairs are in the San Francisco Bay area. and six are in Los Angeles. Transit oriented neighborhoods were built around streetcars or rail stations prior to 1945 and have 8 grid street pattern. Auto-RVTGO.S llOOIONGS PAlSONS ECONOR'IHWEST 129 oriented neighborhoods were built after 1945, with little orientation to transit, and more curving streets and cui-de-sacs. Neighborhood pairs had similar incomes and, as far as possible, similar levels of transit service Six of the seven San Francisco pairs showed the expected results of lower auto ownership and more use of transit and walking for work trips. (In one pair with a large university in the transit neighborhood. the transit neigh borbood had less driving alone but walking substituted for transit.) The difference in the share of drive-alone rates ranged from 2.0 to 17.5 percent of trips The results were more mixed in Los Angeles. 1be authors conclude that neighborhood design matters little in the Los Angeles area because of the overwhe lm ing dom inance of the automobile in this region. 1be results may also be muddled because transit service levels were less closely matched in the Los Angeles pairs. Ewing, R., P. Hallyur, and G. W. Page. 1994. Getting Around A Traditional City, A Suburban PUD, and Everything In Between. Paper presented at the 73rd Annual Meeting of the Transportation Research Board, Washington, D.C. January 9. The authors compared six communities in Palm Beach County, Florida on work accessibility, neighborhood shopping opportunities, and pedestrian accessibil ity. They found little evidence that accessibility to retail affected mode choice or vehicle hours of travel per person. The shortest shopping and recreational trips occurred in a classic 1970s planned unit development (i.e. a suburban auto oriented place) because of ample stores and recreational facilities within the co nun unity. This suggests that the mix of uses is as important as the layout of streets and other design features in determining travel behavior. 'I'II>SJr COOI'fRATIVE RESEAIIOI PROGRAM (rCI!P) fi.IO

PAGE 134

Handy, S. 1992. "Regional Versus Local Aoces.sibility: Neo-Tradltional Develop ment and Its Implications for Non-Work Travel.'' BW/t Environment, Vol. 18, No. 4. pp. 253-167. Handy compared the shopping trip modes of residents of traditional and suburban neighborhoods in the San Francisco Bay Area. She found thai residents of traditional neighborhoods, where shop ping opportunities were located nearby, made 2.75 to 5.5 times as many shopping trips by wallcing as residents of more aut!Kiriented neighborhoods. Residents of both types of neighborhoods made about the same number of auto trips to regional shopping malls, suggesting thai neighborhood shopping trips may have supplemented rather than replaced longer trips. Handy, S. 1995. "Understandlag the Link Bet"""" Urban Form and Travel Behavior.'' Paper presented at the 74h Annual Meeting or the Transportation Research Board, Washington, D.C. Jan uary. Handy makes detailed comparisons of non-work trips in four suburban neigh borhoods in the San Francisco Bay Area. A "traditional" and a "typical" suburban neighborhood were identified in the Sili con Valley, where there are good connections to the rest of the region. Another pair was selected in Santa Rosa on the fringe of the metropolitan area. Data is from original surveys. An analy sis of variance shows thai differences in travel behavior occur because of urban form, controlling for household type (number of adults and number of work ers). People make more shopping trips on foot in the "traditional" neighborhoods where downtowns are connected to residential neighborhoods and offer services to those residents. It is not clear thai these trips replace auto trips, lUIGW ll()()(JNGS PAlSONS ECONOliHWEST 130 however. Secondly, people value choices and on average visit more than one grocery store and regional mall in a month, if they are available. Having choice adds to travel since trips are made to places more distant from home. Kitamura, R., P. L. Mokhtarian, and L. Laldet. 1994. "A Micro-Analysis of Land Use and Travel in Five Neighborhoods In the San Francisco Bay Area.'' Instilu1e or Transportation Studies, University of California, Davis. The authors studied the travel behavior of several hundred families in five San Francisco Bay area neighborhoods. The areas were selected because they bad similar median incomes but were either high or low in density, and varied in mix of use and access to rail transit Three day travel diaries were collected, and site surveys were made to identify urban design characteristics. Models estimated individual travel behavior and, therefore, controlled for individual characteristics such as income, occupation, education, and vehicle ownership. Differences in travel were explained both by individual characteristics and by land-use measures, especially residential density, public transit accessibility, and the presence of sidewalks. Density was most important in explaining the share of non-motorized trips. Acoess to transit influenoed the number of non-motorized trips and the share of transit trips. The mix of uses was not vety powerful at explaining travel behavior, but a dummy variable for place (combining all the land use attributes) was significant. Overall, the models developed had limited explanatory power, explaining only about 15 percent of the variability in the number or share of trips by various modes. TRANSrT COOPERATIVE RESEAAOf PROGRAM (TCRP) fi.IO

PAGE 135

Parsons Brinckerbolt Quade and Douglas, R. Cervero, Howani/Stein-Hudson Associates, and J, Zupan. 1996. Influence of Latul Use Mlz and Neighborhood Deign on Transit Demand. Unpubllshed report for TCRP H-1 project. Transit Cooperative Research Program, Trans portation Research Board, Washington, D.C. March. Three separate studies examined the effects of neighborhood laud-use mix and urban design on the demand for transit and other alternatives to the automobile. One study used Annual Housing Survey data for 1985 for 11 large metropolitan areas to compare mode choices for work nips of residents in areas with and without proximity t o a "comer store" or other commercial activities. A second study of the greater Olicago area used transit and land-use data to identify the factors influencing lrallsit nips per person. The third study compared the mode choice for work and non-work nips in "traditional" and "suburban" neighborhoods in the San Francisco Bay Area using original survey data. All of the studies use multi-linear regression tech niques to control for income and other household characteristics Overall. the studies show that the types and mix of land uses influence the demand for transit as well as the use of nonmotorized modes People who live in mixed-used nei ghborhoods have a lower probability of commuting by car (3 to 4 percentage points), a slightly higher probability of using transit (I to 2 per centage points), and a much higher probability of walking or bicycling (I 0 to 15 percentage points) for work nips. In the Olicago area, a 10 percent increase in residential density is associated with an II percent increase in the number of nips by transit Residents of "traditional" neighborhoods are more likely to use non automotive modes for non-work RUTGEIS HOOKING$ PAISONS ECONOlllfWUT 131 nips than residents of "suburban" neighborhoods The neighborhood com parison study did not find statistically significant differences in mode choice for work nips between the two types of nei ghborhoods. 'These studies found i t difficult to sort out the effects of land-use mix and urban design because these characteristics are strongly correlated with density. When density is included in an equation, mix and design variables generally explain li ttle about mode choice Each of the studies controlled for residential charac teristics such as income and auto ownership. Because the studies are cross sectional, they show only correlation be tween land-use characteristics and mode choice, not causality S. DISPERSED EMPLOYMENT In the studies reviewed in the ftrSt section. Gordon, Richardson, and their colleagues argued that dispersed employ ment has helped to keep work nip times stable even though distances have in creased. The following study finds that a polycentric pattern of development in the San Francisco Bay area bas resulted in longer commutes. Cervero, R. 1996. ''Subcentering and Commuting: Evidence from lhe San Bay Area, 1980-1990." Paper presented at the 1996 TRED Conference on Transportation and Land Use The Lincoln Institute, Cambridge, Massa chusetts. October. This paper examines the growth of dispersed subcenters in the San F rancisco Bay area and the effects of this growth on commuting. Cervero identified 22 em ployment centers with 7 or more workers per gross acre and 9.500 or more TRANSITCOOfERAnVEIISEARCHPROGAAM (TCRP) N-10

PAGE 136

employees in 1990. Downtown San Francisco was the largest and most dense. Other centers were in Silicon Valley, the east Bay core area (Oakland, Berlreley, and Emeryville), and 16 were further out in suburbs. Two did not exist in 1980. Employment in these centers grew on average by 23.6 percent in the 1990s. increasing the regional share of employment in centers from 47.5 percent to 48.2 percent. The growth of centers produced an increase in vehicle miles of travel (VMI) for commuting trips On average, one way VMI' increased from 7.1 to 8.7 miles, a 23 percent increase, with the largest growth in suburban centers This increase in vehicle miles of travel is due to both longer distances and to greater use of single-occupant vehicles. Longer distances between borne and work had the most influence on VMI' since outside of downtown San Francisco and the eastern Bay Area, the vast majority of commuters used cars in both 1980 and 1990. Cervero estimates that more than four-fifths of the growth in VMI' is due to longer distances between home and work. Longer distances were especially important in increasing VMI' in the mone peripberal centers Cervero anributes these longer distances both to regional growth and to mismatches in the jobs and housing market that necessitate long commutes. 6. THE COST OF TRAVEL The previous sections discussed the consequences of various aspects of sprawl on travel times, distances, trip making and modes. The literature on the cost of travel is also relevant for convening these travel behaviors into costs for individuals and society. lUTGQS BaOOI(JNGS PAISONS CONOitllfW(ST 132 The Cost of Travel (An Addition to the Transportation Outcomes or Sprawl Literature Review) A number of recent studies have estimaled the cost of travel using nationwide, average, or location-specific data. All the studies agree that 8lltomobile use is subsidized because of the signifi cant share of costs borne by society at large. There are also significant costs that travelers may not consider when they decide whether to make or trip or what mode to use. This body of literature is relevant to an analysis of the costs of sprawl for several reasons First. it makes use of a "full cost" framework, in which the research ers anempt to quantify and account for a much wider range of costs than may be typically included in discussions of public policies or private activities. These include user costs, governmental costs, and costs to society at large. Regarding the social costs, many are dealt with elsewhere in this literature review Second, the research indicates the magni tude of many of these costs. The esti mates, particularly for social costs, vary widely. Third, the research indicates the incidence of the costs. It introduces the need to consider both public and private costs, for example, into the debate about the costs of sprawl. Fourth, it deals with the costs of different modes of travel. Evidence presented in the previous section of this literature review makes clear that travel behavior varies in different built environments, both with regard to the modes of travel and the distances traveled. In order to estimate the costs of the set of travel behavior associated with these different spatial patterns of settlement. it is TRANSIT COOI'ERADVE WEAIIOI rROGAAM (TCRr) H-10

PAGE 137

necessacy to introduce the findings of the research described below regarding the cost of different modes. Several of the authors mentioned below assert that sprawl is a consequence of lraliSpottation invesunents. Elsewhere, the literature review summarizes how researchers assert that uavel behavior is a consequence of sprawl. Both assertions seem true; the relationship between !rallS ponation and land use is interactive. In the context of this literature review, however, it is the costs and consequences of sprawl that need to hold our attention. Apogee Researcll, Inc:. 1994. Tlu Qlsts of Transportation: Find Report. Conser vation Law Foundation. March. This report reviews the literature on the cost of uansportation and estimates the per mile costs of several modes for Boston, Massachusetts, and Portland, Maine. They divide costs into three types: user costs, governmental costs, and societal costs Extensive data were collected for the case study regions to accurately reflect the cost of uavel in these specific places. Some costs-land loss, water pollution, solid and hazardous waste pollution, and sprawl and social isolation-could not be quantified and are not included in the analysis The report estimates costs for a variety of modes and in different kinds of environ ments For example, they estimale that a peak -period trip in a dense part of Boston using a single-occupant vehicle (SOY) on an expressway would cost $1.05 per mile. Of the $1.05, $0.88 is user costs (including $0.24 for uavel time), $0.05 goes for governmental costs not paid by the user, and $0.12 for societal costs. In the off-peak period, the same lrip costs $0.89 per mile, with $0. 73 for user costs ($0.10 for travel time), $0.05 for governmental costs, and $0.11 for IUTGBS llOOIQNGS PARSONS ECONOIIHWUT 133 societal costs. In a low-density setting the peak and off-peak SOY trips both cost $0. 71. For the SOV mode, user costs, including time, vary the most among different settings. A high-occupancy vehicle (HOY) ex pressway trip in high-density Boston at the peak costs $0.58 per mile, a com muter rail trip $0.58, a rail uansit trip $1.04, a bus trip $1.09, a bicycle trip $0.73, and a walking trip $2.56. Travel time costs add significantly to the costs of rail transit bus, and walking. Costs in the srnaller city of Portland are generally lower for all modes and densities. The authors believe that transportation does influence sprawl, and this should be considered a societal cost of the transpor tation system. They do not, however, measure this cost since srudies neither identify the full range of the costs of sprawl nor the proportion of these costs that are due to the uansportation system. This report documents the ways that travel costs differ with the physical envi ronment and the modes available. As far as possible, costs are based on actual data for the locations studied altho ugh soci etal costs are generally from national studies. Delucchi, M. A. 1996. The Annruzli:.ed &cilzl Cost of Vehide Use in the U.S., 1990-1991: Summary of Theory, Dala, Methods, and Resulls. Davis, California: Institute of Transportation Studies. August. In a 2(}-volume study, Delucchi and his colleagues estimate the total social cost of automobile use in the United States for 1991. The srudy shows that many cost functions are non-linear and dependent upon location Therefore, the studies' estimates cannot be divided by total automobile mileage or some other TRANSIT COOPERATIVE RESEARCH Pi!OGRAM (ICRP} H-10

PAGE 138

measure of use and produce an accuraJe average price for use in a particular study or analysis, although the methods may be applied in other studies. Delucchi divides costs into six categories : I} penonal non -monewy CO
PAGE 139

1be size of cost estimates for some externalities, such as accident costs, is large compared to those reported by others. Utmao, Todd 1995. TrtliUIHJrtalion Cost ANJlysis: Eslim
PAGE 140

automobile use. The largest subsidy is for free employerp rovided parking. The authors estimate that 85 pen:ent of $100 billion cost of employee parking is not paid by the user. The report also estimates that users pay none of the air pollution costs (estimated to be $37 billion), security costs of oil ($25billion), petroleum subsidy ($0. 3 billion), and noise ($9 billion) Also, about 15 pen:ent of accident costs or $55 billion worth, are estimated to be paid by someone other than the responsible party. The authors were unable to estimate some costs, such as the opportunity costs of land devoted to roads Estimates are based on data from pre vious studies. Estimates of etemality costs are more speculative than other costs. Parsons Brinckerboft' Quade & Douglas. 1996. Cost of Trtzel in Bou/Mr. City of Boulder, Colorado. July 15. This study uses the methods of the Apogee study, local data for govern mental and some user costs, and national data for societal and some user costs to estimate the cost of typical trips by various modes within the built environ ment of Boulder. The study utili res actual times to and from spec i fic locations. The study estimates the cost of com muting to Denver (25 5 miles) to be $24.61 by single-occupancy vehicle (SOV) and $15.79 by lnlnsit. The SOV trips breaks down into $19.40 for user costs, $1.16 for governmental costs and $4.04 for societal costs. The lnlnS i t trip includes $10.68 of user costs (mostly time), $4.70 of governmental costs (mostly for lnlnSit provision), and $0.41 of societal costs. In this case, ttaosit is a cheaper trip In contrast, for a multi purpose shopping trip of 9. 75 miles RUTGERS llOOIONGS PARSONS ECONOOTHWEST 136 within the city of Boulder, an SOV trip costs $11.66, whereas lnlnSit costs $29.17 Transit is more expensive be cause of the time involved and the gov emmental expenses for offpeak lnlnSit IJ'avel. For a short 2-mile trip to down town Boulder, more options are avail able. An SOV trip costs $4.02, a lnlnSit trip $3.43, a bike trip $1.74, and a pedestrian trip $5.59 ( l argely because of the longer time). This study shows that IJ'avel costs vary with the environment and type of travel. Transit costs less for long commutes, walking and bicycling are viable alterna lives for short trips in a compact city, and the car is best for linked trips. Voorhees, M. T. 1m. The True Com of the Automobile to Society. City of Boulder, Colorado. January. 1bis study estimates the total annual cost of automobile use in the United States in 1 990. The author divides costs into two main categories: 1) direct expenses of automobile ownership and use, and the cost of highways ; and 2) tluee categories of external costs: direct monetary costs such as for emergency medical care, lost economic gain due to air poll u tion and other externalities and the opportunity costs of us ing land for roads and parking. Us ing data from other studies, he estimates that in 1990 the t otal cost of the automobile system was $1,152 billion. The largest costs are direct expenditures for the automobile ($440 billion, or 38 percent), land use opportunity costs ($246 billion, or 21 percent), congestion costs ($146 billion, or 13 percent), air pollution costs ($1 00 billion, or 9 percent), and highway costs ($80 billion, or 7 percent). V oorbees argues that the automobile has had two land-use impacts: consuming TJWISIT COOPrunVE mEARCH PROGRAM (TCVJ /1.10

PAGE 141

large amounts of land for roads and parlting and encouraging sprawl. H e estimates the opportunity costs of all the land devoted to roads. He does 1101 estimale the costs of sprawl because be laclcs data and because these costs already ate calculaled in the amouot of fuel consumed and other costs of using the automobile that result from a more dispersed pattern of development. 1be external cost estimates are quite subjective and would easily be changed by making different assumptions 1be cost e!'fimate for land opportunity costs is quite large compared to those reported in other studies. l lffGUS lllOOIONGS 'A&SONS ECOHOITHWUI 137

