Socio-Economic Impact Assessment (SEIA) Methodology p j for Urban Transport Projects

Presentation at Santiago, Chile October, 2009

By:

Anvita Arora, PhD

Resident Representative, India Interface for Cycling Expertise (I-CE), Utrecht The Netherlands (I CE) Utrecht, CEO, Innovative Transport Solutions, Technology and Business Incubation Unit, Indian Institute of Technology, New Delhi, India

Urbanization in India
Relativelyslow,yetoneofthelargest R l ti l l t f th l t urbansystems 3050%slumdwellers,unauthorized selfconstructeddwellings,closeto self constructed dwellings close to work Growthofinformalsectoroften fasterthanformalsector faster than formal sector

Bicycleownership3050% Carownership Scooter/Mcycle 313% 4050%

No. Of Cities 200

147 177

150 100 50 0
<1 Town

28

20 6 3 3
1 crore G

1-5 lakh 5-10 lakh 10-20 lakh20-50 lakh 50-100 A&B C D E lakh F

Modal shares in different city sizes(RITES, 1998)

100% 80% % share 60% 40% 20% 0% A:0.1- B0.1- C:.5-1 D:1-2 E: 2-5 0.5 0.5 walk Rick shaw Rick-shaw Cycle City size Public Transport Cars >5

Three MTW

Threattosustainablescenario:IncreasingcarandMTWtrips

Transport Modes of the Urban Poor

ModalShareforthepoor Delhi
bus 8% rickshaw 6% cycle 4% 0thers 5%

ModalShareforthepoor Mumbai
walk 77% ta train 16% others 3%

bus b 14%

walk 61% cycle 6%

Patna Rickshawpolicies?

Jaipur Threewheelersparatransit?

Hyderabad Twowheelers/three wheelers?

Lucknow Rickshaws,cyclespeds?

Urbantransportproblems
Poorrelyonnonmotorizedtransportbuttheirfacilitationis oftenignored Smallchangesinpublictransportfare/servicecansignificantly affecttheirmobility Restraintsoninformaltransportsectorlimitsaffordable Restraints on informal transport sector limits affordable servicestothepoor DominanceofprivatemotorvehiclesmarginalizesNMTs Womenarebadlyservedbytransportsystem Pooraremorevulnerabletoinjuriesandpollution

NationalUrbanTransportPolicy(NUTP)
AsperthedirectivesoftheGOI MOUD UT thevariousproposalsforurban transportbeingpreparedunderJNNURMshouldcomplywithNUTPin transport being prepared under JNNURM should comply with NUTP in ordertobeeligibleforCentralGovt.funding. g g ThefocusofNUTPisonthefollowingstrategies: 1.Equitableallocationofroadspace withpeopleasfocus 2.PrioritytotheuseofPublicTransport 3.Integratedpublictransportsystems 4.Prioritytononmotorisedtransport 5.Promotemultilevelparkingcomplexes 6.Createpublicawareness DelhiCDPprioritiesandprojectshavebeenidentifiedbasedonabove guidelinesofNUTP.

DelhiCityDevelopmentPlan
VisionandInvestment Vision and Investment
1.

Equitableallocationofroadspace withpeopleasfocus
33%modalshareofpedestrians investmentonpedestrianinfrastructure 0.5% oftotalinvestment f l

2.

PrioritytotheuseofPublicTransport
60%ofvehiculartripsbypublictransport Capacitybuildingofpublic transport 3 3projectsLRT,Monorail,HCBS i j L il HC S investment42% oftotal 42% f l investment

3.

Integratedpublictransportsystems
Noinvestment N i

4.

Prioritytononmotorised transport
0.8% oftotalinvestment

5.

Promotemultilevelparkingcomplexes
2% oftotalinvestment

6.

Createpublicawareness
0.2% oftotalinvestment

Whereistheremaining55%investmentbeingmade?

IncreasingRoadLength 32% Flyovers Flyovers 10% RoadWidening 8% Spl.SchemeforCPandoldcity 5%

Investments in flyovers,road expansion and FOBs FOB

DelhiMetroRailSystem
460 km of arterial roads,`10000 buses carrying 6 m trips 256 km by 2021, estimated cost USD~3500 million

Existing Rail Corridors, Delhi M t D lhi Metro Fi t First Phase (2005) 65 km, projected ridership 1.5 15 m/d, actual ridership 0.4 m/d, USD 7.1m , loss/yr, 100% cost overruns Final Phase (2021), 60% residents & 82% area not within walking distance!!!

Overcrowdedbuses

INTRODUCTIONtothestudy
Transportisacriticallinkbetweeneconomicand Transport is a critical link between economic and socialdevelopment Transportisaderiveddemand needbased Transport is a derived demand need based Thebenefitsofimprovingtransport infrastructurehavetraditionallybeenmeasured infrastructure have traditionally been measured byperformancecriteria,likeimproved connectivity,traveltime,speedsandfuelsavings connectivity, travel time, speeds and fuel savings fortheuser.

14

Theproblem

Theusersarenotahomogeneousgroup Someusersmaybenefit,somemaynot,and Some users may benefit some may not and somemaynotbeaffectedatall Alsothenonusersmaybeimpacted an Also the non users may be impacted an externality(+veorve) Benefitsanddisbenefitstousersandnonusers B fit d di b fit t d needtobeunderstoodandinternalizedby transportprojects. transport projects

15

NeedofStudy
Transportinvestmentsadvocateinclusionofsocialassessment intransportprojectsandprioritizepovertyalleviationasan objective.Needtounderstand: bj ti N d t d t d Usersasadisaggregatedmass(differentiatedby income,occupation,gender,age,ethnicity,etc.) , p ,g , g , y, ) Thegapbetweenaccessavailability(transportinfrastructure) andmobilityissues(abilityofdifferentgroupstoutilizethe infrastructure)andtheircorrelationwithpoverty(especially i f ) d h i l i ih ( i ll withrespecttolivelihoodopportunities). Aneedtodevelopamethodologicalframeworkormodelfor A need to develop a methodological framework or model for ensuringtheinclusionofsocioeconomicissuesoftransport planninginpoliciesandprojects.

16

TheContext
Delhi
Populationof13.8million(Census,2001). Modalshare 62%ofthevehiculartrips(33%ofalltrips includingwalk)aremadebybuswithanaveragetriplengthof 10.7Km(RITES,1994). 10.7 Km (RITES, 1994). Heavyinvestmentsintransportinfrastructure,likegrade separatedjunctions,roadwideningandtheDelhiMetroRail.

TheDelhiMetroisarepresentativecasestudyofacapital intensiveurbantransportprojectpromisingtoaccruehigh te s e u ba t a spo t p oject p o s g to acc ue g benefitsofaccessibilityanddecongestion.

17

Objectives&ResearchFocus
Objectives: TounderstandtheimpactofDelhiMetroRailonthe To understand the impact of Delhi Metro Rail on the accessibilitypatternsoftheurbanpoor. Tounderstandtheimpactofchangedaccessibilityon mobilityandthesocio economicstatusofthelow income mobility and the socioeconomic status of the lowincome households. Todevelopindicesofaccessibility,mobilityandSEWBand toformulateanSEIAmethodology. gy Researchfocus: Tounderstandhowaccessibilityandmobilityaffectthesocio To understand how accessibility and mobility affect the socio economicwellbeing(SEWB)oftheurbanpoorandhow indicesofaccessibilityandmobilitycanbeintegratedin SEIAmethods. SEIA methods.
18

Hypothesis

a)

b)

IntroductionoftheMetrorailsysteminDelhi Introduction of the Metro rail system in Delhi haschangedtheaccessibilityfortheurban p poor. Thischangeinaccessibilityhaschangedthe mobilityprofileandthesocio economicwell mobility profile and the socioeconomic well beingoftheurbanpoor.

