Travel Choices in Pedestrian Versus Automobile Oriented Neighborhoods Robert Cervero Carolyn Radlsch Working Paper UCTCNo. 281 The University of California Transportation Center Umversity of Califorma Berke]ey, CA 94720 The University of California Transportation Center The Umverslty of Cahforma Center actlvmes Researchers Transportataon Center (UCTC) at other umversttaes w~thmthe is one of ten regional umts regnon also have opportunlnes mandated by Congress and to collaborate with UCfacuIty estabhshed m Fall 1988 to on selected studies support research, education, and tralmng in surface trans- UCTC’seducational and portation. The UCCenter research programs are focused serves federal RegmnIX and on strategic planmng for is supported by matching lmprowng metropolitan grants from the U S Depart- access~blhty, with emphasis merit of Transportataon, the on the special condlt~ons in Cahforma Department of Region IX Partlcular attenuon Transportation (Caltrans), and is darected to strategies for the Umvers~ty using transportauon as an instrument of economic Based on the Berkeley development, while also ac- Campus, UCTCdraws upon commodatingto the region’s existing capabahttes and persistent expansion and resources of the InsUtutes of while malntalmng and enhanc- TransportaUon Studies at ing the quality of hfe there Berkeley, Davis, Irvme, and Los Angeles, the Institute of The Center distributes reports Urban and Regnonal Develop- on its research m worlang ment at Berkeley, and several papers, monographs, and in acadermc departments at the repnnts of pubhshed articles Berkeley, Davis, Irvme, and It also publishes Access, a Los Angeles campuses magazine presenting sum- Faculty and students on other manesof selected studies, For Umversity of California a hst of publicanons m pnnt, campuses may participate m write to the address below DISCLA!MER Thecontents of this report reflect the wews of the authors, who are responsiblefor the facts and the accuracy of the mformat~on presented University of Caltforma flereia.Thfs decrement is disseminated underthe sponsorshrp of the Transportation Center Departmentof Transpertahon, UmversEty Transportahon Centers Program, in tf]e interestof reformationexchange. TheU S Governmentassumes no 108 NavalArchitecture Building l~abflityfor the contents oruse thereof. Berkeley, Cahforma94720 Tel 510/643-7378 FAX510/643-5456 Thecontents ofth~ report reflect the "clewsof the authorwho is responsible for the facts and accuracyof the data presentedhereto Thecontents do not necessardyreflect the officmi wewsor pohelesof the State of Cahfornmor the U S Departmentof TransportationTh~s report does not constatute a standard, specification,or regulation Travel Choices in Pedestrian Versus Automobile Oriented Neighborhoods Robert Cervero Carolyn Ra&sch Institute of Urbanand RegmnalDevelopment University of Cahfomlaat Berkeley Berkeley, CA 94720-1850 WorkingPaper July 1995 UCTCNo 281 TheUmvers~ty of CahformaTransportation Center Umversltyof Cahformaat Berkeley Abstract The NewUrbanism movement calls for redesigning Americanneighborhoods so that they are less oriented toward automobiletravel and moreconducive to walking, bicycling, and transit riding, especially for non-worktrips. NewUrbanism calls for a remmto compactneighborhoods wlth grid-like street patterns, maxedland uses, and pedestrian amemties.This paper investigates the effects of NewUrbanism design principles on both non-workand commutingtravel by comparingmodal splits between two distinctly different neighborhoodsin the San Francisco Bay Area. The rico-traditional neighborhood, Rockridge, and the nearby conventional suburban cor~mauni~,Lafayette, were chosen as case studies because they have snnilar incomeprofiles, freeway and transit service levels, and geographical locations. Rockridgeresidents averaged arotmd a 10 percent higher share of non-worknaps by non-automobile modesthan did residents of Lafayette, controlling for relevant factors like incomeand transit service levels Thegreatest diffi.~rences were for shop trips under one mile. Modalsphts were moresimilar for worknaps, conJ~rmingthe proposition that neighborhooddesign practices exert their greatest influence on local shopping trips and other non-workpurposes For work trips, compact, mixed-use, and pedestrian-oriented developmentappears to have the strongest effect on access trips to rail stations, in partlcular inducing higher shares of access trips by foot and bicycle. 