PAGE 142


PAGE 143

Burchell, Robert W David Ustoldn. n.d. Land, lntrastnic:ture, HollSing Costs, and Fiscal Impacts Associated with Growth: 1be Uterature on tbe Impacts of Sprawl versus Growth Management. New Brunswick, NJ: Center for Urban Polley Research. 1bis is a short summary paper covering the topics listed in its title, mainly by reviewing the major studies of those topics in the past. It draws beavily upon the research done by the authors for the State of New Jersey and reported on elsewhere in this overall study. The paper examines the implications of planned development versus more traditional decentraliud development in the areas of land consumption, infrastructure costs, housing costs, and fiscal impacts. Most of the studies covered have analyud at least two different develop ment patterns. One is the currently dom inant pattern, or sprawl. "Development of this type typically includes subdivision style residential development and strip nonresidential development consisting of skipped-over, noncontiguous land devel opment, including residential, in the form of 0.33 to 1.0 acre lots, and nonresi dential using floor-area ratios of 0.20 or Jess" (p. 1). In contrast, planned devel opment seeks "to contain most new growth around existing centers and limit development in rural and sensitive environmental areas. 1bis is done by increasing the share and density of development close in to existing development." The growth analyud is assumed to consist of population growth leading job growth, with the former consisting of both immigration and natural increase. Ideally, these trends and the provision of facilities accommodating them are handled in a timely, harmonious manner. IUTGOS IAOOIONG$ PAI$0N$ ECONOitTHWUT 139 Traditional growth departS from the most harmonious possible path by locating residential and other development 'to a new outer ring of the metropOlitan area with access from this new outer ring oriented increasingly to beltway or interstate rather than central core job locations." This results in an increasing unden.ltiliution of core land and infra. structures. 1bis process is associated with the development of new "edge cities." That, in tum, generates a new farther-out ring of bedroom residential subdivisions. 1be core of the metro politan area, absent redevelopment, be comes relatively abandoned by a variety of necessary and blue-chip economic activities and a borne by default for poor residents who cannot follow . or are not allowed to follow upper-income residents to the suburbs (because of z.oning). Even with redevelopment, the central core is a struggling entity with no soft-goods retail anchors, no quality supermarkets or movie theaters a declining upwardly mobile population, public school systems being replaced by private, and increasingly higher property taxes to pay for rising public service costs" (p. 3). Traditional development is costly because of providing new infrastructure for those located far out, and maintaining the old infrastructure for those left behind Yet in the short run this process is not bad for the region. It distributes flflllS and house holds to localities that minimize individual out-of-pocket costs. But no consideration is given to the larger societal costs or impacts of these individual choices. The planned alternative channels growth to more efficient locations over the long run. Mos t of the far-out growth in traditional development is contained closer to existing infrastructures and built-up areas. Thus, "'n the final equation . there is a more orderly and less wasteful relationship between old and new development" (p. 5). TRANSrr COOPERATM RESfARCH l'ttOGAAM (TCRP) 11-10

PAGE 144

"The Com Another goal of planned development is conservation of open space in the forms of agricultural land, forests, and environ mentally-sensitive areas. In the New Jersey study, there was enough land to accommodate a 2(}-year projection of population and fum growth under both basic growth schemes. But the PLAN consumption of land over 20 years was 60 percent lower than the TREND consumption of land. This wquld enable preservation of greater levels of both farmland and frail lands. Concerning infrastructure costs, the New Jersey study showed that the state could save $1.3 billon in infrastructure costs over 20 years for roads, u tilities, and schools, if it followed the PLAN scheme instead of the TREND scheme This saving s occurs mainly through more intensive use of existing infrastructure capacity. as opposed to building more new infrastructure The PLAN approach directs new growth to where such excess capacity is found, rather than directing new growth to virgin territories Also, the PLAN scheme is more compact than the TREND scheme, therefore requiring accommodation of shorter distances of movement. This also applies to sewer and water systems In addition, bringing together larger numbers of people in more compact areas may provi d e for some economies of scale, such as larger educational facilities with lower costs per student. Earlier studies have suggested that road costs per unit might be lower because persons in compact settlements travel Jess and use roads more effi ciently-that is, closer to capacity. Both RAND and the state of Florida had cond u cted s tu dies arriving at that concl u sion. An Urban Land Institute study by James Frank had concl u ded that low-density development was far costlier in terms of all capital costs than a mixture of 30 percent single-family and 70 per cent apartments, plus using contiguous development rather than leapfrogging RUTGERS atOOKJNGS PAlSONS fCONOllHWEST 140 This shift cut costs per unit from $35,000 to less than $ 1 8,000. The Rutgers New Jersey study, using the same increases in population (520,000 persons in 20 years), jobs, and house holds found that PLAN would save: $699 million in roads, a 24 percent savings. $561 million in water and sewer costs, a 7.6 percent savings $173 million in school capital costs, a 3.3 percen t savings An overall savings of $1 4 billion, or just under 10 percent In the James Duncan study, Search for Efficiem Growth Patterns (1989), PLAN achieved savings of 60 percent for roads, 7 percent for schools, 40 percent for utilities, and -2 percent for other capital facilities. In the James Frank study The Costs of Alternative Growth Patterns: A Review of the Literature (1989), PLAN achie ved savings of 27 percent for roads, I percent for school capital spending, and 33 percent for utility capital extensions. In the Burchell studies, PLAN achieved savings of 24 percent for roads. 3 percent for schools, and 8 percent for utilities. Thus, all three major quantified studies of this topic show significant savings through planning and contiguous development for roads and utilities. and smaller ones for schools. The authors of this paper synthesize these studies by concluding that PLAN achieves savings of about 25 percent for roads, 5 percent for schools 15 percent for utilities, and zero for other capital spending. Concerning housing costs, only the New Jersey study bas looked ar them ar a comprehensive scale such as that u s ed for the above analysis of public costs. It found tba1 land prices per acre and housing prioes per unit would rise in areas where low-density units were TRANStr RESEARCH PROGRAM (TCRP)

PAGE 145

pennitted under the PLAN approach. But housing prices would decline (compared to their existing average) in more central locations because higher-density units would be built there. Since more added housing would be built in close-in locations than far-out ones, the net effect was a reduction in housing prices of about 6 percent. Concerning ftseal impacts upon governments, the ability of PLAN growth to use existing excess infrastructure capacity would reduce net fiscal impacts, compared to TREND. The savings would be $112 per year, or2 percent. An overall summary chart is presented at the end synthesized the above findings. Dahl, Thomas E. 1990. Wetlmuls Losses in the United States: 1780s-1980s. Wash ington, D.C.: U.S. Department of tbe In terior, Fish, and Wildlife Service. 13 pages. Dahl documents wetland losses from the colonial period to the 1980s. In colonial America about 400 million acres of wetland existed; as of the 1980s, the wet lands inventory is down to 250 million acres. Wetlands occur in every state in the nation in varying size, shape, and type. Variation occurs around differences in climate, vegetation, soils, and hydrologic conditions. Until recently, wetlands have been con sidered a hindrance. Swamps, bogs, sloughs, and other wetland areas were considered wastelands to be drained, filled, or manipulated to "produce" other than natural services and commodities. Recently, though. wetlands have come to be seen as vital areas that constitute a productive and invaluable public resoUJ"Ce. RUTGERS UOOKINGS PAISONS ECONOIIIWST Ul Ewing, Reid. 1995. Best Devewpment Pt=tU:es: Doing the Right Thing and Making Money At The Same Time. Chi cago, llliDols: American Planning Asso dation. 180 pages. Ewing addresses the need for change in development policy and practice given expected rapid Florida growth l'liiCS (approximately 5 million people during the next 20 years) in Florida's dominant development pattern of urban sprawl. Ewing argues that increasing social and economic costs will occur as a result. In an attempt to minimize these costs, the author advocates community development in which public purposes are weighted against market considerations Such public purposes are listed as affordable housing, energy efficiency, and the preservation of natural areas, among others. Discouraging urban sprawl by creating vibrant communities means an emphasis on population diversity (age and class), the establishment of stteet life, creating a sense of place, and other features contributing to "livability." Reconunen dations to realire these goals are presented. Tbey are meant to be used as a basis for establishing comprehensive plans for new communities and redevel opment projects, land development reg ulations or for the evaluation of de velopment proposals. The use of seven new communities (smaller planned com munities within the 300-to 500-acre range) as case study illustrates the "best" development practices. rR>Str COOP9.411Vf REsu.RCH PROGRAM (TCII;P) 11-10

PAGE 146


PAGE 147

1. POPULAR LITERATURE Fried, Carla, Leslie M. Marable, aad Sheryl Nance-Nasb. 1996. "Best Places to Lin In America.'' Money (July): 66-95. Money magazine publishes an annual article on "the best places to live" that ranks the 300 largest metropolitan areas in the United States. To rank the metropolitan areas. Money first surveys its subscribers to nue 41 quality of life factors They then collect data on specific measures for the 300 cities and assign the data to nine broad categories : crime, economy, health, housing, education, weather, leisure, arts and culture, and ttansportation. The data is then weighted according to readers' preferences to produce the fmal ranking. The top I 0 quality of life characteristics, as nued by Money subscribers, are: low crime rate, clean water. clear air, plenti ful doctors, many hospitals, housing appreciation, good schools, low property taxes, low income taxes, and strong state government. Money points out that the rating of quality of life characteristics differed by gender and type of house hold. This article is one of several annual or semiannual rankings of places in the popular literature. Although Money does survey readers to determine which char acteristics of quality of life are important, it does not reveal enough about its specific measures or scoring method to assess whether its rankinga accunuely reflect the survey results. 1n addition. the survey asks Money subscribers to rate 41 characteristics, and so it is probably not representative of the U.S. population in general, and it may not include every characteristic that readers think are important Overall, however, this article provides insight into how the topic of quality of life is typically treated in the popular literarure. lllfGUS llOOtONG$ PAISONS ECONOIKW$T U3 lhrM Romw Hall, Bob and Mary Lee Kerr. 1991. 1991 1992 Gnen 1mkr: A SUlte-By.SGuide to the Nalion's Enironnuntal Health. Washington, D.C.: Island Press. 162 pages. Drawing from a variety of private and public data sources. the Green Index uses 256 indica1ors to measure and rank each state's environmental health. The indica tors of encompass a broad range of environmental conditions and are grouped into eight major categories: (I) air sickness; (2) water pollution; (3) energy usc and auto abuse; (4) toxic, hazardous and solid waste; (5) community and workplace health; (6) farms, forests, fish and fun; (7) congressional leadership; and (8) state policy initiatives. Based on the indicators, the authors identify the best and worst states overall. Landis, John D. and David S. Sawicki. 1988. "A Planner's Guide to the Ple s Rau Almanac!' Journal of the Anuriaur Planning Association (Summer): 336-346. This 1988 critique of the Places Rated Almanac ( 1985 edition) is relevant to both the 1993 edition and other popu l ar rankings of places. The authors point out that the essential problem with the component measures used to rank places is that they have not been tested against the stated opinions of migrants or observed migration behavior. The authors cite a 1985 study (Pierce 1985) that surveyed a random sample of New York state residents on the importance of the categories used in Places Rated. Respon dents ranked categories in order of importance: personal economics (jobs), housing. crime climate, health, educa tion. recreation, transportation, and arts and culture. An article that compared overall metropolitan scores (not rank ings) in Ples Raled Almanac with a TRANSIT COOPfliATIVf RfSEAROf PROGRAM (TCRP) H-10

PAGE 148

nonrandom sample of households found only four of the nine categories statis tically significant to migration decisions: housing costs, crime, education, and recreation. The authors also compared category ranldngs for 51 mettopolitan areas in Places &ted Almanac with migration between 1975 and 1980; this comparison found rankings of housing cost and economic opportunity signifi cantly correlated with rates of in migration. Landis and Sawicki point out that Places &ted Almanac assumes that a person's quality of life is critically to the qualities of the place in which they Jive or work. Research, however, indicates that most individuals rank personal causes of satisfaction and dissatisfaction as much more important determinants of their quality of life than geographically based factors Precourt, Geofl'rey and Anne Falrclotb. 1996. ''Best Cities: Where the Living Is Easy." Fo111lne (November 11): 126136. This article identifies the 15 best U.S. and 5 best international cities for work and family. Much of the article is devoted to qualitative descriptions of the best cities, with little explanation of the methods used to rank cities. Among the variables considered were the crime rate, quality of schools, availability of culture, traffic congestion, number of doctors, tax rates, price of teal estate, and costs of a martini and a fiTSt-run movie. The article contains a table showing attributes of the cities in six categories. These categories, and the specific measures shown, Demographics: 1996 population, % change in population 1996-200 1, median household income, and % of population with bachelor's degree Cost of living: cost of living index, highlUTGW ll
PAGE 149

Health Care: number of general/family practitioners, specialisiS. general short-tenn hospital beds, and hos pitals. Crime: violent crime and property crime rates. The Arts: number of concert or classical format radio stations, touring anisiS bookings (classical music, dance, professional theatre), resident ans companies (classical music, ballet, professional theatre), non-profit an museums/galleries, and public library collections. Recrearion: public golf, good restauraniS, movie theatre screens. zoos, aquari ums, family these parks, pari-mutual betting, professional and college sports, ocean or Gteat Lakes coast line, national foresiS, national parks, national wildlife refuges, and state parks. Clinulte: number of very bot and cold months, seasonal temperature varia tion, beatingand cooling-degree days, freezing days, zero-degree days, 90-degree days. The scoring system uses a variety of methods to convert each measures into a score, which is then summed to rank metropolitan areas in each category. The scoring method implicitly weight the specific measures and descnl>e the relationship between the measure and quality of life. The ranks in each category are summed for an overall score that is used to rank metropolitan areas; each category has equal weight in the overall ranking. The authors discuss bow the reader can use his or her personal preferences to weight the categories for a personalized overall ranking of metro politan areas. This book is popular literature; i t has a common sense or anecdotal notion of quality of life, with no theoretical underpinning or review of relevant literature. The authors' scoring system RUTGUS llOOICJNGS PAJSONS ECONOitlHWEST 145 implicitly weights the various measures with no apparent basis other than the authors' opinion. Although the book acknowledges that individuals will have different preferences, their suggested method for the reader t o apply weigbiS to the rank by category will not yield a ranking based on preference because it does not change the weighting of specific measures that went into the ranks by category, nor will it address measures that were not considered. Overall, bow ever, tbis book provides insight into bow the topic of quality of life is typically treated in the popular literature. 2. INDICATORS, REPORT CARDS, AND BENCHMARKS Andrews James H. 1996. "Going by the Numbers." Planning (September 14-18). Many states, cities, and bamleiS use indicators to measure their own economic and social health, and to set future goals. Indicators are also called benchmarks or vital signs. Local government often creates the measures, but they have also been created by community groups. All indicator projeciS discussed in this anicle have used public process to identify specific measures. Some indicator projeciS have a specific focus, such as gov ernment performance or the environment, whereas others are comprehensive. Jacksonville, Florida developed a Quality of Life index in 1985 and updates the report annually. A 1991 community re view of the index revealed education as the community's top priority. The other categories in the index are the economy, public safety, natural environment, health, social environment, government and politics, culture and recreation, and mobility. Specific measures include the number of outdoor sign permits issued, cost of 1,000 kwh of electricity, student TIIANSrT COOPRJ.TIVE RESEARCH PROGRAM (TCRP) H-10

PAGE 150

fitness test scores in 50th percen tile or better, and people reponing commute times of Jess than 25 minutes. Jackson ville has recently developed an equity index that provides a neighborhood-level looks at measures from the Quality of Life index related to delivery of public services, such as police response times. "Sustainable Seattle" is an indicator project focused on the regional long-term cultural, economic, environmental, and social health and vitality. The project has developed a set of indicators with the headings environment, population and resources, economy, youth and educa tion, and health and community. Specific measures include wild salmon, VMI' and fuel consumption, work required for basic needs, ethnic diversity of teachers, and asthma hospitalization rate for children. The Upper VaUey 2001 project in the upper Connecticut River valley has developed a list of indicalors with 15 categories, such as citizenship, com munity. communications, education .. recreation, health care, personal and public safety. human services, the arts, transportation, businesses, farms and forests, resource use, and the natural environment. The goal indicators is to change policy to move the measures in positive directions. This change does happen, but on a ad hoc basis. Myers, Dowell 1987. "Internal Monitor ing of Quality of Ufe for Economic Development." &onomie !Hvelopment Quanerly 1 : 238-278. Quality of life is recognized as an important factor in economic develop ment, but its exact role and the methods for measuring it are poorly understood. The author identifies four major 1imitaRUTGERS &IOOICJNGS PAISONS ECONOlTHWtST 1.46 tions to developing quality of life mea sures to compare cities or regions: (I) poor availability of comparable objective data, {2) lack of subjective data necessary for addressing this inherently subjective topic, (3) inability to address unique local features, and (4) difficulties in choosing commonly valued weights for combining different components in overall indexes. This article argues for the monitoring of quality of life within a city or region as an important substitute to e)(temal compari sons Internal monitoring can measure changes in local quality of life over time to guard against deterioration of com petitive advantages in the future. The author uses Austin, Te)(as as an e"ample, because the city has used quality of life to attract high-technology firms, and locals are now concerned that rapid development. particularly suburban "silicon strips," will cause quality of life and thus the city's attractiveness to high tech firms to decline Austin's quality of life was a major factor in the location decision of a high-tech finn. Quality of life was an e"plicit element of a formal offer to the firm to locate in Austin. Nine quality of life advantages were itemized in the e)(ecutive summary of the offer. e)(celJent schools, parks and playgrounds; ease of mobility around the city; close-by lakes for water recreation; other opportunities for hunt ing, fishing, and camping; access within two-hours flying to Colorado skiing and Mexican vacations; abundant cultural and entertainment possibilities; general clean liness of the city; attractive topography and mild year-round climate; and "open receptive social structure, a population long noted for friendliness, and a reputation as a desirable place to live and raise children Accelerated growth triggered by the fum's localion produce negative conse quences for quality of life, including TRANSIT COOI'ERAl!VE RESEAROI PROGRAM (TCRP) fi.IO

PAGE 151

housing affordability, traffic congestion, threats to the area's water quality and natural environment, and the perception that downtown office development threatened the city's music scene. The local perception was that these consequences were caused by unmanaged development In reaction to these consequences, the Austin Chamber of Commerce began a research program to measure trends in the area's quality of life This program interviewed leaders of interest groups to identify significant aspects of Austin's quality of life, developed measures for those aspects, and surveyed residents about the important of these measures in their perceived quality of life Residents placed most iinportance on concerns such as crime, cost o f living, schools, traffic, and jobs, and l ess importance on amen ities such as shopping, restaurants, and entertainment. Sixty-two pen:ent of recent migrants identified quality of life as an important factor in attraCting them 10 Austin. Oregon l'r1lgress Board. 1994. Oregon Benchmarl
PAGE 152