November 2007

19

CaseStudy TargetGroup UrbanpooraffectedbytheDelhiMetroRailProject Urbanpoorastheinhabitantsofslumsinthecity Urban poor as the inhabitants of slums in the city
UrbanDelhipovertylineatRs505.45(USD12.64)percapitaper month,(Saxena,2001) ForDelhislumspercapitaincomeoflessthanRs.600(15USD) F D lhi l it i fl th R 600 (15 USD) permonthfor78%inhabitants(Anand,2006)

Twocategoriesoflowincomehouseholdsselected:
thoselivinginthevicinity(within1km)ofthemetro stations,and stations and thoserelocatedduetotheconstructionofthemetro.

20

THEORETICAL FRAMEWORK (Figure 1.1)

The Poor (urban)

Transport project

Transport System (urban)

Geographic grouping (slums)

Occupational grouping (rickshaw pullers, hawkers...)

Condition of Infrastructure (urban)

Socio-economic profile

Usage of Infrastructure

Social well-being indicators

Mobility indicators

Accessibility indicators

LEGEND
Relationship between mobility and well-being Relationship between mobility and accessibility
Existing System Direct Impact Indirect Impact

November 2007

21

Methodology
Householdsurveybaseddatacollectedfortargetgroup. Datasetusedtoderiveindicatorsofaccessibility,mobilityand SEWB. Theindicatorsaggregatedintoindicesof accessibility,mobilityandSEWBbyusingthePrincipal y, y y g p ComponentAnalysis(PCA)technique. Thechangeinindicatorsandindicesinthebeforeandafter metroscenariosusedtoassessthesignificanceoftheimpact metro scenarios used to assess the significance of the impact ofthemetroprojectontheurbanpoor. Thecorrelationbetweenaccessibility,mobilityandSEWBis modeledusinglinearregressiontoillustratethatthechange modeled using linear regression to illustrate that the change inaccessibilityandmobilityduetoatransportproject changestheSEWBofthecommunity.
22

Structure
1. 2. 3. 4. 5. 6. 7.

Introduction Socioeconomicimpactassessment(SEIA) currentpractices Transportandpoverty SEIAmethodologyforurbantransportprojects Accessibility,mobilityandsocioeconomicwellbeing Casestudy Delhimetrorail Formulationofthesocioeconomicimpactassessment(SEIA) model Conclusions,contributionandscopeforfuturework Conclusions contribution and scope for future work

8. 8

23

SEIA CURRENTPRACTICES
Socialimpacts theconsequencestohumanpopulationsofany publicorprivateactionsthatalterthewaysinwhichpeople live,work,play,relatetooneanother,organizetomeettheir live work play relate to one another organize to meet their needsandgenerallycopeasmembersofsociety.

History
SIArealizedasimportantpartofEIAsince1969to1980s.Partiallyforced byprojectfailuresresultingfrominadequateappraisalofprojectson by project failures resulting from inadequate appraisal of projects on narroweconomicandtechnicalcriteria(Ricksonetal.,1990; Burdge,1998). WHOhaspointedoutthatthecostofsubmittingmajorproposalsfor WHO has pointed out that the cost of submitting major proposals for socialimpactassessmentwasfarlessthanthecostofcorrecting unforeseennegativeimpactsthatoccurredafterimplementation (Giroult,1983,citedinBurdge1990).
24

TheIndianScenario:
TheMinistryofEnvironmentandForests,hasaseparateEnvironment Clearancemanualforlargeconstructionprojects(MoEF,2006). However,thesocioeconomicaspectsmeritonlya3pointwriteup inAnnexureII.Questionstobeanswered: i A II Q ti t b d
7.SocioEconomicAspects 7.1.Willtheproposalresultinanychangestothedemographicstructureof localpopulation?Providethedetails. local population? Provide the details. 7.2.Givedetailsoftheexistingsocialinfrastructurearoundtheproposed project. 7.3.Willtheprojectcauseadverseeffectsonlocalcommunities,disturbance tosacredsitesorotherculturalvalues?Whatarethesafeguards t d it th lt l l ? Wh t th f d proposed?

ThesepointshighlighttheinadequacyofinclusionofSIAinlarge These points highlight the inadequacy of inclusion of SIA in large infrastructureprojectsinIndiaandreiteratetheneedfor comprehensiveworkonit.

25

ImpactAssessmentMethodologies
Themethodologiesreviewedinthissectionare: Thefundingagenciesapproach
TheWorldBank AsianDevelopmentBank

TheSCOPEframework Theimplementingagenciesguidelines
TheFDOThandbook

TheNGOsperspective
QueenslandFamilies,YouthandCommunityCare,Australia
26

Discussion
TheWorldBankapproach:largerpolicyframework,generic applicability,focusoninstitutionalmechanisms.TheADB document:comprehensivebutgenericnotincludethe p g specialproblemsoftransportationprojects. TheSCOPEframework:formulationofasocioeconomic frameworkofacommunity,emphasisontheneedtoquantify framework of a community emphasis on the need to quantify allparameterslistedbutnoholisticassessmentdesign. FDOTGuidelines:focusonlanduseimpactsoftransportation projects,communitiesinfluencetheuseoflandandvice j t iti i fl th fl d d i versaandtransportationprojectsinfluencebothina correlatedmanner. TheAustralianNGOapproach:emphasizesonpeopleand theirneedandreactions,conceptslikecommunitysensitivity indicesandthevulnerablecommunitygroups.
27

Conclusion
TheSEIAofatransportationprojectmustanswerthefollowing: Whatistheimpactareaofthetransportproject(spatialand temporal)? Whoisaffectedbytheproject? Whatistheirsocioeconomicstructure? Whataretheirneeds? Whataretheirdemands? Whatistheirabsorptivecapacity? Whicharethevulnerablegroups? Which are the vulnerable groups? Whatistheincomedifferentialinmobilityandaccessibility? Whatisthegenderdifferentialinmobilityandaccessibility? Whatisthesocio culturaldifferentialinmobilityandaccessibility? What is the sociocultural differential in mobility and accessibility? Whatistheexistingtransportsystemused(formal/informal)? Whatarethepotentialadverseimpacts?

28

TRANSPORTANDPOVERTY
DefiningPoverty amultidimensionalphenomenon,encompassinginabilitytosatisfybasic needs,lackofcontroloverresources,lackofeducationandskill,poor health,malnutrition,lackofshelter,pooraccesstowaterand sanitation,vulnerabilitytoshocks,violenceandcrime,lackofpolitical freedomandvoice.TheWorldBank(a,1999) povertymustbeseenasthedeprivationofbasiccapabilitiesratherthan b h d fb bl h h merelyaslownessofincome(Sen,1999). Povertyimpactsoftransportinterventions Complexbecausetransportisanintermediateservice transport improvementsreducepovertynotthroughincreasedconsumptionof transportpersebutthroughimprovingthequalityandsecurityof t t b t th hi i th lit d it f accesstowork,markets,andservices,andthroughreleaseofscarce resourcesforconsumptionandproduction

29

Issues
EfficiencyvsEquity: Goodtransportpolicycontributestopoverty Efficiency vs Equity: Good transport policy contributes to poverty reductionbyenhancingefficiencyandequity(Gannon,etal,2001). AccessandLivelihoodneedsoftheurbanpoor: Urbantransportinteracts withemploymentissuesforthepoorintwomainways:indirectlyby providingaccesstoemploymentopportunitiesanddirectlythrough employmentoflowincomepeopleinthetransportsector
DEPENDANCY CONSTRUCT: SOCIETY AND TRANSPORT SYSTEM

Access to livelihood
( g (eg. bus, cycle, pedestrian) , y ,p )

SOCIETY
(URBAN POOR)

Means of livelihood
(eg. Rickshaw pullers)

TRANSPORT SYSTEM

Facilitator of livelihood (eg. Hawkers)