1. INTRODUCTION In recent years a movementcalled "NewUrbanism" has surfaced that calls for fundamental changes in how American communitmsare designed and built (Katz, 1994) New urb;mists embracemany of the design principles of late-19th and early-20th Americantowns. Amongthese are compactdevelopment, a mixture of land uses and housing types, a grid-like street pattern well-suited to walking, and prominentcivic spaces. Traditional towns, like Amlapolis, Marylandand Savannah, Georgia, are held as exemplars of old-fashion, pedestrian- oriented communitmswhere thousands of residents live within an easy and pleasant walk of the town center. A central premise of the NewUrbanism movement is that designing American co~mlunities"~’’’ like those of yesteryear will reduce autom age and dependency by making public transit, walking, an~ TM- ~ent federal and state clean air ¯ ~ O~/r-~ . ¯ ¯ reqlarements have ~ ~ ~ - F26~ gr~ lg land-use tmtlatlves as one of (I a dozen or so possib /~ ~ Several cities, notably San Diego, California ant nmurtity design guidelines 0j, f._4,/~X7 [.1/ ~ based on NewUrban1 ucle miles traveled (VMT) ~ v I.~/ /~, and enhancing the qua ~_b~vJ, r- . ~ ~ le NewUrbanism ideas have capturedtile imaginati¢ ~ .~’ I_ ]~ crY/ bU’ there has been Iittle research to suppo th ) t-supportive community ~ v’a9 desi gns will actually red The few commun _ ~ ~eo-traditional in their designs offer httle insight into the. _..,,,uns. Seaside, widely consideredthe nation’s first neo- traditional commumty,is a fairly exclusive beach resort on Florida’s panhandleand outside a metropolitan area; analysis of travel patterns amongSeaside’s residents wouldhave limited applicability elsewhere. Twoother examples-- the Kentlands, a Marylandsuburb of W~shington,D.C., and LagunaWest, outside of Sacramento, California -- do not yet have enoughretail or employmentactivities to qualify them as bonafide mixed-use communities. Both also receive modestlevels of public transit services. In light of there being few good examplesofneo-traditional or ’~NewUrban" communities, this study examinestravel behavior in existing neighborhoodswhich embodythe characteristics of either pedestrian or auto-oriented designs The research focuses on how conWastingresldential built environments influence modechoices, both for work and non-work trips. It probes whetherfactors like higher densities, mixedland uses, pedestrian-oriented street designs, and neighborhoodretail clusters, m combination, encouragepeople to give up their cars and walk, take transit, or travel by someother means. The degree to winch this is demonstrated should either lend credence or cast skepticism over the transportation benefits of NewUrbanism design principles 2. PAST RESEARCH ON THE TRAVEL IMPACTS OF NEIGHBOR.HOOD ENVIRONMENT A considerable body of literature nowexists on the impacts ofbuiIt environmentson travel choices. Muchof this research, however, has focused on commutetrips to large employmentsites (Cervero, 1989; CambridgeSystematics, 1994); until recently, less attention has been given to the effects of neighborhooddesigns on travel demand,particularly for non- work purposes. This secUonbnefly summariessome of the relevant literature on the travel impacts of neighborhood built em4ronments. One of the earhest stu&es on the travel demandeffects of neighborhoodswas by Levinson and Wyrm(1963), whofound that neighborhood density substantially reduces vehicle trip frequency. Their results indicated that if neighborhooddensity and distance-to-CBDwere both to increase by one standard deviation, average household VMTwould drop by roughly one third. Subsequent work by Pushkarev and Zupan (1977) confm~edthat both population density and proximityto CBDare critical factors in justifying investmentsin heavy rail transit systems. To support light rail servmes on five-minute peak headways,for instance, Pushkarev and Zupan concludedthat densities of at least nine dwelling units per acre within a fifteen maleradius of a downtownwould be required. Handy’s (1993) comparison of shop trip-making between "traditional" neighborhoods and more auto-oriented ones in the San Francisco Bay Area is one of the most in-depth analyses 2 on non-worktravel to date. She found those living in traditional neighborhoodsmade two to four morewalk/Ncycle trips per weekto neighborhoodstores than those living in nearby areas thal were served mainly by automobile-oriented, stop retail establishments. Residents of mixed- use neighborhooods,however, averaged similar rates of auto travel to regional shopping mails, suggestingthat internal walktrips did not replace, but rather were in addition to, external driving trips. Fehrs and Peers Associates (1992) found substantially higher rates of foot and transit travel in ~-aditional communitiesversus conventional suburban subdivisions of the San Francisco Bay Area; their study, however,did not control for the influences of differing transit service levels or soci o-economiccharacteristics, like incomes, amongneighborhoods. A more recent study by Ev, n~ng et al. (1994) comparedwork and non-worktravel