Gabriel, Stuart A., Joe P. Mattey, and William L. Waseher. 1996. Compen sating Dif!enntWs and Eolulion of the Qualily-of-Ufe Among U.S. StaUs. San Francisco: Federal Reserve Bank of San Francisco. 96-07. June. 21 pages. This article extends the existiJlg "static" literature on regional differences in quality of life by examining how changes in the quality and quantity of amenities can contribute to the evolution of quality of life over time and across places. 1be article provides estimates of quality of life rankings for U.S. stares over the period 1981-1990. Results indicate thai sparsely populated mountainous western states such as Montana and Wyoming are highly ranked in the estimated quality of life throughout the decade, whereas densely populated midwestern and eastern states consistent ly ranked near the bottom in terms of quality of life. Reduced state and local government spending on highways, increased traffic congestion, and air pol lution, were found to be the most important contributors to the deterioration of quality of life in states that declined in the rankings States that ascended in the quality of life rankiogs did so for a variety of reasons including improved air quality, increased highway spending, reduced commutiJlg times, and reduced state and local taxes Gottlieb, Paul D. 1995 "Residential Amenities, Firm Location and Economic Development." Urban Studies 32 ( 9): 1413-1436. In this article, Gottlieb investigateS whether residential amenities can influ ence the location decisions of high-tech ftrms in New Jersey. In order to determine whether firms evaluate ameni ties on behalf of potential employees, Gottlieb measures a variety of amenities lUIG$ IIOOIONGS PAaSONS ECONotllfWEST at both the potential location of the fum and the residential area where potential employees are likely to Jive. Results of the study suggest thai !inns in the high tech sector are repelled by disamenities like violent crime and high municipal expenditures at the work site. Gottlieb ftnds weak evidence to suppon his hypothesis thai residential amenities such as recreation, low traffic congestion, and public education affect the locational decisions of high tech ftrms. Greenwood, Mkbael J., Gary L. Hunt, Dan S. Rickman, and George L Treyz. 1991. "M.igralloa, Regional Equ!Ubrium, and the Estimation of Compensating Differentials.'' T/u American EconomU: Review 81 (S): 1382-1390 This study examines the p atterns of migration across the fifty stares and attempts to detennioe the relative strengths of two printary motives workers and households have for moving: (I) to earn a higher wage (adjusted for differences among the states in the costs of living); and (2) to have access to the panicular amenities of the individual states. Based on migration patterns for 1971-87, Greenwood and others estimate the amenity-related differ ential in wages for each state, relative to a national average Roback, Jennifer. 1982. "Wages Rents, and the Quality of Life." JoUJ'Nil of Polilical Economy 90 (6): 1257-U78. Roback investigates the role of wages and rents in allocating workers to locations with varying quantities of amenities, both theoretically and empirically. Roback finds that regional differences in wages and land rents are largely explained by regional differences in amenities. Results of her empirical work indicate thai crime, pollution. and cold weather are li!ANSIC COOPfRAllVf RESEARCH PROGRAM (TCRP) H-1 0

PAGE 153

disamenities and that clear days and population density are amenities. Amen ities will decrease wages and increase land rents, whereas disamenities will increase wages and decrease land rents. Rosen, Sherwin. 1979. "Wage Based lnclexes of Urban Quality of Life." ln Cunwat lssiUS in Utbtm EcoMmics. Edited by Peter Mieszkowsld and Mahlon Straszhelm.. Baltimore: Johns Hopkins University Press. Rosen examines the determinants of intercity wage differentials for 19 SMSAs. Rosen fmds that particulates, rain, crime, population growth, and unemployment are disamenities whereas sunny days are amenities. Using his regression estimates, Rosen computes an average quality of life ranking for the 19 SMSAs. He finds that the SMSAs with the highest average quality of life rank ings in general exhibit less pollution, better climate, and lower crime rates than the SMSAs with the lowest rankings. He cautions the reader that the rankings of the SMSAs may be altered depending on the weight given to the various city attributes., especially population density. Salant, Priscilla, Lisa R. Carley, and Don A. Dillman. 19%. Estimating the Con tribution of Lone &gks II> Metro and Nonmetro In-Migl'tllion. Pullman, WA: Social & Economic Sciences Research Center, Washington Slate University. 86-19. June. 34 pages. The main objective of this study is to detennine to what extent decisions about recent interstate migration to Washington state and subsequent employment are influenced by the availability and the use of infonnation technology at the new location. However, the study also investi gates the push and puU factors tbat contribute to a migrant's decision to move to Washington. lUTG!ItS llOOKINGS PARSONS ECONOllHWEST U9 The study estimates that 2,600 so-called lone eagles-individuals who are able to live anywhere and telecommute to work-moved to Washington in 1995 and that many of them did so for quality of life reasons. The most important puU factors tbat lone eagles cited as influ encing their decision to move included the quality of the natural environment, out door recreational opportunities, a de sirable climate, and a safe place to live. Important push factors that influenced the decision to move included urban con gestion, undesirable climate, and fear of crime. von Reichert, Christiane, and Gundars Rudzitis. 1992. "Mnllinomial Logistical Models Explaining Income Changes of Migrants to High-Amenity Counties." The Review of Regionol Studies 22 ( 1): 25-42. This article uses a survey of migrants to, and residents of, I 5 high-amenity wilder ness counties to determine what factors explain the willingness of migrants to accept or not accept declines in income after moving. Survey respondents were asked about their dissatisfaction or satis faction with the previous location (push factors) and the importance the attributes of the destination county in the migration decision (puU factors). On the push side, factors such as environmental quality, pace of life, crime, scenery, and outdoor recreation had higher levels of dissatis faction than employment opportunities and cost of living. In a similar manner. survey respondents placed imponance on puU factors such as environmental quality, scenery, outdoor recreation, and other natural resource amenities and less importance on employment opportunities and cost of living. Results of the study indicate that approximately half of the surveyed migrants received lower incomes and that TRANSIT COOPERATIVE WEAACH PROGRAM (TCRP) H-10

PAGE 154

quality of life and amenities were more imponant factors in attracting migrants to the counties than employment oppor tunities. 4 SOCIOLOGICAL LITERATURE Popenoe, Da>id 1979. "Urbaa Sprawl: Some Sodol oclcal Collsid eratioDB-" Socioloo an4 Sociol R'f1t1rch 63 (2): 255-268. Urban sprawl is defmed by the author as signifying very low-density opment, orieoted to the automobile, wnh detached single-family houses on tela lively large lots. Urban sprawl implies a scatteration of jobs, shops, and services, often in tbe form of strip commercial development; a scarcity of large open or gteen spaces; and a lack of community focus in bolh the physical and social sense. Despite its negative image, most Americans live in environments cbaracterized by wban sprawl. Many Americans, including some sociol ogists, see urban sprawl as desirable when compared to crowded, noisy, violent, and corrupt cities. Urban sprawl gives the individual mote space in eteased safety, more privacy, and a piece ofland to call one' s own. Urban sprawl, however. bas recently been anacked as expensive and an exorbitant user of nat ural resources, especially land and gaso line While most of the attack from sociologists bas focused on the impacts to society, this article examines the on residents of living in low-density, suburban residential environments. Since the positive consequences of suburban living are reasonably weD known, this article is devoeed to the negative con sequences. Most of the wort of sociologists focuses on negative consequences for tbe metro-1t11JG1U 1100iCJHGS 'AUONS C0N0ntWm 150 politan area as a whole and its social, economic, and political functioning F?ur negative consequences have been fairly weU-
PAGE 155

is particularly a consequence for teenageiS. The walking environment of the low-density American suburb is vi.rtually the sole environment for !he teenage resident, yet it typically consists of little else !han homes placed far enough apart that even access to local friends is difficult. There is little diversity or variety of activities: the best that usually is offered is a shopping center where teen agers are made to feel unwelcome if they are just banging out, and perhaps a fast food restaurant. Other potential negative consequences are "sensory underload" and the "fall of public man." The suburban trend of differentiation of residential areas by stages of the life cycle--wilh families smgle adults, and the elderly inhabiting entirely separate up the "round of life" and may have negative consequences for young people. S-PSYCHOLOGICAL LITERATURE Zlnam, Oleg 1989. "Quality of Ufe, Quality of the Individual, Technology and Eronomlc Development." American ]tJu171fll of &onomks and Socw/ogy 48 (1): SS-68. This article relates Maslow 's ( 1970) hierarchy of needs to components of quality of life. These needs and quality of life components are: I. Physical-safety of natural habitat. 2. Peace-security. 3. Physiological-material well-being. 4. Reputation, Love, Belonging ness-social harmony and justice. 5. Independence-freedom. buman rights, and dignity. lUTGfi.S aROOIQNGS PAASON$ ECONOt!HWUT 151 6. Collective Self-actualization cultural heritage and consensus on values. 7. PeJ:SOnal Self-actualization moral perfection. It is generally accepted that !here is a direct positive relationship between qual ity of life and quality of the person; that a higher quality of life improves the quality of the peiSOn in a self-reinforcing man ner. Yet, there is ample evidence of the possibility of an inverse relationship: that a higher quality of life may reduce the quality of the peiSon (i.e. moral decay) or that a lower quality of life may increase the quality of the peiSon (i.e. "adversity builds character"). TRANStr COOf>BIADVf RESEAACH PROGRAM (TCRP) /1.1 0

PAGE 156


PAGE 157

Andrews, Marcellus. 1994. "On lbe Dynamles or Growth and Poverty in Cities," Ciliscape, Vol. 1, No. 1 (August), pp. 53-73. This article presents a model of bow poveny concentrations within cities are related to city growth rates. 'The central theme of this article hold that the logic of meritocracy creates class divisious in lbe urban labor market which may undermine the very conditions that make rapid economic growth possible" (p. 53). The need for high-skilled workers in a modem high-tech economy creates two classes of workers: those with the requisite skills, and unskilled workers. But schools in many large cities are failing to provide their students with the skills needed to be in the first class. This creates a caste-like result, because the primary determinant of the school perfonnance of children is the educa tional level of their parents. The basic dynamic is as follows: The "underclass" within the ctUes suives to attain the standard of living and jobs suitable for high-skilled workers, but are frustrated by their inability to do so because of the poor quality of city schools. The life styles of the middle class have a demon stration effect upon the underclass, encouraging them to want to consume more. The resulting frustration leads to criminal behavior and violence on the part of the underclass. This results because members of this class have only two sources of income in the model-transfer payments and crime. That behavior drives middle-class (upper-tie r) workers and households out of the city into the suburbs where they can escape from crime and violence. The departure of the middle-class weakens the fiscal position of the city government, thereby reducing its aurGEIIS UOOIONGS PAJSONS ECONOlTHWEST 153 ability to provide good quality schooling to the underclass This createS a negative downward spiral-a "vicious circle. A key variable in this dynamic system is the "middle-class ratio"-that is, the percentage of the total population consisting of middle-class residents. Another key variable is the attitude of students towards academic achievement The author argues that membership in the undenclass causes anti-academic attitudes among students. There is a "critical failure ratio" among city students which determines wbether the middle class will grow or decline within the city. H the actual failure lllle among students (which detemtines whether they will become middle-class or under-class members) rises above this critical rate, that causes the middle class ratio to decline because the behavior of the uoderclass, then larger, drives middle-class residents out. If the actual failure lllle is below the critical level, then more students graduate into the middle-class, and the incentives for middle-class residents to leave is reduced--even though greater compe tition in the labor marlcet among the larger numbers of middle-class workers may cause the unemployment rate to rise. The author regards this entire situation as a negative extemality-;m unintended consequence of teChnological change that bas raised the skill requirements for high -wage workers, where society pro vides unequal access to learning among its young people. Thus, "the increasing importance of knowledge capital in economic growth conuibutes to the problem of urban poverty." (p. 63) The futwe of the cily, and panicularly its ability 10 change the way it grows, may COOmATM: RESWCH PROGRAM (TCRP) /1.10

PAGE 158

ultimately depend upon the willingness of the nUddJe cJass to remain in the city despite the difficu1ties of caste division and crime that owe the undmide of the role of knowledge c:apital in economic life. In a national government policy that encourages the exodus of middle-elass citizens from the city may make significant urban reform and I'!ICOOs!niCtion impossible. (p. 63) The Federal Government must recognize the role of knowledge capital in unwittingly eucerbating the Uiban crisis. In panicular any urban policy that intends to malce cities into virtuOus ci:cles must ..oognize the folly of forcing local governments to deal with the negative tspn. New York: HarperColllns. This is a strictly narrative analysis of metropolitan area trends advancing the thesis that U.S. metropolitan settlements are splitting apart into "old cities'' and "new cities." It covers muc h of the same ground as Anthony Downs's New Visions for Metropolitan America but in a much less systematized, tolally non quantified way. He proposes redirecting a lot of future growth into older cities where they bave been "emptied out," and integrating new and old cities with strong public transit nerworks. Barnett's analysis is heavily skewed towards physical design, since he is an axchitect and wban planner. He artacks strip commercial development in suburbs and advances many of the ideas of the "new urbanism." He favors lUTGBS aaOOtONGS PARSONS ECONOntWE.Sr compact development over continued sprawl. He favors tree preservation ordinances and other environmentally sensitive regulations over trying to do away with regulations altogether. A second major section of his book deals with older core areas. He traces their historic development and shows why the desire of the rich to live away from the poor, combined with transportation improvements, caused a withdrawl!} of resources from the center of our metropOlitan areas. Atllaeting new investment to the bypassed areas of the older city is also the other side of the coin of policies to restrict growth at the urban fiinge. One wiU not work without the other. (p. 118) He argues tha! downtowns have been growing, but the rest of older cities has been shrinking. The mMket for a new subUib in derelict parts of an old city is likely 10 consist of people from nearby areas who have Slarted to make a little money, plus people whose other housing cboice is a small bouse or a mobile-home way out on the urban fringe (p. 146) The minimum requirementS [of successful inner
PAGE 159

Reintegrating the metropolitan area is necessary for the survival of cities, suburbs, and the regional oro-system. (p. I 75) Barnett's weakness is that he does not indicate how to do this, or grapple with the political forces involved He claims there have been major changes in the environment for metropolitan development, including the following: The addition of design methods to the practice of planning. Community panicipation 10 plan ning. The rise of the conservation edlic and the concept of sustainability. Environmental conservatism. We need positive planning about how to grow in the future. But: (L)ocal governments ..e not acaJStomed to malcing affirmative decisions about which an:as of the nal\lrlll 1andseapc ougbt to be preserve and which areas should be built up (p. 191) The basic components of any city design are the organization of public open space including strWS. plazas, and parks or gardens---4he architecl\lrlll relationships among buildings. and the composition of building mass in relation to the landscape or the skyline. (p. 193) The most difficult and cenlral problems of urban design today [are] reconciling tall buildings with lower structures, or the need to ineorponote parking and highway viadUCts within a physical fabric defined by streets and buildings. (p. 196) Experience has 1od city desig:nas to seek to reestabtisb the primacy of the s
PAGE 160

Bradbury, Kalharble L, Anthony Downs, and KtDMtb Small. 1982 Urban and the FuJUrc of American Ci&s. Washington D.C.: The Brook lnp Institution. This book presents definitions of the concept of "wban decline" and relaled variables that could be used in our sprawl study. Since this book presents by far tbe most systematic analysis of uroan decline ever done, its melhods of thinkiog about decline may be useful in our owo sprawl study We will need some measure of deciine to compare statistically with measure s of sprawl to determine what relationships exist between them. if any The following malerial is taken mainly from chaprer 3. The defmitions used in this book are based upon tbe concept that every city has certain specific social functions (set forth in the book). and lbetefore changes in its ability t o perfonn those functions coostillltes urban d ecline. In contrast, a low level of ability to perform those same funcdons-a static concept-<:onstillltes urlxln distress. Some cities with high wban distress are not declioiog but may even be growing rapidly-such as cities with high poverty rates but high immigration. The specific index of city urban decline used in this study is based upon changes in four variables They are the unemployment rate, per capita income, the violent crime rate, and the government debt burden Since there are c hanges between two dates, values of these variables must be compiled for two dates. In tbe present study, this would probab ly be 1980 and 1990. The inda of c ity urban decline was based upon ranking all cities for each of these variables, and assigning points to each city based upon its relattve positron in the ranking for each variable. Cities in the lowest third (in terms of desirability) liiTOElS atOOIONGS PAISONS CONO.THWEST 156 received a I for that variable ; cities in the highest third, a +I, and cities in the middle third, zero. Then the scores of each ci ty for all four variables were added up. The highest possible index score was thus +4 and the lowe s t was 4 A similar index was computed for city urban distress. This index used was based upon five variables, each for a single date. They were the UDemployment rate, the incidence o f poverty, the violent crime rate the pen:ent of housing considered old, and the city's tax effon relative to that of its metropolitan area. I t is notable that neither city population change nor city employment cbange was used as part of the decline measure. This was done because not all population declines are bad (if the city is overcro wded to s tart), and bec ause the authors used declining population as a separate measure that they related to the index of decline thus constructed The unemployment rate captured some aspect of employment change anyway. Two other measures were computed in thi s study. City disparity was a m eas ure of the difference between each cenlral city's scores for these variables and the score of its subwban areas. City divergence was a measure of the rate of change in city disparity over time. This book contains a relevant discussion of the future of large cides. It points out that there are both self-reinforcing and selflimldng factors involved in urban decline, but that the former seem to be much more powerful than the latter H ence the concept of a self-reinforcing downward spiral of deciine is validated by its analysis. Therefore. in reviewing the impact of sprawl upon core areas, the chapten of this book on the future of cities and possible policies to be followed to make that future better should be reviewed. TRANSIT COOI'ERA!IVE RfSEMOf rl!OGMM (7CIIPJ If. I 0