30

GenderBias: Womentendtohavedifferenttravelneedsderivingfrom themultipletaskstheymustperformintheirhouseholdsandintheir communities(Greicoetal,1997). HealthImpactsofTransport: Pollution(air,water,noise)effectsthe urbanpoorparticularlyseverely,sincetheyaretheleastabletoavoidor b l l l h h l bl d seekprotectionfromthem(UNDP1998).Pedestrianandcyclistsaremost vulnerabletoroadaccidents. TheSheltertransportlivelihoodlink: Accesstoaffordabletransportis oneofthemostimportantfactorsindetermininglivelihoodsfortheurban p poorTheriseofprivatevehiculartraffichasdecreasedbusspeedsand p p servicelevelsdrasticallyandmadenonmotorizedtransportdangerous anddifficult.Travelforthepoorhasthusbecomeslowerandmore difficultevenasothereconomicandplanningforceshavecausedmanyof themtobedisplacedfromcentralinformalsettlementstomore them to be displaced from central informal settlements to more peripherallocations(Immersetal,1993)

31

Evictionandrelocation
People evicted because of transport projects

Transportation aspects of t f eviction and resettlement

Transport implications for evicted people i t d l (due to any project)

Thecentralconcernoftheprocessofevictionand relocationisthereductioninaccessibilityandmobility optionsoftheurbanpoor,whichdirectlyaffectstheir livelihoodandthussocialwellbeing.

32

ACCESSIBILITY,MOBILITYANDSOCIO ECONOMICWELLBEING ECONOMIC WELLBEING

Reviewdefinitionsanddiscussions Review definitions and discussions DefineAccessibility,MobilityandSEWBforthe study Postulateindicatorsandindications

33

Author Roberts R b t Black

Year Definition/ Discussion 1988 th number of trips made. the b ft i d number of, and/or the ease of making journeys 1981 accessibility is a function of land-use intensity and transport supply y p y 1992 accessibility is a description of how conveniently land-uses are located in relation to each other and how easy or difficult it is to reach these land use activities via the transport network of both public and private transport modes. 2000 Often understood as the ease of access to destinations, amongst other parameters i (accessibility) encompasses id it ( ibili ) ideas of costs in time and money; f i i d extent, comfort and frequency of the public transport system; and the distance to be negotiated to reach destinations such as shops, work places and schools 2001 Access to urban activities for a population presupposes the existence of a public transport service offering all city dwellers, whatever their income dwellers level, age or handicaps, the possibility of getting to work or school, going shopping and enjoying themselves. Accessibility is good when density is high because distances to be covered are low and when public transport is fast.

Accessibility

Ross

Vivier

Accessibility is a description of the proximity of destinations of choice and the facilitation offered by the transport systems (including public transport and non-motorized modes) to reach them.
34

Author Ekeh

Year Definition/ Discussion 1974 mobility is closely linked with personal and individual freedom, and lack of mobility freedom mobility is often associated with the repression of basic freedoms and even human rights 1988 the number of kilometers traveled 2000 The amount of travel people undertake measured by per capita vehicle The undertake per kilometers traveled A positive relationship exists between mobility and such indicators as transport energy use, motor vehicle ownership and use, journey to work distance, journey to work speed and general car speed. 2001 motorized mobility, measured by average annual distances traveled by city dwellers in automobiles, motorized two-wheeled vehicles, taxis and public transport 2003 the movement of people or goods The Th mobility perspective defines transportation problems in terms of bilit ti d fi t t ti bl i t f constraints on physical movement, and so favors solutions that increase motor vehicle system capacity and speed

Mobility

Roberts Ross

Vivier

Litman

Mobility is both the ability to travel to destinations of choice and the amount of movement necessary to do so.

35

Author

Year Definition/ Discussion

1966 Social indicators are statistics, statistical series, and all other forms of evidence that enable us to assess where we stand and are going with respect to our values and goals, and to evaluate specific programs and determine their impact. Social indicators are constructs, based on observation and usually quantitative which tell us something about the aspect of life in which we are interested or about changes in it. Social i di S i l indicators are facts about society in a quantitative form. They involve f b i i i i f h i l interpretation of advance and retrogression against some norm Human Development Index (HDI): The index is composed of three indicators: longevity, educational attainment, and standard of living, Economic and social development can be broadly distinguished but usually interact and should preferably be considered together. i t t d h ld f bl b id d t th National level economic development indicators commonly used are Gross National Product and Gross Domestic product. Others are National accounts Systems and Income distribution The Basic Quality of Life (QOL) Index includes seven variables: purchasing power, power homicide rate fulfillment of basic needs, suicide rate literacy rate, gross rate, needs rate, rate human rights violations, and deforestation Quality of Life (QOL) Index consists of Social, Health, Economic and Environmental indicators The Community Well-being Index (CWB) is composed of four indicators education, education labour force, income and housing. force income, housing

SEWB
Bauer 1975 UN Statistical office, F/18. Hauser 1975 19 UNDP Horn 1990 1993

Ed Diener

1995

Shookner INAC

1998 2004

Socio-economic well-being is defined as the status of a household where the basic social and economic needs for survival are fulfilled and the household has the capacity to improve its quality of life.
36

Indicators of Accessibility Indicator Type Indicator

Public Transport Accessibility (APT) (unit = per user) Dtotal, where D is distance Ttotal, where T is time Ctotal, where C is cost NA + NE , where N is no. of trips p NMLH DA + DE , where D is distance DMLH TA + TE , where T is time TMLH CA + CE , where C is cost CMLH (DA + DE )VEH, where D is distance (DA + DE )PED SDeducation , where SD is spatial distance SDhealth , where SD is spatial distance SDservices , where SD is spatial distance SDbus-stop , where SD i spatial h is ti l distance Sbus , where S is service of buses i.e. time gap between two successive buses

Indication
Lower value gives better accessibility Lower value gives better accessibility Lower value gives better accessibility Lower value gives g better accessibility Lower value gives better accessibility Lower value gives better accessibility Lower value gives better accessibility Lower value gives better accessibility Lower value gives better accessibility Lower value gives better accessibility Lower value gives better accessibility Lower value gives L l i better accessibility Lower value gives better accessibility

Spatio-travel Accessibility (AST) (unit = per household)

Notes on subscripts: A = access, E = egress MLH = main line haul access egress, NMV = non motorized modes including walking, MV = motorized modes 37

Indicators of Mobility Indicator Type Indicator

Household (+) Mobility (MHH) (unit = per household) PCTRwork, where PCTR is the average per capita trip rate HH PCTReducation, where PCTR is the average per capita trip rate of HH PCTRothers, where PCTR is the average per capita trip rate of HH MNMV , where M is modes MALL Dwork, where D is daily travel distance Deducation, where D is daily travel distance Dothers, where D is daily travel distance Twork, where T is daily travel time Teducation, where T is daily travel time Tothers, where T is daily travel time Cwork, where C is daily travel cost Ceducation, where C is daily travel cost d i Cothers, where C is daily travel cost
Notes on subscripts: TR = travel HH = household travel, NMV = non-motorized vehicle

Indication
Higher value higher mobility of HH Higher value higher mobility of HH Higher value higher mobility of HH Higher value higher g g mobility of HH Higher value higher mobility Higher value higher mobility Higher value higher mobility Higher value higher mobility Higher value higher mobility bilit Higher value shows higher mobility Higher value higher mobility Higher value higher mobility Higher value higher mobility

Personal (-) Mobility (MP) (unit = per household) h h ld)

38

Indicators of SEWB Indicator Type

Social Well-being (WBS) (unit ( it = per household) h h ld)

Indicator
NGinschool, where NG is no of girls NGschoolage NAliterate (>5grade), where NA is no. of adults NAall Infrastructure rank score * (Electricity, water, toilet) Ylo-income settlement, where Y is no. of Ydelhi years Nworking , where N is no. people Nall Itotal , where I is income Nall Vehall , where Veh is no. of Nall vehicles hi l

Indication
Higher value shows higher hi h social well being i l ll b i Higher value shows higher social well being Higher value shows higher social well being Higher value shows higher social well being Higher value shows higher economic well being Higher value shows higher economic well being Higher value shows higher hi h economic well i ll being

Economic Well-being (WBE) (unit = per household)

* Infrastructure rank score refers to the additive score of the types of services where the service which is formally provided and operational is given a value of 2 that which is self obtained has a value of 1 and that which is not 2, 1, available is given a value of 0 39

ImpactofTransportProject(DelhiMetroRail)
ChangeinAccessibility: PublicTransportAccessibility(A ) Public Transport Accessibility (APT)
ThedifferencesinindicatorsforbothsetsofBususersandMetro

SpatioTravelAccessibility(AST)
Directimpact changeinindicatorsofA ofhouseholdsinthevicinity. Direct impact change in indicators of AST of households in the vicinity Indirectimpact changeinindicatorsofAST ofhouseholdsrelocated.

