Residential location, land use and transportation

RESIDENTIALLOCATION,LANDUSEANDTRANSPORTATION:THENEGLECTEDROLEOFNONWORKTRAVELBumsooLee,PeterGordon,JamesE.Moore,II,andHarryW.RichardsonSchoolofPolicy,PlanningandDevelopmentUniversityofSouthernCaliforniaLosAngeles,California90089-0626June,2006ThisresearchwassupportedbyagrantfromtheLuskCenterforRealEstateattheUniversityofSouthernCalifornia.11.INTRODUCTIONSimpleintrospectionaswellasaccumulatingevidencefromacademicresearchsuggeststhatacoreideaofurbaneconomists,thatjourneys-to-workdominatehouseholds’choiceofresidentiallocation,issuspect(GiulianoandSmall,1993).Indeed,ourownrecentresearchidentifiesneighborhoodcharacteristics(attractions)thathelptoexplainlongercommutes;wefoundveryfewneighborhoodtypesthatwouldhelptoexplainashorterjourneytowork(Gordonetal.,2005).Also,findingsthatrevealtherelativeimportanceofnon-workactivitiesandtrips,someofwhichmaybeamenabletomoreflexibleschedulingand/orarelessessential,alsoinformthelong-runningdiscussionaboutthepotentialpowerofpeak-loadroadpricingtoreducehighwaycongestion.Inthispaper,wepresentevidencethatcorroboratesbothoftheseideas.Foursetsoffindingsarediscussed.First,weinvestigateworkandnonworktravelforasetofweeklytravelperiodsmoredetailedthanthestandarddivisionintopeakvs.off-peaktravel.Instead,weusetendistinctperiodsoftheweek.Second,weattempttoidentifythevariablesthatexplainnonworktravelfrequenciesanddurations.Third,werevisitthestandardworkvs.nonworktrippurposedichotomyandstudytypesoftours(involvingtripchains)thatcombinethetwo.In2001,approximately25percentofcommutesincludedoneormorestopsalongtheway.Fourth,wetesthypothesesthatexplainthenatureandthefrequenciesofthetripandtourtypes.Mostofouranalysisisbasedondatafromthe1990and1995NationwidePersonalTransportationSurvey(NPTS)andthe2001NationalHouseholdTravelSurvey(NHTS).Theanalysisoftripsandtours,however,isrestrictedtodifferentperiods,19902to2001fortripanalysisand1995to2001fortour-levelanalysis,becauseoflimitationsofeachyear’ssurvey.Intheelevenyearsbetweenthe1990and2001surveys,theU.S.populationgrewby15.8percentbutthenumberofdriversincreasedby16.8percentwhilethenumberofworkersgrewby22.8percent.TheCensusBureaualsoreportsthatthelaborforceparticipationratedippedslightlybetween1990and2000,from65.3percentto63.9percent.2.LITERATUREREVIEWAstheimportanceofdailynon-workactivitiesandtravelcontinuestogrow,urbanresearchershavebelatedlystartedtolookbeyondthejourney-to-work.Someofus(Gordon,Kumar,andRichardson,1988)reportedthattherehadbeenremarkableincreasesinnon-worktravelbythemid-1980s,withparticularlyfastergrowthinthepeakperiods.Theobservationofafasterincreaseinthefrequencyofnonworktravelamongsuburbanitesledustosuggestthattripcostsavingsresultingfrommoreefficientspatialstructure(suburbanization)helpedtoexplainmoretripmaking.Laterresearchattributednon-worktravelgrowthtochangesinhouseholdstructure(Strathman,Dueker,andDavis,1994)orchangesinlifestyleandretailindustrystructure(NelsonandNiles,2000).Becausetraveldemandisaderiveddemand,theincreaseinnon-worktravelmustultimatelybeassociatedwithindividual(orhousehold)preferredactivitypatterns.Inthiscontext,non-worktravelbehaviorcanbestbeunderstoodviatheperspectiveof“humanactivity”approaches.Webrieflyreviewactivity-basedstudiesofnon-worktravelbehaviorfocusingontrip-chainingandonthelinksbetweenurbanformandnon-work3travel.Formorecompletesurveys,seeKitamura(1988),Fox(1995),andBhatandKoppelman(1999).Activity-basedapproaches“studytravelinthecontextofdailyhouseholdactivitypatterns,asalinkintheprocessoffulfillingtraveldemandsthroughtheformationofdailysets,orchains,ofactivities”(Fox,1995,p.105).Twokeyideasassociatedwiththisapproacharethattraveldemandisderivedfromthedemandforactivitiesandthatpeoplemakedecisionsaboutactivitiesandtravelinthefaceofspatio-temporalconstraints(BowmanandBen-Akiva,2000).Researchersusingthisapproachhavemadesignificantcontributionstoabetterunderstandingofactivityparticipationandscheduling,spatio-temporalandinterpersonalconstraints,andinteractionsbetweentraveldecisionsandhouseholdstructure(Kitamura,1988).Intheactivity-basedapproach,therapidgrowthinnon-worktravelisexplainedbysocio-economicchange.Forexample,risingfemalelaborparticipationhasincreasedtheopportunitycostassociatedwithactivitiesathome,andincreasingincomeshaveenabledmoreactivitiesoutsidethehome,oftensubstitutingforat-homeactivities(LevinsonandKumar,1995).Diningoutisagoodexample.Thedispersionofworktripsoveroff-peakperiodsandincreasednon-worktripsduringthepeakhourscanalsobeexplainedbysubstantialincreasesinthechainingofworkandnon-worktrips(LevinsonandKumar,1995).Indeed,trip-chainingbehaviorisacriticalfactorforunderstandingdailyurbantravelandempiricalstudieshaveshownhowindividualsorhouseholdslinkworkandnon-worktripswithintypesoftripchains(tours)givenspatio-temporalconstraints.4Ingeneral,peoplearemorelikelytolinkmultipletripsinachaintoeconomizeontimewhentheirdemandfortravelintensifies,whilelonger-durationactivitiesarelesslikelytobechained(Recker,McNally,andRoot,1987;Strathman,Dueker,andDavis,1994).Workerswithlongercommutedistances(Nishii,Kondo