PAGE 161

Callhorpe, Peter. 1993. Tlu Nert Amer can Mtropqlis: Ecology, Community, and the Americtzll Dream. New York: Prlnc:eton Arcbltectural Press. This is a book by an architect and urban planner about the spirit of American communities. and the "new w:banism" approach to altering that spirit The primary approach is through urban design rather Chan quantified analysis, lhougb there is some of the latter too. As the author says, "Social integration, economic effic i ency, political equity and environmental sustainability are the imperatives which order my thinki11g about the form of community" (p. II). He contrasts those themes to the excessive privatization and individual ism be believes bave been embodied in the suburban development process in the post1945 period. "The scale of our environment is now set in proportion to large institutions and bureaucracies rather Chan community and neigh borhood" (p. 11 ) "Tbe suburb was the physical expression of the privatization of life and specialization of place which marks our time" (p. 9). The alternative to sprawl i s simple and timely : neighborhoods of housing, parks, and school$ pbloed withln walking diswlce of shops, civic services jobs, and transit-a modem version of the traditional cown. (p. 16) At the core of thls alternative, p hi losophi c ally and practically, is th e pedestrian. . Pede$1rians are the lost meosure of a community, they set the seale for both center and edge of our neighborhoods. Two complementary strategies are ncc:ded. A tough regio n al plan which limits spnowl ard channels development bock to the city or around suburban transit stations; and a matching greenbelt straJegy to preserve open space at the edge of !be region. We cannot revitalize inner cities without changing !be OlllGaS llOOIQNGS PARSONS ECONOllHW(Sf 157 patterns of growth at !be periphery o f metropolitan regions: i t i s a simple matter of the finite distribution of resources (p. 20) Like most planners, be dislikes the automobile and the scaling of the urban landscape to accommodate it. He wants to change the scale to base it on pedestrians walking to suburban transit stops and linked to downtowns by ttansit. He wants to make housing units and lots smaller, link them by walking, and encourage in-law accessory housing units. He strongly supports regional growth management channeling growth inward to infi11 sites and limiting o u tward extension. nus calls for regi<>nal policies IIlii govcrnan<:e which can both educate and guide the complex interaction of economics, ecology, jurisdiction, and soc i al equity. (p. 32) Adding transit oriented new towns and new groW1h areas can the city' s role as !be region's cultural and economic center (p. 32) Three enlightened developers, and inn er-city advocate&--nd common purpose in regional planning goals 'J'bey ean f orm a powerful coalitio n ." (p. 36) Identifying rational infill and revitalization districts. New Growth Areas and potential New Town sites should be the work of an agency wh.ich $pans the numerous cities aod counties withln a meu-opolitarl area. Lacking such entities_ counties. air quality boards, and regional transponation agenci es often take on !be taSks w i thout legal power to fully implement the results. Regional governxne
PAGE 162

Suburbs are built upon a fundamentally wrong spirit and orientation: The rise of the modem suburb is in part a manifestation of a deep cultural and political wft away from public life . . Soci ally. the bouse fonress represents a sclf-tulfil1lng prophecy. The more isolated people become and the less they sham with others unlike themselves, the more they do bave to fear. . The private domain, whether in a c:ar, a home, or a subdivision. sets the direction of tbe D>Odem suburb . . In one of the primary obstacles to innovations in community planning temoins the impulse towatd a more gated and priv.ue world. (p 37) His design stralegy is based upon three major principles: FlfSt . the regional stnlct= of growth should be guided by the expansion of transit and a mme compact urban form; second, ... our ubiquitous singleu se zoning sbould be replaced with standards for mixed-use, walkable neighborhoods; and third, . our urban design policies should aoate an architeclllre oriented toWatd the public domain and human dimension rather than the private domain and auto scale. (p. 41) He adv ances tbe concept of tbe TOD, or Transit Oriented Development-a basic building block in his regional development scheme. I t features "pedestrian pockets" around transit stops, within one-ijuarter of a mile-an easy walkable distance. These poc k ets contain mixed uses including commercial centers and public services. Farther out from tbe stations are secondary areas containing mostly housing He believes automobile usage in such communities would be much lower tban it is now, because more people would walk to things. Tbere would be both Uiban TODs and Neighborhood TODs (for lower-density tUTGUS n()()(INGS PAISONS ECONOliHWUT 158 literature Review areas). Average residential densities of 1 0 units per acre should be maintained to support bus service, witb higher densities to support rail transit. In another spot, he =mmends net densities of 18 units per acre. He would like a 40-60% split between transit and autos, but tba! still implies a majority of travel by autos. His larger regional scheme show s transit s tops one mile apart. Each TOD around such a stop contains 288.5 acres-a cirole of 2,500 feet in radius A key is what fraction of tbe land should be used for housing. At 40%, tba! would be 115.4 acres; a1 65%, it would be 187.5 acres. Then what average density of housing would pre v ail? He suggests a range from 10 to 25 units per acre. In another place, Caltborpe says Neighborhood TODs should have minimum densities of 7 units per net acre {5,600 persons per square mile) and a minimum average of 10 units (8,000 persons per square mile-just a bit higher tban tbe city of Los Angeles). In Urban TODs, tbe minimum should be 12 and tbe minimum average 15, witb maximums set by local plans. AI 15 units per net acre, tbe gross density would be 15,600 persons per square mile witb a residential land coverage of 65 percent, and 12,000 witb coverage of 50 peroent-the coverage used for tbe other statistics earlier in this paragraph. Secondary areas should have a minimum average density of 6 units per net acre, or 4,800 persons per square mile witb 50 percent residential land coverage. This is therefore tbe minimum permissible density anywhere in tbe developed region. Much of tbe book sets forth design guidelines for parks, oommercial areas, transit stops, and a set of specific projects done by Caltborpe embodying TRANSIT COOI'fi!AT1Vf RESEARCH PROGRAM (rCRP) H-10

PAGE 163

his ideas. Clark, Charles S. 1995. "Revitalizing the Cities: Is Regional Planning the Answer?" CQ Rtseio. New York: Basic Books. This book presents two different views of suburbia as phases in the historic development of modem urban life. They involve the original suburb, and the post-industrial "tech.noburb." The original suburb as deftned by Fishman, was a retreat from the diversity of the city into an exclusively residential community located outside of the complexity of industry and commerce and high-density residences that bad previously been the modem industrial city It flrst appeared in the London area in the late eighteenth century, but became more prominent in the nineteenth and twentieth centuries, both in England and in America. Suburbs of this type were exclus i vely residential areas occupied almost entirely by the middle-clas s elite, excluding all industry and commerce, and all lower income households. Tbey were a retreat from the ills of city life into a more utopian scene linked to oature through TRANSIT COOI'ERADVE R$EARCH PROGRAM (TCRPI H-10

PAGE 164

the use of single-family homes with their own yards. Suburban life was direclly family-antered, and separated middle-i:lass women from the world of work into a world exclusively focused on the family. The suburb in this view was a specialiud bedroom community, the employed residents of which commuted into the central city downtown or its industrial areas and never worked in suburos themselves. Exclusion was oz the he4Jt of the suburbs as thus conceived. That means exclusion of industry, commerce, diversity, jobs for women, and low income households-all conceived as potential threats to the primacy of the family-centered, nature-linked single family home in a park-like setting. This situation bas gradually evolved into a completely different urban arrange ment structured around what Fisbman calls the tecbnoburb, but this member of the research team prefers to call the urban network form. What most people conceive of as the suburbanization of America be considers a shift into a radically different form that totally undermines the suburbs-and the old central city-using the original defmition of suburbs as given above. Although suburbs maintained their specialized roles as bedroom communities into the 1950s, after that the migration of all other types of activities besides housing into suburban areas changed the basic nature of those communities. As they acquired first shopping facilities, then warehouses, then industrial ftrms, and fmally offices, they lost their exclusively residential character. They became transformed into fully urban communities but without any single center, and with very low densities. lbis transformation was made possible by automobiles, roadways, and communications innovations that made decentralization feasible. lUTGaS llOO
PAGE 165

household from all others, although it does provide some commonality of experience across multitudes (which may be undennined by the multiplication of channels). Fishman believes we are still worldng out the cultural impli cations of this new form, although it does seem to have the drawback of isolating the poorest people in older core areas with low fiscal resources. He says: The new cil)l wjJI probably never be able to compete culruntlly wilh lhc old centers. Thc!:e will be for lhe fore=al>le future a division founded on choice between !hose who seek out even at great cost lhe kind of culruntl excirtment that can only be found in the center, and !hose who choose the familycenl<:led life or the ourtr eil)l. (p. 202) Fishman bas underestimated the degree to which cultural activities can taJce place in the outer regions of such networks, because people with common cultural interests can still gather together in outlying locations in sufficient numbers to support cultural activities like symphonies, !beaters, etc. Seen in historical pen.pective, suburbia now appears as the point of transition between two dccentraliud ens: lhe p
PAGE 166

the social benefits of accumulated knowledge. Lucas focused this idea on returns to bwnan (private) capital, but the truth must be that both private capital and general social knowledge gain from innovations in the long run. 1bese ideas are related to cities because people living and working close together can more easily tap into the store of accwnulated knowledge, and exchange ideas with each other The externalities of knowledge exchange are clearly facilitated by urban proximity, as opposed to its alternatives. Barro regressed growth in per capita GOP against a whole lot of other variables across countries, and discovered that poor governmental qualities are negatively correlated with rapid growth. His basic findings were that education and absence of regulation were positively correlaled with rapid growth. Rauch found that SMSA cities with high levels of human capital had both higher property costs and higher w a ges than other cities, holding individual traits constant. Glaeser and others arrived at the following findings: (I) initial concen tration in an industry does NOT seem to foster later creativity, therefore scale economies in a local industry do not really create growth, (2) urban diversity is positively related to later growth, and (3) more competitive industries grow more quickly. In general equilibrium theory, real differences in incomes among cities should be quickly eliminated by migration of workers and capital-any remaining differences should reflect negative amenities in the higher-income cities that must be offset by higher incomes lUTGEIS ll
PAGE 167

into account, because they can retaliate against you in the future, and you must live with them for a long time This enables residents in stable neighborhoods to more strongly reinforce good behavior than residents in unstable areas. It is also an argument for sub sidizing homeownership, which creates greater residential stability. Cities also foster proximity to political power, which is concentrated there. That may influence people to undertake actions to change the behavior of key authorities located in cities. Political agitation is much more likely to work in cities than in rural areas for that reason-also there are more people to get agitated per erg of effon than in rural areas. One of the most critical challenges in the future is reducing infonnational barriers between ghettos and downtown power centers. Subwbanization provides many of the benefits of urban agglomeration while avoiding many of the negative impacts thereof, such as high rates of crime and greater probability of rioting and less residential stability in local neigh borhoods to inhibit negative behaviors. lblanfeldt, Keith R. 1995. "The Impor tance of tbe Central City to the Regional and National Economy: A Review of the Arguments and Empirical Evidence;' Citiscape, Vol. 1, No. 2 (June), pp. 125-150. This article reviews most of the literature on the linkages between central cities and suburbs. The author cites five basic linlcages as follows: (1) outsiders' perceptions of the attraction of an entire metropolitan area are influenced by conditions prevailing within its ceotral city, (2) cities contain many amenities RUTGSlS lltOOIIINGS PAISOtiS ECONOmWfST 163 valued throughout their regions, (3) individual cities may provide a "sense of place" valued by both their residents and outsiders, (4) fiscal problems in central cities may eventually raise taxes on suburbanites and thereby reduce suburban economic development, and (5) agglomeration economies create special roles for central cities in their regional economies. Tbe author does not cite two other linkages that are believed to be i.mponant: cities provide low-cost housing for low-wage workers employed in-and necessary foractivities in suburbs where those workers cannot afford to live, and cities provide many jobs held by suburban residents that increase suburban incomes. The author claims thai there is no empirical evidence either supponing or denying the first four factors he cites; therefore he deals with them in two pages He does not deny that these linkages exist, but says that no one knows bow strong or important they are because there have been no studies measuring them. So he devotes most of his anicle to agglomeration economies, which have been Studied at length and by many persons. Agglomeration economies are essentially increasing returns to scale in processing activities they are "the economies of large-scale production, commonly considered, [and] the cumulative advantages accruing from the growth of industry itself-the development of skill and no-how; the opponurtities for easy communication of ideas and experience; the opponurtity of ever-increasing differentiation of processes and of specialization in human activities (P. 128, quoted from Nicholas Kaldor1970) JRANSrT COO!ftAnVE RfSEAAOI PROGRAM (IO!P}If.IO

PAGE 168

Agglomeration economies are divided into two types. Locali
PAGE 169

Cal thorpe's essay is a very condensed version of his book, discussed earlier. Andres Duany and Eliza beth PlaterZ yberk have an essay about the neighborhood, the distric t, aod the corridor I t is only a few pages long and is not relevant to this study. Elizabeth Moule and Stefanos Pol yzoides have an essay about the street, the bloclc and the building This scale is too micro for thi s study. Todd W Bressi has an essay entitled " the American Dream" (pp. xx v xlli) It discusses the overall approach of the ''new lllbanism" practitioners repeating much of what is in Caltborpe's book. He claims thai the suburban explosion after World Wars I and D achieved certain desirable out com es, at heavy costs It "re inforced the Victorian notion thai a neigh borhood was a pro!edive eDC!ave requiring insulation from commeroe work, aod traffic and held thai the functional and literal ce nter of a neighborhood should be an elementary s c h ool." (p. Xxvil) The suburbaoiza tion movement "liberaJed many people from crowded, unhealthy living conditions." But il created the foll owing problems: (I) raising the cost of homeownership and decent housing too high for many households, (2) forcing people to spend more and more lime commuting [thi s point is false], ( 3 ) undermining the mobility o f people who cannot afford cars or cannot drive them, ( 4) creating air pollution, (5) absorbing atlr.ICI:ive rural landscape into urban uses, and m ost important of all but most problematic--{6) underm i ning civic life Tbe main principles of the new urban ism as he describes them are as follows : Tbe c enter of e;ocb neigbborbood should be defined b y a public space and activated by locally oriented civi c and comme rcial facilities Each neighborhood should accommodate a range of h o usehold typeS and laod U!eS. Cars should be kept in per>pective Architecture should respond to the surrounding fabric of buildings and spaces and to local traditions New urbanists draw upon several past tradition s includin g the City Beautiful and Town Plann ing mov ements. Caltborpe has wriueo lhat in lheory 2,000 bomcs, a million square fed of COIJIIDOR:ial parks, scl>ools and cloy ca.: could fit within a qtWit:rmile walk o f station, or about 120 acres (p. xxxi) Calthorpe introduced the secondary resi dential district only when others complained about the lack of space for single-family housing in his other TODs The strategy of the new urbanists is to change local c odes in detail to compel the following of their principles, or at least to pennit it. It has in fact been difficult to integrate TODs with transit, since most areas do not have rail ttansit systems. Some critics claim that the new wbanists emphasize visual style over substan ce Tbeir large-scale proposals seem to continue sprawl rather than changing it. And the imPact of this approach at a large scale will be minimal unless some type of regional governance is more widely adopted Finally, the new urbanists have largely ignored the growing divisions of wealth and power among h ouseholds The New Urbanism is a welcome step forward. but it is only a step. (p. dii)

PAGE 170

The rest of the book is a series of illustraled case srudies thai are too detailed to go into in these notes. KUDSder, James Howard. 1993. The Geography of Nowhere: The /Use and Decline of America's Man-Made Land"'P' New York: Simon aDd Schuster. Kunstler has written a polemic-a true "exagger-book"-about the aesthetic and other qualities of metropolitan development in the United States, especially during the post world war n era. 'The tone of this book can be conveyed by the following quotations from chapter I: More and more we appear to be a nation of overfed clowns living in a hostile cartoon environment Eighty peroent of everything ever built in America has been built i n lhe last fifty years. and most of it is depn>ssing, brulal, ugly unhealthy, and spirirually degrading. To me, it is a lAndscape of SJ:MY. places, lhe geography of nowhere, that has simply cased to be a credible human habitat. These are not objective or even believable statements about reality, but they convey the spirit in which Kunstler denounces everything American. In fact, there is nothing about American life thai appeals to Kuostler. He attacks individualism, low-density develop ment, business, you name it, he hates it. Another quotation: Riverside seems a remplare for all the ghastly automobile suburbs of the postwar ..,._ individual houses on big blobs of lAnd along curvy slreet.S. (p. 49) Yet, for all !heir artificiality and impermanence, the early railroad suburl>s wore lovely places to live. RIITGERS &ROOIONGS ARSONs ECONOI!IHWUT 166 He decries artchitectural modernism and the art..
PAGE 171

We should adopt !he approach of the new urbanism in designing small towns He specifically discusses Seaside and Peter Cal thorpe's pedestrian pockets as cures for all the ills he has been blasting. Mandatory open space zoning is also praised. Until we do these things, "the standard of living in the United States is apt to decline sharply, and as it does the probability of political ttouble will rise (p. 274) We'U have 10 sive up our fetish for extreme individualism and rediscover public life .... We will have to down scale our gigantic enterprises and institutions--corporations, governments, banks, schools, hospitals, marlcel$, farms-and learn to live locally, heoce responsibly He offers no guidance concerning how to go about doing these things. Ledebur Larry C.. llDd WUiiam R. Barnes. 1992. MtroJH1lit4n and &onomil: Growth: Cily Distnss and th. Ned for a F .Ural Loeal Growth Pacluzg. Washington D.C.: National League of Cities. Marcb. This is a swistical srudy of the rela tionship between income disparities between central cities and their suburbs on the one band, and metropolitm area growth rates on the other. 1be basic conclusion is that "During the period 1988-1991, metropolitan aneas with greater internal disparities tended to perfonn less weU economically than metropolitan aneas with lesser disparities" (p. 1) Overall. central city per capita income as a percentage of subwban per capita income bas declined from 105% in 1%0 to 96% in 1973, 89% in 1980 and 59% in 1987. Much of this article aims at justifying a substantial federal aid lUfOIU llaOOICIHG$ PA.UONS ECOHOilliWm t67 pacb8e to cities, especially cities in distress. Data on children being raised in poverty. by race, are presented. In 1990, 45% of all black children under the age of 4 were being raised in poverty, compared to 38% of Hispanics and 20.6% of all children These proportions were higher in central citie s and lower in suburbs Ra\'ll<:b, Diane. 1996. ''The Problem of the Schools: A Pl-oposal for Renewal in JuUa VIIWio-MU'tia, ed., Away : Tlu Fumn of Ciliu. New York: Twentieth Century Fund. pp. 77. The author criticizes New York' s schools because they are run by a topheavy bureaucracy thai makes all decisions centrally, and leaves almost no authority within individual schools themselves The results are teniblt>oDiy about SO% of all students who enter high school graduate, even within 5 years We are now demanding that our schools educate all young people, something that was never done in the past, but must be done in order to prepare them for life in a higb-th world. To do this, we must abandon centralized control and change to a system in which "each school must be managed by a group of adults who have direct, personal, and professional responsibility-and accountability-for the success of their students." (p. 81) It may be that the be$1 direction for reforming the schools is to see1: a c!ivetSity of providers that are publicl y monitored, rather than a bureauaatic system controlled by the mandltes of a single SO>
PAGE 172