ChangeinMobility:
Directimpact changeinindicatorsofHouseholdMobility(M )andPersonal Direct impact change in indicators of Household Mobility (MHH) and Personal Mobility(MP )ofhouseholds. Indirectimpact changeinindicatorsofMHH andMP ofhouseholdsrelocated.

ChangeinSEWB:
Directimpact changeinindicatorsofSocialWellbeingandEconomicWell beingofhouseholdsinthevicinity. Indirectimpact changeinindicatorsofhouseholdsrelocated.

40

CASESTUDY:DelhiMetroRail

Legend In vicinity Relocated R l t d

Part map of Delhi showing Case Study Area of Metro Rail line and locations of household survey
41

BususersandMetrousers
Indicator Comparability platform Percentage of Bus users 33% 57% 42% 4% 3% 16% 76% 5% 77% 5% Percentage of metro users 19% 80% 2% 48% 27% 67% 22% 19% 19% 37%

Dtotal (Km) Ttotal (min) Ctotal (Rs.) (Rs ) Na +Ne / Nmlh Da +De / Dmlh Ta +Te / Tmlh Ca +Ce / Cmlh (Da+De)veh / (Da+De)ped

Upto 20 Km of total daily travel distance Upto 2 Hours of total daily travel time Upto Rs 15 (0 38 USD) of total daily travel Rs. (0.38 expenditure > 2 access trips for every MLH trip More distance of Access than MLH More time of Access than MLH No cost of Access More cost of Access than MLH No Vehicle used for Access > 10 times access distance by Vehicle than by foot

42

HouseholdSurvey
InvicinityofMetroline: In vicinity of Metro line: Nosignificantimpactontheirsocioeconomicandtravelprofile. Decreaseintheavailabilityofbusessinceseveralbusrouteswererealigned bypolicytoimprovemetroridership. by policy to improve metro ridership. Consideringthatonly8%oftheirtripsareonbusand77%bywalk,4%by cycleand6%byrickshaw,itisunlikelythatthesetripswillbereplacedby metrotrips.

43

Relocatedduetothemetroline: Significantchangeintheiraccessibilityandtravelprofileandincome. Significant change in their accessibility and travel profile and income Theincreasingdistance,timeandcostofdailytravel,alongwithreduced incomeshasanegativeimpactonthehouseholds. incomes has a negative impact on the households Thelanduseaccessibilityhasdeterioratedasdistanceto educa o , ea se ces a d o e u ba se ces as c eased o education,healthservicesandotherurbanserviceshasincreasedfor 52%,63%and52%ofthehouseholdsrespectively.Thetransport accessibilityhasdeterioratedevenmoreasdistancetobusstophas increasedfor72%ofthehouseholdsandthebusfrequencyhasseenan averagedecreasefrom5minto63min(almost13times) average decrease from 5 min to 63 min (almost 13 times)

November 2007

44

FormulationofSEIAModel
TheSEIAmodelisformulatedin3steps StepI:EstimatingIndicators StepII:DevelopingIndices Step II: Developing Indices StepIII:FormulatingtheModel

DEVELOPMENTOFINDICATORS
Illustratedvaluesofindicators,theirchangeandsignificanceofthatchange duetotheintroductionofthemetro ACCESSIBILITY(A) Vicinity: littlechangeindistancetoeducationandhealthservices. Distancetourbanserviceslikevegetablemarkets,dailyneedsshops increasedfor23.6%ofthehouseholds.Thebusservicetimegaphas increased for 23 6% of the households The bus service time gap has decreasedfor34%ofhouseholds Relocated: alltheindicatorshavechangedforthemajorityofthe g g y households.Valueshighershowingdeteriorationofaccessibility
45

Change Category

Deducation (diff)

Dhealth (diff)

Dservices (diff)

Dbusstop(diff)

Sbus (diff)

Households in Vicinity of metro line Total Decrease No change Total I T t l Increase

0.0% 98.0% 2.0% 2 0% 40.8% 7.5% 7 5% 51.7%

3.0% 93.1% 3.9% 3 9% 33.8% 3.5% 3 5% 62.7%

4.9% 71.4% 23.6% 23 6% 36.3% 11.9% 11 9% 51.7%

0.5% 80.3% 19.2% 19 2% 13.9% 14.4% 14 4% 71.6%

34.5% 65.0% 0.5% 0 5% 1.5% 0.0% 0 0% 98.5%

Households relocated due to metro line Total Decrease No change Total Increase

Significance of change
No. Indicators Significance of change for HH in metro vicinity At 5% confidence level 1 2 3 4 5 Deducation Dhealth Dservices Dbusstop Sbus Not significant Not significant Not significant Significant Significant At 1% confidence level Not significant Not significant Not significant Significant Significant Significance of change for HH relocated At 5% confidence level Not significant Not significant Significant Significant Significant At 1% confidence level Not significant Not significant Significant Not significant Significant
46

MOBILITY HouseholdMobility(MHH) Vicinity: somechangeintheindicatorsofPCTRforworkandotherpurposes butlittlechangeinthePCTRforeducationandtheshareofNMVsinthe modes Relocated: all the indicators have changed for the majority of the alltheindicatorshavechangedforthemajorityofthe households.For49%households,thePCTRforworkhasincreasedandfor 30%ofthehouseholdsithasdecreased.For71%ofhouseholds,thePCTR foreducationdoesnotchangeThePCTRforotherpurposeshasincreased anddecreasedequally.TheshareofNMVsinthemodeusedhasdecreased and decreased equally The share of NMVs in the mode used has decreased for59%ofthehouseholds.
Change category PCTRwork (diff) PCTRedu(diff) PCTRothers(diff) Mnmv/Mall(diff) Households in Vicinity of metro line Total Decrease No change Total Increase

9.4% 77.8% 12.8% 29.9% 21.39% 48.8%

3.9% 91.1% 4.9% 10.4% 70.65% 18.9%

13.8% 81.8% 4.4% 35.3% 29.35% 35.3%

5.4% 87.2% 7.4% 58.7% 21.89% 19.4%

47

Households relocated due to metro line

Total Decrease No change Total Increase

PersonalMobility(MP) Vicinity: minimumchangeinthemobilityindicatorsregardingtravelfor i i i i i h i h bili i di di lf education(distance,time,cost).Thedistance,timetoandcostoftrips madeforotherpurposeshaschangesalittle. Relocated: mobility indicators for travel to work distance time and cost mobilityindicatorsfortraveltowork distance,timeandcost haveincreasedfor83%,82%and61%ofthehouseholdsrespectively.The distance,timeforeducationhavebutnotthecost.Similarlyforother purposesthereismorechangeindistanceandtimethanthecostofthe trip. trip
Change category Dwork (diff) Deduc ation (diff) Dother s (diff) Twork (diff) Teduc ation (diff) Tother s (diff) Cwork (diff) Ceduca tion (diff) Cother s (diff)