Auronomy-Eacb school shoul d control its own budget and hite (and fire) its own teachers and other personnel. Each should be told bow much money it bas (based upon enrollments plus allow ances for disadvantaged students) and allowed to allocate that money as it saw fit-knowing that it would be rigorously audited by public officials. ChoiceTeachers should be able 10 freely decide where they will work. and students and parents should be able 10 decide where they want 10 send their cbildten 10 school. Quality-The centraliml authorities will set standards for perfonnance, periodically assess perfoll11llllCes of every school, and constandy inform parents and the public of the results. Central authorities would also oversee large capital improvements. negotiate union conttaCIS (without inhibiting schools from hiring whomever they wish), approve the creation of new schools, and audit perfOll11llllCe and fmaoces. Schools that want to manage their own affairs shoul d be allowed to conduct elections among staff and parents 10 become chartered schools. and immediately be given autonomy. This would pemlit successful schools 10 become self-governing right away. A second element of the strategy would be contracting of the management of several or many schools by specifi c organizations. A basic idea is encour aging as many new schools to be formed as possible A third element in the strategy is to provide means-tested scholarships 10 poor student s who could then choose to use them in whatever schools the y wanted. These would essentiaUy be vouchers paid to the students or their parents, and therefore not to the institutions themselves thereby finess in g the re ligious school issue, as is now done in many other programs. IUTGEl$ OROOKINGS PAISOHS fCOHOCIHWEST 168 Richmond, Henry R-1 995 Rgionalism: C/Uazgo As An Amorie<1n Rgion. CbJc:oo1o: Tbe Jobn D And Catherine T. MacArthur Foun dation. December 6. Tbls reviewer believes that this is the most comprehensive intellectual attack on sprawl yet launched. Henry Rich mond, one of the architects of the Oregon s!JIIe planning system, has coUected every known argument against sprawl into one long polemic--but a re latively sensible one, not a totally irrational one. Among the arguments be marsbaUs against sprawl are the following: Sprawl concentrates poverty in areas, undermining their flSCal viability. Tbls concentration also produces a host of other negative conditions It undermines the transition of the inner-city mskilled worlcforce to a high -tech worlcforce. It thereby weakens the international compe titive positions of U.S. metropolitan areas. It reduces the efficiency of businesses and the productiv ity of land. It undermines equality of opportunity within metropolitan areas, thereby raising in.ner-city unemployment with aU the resulting pernicious effects. It deslroys the viability of inner-city schools, which contributes t o the failure to make the proper labor force transition. It breeds crime that drive s viable firms and households out of cities, and weakens the ability of young people raised there to sustain themselves economically Sprawl undermines security, especiaUy of ho useholds whose investments in home equities are jeopardized by racial transition

PAGE 173

Sprawl damage;o; the environment in tenns of air pollution, water pollution, ruining historic buildings. and wrecking environmenlally sensitive sites. Sprawl undennines the community of life in subwban areas. and the solidarity of our entire society by separating s u burban residents from city ODC;'l. Sprawl makes urban development inefficient by generating indecisive governments, disputes, delays that add to costs. etc. Ri c hmond believe;o; that a huge number of public policies 81 aD levels have generated sprawl, and maintain it in existence. He catalogs these 81 length. He tben presents a political analysis of why these forces are not likely to change. After having set fonb aD these points in general, he applie;o; the argument t o the Chicago region in detail. He tben sets fonb his J'CCOIIUJlendations for how to anack sprawl and the many institutional suppons underlying it Here be bas also comes up with a more comprehensive set of ideas than anyone else 'This document is an invaluable refetence for both arguments against sprawl and possible tactics to remedy it. lt bas not been given wide;o;pread publicity, but is one of the most important documents in the entire field Rusk, Da vid. 1 995. Ciliu WIJhout Sub urbs. 2d eel Washington, D.C.: The Woodrow Wilson Centu This book is a detailed and compre hensive look a1 sprawl and a1 least one of its altemalives by the fonner mayor of Albuquerque, New Mexico Its basic tbe;o;is is that cities which have elastic boundaries-thai is, they can annex lUIGUS IIOOICIHGS PAUONS ECONOlllWfST 169 snrroundi:ng teaitories-are much healthier than cities which hav e inelastic boundaries-that is, their boundaries are because they are surrounded by mcorporated suburban muni c ipalitie s. The elastic cities can expand outward as metropolitan areas grow. thereby reta1D10g access to the new taxable bases created outside the original boundaries of those cities by such growth. In contrast, inelastic cities cannot reach out to new taxable resources as growth expands beyond their borders Both elastic and inelasti c cities have disproportionale shares of poor people within their original boundaries but the former can counteract this by expanding boundaries. Inelastic cities are stuck w i th rising percentage of poor residents and falling tax bases. 'This causes them to have faDing taxable resources per capita 81 the same time thai they need more such resources to cope with rising percentage of poor residents. Rusk presents a grea1 deal of statistical information to support his claim that elastic cities are bealthier cconomicaDy and social l y than i nelastic ones He docs not use regression analysis, but ratbet presents paired city com paris ons and compares average s of groups of c i ties with different degrees of elasticity, as be measures it This book is one of the most comprehens i ve and intelligent analyses of the spra w l problem and othe r urban problems yet written However, it has one serious flaw. Its author believes that unified metropolitan government is th e best solution for inelasticity. whereas I believe thai there is no political support for that arrangement whatever. Even so, Rusk's analysi s is definitely one of the best studies of urban problemsespe<:iaDy concerning three aspects. The first is Rusk s fearless confron tation of racial aspects of these 1lANSlT COOI'aAJ!Vl USEAIICH PaOGAAM (TCIIrJ H-1 o

PAGE 174

problems. The second consists of tbe four ingiedieolS be recommends to solving the problems thai be describes. They are (I) n:giooal governance of land-use planning, {2) regional tax-base sharing, (3) a regional program of crealiog desegregated affordable hous ing for the poor throughout each region, and (4) promotioo of regioo-wide economic developmeoL The third is bis analysis of the "point of no rerum'' concx:ming three variables for cerural cities: (I) a low ratio of per capila income in them to thai in their suburbs (70 pereeot or less), (2) a bigh fractioo of minority-groups in total population (30 percent or more), and (3) substantial and sustained population loss (of 20 percent or more). He claims thai no city that bas passed beyond aU three of these thresholds bas ever even begun to recover. Sdar, Elliot, 8Dd Waltu Hook. 1993 "lbe lmportaDce of ClUes to tbe National Econom y," In Htlll')' G. Cisn eros, ed., Inurwov111 Dntinks: Ci&s tmd the Nlllion. New York: The American Assembly of Co iiiDlbia University. pp. 1-26 'This is the lead article in a volume of essays presented at tbe 82nd American Assembly held at Arden House, in Haniman New York, in April 1993. The authors clearly regard central cities as vital center of production in the American economy. They complain thai most policy analysts in recent decades have viewed cities mainly as homes for the poor. They cite the foUowing facts as suppott for their view oo the importance of central cities: Central city jobs constitute the bigher paying jobs in most metro areas. Such jobs form 32.2% of all jobs nationally, but gamer 37.7% of nationwide earnings (no source for KOOJCINGS rAISONS EOOHOilHWESI t70 this datum is cited ). Wages of central city jobs average 20% higher than those of suburban jobs. and thi s gap has been widening. Many suburban residents have jobs in central cities. A survey by Arthur Goldberg of the suburban areas of the nation's I 00 largest cities showed that half of suburban famil ies had at least one worker in the central city. Tbc same survey showed heavy suburban depeodence on central city services 67% of suburban residents depeod on the city for major medical care, 43% have family members attending or planning to attend an inst i tution of higher l earning in the city, 46% believed their property values would be hurt by a serio us decline in their central city The top 24 counties accounted for 39% of aU jobs in information intensi ve industries but bad only 27 percent of total jobs. Downtown job pay for jobs in the same categories as the sub urbs was 3.55 times higher in Boston and 2 .37 times highe r in New York City The production advantages of central cities include (I) transponation and communication s cost for both workers and customers are minimized in central location, (2) the best analysis is conducted in diverse environments where experts have easy face-to-face contact willl one anolller, {3) central location s facilitate international transactions because of superior telec -ornmunicalions infrastructures, and ( 4) more specialized producer services can be located where the size of the market is greatest. One reason suburban locations continue to grow faster is thai the costs of moving are not fully borne by the moving flrDIS, but by their employees and public taxpayers. If subwbanization were so efficient, we would see more of 7aANSfT COOPfRAllVf RESEARCH (7Clr)lf.JO

PAGE 175

it in our international competitor nations. U.S. urban policy was more concerned with stimulating demands for consumer products-such as housing and autosthan with productive efficiency. Subwbanization has been encouraged by biased public policies, such as home taX deductions and federal highway finance-not reflected in public transit aids until very recently. 1be nature of pricing of telephone and other services allowed higher..::ost suburban service to be priced at the same rates as lower..::ost city services. Continued dispersal bas major costs to society, especially on the inputs of private fltms. Ooe is the need for employees to own automobiles, which increases wage demands. Auto dependence also increases our trade deficit because we must impon so much oil. We spend more on 1ravel and telecommunications thai rival nations do. 1be Japanese spend far less on transponation than we do----9.4% vs. 15-22% Traffic congestion imposes high costs on production. These authors claim thai most metrOpOlitan areas devote over half of their available land to road infrasttucture (I doubt it). By undermining the taX base of central cities. our society has been unable to invest properly in the education and training of the labor force, or in the infrastructures outside the downtown that are critical to productive efficiency. U.S. invesunent in education through the high school level is the lowest among the seven most industrialized nations-4.1% of GNP compared to 4.6% in West Germany and 4.7% in Japan. We need much more investment in lbe labor force and infrastructures in central cities to remain competitive. RUTGO.S lllOOKINGS PAlS""' ECONOllliWEST 171 Thompson, J, Phillip. 1996. ''Urban Poverty and Race," in Julia VitulloMartia, Editor, BNaking Away: The Futun of C:Ws. New York: Twentieth Century Fund, pp. 13-32.. This author discusses the status of poverty and its relationship to race in inner..::ity areas, with main reference to New York City. He points out thai the middle-<:lass is still dominant in most large American cities, but it has become a minority-group middle class as whites leave the city.ln six of the nation's eight largest cities, a majority of lbe popu lation in 1990 consisted of minority group members-only Philadelphia (48% minority) and San Diego (42%) are exceptions. In New York the number of persons with incomes above lbe median remained about the same in the 1980s, but changed ethnic compo sition to become minority-dominated because the white group in this category fell by 432,000. He reviews various theories of why poverty persists in inner-city neigh borhoods. The cu/Jura/ deprivation theory stresses that some families are less intelligent than others, and a deprived culture is panly a genetic phenom enon. A newer view is that poor families are stuck in poor com munities, where conditions are ripe for a negative subculture to develop around excessive teenage sexual promiscuity, a separate street lan guage, and a depreciation of academic achievement. Thus both views stress deviancy and immorality of behavior among many poor people, with the newer theory attributing it to spatial isolation of the poor and especially of poor blacks from white culture. Christopher Jencks claims that cennrries of racial subordination and prejudice have TRANSIT COOPfRAIIVE IIESEAROf PROGRAM (IOtPJ fl. I 0

PAGE 176

created a black unwillingness to do certain types of work or to work in white cultunl environments Black alienation from cenain types of jobs is rarely discussed in analyses of poverty. The racilll discriniiMIWn theory says that black poverty in particular is caused primarily by continued racial discrimination and resulting spatial segregation Massey and Denton are ad vocates of this view, since housing discrimination isolate$ poor blacks with eacb other in conceoltaled poverty neighborhoods But discrim ination itself is not new; so how can it exp lain rising crime ntes or family instability, which are recent devel opments? Massey and Denton claim that cultural deprivation results from spatial isolation, which in turn is caused by white pre judice and discrimination The structvrrJI tTtuos/ontullion tluory claims tbat blaclc u nemployment results from a change in labor markets and industry that shifted jobs to higher-skill requirements and moved industrial jobs out of big cities where racial minorities lived. William Julius Wilson is a leading proponent of this view. But unemployment does not explain many of the olher pathologies of inner-city poverty areas. Wilson claims the departure of middle-class blacks from poverty areas removes good examples, and the resulting negative cu lture comes from economic deprivation and lack of jobs. But is it not c lear !hat culrural traits of blacks, rather lhan discrim ination by whites, causes whites not to bire black worl
PAGE 177

Local-iented strategies include enter prise and empowennent zones 10 improve conditions where the poor live now. The purpose is to create "vibrant" businesses where poor unemployed people already live. Community based efforts fit into this view and many such efforts are now underway across the nation. Building local housing is one of their major activities A whole host of questions is raised by the author that might be answered by more careful study of community activities now underway. The author then explores why the election of black mayors and city officials has not improved conditions in inner-<:ity neighborhoods very much, if at all. And black political participation in politics has not been raised by black leadership. Why not? Among the reasons he cites are ( 1) black mayors have no control over national trends for decentralization of jobs, (2) the shift of population to the suburbs has reduced the national political power of big-<:ity mayors of all typeS in Congress and in state legislature, reducing their willingness 10 aid cities, (3) city needs 10 maintain favorable tax rates and bond ratings prevent mayors from engaging in redistributive activities-as observed by Paul Peterson in City Limits, (4) fear of being charged with racism has prevented criticism of black local leadership by either whites or blacks, and (5) the civil rights movement has become conservative and did not shift from national issues to local ones to support black local leaders. lflJD's rules against building public housing in poor communities have blocked the efforts of black mayors 10 put new low-rise public housing units in inner-<:ity poverty areas, thereby up grading those areas. In New York City, court actions have prevented giving preference in public housing projects 10 lUTGO$ UOOKING$ PAl$0NS ECONOI.ntWUT 173 persons living in nearby communities, thereby strengthening those commu nities Voting district formation bas reduced representation by minorities on city councils and in Congress. Struggles over crime rates pit civil rights advocates-who want less incarceration of blacks-against local residents who want more secure neighborhoods. Similar struggles have occurred con cerning schools, where those wanting better schools have tried to shift disruptive students into separate "academies" that are opposed by traditional civil rights advocates. The problems of inne<-<:ity poverty d
PAGE 178


PAGE 179

Aberger, Will, and L Propst. 1992. Successful Communicies: Managing Growrh co Protect Disrinctive Local Resourcts Washington, DC: The Conservation Foundation Abrams, Charles. I !n I. The Longuage of Cities. New York: Viking Press. Adams, Thomas, and Edward M Bassett, and Roben Wbitten. 1929. Neighborhood and Community Planning: Regional Survey, Vol VII Committe on Regional Plan of New York and Its Environment. Al-Mosaind, M.A., K. 1 Dueker, and 1. G SualhmaD. 1993. "Light Rail Transit Stations and Propeny Values: A Hedonic Price Approach." Transponation Research Record 1400: 90-4. Altshuler, Alan A. 1977. "Review of 'The Costs of Sprawl."' JoumJJl of the American Planning Association 43, 2: 207-9. Altshuler, Alan A., and Jose A. Gomez-Ibanez. 1993. Regulation for Revenue: The Political Economy of Land Use Exactions. Washington, DC: Brookings Institution. American Farmland Trust (AFT). 1986. Density-Related Public Costs. Washington, DC: American Farmland Trust. American Farmland Trust. 1992a Dots Fannland Protection Pay? The Cost of Community Services in Three Massachusens Towns. Washington, DC: American Farmland Trust. American Farmland Trust. 1992b. The Cost of Community Services in Three Pioneer Connecticut Valley Towns: Agawam, Deerfield, and Gill. Washington, DC: American Farmland Trust. Anderson, WiUiam P., Pavlos S. Kanaroglou. and Eric J Miller. 1996. "Urban form. Energy and the Environment : A Review of Issues, Evidence and Policy." Url>an Studies 33, 1: 7-35. Andrews, James H 1996 "Going by the Numbers." Planning (September 14-18). Andrews, Marcellus. 1994. "On the Dynamics of Growth and Poveny in Cities." Citiscape I, I (August): 53-73. ll/IGOS I(OOOIG$ PAlSON$ K'ONOOtlHWEST I 75 Apogee Research, Inc. 1994. The CosiS of Transportation: Firull Report. Conservation Law Foundation. March Archer, R. W 1973. "Land Speculation and Scattered Development: Failures in the Urban Fringe Market. Orban Studies 10 3 : 367-72. Arcbimore, A. 1993. "Pulling the Community Bac k into Community Retail." Urban Land 52, 8: 33-8. Arendt, Randall. 1994 Designing Open Space Subdivisions: A Practical Step-by-Step Approach. Natural Lands Trust, Inc. Armstrong, R. J Jr. 1 994. "Impacts of Commuter Rail Service as Reflected in Single Family Residential Propeny Valu e s Pape r presented at the 73rd Annual Meeting of the Transponation Research Board, Washington, D.C. Arrington, G. B., Jr. 1995. Beyond the Field of Dreams: Ligh t Rail and Growth Management in Ponland. Portland, OR: Tri-Met. Audirac, L, A H. Shennyen, and M. T. Smith 1990. "Ideal Urban Fonn and Visions of the Good Life: Florida's Growth Management Dilemma." Journal of the American Planning Association 56 (Autumn): 470-482. Audirac, Ivonne, and Maria z;rou. 1989. Urban Development Issues: What is Contro versial in Urban Sprawl? An Annotated Bibliography of Often-O v erlooked Sources Council of Planning Librarians. CPL Bibliography 247. September. Avin, Uri P. 1993. A Review of the Cost of Providing Government Services to Alternative Residential Panerns. Columbia MD: LDR International. Babcock, Richard F. 1966. The Zoning Game. Madison WI: University of Wisconsin Press Bah!, Roy M 1968. A Lond Speculation Model: The role of the Propeny Tax as a Constraint to Urban Sprawl. Journal of Regional Science 8, 2: 199-208. Baldassare, Mark. 1986. Trouble in Paradise: The Suburi>an TransfonnaJion of America. N ew York : Columbia University Press. TRANS" COOI'fAAJJVz; RESEARCH PROGRAM (TOIP)II-10

PAGE 180

Baltimore County. 1978. Growth Management Program Technical Memorandum No. ll, Environmental Asse ssment Towson, Baltimore County, MD. Bank of America, California Resources Agency, Greenbelt Alliance, Low-Income Housing Fund. 1995. Beyond Sprawl: New Paaerns of Growth to Fit tht New California. San Francisco: Bank of America et al. Barnett, Jonathan. 1995. The Fractured Met ropol is: Improving the New City, Restoring the Old City, Reshaping the Region. New York : HouperCollins). Beaumont, Constance. 1993. Vermont. Washington, DC: National Trust for Historic Preservation. Beaumont, C()nstance. 1994. How Supentore Sprawl Can Harm Communities-And What Citizens Can Do About It. Washington, DC: National Trust for Historic Preservation. Beaumont, Constance. 1996. Historic Preservmion. Washington, DC: National Trust for Historic Preservation. Bernick M., and R. Cevero. 1994. "Transit based Residential Development in the United States : A Review of Recent Experiences. Working Pape r 611. Institute of Urban and Regional Development, University of California at Berkeley March. Berry, David, and Thomas Plaut. 1978. "Effects of Urban izatio n on Agricultural Activities." Growth and Change 9, 3: 2-8. Black, J Thomas. 1996. ''Th e Economics of Sprawl." Urban Land 55, 3: 6-52. Blomquist, Glenn C., Mark C. Berger, and lobo P Hoebo. 1988. "New Estimates of Quality of Life in Urban Areas." American Economic Review 78, I: 89107. Bohi, Douglas R., and Joel Darmstadter. 1994. "Twenty Years after the Energy Crisis: What Lessons Were Learned?" Resources 116: 16-20. Bookout, L. W. 1992. "Neotraditional Town Planning: The Test of the Marketplace Urban Land 51, 6: 12-17. lUTGBS llOOt
PAGE 181

Burchell, Robert W ., and David Listoltin. 1978. The Fiscal Impact Handb()()k: Estimating Local Costs and Revenues of Lond Development. New Brunswick, NJ: Center for Urban Policy Research. Burchell, Robert W., and David Listoltin. 1982. Energy and Lond Use. New Brunswick. NJ: Center for Urban Policy R-=h. Burchell, Robert W. and David Listokin. 1990. Fiscal Studies. Report to the Governor's Commission on Growth in the Chesapeake Bay Region, Annapolis, MD: 2020 Commission. Burchell, Robert W .. and David Listoltin. 1994a. "Fiscal Impact Procedures and the Fiscal Impact Hierarchy." Paper prepared for the Association for Budgeting and Financial Management, Annual Conference on Public Budg eting and Finance, Washington D.C. Oc tober. Burchell, Robert W., and David Listoltin. 1994b. "The Economic Effects of Trend versus Vision Growth in the Lexington Mettopolitan Area." Report prepared for Bluegrass Tomorrow, LexJDgton, Kentucky. November. Robert W., and David Listokin 1995a. "Land, Infrastructure, Housing Costs, and FJSCal Impacts Associated with Growth: The Literature on the Impacts of Traditional Versus Managed Growth.'' Paper prepared for "Alternatives to Sprawl" Conference, The Brookings Institution. Washington. D.C. March. Burchell. Robert W and David Listoltin. 1995b. The Economic Impacts of Trend versus Vision Development in the Lexington (Kentucky) Metropolitan Area. Report prepared for Blue Gras s Tomorrow. January. Burchell, Robert W., and David Listoltin. 1996 "Determinants of Municipal and School District Costs." Report prepared for Sterling For est Corporation, Tuxedo, New York. Burchell, Robert W., David Listoltin, and William R. Dolphin 1993. The Development Impact Handbook. Washington, D.C.: The Urban Land Inst itute. RUTGRS IIOOkiNGS PAI.SONS ECONORlHWEST 177 lleratvre Revirw Burchell, Robert W., and Harvey S. Moskowitz. 1995. Impact Assessment of DELEP CCMP versus STATUS QUO on Twelve Muni cipalities in the DELEP Region. Report prepared for the Local Governments Comminee of the Delaware Estuary Program, Philadelphia, Pennsylvania August 15. Burchell Robert W ., and Emilie Schmeidler. 1993. "The Demographic and Social Difference Between Central Cities and Suburbs as They Relate to the Job Ful fillment of Urban Residents." Paper presented at the National Conference on Metropolitan America in Transition: Implications for Land Use and Transportation P lann ing. Cambridge, MA: Lincoln Institute of Land Policy Burnell, James, and George Galster. 1992. "Quality of Life Measurement and Urban Size." Urban Studies 29, 5: 727-35. "Bye -bye Suburban Dream: Fifteen Ways to Fix the Suburbs." 1995. Newsweek. Vol. CXXV, No. 20. April. Calthorpe, Peter G. 1993. The Nex.t American Metropolis. Princeton: Princeton University Press. Cambridge Systematics. 1994. The Effects of Lond Use and Travel Demand Strategies on Commuting Behavior. Washington, DC: U.S. Department of Transportation, Federal Highway Administration. Cameron, Michael W. 1994. Elficitncy and Fairness on the Road: Strategies for Unsnarling Traffic in Southern California. New York: Environmental De-fense Fund. Carlson, Allan 1996. "Two Cheers for the Suburbs." The American Enterprise 7, 6. Cervero, R. 1991. "Land Use and Travel at Suburban Activity Centers." Transportation Quarterly 45, 4 : 479-491. Cervero, R. 1996. "Jobs-Housing Balance Revisited." Journal of the American Planning Association 62, 4: 492-511. Cervero, R. 1 996. "Subcentering and Commuting: Evidence from the San Francisco Bay Area. 1980-1990. Paper presented at the 1996 TRED Conference on Transportation and Land Use, The Lincoln Institute, Cambridge Massachuserts (October). TRANSIT COOPfRATM' RESEARCH PROGRAM (TCIIP) ff.IO

PAGE 182

Cervero, Robert. 1986. Suburban Gridlock. New Brunswick, NJ: Center for Urban Policy Resean:h. Cervero, Robert. 1989. Amuica 's Suburban Activ ity Centers: The lAnd Use-Transportation Link. Boston: Unwin-Hyman. Cervero, Robert 1991. "Congestion Relief: The Land-Use Alternative." Journal of Plmuring Education and Research 10: 119 -29. Cervero. Robert. 1994a. ''Transit-Focused Development: Does it Draw People into Transit and Buses?" /URD Universe 4: 3-S. Cervero, Robert. 1994b. ''Transit Villages: From Idea to Implementa t ion." Access 5: 8-13. Cervero, Robert, M. Bernick, and J. Gilbert. 1994. "Market Opportunities and Barriers to Transit-Based Development in Cali fornia." Worl
PAGE 183

and Margery A. Turner, eds., Housing Markets and Residential Mobility. Washington, DC: The Urban Institute Press. pp. 243-252. DeChiara. Joseph, and Lee E. Koppelman. 1975 Manual of Housing/Planning and Design Criteria. Englewood Cliffs. NJ: Prentice-Hall. DeCorla-Souza, P. 1992 "The Impacts of Alternative Urban Development Patterns on Highway System Performance." Paper presented at Institute of Transportation Engineers International Conference, Washington, D.C. Delucchi, M. A. 1996. The Annualized Social Cost of MawrVehicle Use in the U.S . 19901991: Summary of Theory, Data, Methods, and Results. Davis, California: lnstirute of Transponation Studies. August. Dougbarty, L. A., et al. 1975. Municipal Service Pricing Impacts on Fiscal Position. Santa Monica, CA.: RAND. D owall, David E 1984 The Suburban Squeeze Berkeley, CA: University of California Press. Downing, P. B., and R. D. Gustely. 1977. "The Public Service Costs of Alternative Development Patterns: A Review of the Evidence." In P .B. Downing. ed., Local Service Pricing Policies and Their Effect on Urban Spatial Structure Vancouver, B C.: University of British Columbia Press. Downing, Paul B., ed. 1977. Local Services Pricing and Their Effect on Urban Spatial Structure. Vancouver: University of British Columbia Press. Downs, Anthony. 1970. "Uncompensated Nonconstruct i ve Costs Which Urban H i ghways and Urban Renewal Impose Upon Resident ial Households." In Julius Margolis, The Analysis of Public Outpur New York, NY: National Bureau of Economic Research, Inc Downs, Anthony. 1973. Opening Up the Suburbs: An Urban Stra1egy for America New Haven, CT: Yale University Press. Downs, Anthony. 1985 The Revolution in Real Estate Finance. Washington, DC: The Brookings Institut i on. IUTGOS lllOOOCINOS PJIO$ONS ECON
PAGE 184

Dzurik, Andrew. 1993. "Transportation Costs of Urban Sprawl: A Review of the Literature," State Transportation Policy Initiativ e Center for Urban Transportation Research. November. ECONonhwest. 1994. Evaluation of No Growth and Slow Growth Policies for the Portland Region. Portland, OR: Metropolitan Portland Government. ECONonhwest. 1996. Summary: Technical Repons for 2040 Means Business. ECONonhwest. November ECONonhwest, Parsons Brinckerboff Quade and Douglas, and COMSIS Corporation 1995. Least-Cost Planning: Principles, Applications, and ISJues. Washington, DC: U.S. Depanment of Transportation and Federal Highway Administration. Ehrenberg, R. G., and R. S. Smith. 1994. Modem lAbor Economics: Theory and Public Polic. 5th ed. HarperCollins. El lwood, David T. 1986. "The Spatial Mis match Hypothesis: Ate There Teen-ag e Jobs Missing in the Ghetto?" In Richard B. Freeman and Harry J. Holzer, The Black Youth Emplayment Crisis. Chicago, IL: University of Chicago Press. Environmental Protection Agency. 19 71. Community Noise. Washington, D.C.: U.S. Government Printing Office. Environmental Protection Agency. 1973. The Quality of Life Concept: A Potential New Tool for Decision Makers. Wasbin$10n. DC: EPA, Office of Resear ch and Morutoring. Environmental Studies Divi sion. Etzioni, Amitai. I 993. The Spirit of Community: Rights, Responsibilities. and the Communiurrian Agenda. New York: Crown Publishers. Evaluation of Ciry of San Diego Growth Management Program. 1978. Baltimore County, Maryland. Ewing, R. 1991. Developing Successful New Communities. WashingtOn, DC: Urban Land Institute. Ewing, R. 1994. "Characteristics, Causes, and Effects of Sprawl: A Literature Review." Environmental and Urban Issues (Winter): 115 1.\ITGERS NOOIONGS PARSONS KOHotTHWEST 180 Ewing, R 1995. Best Development Practices: Doing the Right Thing and Making Money at the Same Time. Chicago: American Planning Association. Ewing, Reid, 1995a "Measuring Transportation Performance." Transportation Quarterly 49,1:91-104. Ewing, Reid, 1995b. "Beyond Density, Mode Choice, and Single-Purpose Trips." Transportation Quarterly. 49,4:15-24. Ewing, Reid. 1997 "Is Los Angeles-Style Sprawl Desirable?" Journal of the American Planning Association 63, 1 (Winter): 107-126 Ewing, R., P. Haliyur, and G. W. Page. 1994. "Gelling Around A Traditional City, A Suburban PUD, and Everything In-Between." Transportation Research Record 1466: 53-62 Ewing, Reid, MaryBeth DeAnna, and Sbi Chiang Li. 1996. "Land-Use Impacts on Trip Generation Rates." Transportation Research Record 1518: 1-7. Fannie Mae. 1985, 1989, 1992, 1994, 1996. Survey of Residential Satisfactwn of Housing Occupants. Washington, DC : Federal Nati onal Mortgage Association. Fannie Mae. 1996. National Housing Survey. Washington, DC: Federal National Mortgage Association. Feagin, Joe R., and Robert Parker. I 990. Building American Cities: The Urban Real Estate Game. 2d ed. Englewood Cliffs NJ: Prentice Hall. Federal Bureau of In vestigation. I 996. Crime in the United States. U.S. Department of Justice, Federal Bureau of Inves tigation Press Release. October 13. Fishman, Robert. 1987. Bourgeois Utopia : The Rise and Fall of Suburbia. New York: Basic Books. Fischel, William A 1985 The Economics of Zoning LAws: A Property Rights Approach to American Land Use Controls. Baltimore: The Johns Hopkins University Press. Fischel, William A. 1990. Do Growth Controls Maner? Cambridge, MA: Lincoln Institute of Land Policy. May. JRANSIT COOPERAIIVE RSEAIICH I'ROGl!AM (TCRP) fi.IO

PAGE 185

Fischer, Paul B. 1991. "Is Housing Mobility an Effective Anti-Poverty Strategy? An Examination of th e Cincinnati Experience. Cincinnati OH: The Stephen H. Wilder Foundation. Florida Advisory Council on Intergovernmental Relations. 19%. Impact Fees in Florida Tallahssee, FL. November. Fodor. Eben V. 1995. "The Real Cost of Growth in Oregon." Preliminary Repon. Prepared by Energy & Environmental Planning Associates, Eugene, Oregon. October. Foster Kathryn A 1996. "Special i zation in Government : The Uneven Use of Special Service Districts in Metropolitan Areas ." Urban Affmrs Review 31, 3 : 283-313. Frank, James E. 1989. The Costs of Altemalive Development Panems: A Review of the Liter ature. Washington, DC: Urban Land Institute. Frank, L. D., and Gary Pivo. 1994. The Re/otionship Betwun Land Use and Travel Behavior in the Puget Sound Region. WA-RD 351.1. Olympia, W A: Washington State Depanment of Tran s portation Freilich, Roben H 1980. Saving the Land: The Utilization of Modem Techniques of Growth Management to Preserve Rural and Agricultural America. Illinois Fanners Home Adminisuation. Freilich, Roben H., El i zabeth A. Garvin, and S. Mark White. 1993. "Economic Development and Public Transit: Making the Most of the Washington Growth Management Act." 16 University of Puget South Law Review 949 (Spring) Freili ch, Roben H., and S Mark White. 1991. "Transportation Congestion and Growth Management: Comprehensive Approaches to Resolving America's Ma j or Quality of Life Crisis 24 Layola University of Los Angeles Low Review 915. Fried, Carla, Lesl i e M Marable, and Sberyl Nance-Nash 1996. "Best Places to Live In America." Money (July): 66-95. Gabriel, Stuan A., Joe P Maney, and WiUiarn L Wascher. 1996. Compensating Differentials and Evolution of the Quality-of Life Among U.S. States. San Francisco: Federal Reserve Bank of San Francisco. 96-07 June. 21 pages. ttJIGEIS BI.OOO
PAGE 186

Gordon, Peter, and Harry W. Richardson. 1994a. "Sustainable Congestion," in J. Brotchie ed .. Cities in Competition: The Emergence of Productive and SllSiainoble Cities for the 2I st Century. Gordon, Peter, and Harry W. Richardson. 1994b. "Geographic Factors Exp laining Worktrip Length Changes." Prepared for the U.S. Department of Transportation, Federal Highway Administtation. Gordon, Peter, and Harry W Richardson. 1994c. "Congestion Trends in Mettopolitan Areas." In Curbing Gridlock: Peaking Period Fees w Relieve Traffic DJngestion. National Research Council Washington, DC; National Academy Press. 1-31 Gordon, Peter, and Hany W. Richardson 1995. "Beyond Polycentricity: The Dispersed Mettopolis, Los Angeles, 1970-90." Jourtu:Jl of the American Planning Association 61. Gordon, Peter, and Harry W. Richardson. 1996b. "Los Angeles Among Other CMSAs: Outlier or the Nonn?" Environment and PlaMing A. Gordon, Peter and Hany W. Richardson. 1997. "Are Compact Cities a Desirable Planning Goal?" Journal of the American Planning Association. Vol. 63, No. I Winter 1997. American Planning Association, Chicago, IL. Gordon, Peter, Hany W. Richardson, andY. Choi. 1992. "Tests of the Standard Urban Model: A Micro (Trade-off) Alternative." Review of Urban and Regional Development Studies 4, I : 50-66. Gordon, Peter Harry W. Richardson and Myung Jin Jun. 1991. "The Commuting Paradox: Evidence from the Top Twenty." Journal of the American Planning Association 57: 416-420. Gondiener, Mad; 1977. Planned Sprawl : Private and Public Interests in Suburbia Beverly Hills, CA: Sage. Gottlieb, Paul D. 1995. "Residential Amenities, Finn Location and Economic Development." Urban Studies 32, 9): 14131436. lii!GEltS IWOIONGS ,ARSONS ECONOITHWEST 1 82 Governor's Commission on Growth in the Chesapeake Bay Region. 1991. Fiscal Studies. Baltimore MD: Maryland Office of Planning Greene, David L., and Donald N. Jones. 1995. "The Full Costs and Benefits of Transportation : Conceptual and Theoretical Issues." Paper presented at the Conference on the Full Social Costs and Benefits of Transportation. Greenwood, Michael J., Gary L. Hunt, Dan S. Rickman, and George I. Treyz. 1991. "Migration, Regional &juilibrium, and the Estimation of Compensating Differentials." The American Economic Review 81, 5: 1382 1390. Growth Management Planning and Research Clearinghouse (University of Washington). 1993. Literature Review of Community Impacts and Costs of Urban SprawL Washington, DC: National Trust for Historic Preservation. Haines, V. 1986. "Energy and Urban Fonn: A Human Ecological Critique. Urban Affairs Quarterly 21, 3: 337-53. Hall, Bob and Mary Lee Kerr. 1991. 199I-1992 Green Index: A Stare-By-Stale Guide to the Nation's Environmental Health. Washington, D.C.: Island Press. Handy, S. 1992. "Regional Versus Local Accessibility: NeoTraditional Development and Its Implic ations for Non-Work Travel." Paper prepared for the 1992 Annual Meeting of the Association of Collegiate Schools of Planning. Handy, S. 1994. "Highway Blues: Nothing a Little Accessibility Can't Cure.'' Access 5 : 3-7. Handy, S 1995. "Understanding the Link Between Urban Fonn and T r avel Behavior Paper presented at the 74th Annual Meeting of the Transportation Research Board (January). Hanson, M E. 1992. "Automobile Subsidies and Land Use: Estimates and Policy Responses." Journal of the American Planning Association 58, 1: 60-71. Harris, C. 1992. "Bringing Land Use Ratios into the 90s." PAS Memo. Chicago: American Planning Association. Hamson, Bennett. 1974. "Discrimination in Space: Suburban i zation and Black Unemployment in Cities In George M. von Furstenburg, Bennett Hamson, and Ann lRANSIT COOPERATIVE RESEARCH AAOGRAM (tCRP) H-10