Households in Vicinity of metro line Total Decrease No change Total Increase

10.3% 72.9% 16.7% 14.9% 2.5% 82.6%

3.9% 90.6% 5.4% 22.9% 43.3% 33.8%

15.3% 72.4% 12.3% 58.2% 9.0% 32.8%

13.8% 69.5% 16.7% 14.4% 3.5% 82.1%

4.4% 88.7% 6.9% 21.9% 42.8% 35.3%

16.3% 71.9% 11.8% 52.2% 8.0% 39.8%

3.4% 91.1% 5.4% 10.4% 28.4% 61.2%

0.0% 100.0% 0.0% 2.5% 93.5% 4.0%

4.4% 93.6% 2.0% 12.4% 65.2% 22.4%

48

Households relocated due to metro line Total Decrease No change Total Increase

Significanceofchangeofmobilityindicators
No. Indicators Significance of change for HH in metro vicinity At 5% confidence level 1 2 3 4 5 6 7 8 9 10 11 12 13 PCTRwork PCTRedu PCTRothers Mnmv/Mall Dwork Deducation Dothers Twork Teducation Tothers Cwork Ceducation Cothers Not significant Not significant Not significant Not significant Not significant Not significant Significant Not significant Not significant Significant Not significant Not significant Not significant At 1% confidence level Not significant Not significant Not significant Not significant Not significant Not significant Not significant Not significant Not significant Not significant Not significant Not significant Not significant Significance of change for HH relocated At 5% confidence level Not significant Not significant Not significant Significant Significant Significant Not significant Significant Significant Not significant Significant Not significant Significant At 1% confidence level Not significant Not significant Not significant Significant Significant Not significant Not significant Significant Significant Not significant Significant Not significant Not significant
49

SOCIOECONOMICWELLBEING(SEWB) Vicinity: onlytwoindicatorsIRSandHouseholdincomeshowchangewith theintroductionofthemetro. Relocated: alltheindicatorshavechangedforthemajorityofthe households.Theindicatorsmostaffectedarefemaleliteracy(21% decrease),residency(100%decrease),Householdincomeperperson(66% decrease) residency (100% decrease) Household income per person (66% decrease),Infrastructurerankscore(33%decreaseand61%increase),and employment(8%decreaseand14%increase).
Change category NGinschl/ Ngschage (diff) NAdults>=5/ Nadults (diff) IRS (diff) Yslum/ Ydelhi (diff) V/N (diff)

W/N (diff)

I/N (diff)

Households in Vicinity of metro line Total Decrease No change Total Increase NA 0.0% 0 0% 55.67% 0.0% 44.33% 0.0% 0 0% 100.00% 0.0% 3.4% 3 4% 78.3% 18.2% 0.0% 0 0% 100% 0.0% 0.0% 0 0% 100% 0.0% 9.9% 9 9% 66.01% 24.1% 0.0% 0 0% 100% 0.0%

Households relocated due to metro line Total Decrease No change Total Increase NA 20.9% 41.79% 4.5% 32.84%
50

3.5% 82.09% 14.4%

32.8% 5.97% 61.2%

100% 0.00% 0.0%

8.0% 78.11% 13.9%

65.7% 19.4% 14.9%

5.0% 94.53% 0.5%

SignificanceofchangeofSEWBindicators g g

No.

Indicators

Significance of change for HH in metro vicinity At 5% confidence level At 1% confidence level Not significant Not significant Significant S f Not significant Not significant Not significant Not significant

Significance of change for HH relocated At 5% confidence level Significant Significant Significant S f Significant Not significant Significant Significant At 1% confidence level Significant Significant Significant S f Significant Not significant Significant Significant

1 2 3 4 5 6 7

NGinschl/ Ngschage NAdults>=5/ Nadults IRS S Yslum/ Ydelhi W/N I/N V/N

Not significant Not significant Significant S f Not significant Not significant Not significant Not significant

51

DEVELOPMENTOFINDICES
PrincipalcomponentsarecalculatedusingPCA Principal components are calculated using PCA Differentrotationsaretriedtomaximizeloadingontheprincipal components(PC1,PC2,PCn)sothattheyexplainmaximumpercent ofthetotalvariance.Theoreticallythe varimax rotation maximizes variance of the total variance. Theoretically the varimaxrotationmaximizesvariance explainedwhileincreasingthelargeloadinganddecreasingthesmallerloadings. The higherloadingsineachPCareretainedandthesmallerloadingsare discardedinamannersothateachPCclubstogethersimilar/ correlatedindicatorsinalogicalmanner.EachPCbecomesatypeof l d d l l h b f factorexplainingtheaggregateindexandeachPCisindependentof theothers. TheloadingsoftheretainedvariablesineachPCaretakenas Th l di f th t i d i bl i h PC t k indicativeweightsfortheindicatorsandcalculatedasafractionof1. ThevarianceexplainedaretakenasrelativeweightsforeachPCto aggregatethemasanindex. aggregate them as an index Thevalueoftheindexiscalculatedforeachhousehold.

52

Accessibility
A=E1(PC1)+E2(PC2) A = E1(PC1) + E2(PC2)
WhereE1andE2aretheeigenvalues And PC1=d(Dbusstop)+e(Sbus) PC2=a(Ded)+b(Dhealth)+c(Dser) Wherea,b,.earecomponentloadings.

ThePC1explainsaccessibilityprovidedbythebussystemandthePC2explains thelanduseaccessibility.ThePC1andPC2explainapproximately55%ofthe totalvariance. Theaggregatedindexreadsasfollowsforthe4datasets:

InVicinitybeforemetro A=0.49(Ded)+0.57(Dhealth)+0.62(Dser)+0.63(Dbusstop)+0.62(Sbus) 1a InVicinityaftermetro A 1.07(D d)+0.17(Dhealth)+0.35(D )+0.52(Dbusstop)+0.52(Sbus 1 b A =1.07(Ded) + 0.17(Dh lth) + 0.35(Dser) + 0.52(Db t ) + 0.52(Sb ) 1b Relocatedbeforemetro A=0.91(Ded)+0.27(Dhealth)+0.49(Dser)+0.54(Dbusstop)+0.52(Sbus) 1c Relocated aftermetro A=0.34(D )+0.39(D A 0 34(Ded) 0 39(Dhealth) 0 53(Dser) 0 72(Dbusstop) 0 50(Sbus) 1d )+0.53(D )+0.72(D )+0.50(S
53

Mobility
M=E1(PC1)+E2(PC2)+E3(PC3)+E4(PC4) M = E1 (PC1) + E2 (PC2) + E3 (PC3) + E4 (PC4)
WhereE1,E2,E3andE4aretheeigenvalues And PC1=b(PCTReducation)+e(Ded)+h(Ted)+k(Ced) PC2=c(PCTRothers)+f(Dothers)+i(Tothers)+l(Cothers) ( ) ( ) ( ) ( ) PC3=a(PCTRwork)+d(Dwork)+g(Twork)+j(Cwork) PC4=Mnmv/Mall Wherea,b,.larecomponentloadings.

ThePC1explainsthetripforeducation,PC2explainsthetripforother purposeslikesocial,health,religiousandPC3explainsthetriptoworkand PC4explainsonlyasingleindicatorofuseofnonmotorizedmodes.The PC4 explains only a single indicator of use of non motorized modes The PC1,PC2,PC3ANDPC4explainapproximately65%ofthetotalvariance. TheweightagesofthePCsimplythatthetripforeducationandother y g y pp g p reasonslikebuyingdailyneedsupplieswouldhaveahigherimpactonthe mobilityindexthantheworktrips,thoughthedifferenceisnotsignificant. SinceMhh indicatorsareseenasdesirablemobilityandMp asundesirable mobilitytheyareascribedopposingsignsintheindex.

54

InVicinitybeforemetro M=[0.53(PCTR )+0.79(PCTR M = [0 53(PCTRwork) + 0 79(PCTReducation) + 0 55(PCTRothers) + 1 68(Mnmv/Mall)] )+0.55(PCTR )+1.68(M )] [0.65(Dwork)+0.85(Deducation)+0.74(Dothers)+0.62(Twork)+0.85(Teducation)+0.75(Tothers)+ 0.25(Cwork)+0.17(Ceducation)+0.63(Cothers)] ...2a InVicinityaftermetro M=[0.53(PCTRwork)+0.78(PCTReducation)+0.63(PCTRothers)+1.39(Mnmv/Mall)] [0.64(Dwork)+0.85(Deducation)+0.65(Dothers)+0.62(Twork)+0.85(Teducation)+0.69(Tothers)+ 0.25(Cwork) ( )+0.18(Ceducation) ( )+0.38(Cothers)] ( ...2b Relocatedbeforemetro M [0.67(PCTR k)+0.75(PCTR d ti )+0.55(PCTR th )+1.58(M M =[0.67(PCTRwork) + 0.75(PCTReducation) + 0.55(PCTRothers) + 1.58 (Mnmv/Mall)] ll [0.74(Dwork)+0.80(Deducation)+0.61(Dothers)+0.73(Twork)+0.80(Teducation)+0.70(Tothers)+ ...2c 0.53(Cwork)+0.22(Ceducation)+0.31(Cothers)] Relocatedaftermetro l f M=[0.73(PCTRwork)+0.54(PCTReducation)+0.28(PCTRothers)+1.23(Mnmv/Mall)] [0.83(Dwork)+0.84(Deducation)+0.89(Dothers)+0.78(Twork)+0.80(Teducation)+0.86(Tothers)+ 0.84(C )+0.78(C 0 84(Cwork) + 0 78(Ceducation) + 0 86(Cothers)] )+0.86(C ...2d 2d
55

SEWB
SEWB=E1(PC1)+E2(PC2)+E3(PC3) SEWB = E1 (PC1) + E2 (PC2) + E3 (PC3)
Where,E1,E2andE3aretheeigenvalues AndPC1=e(W/N)+f(I/N)+g(V/N) PC2=c(IRS)+d(Yslum/Ydelhi) PC3=a(NGinschl/NGschage)+b(Nadults>=5/Nadults) PC3 = a(NGinschl/ NGschage) + b(Nadults>=5/ Nadults) Wherea,b,.garecomponentloadings

PC1explainseconomicwellbeing,PC2explainsconditionofphysicalinfrastructureand PC3explainssocialwellbeing.Together,thethreePCsexplain60%ofthevariance. The aggregatedindexreadsasfollowsforthe4datasets: aggregated index reads as follows for the 4 data sets:
InVicinitybeforemetro SEWB=0.61(NGinschl/NGschage)+0.42(Nadults>=5/Nadults)+0.83(IRS) +0.61(Yslum/Ydelhi)+0.66(W/N)+0.65(I/N)+0.14(V/N) + 0.61(Yslum/Ydelhi) + 0.66(W/N) + 0.65(I/N) + 0.14(V/N) InVicinityaftermetro SEWB=0.57(NGinschl/NGschage)+0.46(Nadults>=5/Nadults)+0.71(IRS) +0.62(Yslum/Ydelhi)+0.63(W/N)+0.63(I/N)+0.19(V/N) Relocated beforemetro Relocatedbefore metro SEWB=0.68(NGinschl/NGschage)+0.68(Nadults>=5/Nadults)+0.93(IRS) +0.14(Yslum/Ydelhi)+0.62(W/N)+0.62(I/N)+0.22(V/N) Relocated aftermetro SEWB 0.68(NGinschl/ NGschage) + 0.66(Nadults>=5/ Nadults) + 0.60(IRS) SEWB =0.68(NGinschl/NGschage)+0.66(Nadults> 5/Nadults)+0.60(IRS) +0.65(Yslum/Ydelhi)+0.72(W/N)+0.67(I/N)+0.06(V/N)

3 a 3a

3b

3c

3d
56

SignificanceofchangeintheIndices g g
No. Indices Significance of change for HH in metro vicinity At 5% confidence level 1 2 3 Accessibility Mobility SEWB Significant Not significant Not significant At 1% confidence level Significant Not significant Not significant Significance of change for HH relocated At 5% confidence level Significant Significant Significant At 1% confidence level Significant Significant Significant

57

THESEIAMODEL
CorrelationbetweenAccessibility,MobilityandSEWBmodeledintwoways Correlation between Accessibility Mobility and SEWB modeled in two ways 1. Correlationbetweentheindices 2. Correlationofdependentindexwithindependentindicators

Correlationbetweenindices
Methodsforlinearcorrelation: 1. 1 parametric:Pearsoncorrelation(Continuousdata) parametric: Pearson correlation (Continuous data) 2. nonparametric:Spearmancorrelation(Rankorderdataassumed)

Data Set
In Vicinity- b4 metro In Vicinity- aft metro In Vicinity- change Relocated- b4 metro Relocated- aft metro Relocated- change TOTAL

A&M
Parametric Nonpara

M&S
Parametric Nonpara

A&S
Parametric Nonpara

-0.001 0.128 -0.157 -0.034 0.001 0.026 -0.223

0.004 0.108 -0.202 0.055 -0.049 -0.027 -0.335

0.176 0.112 0.014 0.169 -0.039 -0.219 0.122

0.180 0.089 0.114 0.134 -0.090 -0.229 0.115

0.035 0.277 -0.170 0.057 -0.065 0.016 0.020

0.084 0.280 -0.177 0.140 -0.125 0.045 0.034

58

Linearregressionofdependentindexwithindependentindicators
Thishasbeentriedforthefollowingequations(forall4datasets,andall Thi h b t i d f th f ll i ti (f ll 4 d t t d ll repeatedforeachset) Indexofmobilityandindicatorsofaccessibility M=a+b(AIi)+c(AIj)++x(AIn) ..4 IndexofSEWBandindicatorsofmobility SEWB=a+b(MIi)+c(MIj)++x(MIn) IndexofSEWBandindicatorsofaccessibility Index of SEWB and indicators of accessibility SEWB=a+b(AIi)+c(AIj)++x(AIn)

..5

..6

IndexofSEWBandindicatorsofbothaccessibilityandmobility Index of SEWB and indicators of both accessibility and mobility SEWB=a+[b(AIi)+c(AIj)++x(AIn)]+[b(MIi)+c(MIj)++x(MIn)]

..7

59

SummaryofResultsofLinearregression
No. No 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 Model used Data set In Vicinity- before metro In Vicinity- after metro Relocated- before metro Relocated Relocated- after metro TOTAL R2 value 0.022 0.020 0.025 0.051 0 051 0.103 0.283 0.257 0.200 0.283 0.202 0.157 0.130 0.011 0.012 0.037 0 037 0.361 0.331 0.231 0.295 0.234 P value for F test F-test 0.49 0.55 0.43 0.07 0 07 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.83 0.81 0.00 0 00 0.00 0.00 0.00 0.00 0.00 60

Equation 4

Equation 5

In Vicinity- before metro In Vicinity- after metro Relocated- before metro Relocated- after metro TOTAL

Equation 6

In Vicinity- before metro In Vicinity- after metro Relocated- before metro Relocated- after metro TOTAL

Equation 7

In Vicinity- before metro In Vicinity- after metro Relocated Relocated- before metro Relocated- after metro TOTAL