PAGE 187

Horowitz (eds. ), Patterns of RacU.I Discrimination Vol 1: Housing. LexingtOD9 MA: Lexi ngton Books Hartshorn, Truman A and Peter 0. Muller. 1989. "Suburban Downtowns and the Transformation of Metropolitan Atlanta's Business Landscape," Urban Geography I 0: 375-395. Hartshorn, Truman A., and Peter 0. Muller. 1992. "The Suburban Downtown and Urban Economic Development Today." In Edwin S. Mills and John F. McDonald (eds.), Sources of Metropolitan Growth. New Brunswick, NJ: Rutgens University, Center for Urban Policy Research Harvey, G. 1990. Relation of Residential Density to VMT per Resitknl: Oalcland Metropolitan Transponation Commission Harvey, Robert Q and W.A.V. Clark. 1965. "The Nature and Econ o mics of Urban Sprawl." Land Economics 41, 1: 1-9 Heildcila, E et al. "Wbat Happened to tbe CBD Distance Gradient? Land Values in a Polycentric City." Enviro1111U1nl and Planning A 21: 221-32. Heiimlich, Ralph E 1989. "Metropolitan Agriculture Fanning in the City's Shadow." Journal of the American Planning Association 55, 4: 457-66. Herbers, J. 1982 The New Heartland: America's Flighl Beyond the Suburbs and How It Is C/uJngin g Our Future. Herbers Publishers. Herson, Lawrence J R. 1992. ''The New Suburbanization: Challenge to the Central City Annals of the American Academy of Political and Social Science 522: 184 Hidlebaugb, Allen R. 1981. Agricultural lAnd Data Sheet: America's Agricultural Land Base. National Agricultural Lands Srudy, Washington, DC. Hiss. Tony. 1990. The Experience of Place. New York : V i ntage Books. Hodge, G. 1981. "The Citification of Small T owns: A Challenge to Planning." Plan Canada 21: 43-7. RUTGERS PAISONS ECONOllKWEST 183 Holtzc law, J 1990. Explaining Urban Density and Tronsit lmpocts on Auw Use Paper presented to tbe State of California Energy Resouroes Conservation and Development Commission by Natural Resources Defense Council and the Sierra Club. April 19. Holtzclaw, J. 1994. Using ResidentU.I Patterns and Transit to Decrease Auto Dependence and Costs San Francisco, CA: Narural Resources Defense Council. Howard County (Maryland) Planning Commission 1967. Howard County: 1985. Howard County, MD: Howard County Plann i ng Commission April. Hughes, Mark A. 1991. "Employment Decentralization and Accessibility-A Strategy for Stimulating Regional Mobility." Journal of the American Planning AssocU.tion 57, 3: 288-98. Hurnmon, D. 1990. Commonplaces: Commu nity Ideology and ldenlity in American Cui ture. B i nghamton NY: SUNY Press. lblanfeldt, Keith R. 1995. "The Importance of the Central City to the Regional and National Economy: A Review of the Arguments and Empirical Evidence," Citiscape 1, 2: 1 2 5-150. June. Ihlanfeldt, Keith R., and David L. Sjoquist 1990 "Job Accessibility and Racial Differences in Youth Emp l oyment Rates American Economic Review 80: 267-76 Institute of Transponation Engineers. 1984 Recommended Guidelines for Subdivision Streets. Washington, D.C. : ITE. lsard, Walter, and Robert E Coughlin. 1957. Municipal Costs and Revenues Resulting from Growth Wellesley, MA: Chandler-Davis Jackson, Kenneth T. 1976 "The Effect of Suburbanization on tbe Cities." In P. Dolce, ed Suburbia: The American Dream and Dilemma. Garden City, NY: Doubleday. Jackson, Kenneth T 1985. Crabgrass Frontier : The Suburbani1.Qlion of the United States. New York: Oxford University Press. Jacobs, Jane. 1961. The D e ath and Life of Great American Cities. New York: Vintage Books. lRANSrr COOI'fl!ADVE RESEARCH PROGRAM (TCRP) /1.10

PAGE 188

Jarrell, Stephen and Roy M. Howsen. 1990 "Transient Crowding and Crime: The More 'Strangers' in an Area. the More Crime Except for Mwder, Assault and Rape American Joui'Ml of Economics and Sociology 49, 4: 483-494. Johnson, Thomas G. 1987. "Fiscal Impac ts of Economic Development." in Responding to the Crisis in the Rural South: Highlighls of Selected Publicand Privale-Sector lnitialives. Southern Rural Development Center, Mississippi State Univers ity. Jon es. P. 1989. "Regional Shopping CentersPlanning Issues." Town and Country Planning 58, 10 : 280-2. Jones, L. L., and M. D. Woods. 1985. "Deter mining the lmpacts of New industry on Small Towns in Texas Small Town (September Oc tobe r) : 22-5. Joseph Nadine. 1992. "Mommy Myths; Altitudes Toward Working Mothers." Redbook (October): 130. Kelbaugh, Douglas. 1993. "The Costs of Sprawl." Cascadia Forum 1: 20-26. Kain,John. 1967. Urban Form and the Costs of Urban Services. Cambridge, MA: Harvard University. Kain, John. 1988. "Choosing the Wrong Technology: Or How to Spend Billions and Reduce Transit Use." Journal of Advanced Transportalion 21. Kaio. John 1991. "Trends in Urban Spatial Structure, Demographic Change, Auto and Transit Use, and the Role of Pricing." Statement prepaned for the United States Senate, Committee on Environment and Public Works. Kain, John F. 1992. "The Spatial Mismatch Hypothesis: Three Decades Later." Housing Policy Debate 3, 2: 371-459. Kaman Sciences Corporation. I 972. lAnd Use Planning for Air Quality in the Pikes Peak Area. Colorado Springs. CO: Kaman Sciences Corporation. August. Kasarda, John D. 1990. "City Jobs and Residents on a Collision course: The Urban Underclass Dilemma." Economic Development Quarterly 4, 4: 313 19. RUTGBS llOOKINGS PAISONS ECONOll>iWEST ISA Katz, Peter 1994. The New Urbanism: Towards an Architecture of Community. New York: McGraw Hill. Katz, L . and Kenneth Rosen. 1987. "The lnteljurisdictional Effects of Growth Controls on Housing Prices." Journal of Law and Economics 30 (April): 14960. Kingsley, 0. Thomas. and Margery Austin Turner, eds. 1993 Housing Markets and Resid e ntial Mobility. Washington, DC: The Urban institute Press. Kitamura, R., L. Laidet, P Mokhtarian, C. Bucltinger, and F. Gianelli. 1994. Mobility and Livable Communities. State of California Air Resources Board. Kitamura, R., P. L. Mokhtarian and L. Laidet. 1994. "A Micro-Analysis of Land Use and Travel in Five Neighborhoods in the San Francisco Bay Area." Institute of Transporta t ion Srudies, Universily of California, Davis November. Knapp, Thomas A . and PhiliJ? E. Graves. 1989. "On the Role of Amemties in Models of Migration and Regional Development." Journal of Regional Science 29 (1): 71-87. Koslowsky, Meni, and Moshe Krausz. 1994. "On the Relationship Between Commuting, Stness Symptoms. and Anirudinal Measures: A LISREL Application Journal of Applied Behavioral Science 29, 4: 485-492. Kotkin Joel. 1996. "Make Way for the Urban Confederates." The American Enterprise 7, 6. Kotval, Zenia, and John R. Mullin. 1992. A Fiscal Impact Analysis Technique: The Pawling ModeL Hadley, MA. Krieger, Alex, ed. 1991. Towns and Town making Principles: The Work of Andres Duany and Elizabeth Plater-Zyberk. New York: Rizzoli International Publishers. Krote, lames J 1992. "Is Downtown Worth Saving?" Planning 58, 8: 148-62. Kunstler, James Howard. 1993. The Geography of Nowhere: The Rise and Decline of America s Man-made lAndscape. New York: Simon & Schuster. !RANSIT COOPfRAJM: RfSfAAOf PIIOCRAM (TCJIP/11-1 0

PAGE 189

Kunstler, James. 1996a "Home From Nowhere: Zoning Laws' Effects on Quality of Life." The Atlantic 278, 3: 43. Kunstler, James. 1996b. Statement made at Michigan Society of Planning Officials Annual Conference. Lansing, Michigan. Spring. Ladd, Helen F 1992. ''Population Growth Density, and lhe Costs of Providing Public Services." Urban Studies 29, 2: 273-96. Ladd, Helen F., ed. 1991. "Causes and Consequences of lhe Changing Urban Fonn." Regional Science and Url>an Economics 21 (July): 157-315 Special Issue. Ladd, H. F., and J. Yinger. 1989. America's Ailing Cilies: Fi.scal Heallh and the Design of Urban Policy. Baltimore, MD: Johns Hopkins University Press. Lafferty, R. N., and H. E. Frech. 1978. "Community Environment and the Market Value of Single Family Homes: The Effect of the Dispersion of Land Uses." Journal of Law and Economics 21: 380-393. Landis, John D. 1994. "The California Urban Failures Model: A New Generation of Metropolitan Simulation Models." Environment and Planning 21: 399-421. Landis. John D. 1995. "Improving Land Use Futures: Applying the California Urban Futures Model." Journal of the American Planning Association 6!, 4 (Autumn): 438-457. Landis, John D and David S Sawicki. 1988. "A Planner's Guide to the Places Rated Almanac Journal of the American Planning Association (Summer) : 336-346. Lang, Roben E . and Steven P Homburg 1997. "Planning Ponland Style: Pitfalls and Possibilities." Housing Policy Del>ate 8, 1 (April) : 1-16 Langdon, Philip. A Bttter Place to Live: Reshaping the American Suburb. !994 Amherst: University of MassachusettS Press. Lansing, John B., et. al. 1970. Planned Residential Environments Ann Arbor, Michigan : Institute for Social Research, University of Michigan lllrGElS U.OOt
PAGE 190

Logan, J R . and H M oloteh 1987 Urban Forrwu:s: The Poli tical Econo"!)'_ of Place. Berkeley. CA: Unive rsity of Cahfo mia Press. Lope:t, Rigobeno A .. Adesoji 0 Adelaja, and Margaret S. Andrews. 1988. "The Effects of Suburbanization on Agricullure." American Journal of Agriculture Economics 7 0 2: 34658 Lowry. I. 19 88 "Planning for Urban Sprawl. n In Transporrmion Researr:h Boord Special Report Z21: Looking Ahead. the Year 2()(}(). Washington, DC : T ransportation Research Board. MacKenzie. J. J . R C Dower, and D. D T. Chen 1992. The Going Rate: What It Really Co sts to Drive. Washington, DC: World Resources Institute M c Dooald, John, Charles Orlebeke. Ashish Sen and Wim Wiewel. 1992 "Real Eswe Development and Property Taxes in DuPag e County : Fmal Repott Project 11342. Cbiicago. U.: U niversity of lllinois, School o f Urban Planning and Policy, Center for Urban Economic Development February. M c Kee, D L., and G H. Smith 1972. "Environmental Diseconomies in Suburban Expansion." The Ameri can Journal of Economics and Sociology 31, 2: 181-88. McMiUan, David W and David M Chavis. 1986. "Sense of Community : A Deflllition and Theory Journal of Community Psychology 14, 1: 6-23 Maher, Ian. 1992 Commuling C a lculations." Journal of the American Plan nin g A ssociatio n 58: 386-8 7 Maryland Office of Stat e Planning. 1989 E nvironmental and Economic Impacts of Lot Site and Other Devewpm ent Standa rds. Baltimore, MD: Office of Sta t e Planning. M aryland -National Park and Planning Commission and Mont gomery County P lanning Depanment. 1989. Comprehensive Growth Policy Study A Policy Vision : Centers and Trails. Silver Spring, MD. Mas low, Abraham Harold. 1 970 Motivation and Personality. 2d ed New York: Harper & Row. lVTGEIS U.OOICJ'r4GS ,AISONS ECONOmiWIST 116 Meek, Stuart 1996. "Pl anning Practice P lanni n g (Novembe r) M etro. 1994 Region 2040 Recommended Alternative Decision Kit. Portland OR. Meyer I., and J A. Gomez-lbane :t 1981. Awos, Transit, and Citi tS. Cambridge : Harvard Univenity Press. Meyer S M 1992. Environmenralism and Economic Pro>perity: Testing the Enviro n menrallmpact Hypothesi> Cambridge MA: MIT Press M ichigan Society of P l anning Officials 1995. Patt ems on the Lond: Our Choicts, Our Fwure. Roche s ter, MJ: Michigan Society of Planning Offi cia ls Middle sex-som erset Mercer Regiona l Council and The Regional Plan Association 1994. Redesignin g the Suburbs: Tumlng Sprawl into Prinoeton. New Jersey M iddlesexSomerset M ercer Regional P lanning B oard 1990 MixedUse Devewpment : Traffic and Other lmpocts. Princeton, New Jersey. Mieszlcow s ki. Peter, and Edwin S. Mill s. 1993. "The Causes of Metropolitan Suburbaniza tion." Journal of Economi c Perspec tives 7, 3: 1 35 47 M ills, David E 1981. "Growth, Specu l ation, and Spraw l in a Monocen lric City Journal of Urban Economics 10: 201-26. Mills, Edwin S. 1991. "Criti q ue o f Impacts of Development on DuPage County Prorny Taxes."' Kellogg G raduate School o Management, Nottbwestem University June. Mills, Edwin S and Bruce W. Hami lton 1994. Urban E co nomics New York : HarperCollins. Mills, Edwin S., and John F. M cDonald. 1992. Sources of Metropolitan Growth New Bruns wick, NJ: Center for Urban Policy Research Mioess, D., and R C. Einswciler 1 992 Managing CO!IIIIUlllily Gr(llllth and Cluutge. Cambridge, MA: Lincoln Institut e o f Land Policy M obility f o r the 21st Century : A Blueprint for th e Future. 1996. The American Public Transit Associa tion's Mobility for the 21st Century TRANSIT COOI'fRAJIY lESENCH ,_OGAAM CTCJ'l H-to

PAGE 191

Task Force and Robert L. Olson, Institute for Alternative Futures October. Moe, Richard. 1995. Quoted in Dwight Young, Alternatives to Sprawl (Cambridge: Lincoln Institute of Land Policy, the Brookings Institution, and the National Trust for Historic Preservation). p. 4. Montgomery County, Maryland. 1989 "Alternative Scenarios: Analysis and Evaluation." Moore, Terry, and Paul Thorsnes. 1994. The TratUport4lion-Land-Use Connection: A Framework for Practical Policy. Planning Advisory Service Report 448/449. Chicago: American Planning Association Muller, Peter 0. 1986. "Transportation and Urban Form: Stages in the Spatial Evolution of the American Metropolis." In Susan Hanson, ed., Geography of Urban TranspoTU1tion. New York: Guilford Press. Myers, Dowell. 1987. "Internal Monitoring of Quality of Life for Economic Development Economic Devtlopment Quarterly 1 : 238-278. Myers, Dowell. 1989. "The Ecology of 'Quality of Life' and Urban Growth." In Understanding Growth Management: Critical Issues and a Research Agenda. Washington, DC: Urban Land Institute. Nasar, Jack L., and David A. Julian. 1995. "The Psychological Sense of Community in the Neighborhood." Journal of the American Planning Association 61, 2: 178-184. Nathan, Richard P., and Charles Adams. 1976. "Understand ing Central City Hardship." Political Science Quarterly 91, I. National Trust for Historic Preservation. 1993. "Potential Effects of a Mega-Store in Kent County." Report prepared by James Hendry. National Trust for Historic Preservation, Washington, DC. Nelessen, Anton C. 1994. Visions for a New American Dream: Process, Principles, and an Ordinance to Plan and Design Small Urban Communities. Chicago: American Planning Association. RUTGUS HOO
PAGE 192