InterpretationofResults
Equation4: nosignificantcorrelationbetweentheindexofmobilityandthe Equation 4: no significant correlation between the index of mobility and the indicatorsofA, A doesnotaffectM significantly. Equation5: thereisasignificantcorrelationbetweentheindexofSEWB andthe indicatorsofM, M affectsSEWB significantly. Equation6: thereisasignificantcorrelationbetweentheindexofSEWB andthe indicatorsofA forthehouseholdsresidinginthevicinitybutthecorrelation isnotsignificantforthehouseholdsrelocated Equation7:thereisasignificantcorrelationbetweentheindexofSEWB andthe Equation 7 there is a significant correlation between the index of SEWB and the combinedindicatorsofA andM, A andM affectSEWB significantly. ComparingtheR2 values ofallthemodels,thebestresultsaregivenbyEquation , py g p / 7,implyingthattheSEWB isexplainedbestwhentheaffects/contributions ofindicatorsofbothA andM areconsidered.However,itisobservedthat theR2 valueschangeforthehouseholdsaftertheintroductionofthemetro. Forthehouseholdslocatedinthevicinity,theaffectsifA andM onSEWB becomelesssignificantafterthemetroandforthehouseholdsrelocated, become less significant after the metro and for the households relocated theybecomemoresignificant.

61

SignificanceofCoefficients(Eqn7)
Indicator CONST A1 A2 A3 A4 A5 M1 M2 M3 M4 M5 M6 M7 M8 M9 M10 M11 M12 M13 SDeducation SDhealth SDservices SDbus-stop Sbus PCTRwork PCTReducation PCTRothers MNMV/Mall Dwork Deducation Dothers Twork Teducation Tothers Cwork Ceducation Cothers Description In Vicinity-b4 Vicinity b4 metro Coeff 435.2 -81.3 -15.7 -69.9 65.6 -0.1 102.5 45.3 31.9 59.3 -4.7 2.5 -1.5 0.0 -0.6 -0.4 -0.9 1.1 1.2 P(2Tail) 0.006 0.041 0.353 0 0.118 0.929 0 0.151 0.224 0.675 0.013 0.814 0.721 0.909 0.29 0.371 0.558 0.9 0.594 In Vicinity-aft Vicinity aft metro Coeff 308.1 -43.8 -23.0 -17.6 30.9 1.0 89.7 54.0 45.8 25.0 -2.7 4.0 -2.3 0.0 -0.6 -0.4 -2.0 1.5 2.1 P(2Tail) 0.019 0.123 0.153 0.477 0.037 0.099 0 0.068 0.054 0.831 0.063 0.704 0.62 0.88 0.274 0.443 0.135 0.861 0.619 Relocated b4 Relocated-b4 metro Coeff 318.2 -2.6 -27.3 -1.1 295.9 4.1 126.4 53.5 56.2 -37.9 -1.8 -16.2 -3.3 -0.8 0.1 -0.8 -1.6 7.7 3.3 P(2Tail) 0.013 0.812 0.059 0.958 0.088 0.51 0 0.344 0.004 0.746 0.426 0.323 0.454 0.038 0.912 0.28 0.364 0.631 0.384 Relocated aft Relocated-aft metro Coeff 515.5 -10.7 -11.0 -4.6 5.3 -0.2 105.6 -1.4 31.0 -280.3 0.3 4.3 6.4 0.0 -0.3 -0.9 -3.0 -9.1 -6.0 P(2Tail) 0 0.736 0.484 0.238 0.704 0.57 0 0.966 0.042 0 0.581 0.567 0.005 0.844 0.479 0.038 0.012 0.485 0.045 62

Note: The indicator coefficients with P value significant at 90% confidence levels have been highlighted as the coefficients are significant can be included in the models.

InterpretationofResults
Comparativestudyofthecoefficientsshowsthat: Differentcoefficientscontributetothemodelsignificantlyfordifferent datasets. Thenumberofsignificantcoefficientsincreasesaftertheintroductionof h b f i ifi ffi i i f h i d i f themetrointhehouseholdsbothlivinginthevicinityandrelocateddue tothemetro. ThePCTRforworkistheonlyindicatorthatissignificantlyconsistent The PCTR for work is the only indicator that is significantly consistent acrosstheboard. ThecostoftravelhasnosignificanceinexplainingSEWBifrelocationnot g y therebutitbecomessignificantwhentheyarerelocated. Astudyofthecoefficientsofthecombineddatasettogetanoverviewof whetherthecoefficientsare+veorveshowsthatapproximately90%of thesignificantindicatorsand72%ofallindicatorsarecorrelatedtothe SEWBindexinaccordancewiththeempiricallyobservedbehavior SEWB i d i d ih h i i ll b db h i (expectedindications)

63

FinalEquations
ThefinalequationsderivedfromtheapplicationofEquation7usingsignificant indicatorsareillustratedbelow:
SEWBVb4 =435.2 81 3(SDeducation) 69 9(SDservices) 102 5(PCTRwork) 435 2 81.3(SD ) 69.9(SD )+102.5(PCTR 4.7(Dwork) SEWBVaft =308.1+30.9SDbusstop) 1 0(Sbus) 89 7(PCTRwork) 308 1 30 9 SD )+1.0(S )+89.7(PCTR ) +54.0(PCTReducation)+45.8(PCTRothers) 2.7(Dwork)

8a

8b

SEWBRb4 =318.2 27 3(SDhealth) 295 9(SDbusstop) 126 4(PCTRwork) 318 2 27.3(SD ) 295.9(SD )+126.4(PCTR ) +56.2(PCTRothers) 0.8(Twork)

8c

SEWBRaft = 515 5 + 105 6(PCTRwork) + 31 0(PCTRothers) 280 3(MNMV/Mall) =515.5+105.6(PCTR )+31.0(PCTR )280.3(M ) +6.4(Dothers) 0.9(Tothers) 3.0(Cwork) 6.0(Cothers) 8d

64

Interpretationofresults
ThePCTRforworkmostimportantpositivedeterminantofSEWB.This Th PCTR f k ti t t iti d t i t f SEWB Thi impliesthetripstoworkmadebyahouseholdensuretheSEWB, Thedistancetoworkisconsistentlyanegativeindicatorforhouseholds implyingthatincreaseindistancetoworkwillnegativelyaffectSEWB. implying that increase in distance to work will negatively affect SEWB. Theintroductionofthemetrochangestheindicatorswhichaffect SEWB.Also,morenumbersofindicatorshaveasignificantimpacton SEWBaftertheintroductionofthemetro.Thisimpliesthatthe introductionofanewtransportsystemrestructuresthedeterminants ofSEWB,makingthehouseholdsmorevulnerablebyincreasingthe numberofsignificantindicators. HHinVicinity: HH in Vicinity: Sincebusroutesandserviceshavebeenaffectedbytheintroductionof themetro,theybecomesignificantindicatorsaffectingSEWB.This impliesthattheintroductionofanewtransportsystemmakesthe implies that the introduction of a new transport system makes the existingtransportsystemimportantindeterminingSEWB.

65

HHRelocated: Travelforpurposesotherthanworkandeducationisaffectedbythe Travel for purposes other than work and education is affected by the relocation.Whilethedistanceforthesetripscontributespositively toSEWB,thetimeandcostofthesetripscontributesnegativelyto it. ThecommutingcosthadnosignificantcorrelationwithSEWBbefore relocation,afterrelocationithasasignificantnegativeimpacton SEWBofthehouseholds. RatioofNMVtoallmodesusedhasbecomeasignificantindicator afterrelocation.Thehighnegativevalueofthisindicatorimplies thatthereductioninthisratio(implyingreductioninuseofNMVin thehousehold)hasaseverenegativeimpactontheSEWBofthe the household) has a severe negative impact on the SEWB of the households.Sincetheprocessofrelocationhasincreaseddistances todestinationsofchoiceforthehousehold,beyondcomfortable , p NMVdistances,thisindicatorimpliesthatthemodalshiftfromNMV tomotorizedmodeshashadanegativeimpactontheSEWBofthe relocatedhouseholds.