Novaco, R. W., D. Stokols, J. and J. Stolcols. 1979. ''Transportation, Stress, and Community Psychology." American Journal of Community Psychology 7: 361. Novaco, Raymond W ., Daniel Stolcols, and Louis Milanesi. 1990. "'bjective and Subjective Dimensions of Travel Impedance as Determinants of Communting Stress." American Journal of Community Psychology 18, 2: 231-257. Nowlan, David, and Greg Stewart. 1991. "Downtown Population Growth and Commuting Trips : Recent Experience in Toronto." Journal of the American Planning Association 57: 165-82. O'Connor, Kevin. 1988. Suburbia Makes the Central City: A New lnterpreiDiion of the CilySuburb Relation.ship. Berkeley, CA: Instirute of Urban and Regional Development, University of California at Berkeley Office of Planning and Research. 1982. Paying the Piper: New Ways to Pay for Public Infra structure in California. Sacramento, CA. December Obis, James C., and David Pines. 1975. "Discontinuous Urban Development and Economic Efficiency." Land Economics 3 (August): 224-34. 1000 Friends of Oregon. 1995. LUTRAQ. Vol. 6. Ponland, Oregon. 1000 Friends of Oregon. 1996. Making the Land Use Tran.sportation Air Quality Connection: Analysis of Alternatives. Vol. 5. Ponland, OR. May. Oregon Progress Board 1994. Oregon Benchmarks: Standards for Measuring Statewide Progress and Institutional Performance. Salem, OR: Oregon Progress Board. Repon to the 1995 Legislarure December. Orfield, Myron. 1995. Metropolities : A Regional Agenda for Community and Stability. Minneapolis-St. Paul: Metropolitan Area Research Fund Orlando, Florida. 1981. Urban Area Growth Manaagemenl Plan. RUTGERS IROOIQNG$ PARSONS ECOtllTHWUU' I 88 Orski, C. Kenneth. 1992. "Congestion Pricing: Promise and Limitations." Transportation Quarterly 46, 2: 157-67. O'Sullivan, Anhur M. 1993. Urban Econ omics. 2d ed. Homewood, IL: Irwin. Parsons, George R. 1991. "The Effect of Coastal Land Use Restrictions on Housing Prices." Journal of Environmenral Economics and Management. February Parsons Brinckerhoff Quade Douglas. Inc. 1993. The Pedestrian Environment Ponland, OR: 1000 Friends of Oregon. pp. 30. Parsons Brinckerhoff Quade & Douglas. 1996. Cost of Travel in Boulder The City of Boulder. July 15. Parsons Brinckerboff Quade and Douglas, R. Cervero, Howard/Stein-Hudson Associates, and J. Zupan. I 996. "Commuter and Light Rail Transit Corridors: The Land Use Connection" in Transit and Urban Form. Vol. 1. Transit Cooperative Research Program, Transportation Research Board, Washington, DC. October. Parsons Brinckerhoff Quade and Douglas, R. Cervero, Howard/Stein-Hudson Associates, and J. Zupan. 1996 Influence of Land Use Mix and Neighborhood Design on Transit Demand. Unpublished repon for TCRP H-I project T rans it Cooperative Research Program, Transponation Research Board, Was hington, DC. March. Parsons Brinckerhoff and ECONonbwest. 1996. A Framework for Evaluating the Impacts of Alternative Urban Forms. Washington, DC. Pastore, Ann L., and Kathleen Maguire, eds. 1996. Sourcebook of Criminal Justice Statistics 1995. Albany, NY: H inde lang Criminal Justice Research Center, University of Albany. Peiser, Richard B. 19&4. "Docs It Pay to Plan Suburban Growth?" Journal of the American Planning Association 50: 419. Peiser, Richard B. 1989 "Density and Urban Sprawl." Land Economics 65: 193-204. Petersen, John. 1992. Statement made at Lincoln Instirute for Land Policy Conference. April. Reaffirmed Summer 1996 TRANSrr COOPERATIVE RESfAACH PIIOGAAM (TCRP) H-10

PAGE 193

Pickrell, Donald H. 1989. "Urban Rail Transit Projects: Forecast vs. Actual Ridership and Costs." Urban Mass Transportation Adminis trmion Report. United States Depanment of Transponation. Washington, DC: United States Government Printing Office. Pigou, A. C. 1920. The Economics of Welfare. Macmillan Pisarski, Alan E. 1987 Commuting in America: A Nmional Report on Commuting Patterns and Tr.nds. Westpon, CT: Eno Foundation for Transponation. Pisarski, Alan E. 1992. Ntw Puspectives in Commuting. U.S. Department of Transpor tation. July. Pisarski, Alan E. 1992. Travel Behavior Issues in the 90s. Office of Highway Information Management, Federal Highway Administration. Pivo, G. 1990. ''The Net of Beads: Suburban Office Development in Six Metropolitan Areas." Journal of the American Planning Associarion 56, 4: 457 -69. P lotkin, Sidney. 1987. Keep Out: The Struggle for Land Use ControL Berkeley, CA: University of California Press. Popenoe, David. 1979. "Urban Sprawl: Some Neglected Sociological Considerations." Sociology and Social Research 63, 2: 255-68. Popper, Prank J. 1981. The Politics of Land-Use Reform. Madison, WI: University of Wisconsin Press. Power, T. M. 1980. The Economic Value of the Quality of Life. Boulder, CO : Westview Press. Precoun, Geoffrey, and Anne Faircloth 1996. "Best Cities: Where the Living is Easy." Fortune (November 11): 126-136. Propst, L., and M. Schmid. 1993. The Fis cal and Economic Impacts of Local Conservarion and Community Development Measures: A Review af Litermure. Tucson, AZ: The Sonoran Institute. Pucber, John. 1995. "The Costs of Transit versus the Automob ile." Paper presented at National Trust for Historic Preservation "Costs of Sprawl" conference, Spring 1995. lUTGEitS 611001C1NGS PAlSOtiS ECONOlliWESt 189 Pusbkarev, B and J. Zupan. 1977 Public Transportation and lAnd Use Policy. Bloomington, IN: Indiana University Press Pushkarev, B., and J. M. Zupan. 1982. "Where Transit Walks: Urban Densities for Public Transponation." In H. S. Levinson and R. A. Weant, eds., Urban Transponation: Perspecrives and Prospects. Westpon, CT: Eno Foundation. Ravitch, Diane. 1996. "The Problem of the Schools: A Proposal for Renewal," in Julia Vitullo-Martin, ed., Breaking Away: The Future of Cities. New York: Twentieth Century Fund. pp. 77-87 Real Estate Research Corporation. 1974a. The Costs of Urban Sprawl. Detailed Cost Analysis. Washington, DC: U.S Government Printing Office. Real Estate Research Corporation. 1974b. The Costs of Sprawl: Environmemal and Economic Costs of Altem41ive Residemial Development Pantms at tht Urban Fringe. Washington, DC: U.S. Government Printing Office. Regional Planning Boasd. 1956. "Urban Sprawl." Lower Mainland New Westminster, B.C. April. Resource Management Consultants, Inc. 1989 Development in Wright County, Minnesota: Cost-Revenue Relations/Up. Minneapolis: Resource Management Consultants Rice Center for Urban Mobility Research. 1987 Hou.srons Major Activity Centers and Worker Travel Behavior. Houston: Joint Center for Urban Mobility Research Richmond, Henry R. 1995. Regionalism: Chicago As An American Reg/en. Chicago: The John D. and Catherine T. MacAnhur Foundation December 6. Richardson, H. W. and P. Gordon. 1989. "Counting Nonwork Trips: The Missing Link in Transponation, Land Use, and Urban Policy." Urban lAnd (September): 6-12. Richardson, H. W., and P. Gordon. 1993. "Market Planning: Oxymoron or Common Sense?" Journal of the American Planning Association 59: 347-52. TRANsrrCOOPfRAnVE RESEARCH PROGRAM (TC'RP) 11-10

PAGE 194

Roback, Jennifer. 1982 "Wages. Rents, and the Quality of Life. JouT7UII of Political Economy 90, 6: 1257-1278. Roback, Jennifer. 1988. ''Wages, Rents, and Amenities: Differences Among Workers and Regions." Economic Inquiry 26: 23-41. Robinson, Susan ed 1990 Financing Growth: Who Bonefits? Who Pays? And How Much? Chicago: Government Finance Officers Association, Government Finance Research Center. Rockerfeller Brothers Fund. 1973. In William K. Reilly, ed., The Use of Land: A Citizen's Policy Guith to Urban Growth. New York: Thomas Y. Crowell Company Rosen, Sherwin 1979. "Wage-Based Indexes of Urban Quality of Life." In Currefll Issues in Urban Economics Edited by Peter Mieszlcowslci and Mahlon Straszheim. Baltimore, MD: Johns Hopkins University Press. Rossetti, M. A and B. S Eversole 1993 Journey to Work Trends in the United States and Irs Major Metropclitan Areas, 1960-1990. Washington. DC: U.S. Depanment of Transportation. Rudel Thomas K 1989. Situations and Strategies in American lAnd-Use Planning. New York: Cambridge University Press. Rusk, David. 1993. Cities without Suburbs Washington, DC: Woodrow Wilson Center Press. Rusk, David 1995. Cities without Suburbs. 2d ed. Washington D.C.: The Woodrow Wilson Center Press. Salant, Priscilla, Lisa R. Carley, and Don A Dillman. 1996. Estimating the Contribution of Lone Eagles to Metro and Nonmetro In Migration. Pullman, W A: Social and Economic Sciences Research Center, Washington State Un i versity. 86 19. June Savageau, David, and Richard Boyer. 1993. Places Rated Almanac. New York: Macmillan Travel. Savitch, H. V . et at 1993. "Ties That Bind: Cenual Cities, Suburbs, and the New MettopoUtan Region Economic Developmefll Quarterly 7, 4: 341-57. RUTGW II.OOKINGS PAJtSONS KONOU'HWEST 190 Schaeffer, M. H., S. W. Street, J. B. Singer, and S. Baum. 1988. 'Effects of Control on Stress Reactions of Commuters." JouT7Ull of Applied Social Psychology 18: 944-957. Schafer. Robert. 1975 Exclusionary Land Use Controls: Conceptual and Empiric
PAGE 195

United States Problems and a Major Answer to Poverty Journal of the American Planning Association 61, 4: 496 503 Siemon, Larsen and Purdy. 1990. Crossroads: Two Growth AltemaJives for Virginia Beach. Virginia Beach (VA) Growth Management Sllldy. Chicago, Illinois Skogan, W 1986 ''Fe.v of Crime and Neighborhood Change In A. Reiss and M. Tomy, eds., CommuniJies and Crime. C hicago, D..: University of Chicago l'less Small. K. 1980 "Energy Scarcity and Urban Development Patterns. International Regional Science Review 5, 2 : 97-117. Soli, Michael, and Fritz Wagner. 1991. "Coping with Growth in a Subwban Town: The Case of Mandeville, Louisiana." Small Town (July August) : 18 24. Solomon, Arthur P., ed. 1981. The Prospective City MIT Press. Cambridge, MA Sorensen, A. Ann Richard P. Greene, and Karen Russ. 1997. Farming on the Edge American Farmland Trust (AFr), Center for Agriculture in the Environment DeKalb, D..: Nortbem Dlinois University. Spain, Daphne. 1988. "An Examination of Residential Preferences in the Suburban Era. Sociologi cal Focus 2 1 1: 1 8. Sperling, Daniel. 1995. Future Drive: Electric Vehicles and Sustailulble Transportation Washington, DC: Island Press Steiner, Bill. 1992. 'The Future of Downtowns : Issues to Consider As We Approach 2001." Small Town (November-December) Steiner, R. L. 1994 "Residential Density and T ravel Patterns: A Review of the Literature and Methodological Approach." Paper presented at 73rd Annual Meeting Transportati o n Research Board, Washington, DC. Stemlieb, George, and W. Patrick Beaton 1972. The Zone of Emergence: A Case Study of Plainfield, New Jersey New Brunswick, NJ : T ransaction Books Stemlieb, George, and Robert W. Burchell. 1 977. Residential Aban:dcnment: The Tenement RUTGUS IROOIONGS PAJSONS ECON0
PAGE 196

Tischler & Associates, Inc 1989 "Howard County Maryland : Development Trends and Their Fiscal Impacts." Bethesda, MD: Tischler & Associates Tischler & Associates, Inc 1990. Service Level, Cost and Revenue Assumptions : Evaluation of Development ConceptsHoward County, Mary/4nd. Bethesda, MD: Tischler & Associates. Ti schler & Associates. Inc. 1994. Marginal Cost Analysis of Growth Alternatives King County, Washington. Bethesda, MD. Transit Cooperative Research Program. 1995 "An Evaluation of the Relationship between Transit and Urban Form. Research Results Digest 7 (June). Turque, Bill, and F rank Washington. 1991. "Are Cities Obsolete?" Newsweek (September 9), pp. 42-45. Ulrich, R.S., R.F. Simons. B.D. Losito, E. Fiorito, M.A. M iles, and M. Zelson. 1991. "Stress Recovery During Exposure to Natural and Urban Environments." Journal of EnvironmenUJl Psychology II: 201-230. Urban Land Institute 1963. Innovations vs Traditions in Community Development. Washington, D.C.: ULI Technical BuJletin No. 47. Urban Land lnstitue, National Association of Home Builders, and American Society of Civil Engineers 1976. Residential Streets: Objectives, Principles, and Design Considerations Washington, D.C.: Urban Land Institute. U S Department of Commerce. Bureau of the Census I 982. U.S. Census of Governments. U. S. Department of Commerce Bureau of the Census U S Census reports, 1 980, 1990. U .S. Department of Housing and Urban Development. 1972 Total Energy and Pneumatic Waste Collection Demonstrations Ref. No. 07-028. U S Department of Housing and Urban Development. 1980 Metropolitan Development Patterns: What Difference Do They Make? Washington, DC: U.S. Department of Housing and Urban Development, O ffice of Policy Development and Research. lUIGEU UOOICJNGS PAMONS EOONCllHWEST 192 U S Department of Housing and Urban Development 1995 Cityscape: A Journal of Policy Development and Research Vol I, No. 2 Washingto n D.C. U S. Department of Housing and Urban Development, Office of Policy Development and Research U. S. Department of Transponation, Bureau of Transponation Statistics. 1994 Transportation SUJtistics Annual Report /994. Washington, DC: U .S. Department of Transportation Van dec Ryn, S and P Caltborpe. 1993. Sustainable Communities. San Francisco: Sierra Club Books. Vermont League of Cities and Towns and Vermont Natural Resources Council. "The Tax Base and The Tax Bill Montpelier, VT: Vermont League of Cit i es and Towns. September Vincent, Mary Jayne, et al. 1994 NPTS Urban Travel Panerns: 1990 NPTS. Washington, DC: OffiCe of Highway Information Management, Federal Highway Administration, U .S. Depart ment of Transportation VNI Rainbow Appraisal Service. 1992. Analysis of the lmpacl of Lighr Rail Transit on Real Estate Values San Diego, CA: MTDB. Voith, R. 1993 "Changing Capitalization of CBD-Oriented Transponation Systems: Evidence from Philadelphia 1970-1988 Journal of Urban Economics 33: 361-76 von Reichert, Christiane and Gundars Rudzitis. 1992 "Mul tinomial Logisti c al Models Explaining Income Changes of Migrants to High-Amen ity Counties." The Review of Regional Studies 22, I: 25-42. Voorhees, M T 1991 "The True Costs of the Automobile: A Basis for Comparison.'" Grass Roors to Green Modes Proceedings of the 12th International Pedestrian Conference, Bou lder co. pp. 231-6. Wachs, Martin. 1989. "United States Transit S u bsidy Policy: In Need of Reform.'' Science: 244. Voorh ees, M. T 1992. The Tnu Costs ofrhe Automobile to Society City of Boulder, CO. January lRANSrr COOPfAAnVE RESfAACH PI!OGRAM /TOIP) fl. I 0

PAGE 197

Webber. Melvin M. 1976. ''The BART Experience-What Have We Learned?" The Public truerest 45. Webber. Melvin M. 1993. ''The Marriage of Autos and Transit: How to Make Transit Popular Again." Presented to the Founb International Research Conference, Center for Transportation Studies, University of Minnesota Wendell Cox Consultancy. 1996. Population and Land Area for Urbanized Areas with More Than I Million Population in 1990. Wheaton, William L., and Morton J Scbussheim. 1955. TM Cost of Municipal Services in Residential Areas. Washington, D.C.: U.S. Depanment of Commerce. Whyte, William H. Jr. 1957. "Urban Sprawl." In The Exploding Metropolis. Garden City, NY: Doubleday. Wiewel, Wim. 1993. ''The Fiscal Impact of Commercial Development." Land Development (Spring-Summer). Windsor, Duane. 1979. "A Critique of 'The Costs of Sprawl,"' Journal of the American Planning Association 45, 2: 279-92. Wish, Naomi, and Stephen Eisdorfer 1996. The Impact of the Mount Laurel Initiatives: An Analysis of the Characteristics of Applicants and Occupants. Soutll Orange, New Jersey: Seton Hall University. Wohl, Martin. 1976. ''The Case for Rapid Transit: Before and After the Fact" Transpor tation Alternatives in Southern California Los Angeles: University of Southern California, Institute for Public Policy R e search, Center for Public Affairs. altOOI(JNGS PAlSONS ECONOlliiWEST 193 York, Marie L. 1989. Encouraging Compact Development in Florid

xml version 1.0 encoding UTF-8 standalone no
record xmlns http:www.loc.govMARC21slim xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.loc.govstandardsmarcxmlschemaMARC21slim.xsd
leader nam 2200457Ia 4500
controlfield tag 001 028898375
005 20110923132147.0
006 m o d
007 cr bn|||||||||
008 110923s1997 nju obt 000 0 eng d
datafield ind1 8 ind2 024
subfield code a C01-00057
0 4 245
The costs of sprawl--revisited
h [electronic resource] :
b TCRP project H-10 literature review /
Rutgers University, The Brookings Institution, Parsons Brinckerhoff Quade and Douglas, ECONorthwest.
1 7 246
Transit Cooperative Research Program (TCRP) H-10
[New Brunswick, NJ :
Rutgers University, Center for Urban Policy Research,
1 online resource (i, 193 p.)
"Principal investigators, Robert W. Burchell ... [et al]"--P. [2] of cover.
"April 22, 1997."
"Work in progress--not for quotation or attribution. April 23, 1997"--P. [2] of cover.
Includes bibliographical references (p. 175-193).
Description based on online resource; title from e-book cover (viewed Sept. 23, 2011).
Land use
z United States.
Local transit
United States.
Municipal services
United States
x Finance.
Urban policy
United States.
Cities and towns
United States
United States.
Infrastructure (Economics)
United States
Burchell, Robert W.
2 710
Rutgers University.
Center for Urban Policy Research.
Brookings Institution.
Parsons, Brinckerhoff, Quade & Douglas.
ECO Northwest, Ltd.
Transit Cooperative Research Program.
t Center for Urban Transportation Research Publications [USF].