66

8.Conclusions
ImpactofMetroonthepoorhouseholdinitsvicinity
NosignificantimpactontheSEWBandMobility Whilethelanduseaccessibilityremainsunchangedtoo,thetransport While the landuse accessibility remains unchanged too the transport accessibilityhaschangedasdistancetothebusstopshasincreasedfor 19%ofthehouseholdsandbusserviceshavebecomenonexistentfor 33%ofthehouseholds. 33% of the households

ImpactofMetroonthepoorhouseholdsrelocated
ThereissignificantimpactonAccessibility,MobilityandSEWB Thelanduseaccessibilityhasdeterioratedasdistancetoeducation, healthservicesandotherurbanserviceshasincreasedfor52%,63%and 52%ofthehouseholdsrespectively.Thetransportaccessibilityhas deterioratedevenmoreasdistancetobusstophasincreasedfor72%of thehouseholdsandthebusfrequencyhasseenanaveragedecreasefrom 5minto63min(almost13times) ( l )
67

Themobilityofthehouseholdshaveincreasedsignificantly.ThePCTRforwork Th bilit f th h h ld h i d i ifi tl Th PCTR f k hasincreasedfor49%ofthehouseholdsanddecreasedfor30%,implying changeinnumberoftripsmadeforworkinthehouseholds.Theshareof NMVsamongstthemodeusedhasdecreasedfor59%ofthehouseholds.The NMVs amongst the mode used has decreased for 59% of the households The mobilityindicatorsfortraveltowork distance,timeandcost have increasedfor83%,82%and61%ofthehouseholdsrespectively TheSEWBindicatorsmostaffectedarefemaleliteracy(21%decrease), The SEWB indicators most affected are female literacy (21% decrease) residency(100%decrease),Householdincomeperperson(66%decrease), Infrastructurerankscore(33%decreaseand61%increase),andemployment (8%decreaseand14%increase).Theindicatorsofadultliteracyandvehicle (8% decrease and 14% increase). The indicators of adult literacy and vehicle ownershipshowleastchangewith82%and94%respectivelyinthenochange category. Theresultsimplythatrelocationduethemetrohashadasignificantnegative impactontheSEWBofthepoorhouseholds.

68

CorrelationofSEWBtoAccessibilityandMobility C l i f SEWB A ibili d M bili

SEWBisaffectedbyindicatorsofbothaccessibilityandmobility SEWBisnegativelycorrelatedtospatialdistancetoeducation,healthand g y p otherurbanservices ItispositivelycorrelatedtoPCTRforwork,educationandotherpurposes Itisnegativelycorrelatedtotraveldistance,timeandcost It is negatively correlated to travel distance, time and cost Thesignificanceofindicatorschangeswithchangeinsituationlikethenew metrolineorrelocationduetoit PCTRforworkispositivelycorrelatedwithSEWBandhasthehighest PCTR for work is positively correlated with SEWB and has the highest coefficientinalldatasets,indicatingthemobilityforworkisimportantin ensuringtheirSEWB,whateverbetheirsituation CostoftravelhasnosignificanceinexplainingSEWBoftheurbanpoorbut Cost of travel has no significance in explaining SEWB of the urban poor but itbecomessignificantwhentheyarerelocatedandnowhavetopayheavily forthetravel

69

Inconclusion Thisstudyillustratesthattheaccessibilityandmobilityandhencethesocio economicwellbeingoftheurbanpoorisaffectedbyitsintroductionin theurbantransportsystem. p y Whiletheymaynotbeexpectedbeneficiariesoftheproject,thedisbenefits accruedtothemduetotheprojectneedtobeassessed,andhence mitigationmeasuresplannedwhenproposingtheproject. g p p p g p j Hence,itisimportanttoconductSocioEconomicImpactAssessment(SEIA) studiesforanewprojectoverdisaggregatedgroups,specificallyincluding impactsonthemostvulnerablegroup theurbanpoor. p g p p

70

Policyrecommendations
Thedefinitionoftheimpactedpopulationforatransportprojectshould includenotonlytheexpectedusersbutthenonusersaffectedbyittoo. Theaccessibilityandmobilityneedsoftheurbanpoorneedtobe studiedandtheurbanpoorshouldbeseenascaptivesofthesystems t di d d th b h ld b ti f th t theyareusing.Introductionofanypolicyorprojectthatchangestheir statushastobecarefullymonitoredforimpacts. Thecostbenefitanalysisofatransportprojectshouldincludethedis benefitstononusergroupsandthecostsofcompensation/mitigation measuresinbuiltaspartofprojectcost.Onlythenshouldaprojectbe declaredfeasible. The Government should constitute a statutory body responsible for the TheGovernmentshouldconstituteastatutorybodyresponsibleforthe SEIAofallinfrastructureprojectsbeforetheyaregivenapprovalfor implementation.Thisisinkeepingwiththesocialwelfarefunctionof theGovernment. Allfundingmechanismsfortransportprojectsshouldhaveinbuilt All funding mechanisms for transport projects should have inbuilt monitoringandevaluationprotocolswithstringentSEIAguidelines.

71

Contributionofresearch
ThisdissertationtriestounderstandhowtheSEWBoftheurbanpoorisimpacted bylargetransportprojects.Theimpactontheaccessibilityandmobilityofthe nonusersofthenewsystemisasimportantastheimpactontheexpectedusers andneedstobeinternalizedbytransportprojects. Thedissertationprovesthattherelocationofthepoorisoneofthemostsevere negativeimpactsofatransportprojectsandneedstobetakeninaccountin impactassessmentstudies. impact assessment studies Thedissertationhasredefinedtheconceptofmobilityintoitspositiveand negativeaspects.Ithasformulatedindicatorsofaccessibility,mobilityandSEWB andaggregatedthemintoindices. gg g IthasmodeledhowSEWBisaffectedbyaccessibilityandmobilityand,indoing so,hasformulatedagenericmethodologyofSEIAwhichisapplicablein understandingtheimpactoflargeurbantransportprojectslike expressways,flyoversetcontheurbanpoor..Differentinterventionscenarioscan becomparedfortheirimpactsandmitigationmeasuresplannedaccordingly.This wouldleadtointernalizingtheexternalcostoftheimpactoftransportprojectson theurbanpoor. the urban poor.
72

Scopeforfuturework
Literaturereviewhasshownthatevenamongsttheurbanpoor,thewomen arepoorerthatthemen,sufferingfrompovertyofmoney,timeand resources.Assessingthegenderedimpactsoftransportprojectswouldgive additionaldepthtotheprocessofSEIA. TheWHOhasdeclaredroadaccidentsasthenumberonediseaseinthe world.Thehealthimpactsoftransportneedtobeincludedmore comprehensivelyintheSEIAmethodafteranecessaryreviewoftheliterature onthesame. Thequalitativedataaboutsocioeconomicconditionsandtheopinionsand choicesofpeopleareanotheraspectofSEIAwhichrequiresfurtherresearch. Differenttechniqueslikestatedpreferencemodelscanbeusedtoinclude qualitativedata. Thebenchmarkingofthevariousparameters/indicatorsneedstobecarried outtoidentifyacceptablelevelofadverseimpactsoftransportprojects. Theimpactsonaccessibility,mobilityandSEWBneedtoascribedvaluein The impacts on accessibility, mobility and SEWB need to ascribed value in termsofmoneyandresourcestoformulatecompensationpackageswhere necessary.Thisstudyshouldfurtherleadtomitigationmeasuresand alternativerecommendationstominimizeadverseimpactsoftransport p j projectsontheurbanpoor. p
73

THANKYOU THANK YOU

November 2007

74